Achaeta hanagarthi Schmelz

Schmelz, Rüdiger M., Collado, Rut & Römbke, Jörg, 2008, Mata Atlântica enchytraeids (Paraná, Brazil): The genus Achaeta (Oligochaeta, Enchytraeidae), Zootaxa 1809, pp. 1-35 : 9-21

Achaeta hanagarthi Schmelz , sp. nov.

Figure 3, Tables 1 , 2 , 4

Holotype: MZUSP #1364, mature spcm, stained whole mount; Cachoeira Natural Reserve, Paraná, Brazil, March 2004.

Paratypes: MZUSP #1365, 41 spms stained and whole-mounted, 21 of them fully mature, same locality as holotype, May 2003 (20 spms), March 2004 (18 spms), Oct. 2004 (3 spms); further 86 spms ethanol-preserved, March 2004 (73 spms), Oct. 2004 (13 spms).

Further material: Ca. 150 spms, identified in vivo, not conserved.

Etymology: Named in honour of the late Werner Hanagarth, zoologist, geographer, all-round ecologist of the old school, who dedicated his life's efforts to nature and people in South America, first project coordinator of SOLOBIOMA (obituary in Beck & Höfer 2003).

Description: Living specimens 5–7 mm long, ca. 0.2 µm wide, whitish opaque under top light. Slow body movements. Preserved specimens 4.5–7 mm long, 0.14–0.2 mm wide. Segment number mostly 32–43 (N=26; 13 specimens with 38–42 segments), rarely below 30 (13–24, N=4, partly with regenerated posterior end).

No pyriform glands. Epidermal gland cells ( Fig. 3 A) 2-8 per segment, 1-4 on each side; two different types present. Type I cells hyaline in the center, and framed anteriorly and posteriorly by a granulated field; this field apparently located between ring muscle and longitudinal muscle layer; two cells per segment, one on each side, situated postero-ventrally right before septum, first pair at II or III. Cells conspicuous only in preclitellar segments, but present throughout except in clitellar region. Type II cells inconspicuous, hyaline without granular framing, not intervening between muscle layers, smaller than Type I cells, situated in one transverse row in the posterior half of a segment, slightly anterior to Type I cells; on each side 2–3 cells dorsally of lateral line, one occasionally ventrally of it. Type I cells only seen in some of the preserved specimens.

Body wall ca. 5 (-10) µm thick, cuticle variable: less than 1 µm thick, inconspicuous, or up to 3.5 µm thick, especially dorsally, or intermediate; thickening more pronounced dorsally than ventrally. Ring muscles inconspicuous, single-layered, ca. 23 rings per segment. Septa 4/5–6/7 thickened.

Head pore on prostomium. Frontal prostomial epithelium ( Fig. 3 B) irregularly thickened between head pore and mouth opening, with recess immediately below head pore. Latero-dorsally two pairs of small isolated patches of thickened epithelium, connection with prostomial nerve not evident. Prostomial musculature well-developed. Brain 1.5x as long as wide (e.g., 83: 55 µm, top view, fix), 2.5x as long as high (side view), bluntly triangular in dorsal view, with a marked anterior convexity; sides merging anteriad, rounded and truncate posteriorly. No perikarya seen on prostomial nerve. Ventral nerve cord ( Fig. 3 B,I) with suboesophageal ganglion in II–IV, not lobed, sides straight, and segmental ganglia from V on, with connectives at level of septa. Gradual change in shape of ganglia from anterior to posterior: in anterior segments spindle-shaped, not lobed; in posterior segments bilobed with a smaller anterior and a larger posterior part, marked by a lateral constriction; ganglionic subdivision more and more conspicuous towards growth zone; here lobes almost separate from each other so as if each segment had two ganglia.

Oesophageal appendage ( Fig. 3 B–E) consisting of an unpaired strand in IV and V dorsally on oesophagus, with winding canal (only seen in living specimens), and two large ovoid or spherical bodies attached dorsolaterally in V, directly anterior to pharyngeal glands; anterior opening of canal into lumen of oesophagus not seen. Four large post-pharyngeal bulbs present ( Fig. 3 E). Pharyngeal glands ( Fig. 3 B-E) in IV–VI; all widely united dorsally; ventral lobes conspicuous, largest in IV and VI, widely connected with dorsal lobe in IV and V, separate from dorsal lobe in VI. No secondary ventral lobes. Chloragocytes dark-grey in living specimens (transmittent light), few single cells in IV-VI, dense layer in VII–IX, here up to 25 µm high, posteriorly increasingly thinner; behind clitellum cells flatter, less opaque; in posterior body half cells rather flat, scattered, inconspicuous in whole mounts. Occasionally cells flat throughout. Gut dilatation gradual in VII-VIII, marked only in contracted specimens. Dorsal blood vessel from VII, pulsating ( Fig. 3 B–E).

One pair of preclitellar nephridia, at 7/8 ( Fig. 3 B); anteseptale about as long as high, truncate; postseptale ca. 4x as long as anteseptale, gradually tapering distad; no efferent duct, only a short ventral bend of narrowed postseptale distally before opening to exterior; no constriction between anteseptale and postseptale; nephridial canal very finely winded; terminal vesicle small. First postclitellar nephridia at 12/13 (!) or 13/14, shape similar to preclitellar nephridia, but usually straight. Few nephridia in following segments and in variable positions, about 1/4 of all possible positions occupied ( Fig. 3 F).

Coelomocytes ( Fig. 3 G) numerous, with strong brownish tint conspicuous in cell aggregations, faint but distinguishable in single cells, rarely absent in very young specimens, tint diffusely distributed in cell cytoplasm. Cells flat, almost as wide as long, comparatively large, diameter 16–45 µm in vivo, margin slightly wavy, cytoplasm filled with pale, regularly spherical vesicles and occasionally a few small interspersed refractile granules. Pars tumida of midgut from XX–XXVIII, extending over 5–7 segments; entire epidermis thickened, not only ventrally, with a tint similar to gut contents (brownish, not present when gut is empty), staining less intensely than adjacent intestinal epithelia (fix). Pygidium short, about as long as wide.

All reproductive organs except spermathecae shifted one segment forward: testis and sperm funnels in X, ovary, vasa deferentia and male pores in XI.

Clitellum ( Fig. 3 H–I) at about XI-1 / 2XII (beginning three ring muscle rows posterior to septum 10/11, ending ca. 10 ring muscle rows before septum 12/13), developed only laterally; cell-free area wider ventrally than dorsally, ventral border of clitellum in longitudinal line with male pores. Cells arranged in ca. 24 regular transverse rows. Each lateral clitellar field distinctly subdivided into a dorso-lateral area with only hyalocytes (single interspersed granulocytes may be present) and a larger latero-ventral field with only granulocytes. Hyalocytes large and conspicuous, often more than twice as wide as long, 20–25 µm high, usually arranged in two longitudinal rows that merge into one row at both ends; slight deviations from this pattern occur. Granulocytes not arranged in longitudinal rows, less then half the size of hyalocytes, decreasing in height from ca. 20 µm laterally to 10 µm ventro-laterally. Dorsal longitudinal borderlines of clitellum conspicuous, ventral borderlines inconspicuous.

Testis and ovary anteriorly in X and XI, respectively, unpaired ventrally below nerve cord ( Fig. 3 I). Seminal vesicle absent, cysts free in X. Spermatozoa 40–50 µm long, measured head lengths 15–18 µm (viv). Sperm funnel in X, ca. 2/3 as long as body diameter, length 85–90 µm (viv, fix), asymmetrical in cross section: either pear-shaped, 1.4x as long as wide, collar narrower than funnel body, or cylindrical, 3x as long as wide, collar as wide as funnel body - depending on the angle of vision ( Fig. 3 I). Vas deferens long, most often in dense, more or less regular coils, and with same diameter throughout (6 µm, fix). Male copulatory organs ( Fig. 3 I) in XI, widely separate, consisting of a small, distinct glandular body (50 µm long, 30 µm high, fix) embedded in muscle tissue surrounding male pore, and of most distal stretch of vas deferens; no bursa (epidermal invagination) seen, male pore apparently on body surface, glandular body nonetheless protrusible. Copulatory body muscles present, inserting latero-dorsally between transverse clitellar gland cell rows. No accessory glands near male pore.

Spermathecae ( Fig. 3 B–E) simple, blind-ending, tube-like, extending into VI or VII, occasionally bent upwards and confined to V, with a club-shaped proximal end. Ectal pores in lateral position, ectal duct ca. 30 µm long (fix), here lined with cuticle, canal almost as wide as walls, no outer muscles detectable. Ampulla ca. 220 µm long (fix); ectal dilatation not or only slightly wider than ectal duct (15–20 µm), recognized mainly by a small sperm bundle and the absence of cuticular lining, not further modified, no glandular granulation, no bend of axis, no asymmetrical enlargement; following connecting tube thin and without distinct lumen; ental reservoir ca. 30 µm wide, of differing length; almost spherical when without sperm, then filled with an amorphous mass. One mature oocyte at a time, extending over 2–3 segments when fully developed.

Remarks. Species diagnosis: Peculiar traits of Achaeta hanagarthi are: (1) Reproductive organs except spermathecae shifted one segment forward, (2) coelomocytes brown, (3) pars tumida of midgut distributed over entire cross-section, not confined to ventral region. A forward shift of sexual organs by one segment (spermathecae excepted) is known in five more species of Achaeta ; they are compared in Table 1 . However, a forward shift of sexual organs may have been overlooked more than once (see above, A. piti ); therefore we also present a comparison with Achaeta species without pyriform glands that are described as having the sexual organs in normal position ( Table 2 ). All species have at least one further distinguishing trait. The comparison is made difficult by several insufficiently detailed original descriptions and by the lack of redescriptions in all but one species ( A. camerani ). The description of A. silvatica Nurminen, 1973 is especially poor; types A. hanagar- A. pannonica A. diddeni A. piti Bitten- A. abulba A. bibulba thi sp. nov. Graefe, 1989 Graefe, 2007 court, 1974 Graefe, 1989 Graefe, 1989

Segments 32–43 25–29 22–24 31–36 20–25 24–27 Secondary ven- absent present present absent absent absent

tral pharyngeal in V, VI in V, VI

gland lobes

Spermathecal lateral lateral lateral lateral ventral ventral

pores

Finds Brazil Estonia, Ger- Germany Brazil, Ger- Germany Sweden, Germany, Austria, many many, Czech Hungary Rep.

are not traceable and probably lost. The species cannot be identified properly, the known characters correspond also to A. singularis sp. nov., described in this paper. Pending a rediscovery of the types or a full description of topotypic material (from forest soils near Montreal, Canada), we consider A. silvatica as a species inquirenda.

Further observations: The whitish colour of the animals is caused by light refractions of the abundant coelomocytes, and in parts also of the chloragocytes, the latter are dark-grey in transmittent light. The coelomocytes are comparatively large, up to one fourth of the worm body diameter. Light refractions are strongest in packages of cells in upright position ( Fig. 3 G, right). The species was most easily recognized by the coelomocytes, this being the most conspicuous character in routine identifications of living specimens. In Bouinfixed specimens the coelomocytes take up a diffuse paracarmine staining. In small juveniles, the brown tint of the cells is less conspicuous and occasionally completely absent. One specimen in the type collection is without pharyngeal glands in VI and another one without glands in V and in VI.

Achaeta neotropica Č ernosvitov, 1937 Figure 4, Tables 3 , 4

Achaeta neotropica Č ernosvitov, 1937: 154–157, Figs 16–22.

Achaeta neotropica Cern. — Righi 1974: 127 –129, Figs 1-6 ; Christoffersen 1979: 153 –154, Figs 1 –10; Righi 1981: 427.? Achaeta becki Schmelz & Collado, 2005: 49 -57, Figs 1–2 , Table 1 .

Collected material: MZUSP #1366, 53 spms, investigated in vivo and as stained whole mounts, Cachoeira Natural Reserve, Paraná, Brazil, May 2003 (9 spms, fixed in formol), March 2004 (36 spms, fixed in hot Bouin's fluid); remaining 9 spms ethanol-preserved. Further 35 spms, identified in vivo, not preserved.

Syntypes: BMNH #1949.3.1.945–948, 7 spms: 1 whole mount on slide (in unidentified medium, probably glycerine, coverslip framed with black varnish; preparation partly dried); 1 fully mature in cross sections on slide, anterior end, down to ca. segment XVI; 1 complete spcm in horizontal to transverse sections (worm bent); 4 ethanol-preserved spms. One further syntype spcm—not investigated here—present at the Museo Argentino de Ciencias Naturales (A. Tablado, Jefe División Invertebrados, pers. com.; see also "Note de la Direction" in Černosvitov 1937: 157).

Further reference material investigated: MZUSP #350, 11 spms, whole-mounted on one slide each, Brazil, São Paulo, Evangelista de Souza, rain forest region, on the banks of a small brooklet, coll. M. L. Christoffersen (Christoffersen 1979: 153: "variety B"). MZUSP #355, one formol-preserved spcm, fully mature, Brazil, Mato Grosso, Sidrolândia, Fazenda Lagoinha, moist soil, coll. G. Righi (Righi 1981), wholemounted for this study, later retransferred to ethanol. MZUSP #356A, three formol-preserved spms, Brazil, São Paulo, Engenheiro Marsilac, rain forest region, on the banks of a small brooklet, coll. M. L. Christoffersen (Christoffersen 1979: 153, "variety A"). MZUSP #359, six formol-preserved spms, Brazil, São Paulo, Evangelista de Souza, rain forest region, on a damp earth slope, coll. M. L. Christoffersen (Christoffersen 1979: 153: "variety D"). ZMH OL #14518: one spcm, Mexico, fixed in hot Bouin's fluid, stained with Boraxcarmine; whole-mounted for this study; Mexico, mount Iztaccíhuatl, NE-slope, ca. 3000 m a.s.l., soil type humic Andosol, leg. Miehlich, 10.10.1978, spcm preserved and donated by U. Graefe.

Description of new material: (Differences in other descriptions, confirmed by reinvestigation of the reference material, are added, marked by "COD" for Černosvitov (1937), original description, and " CHR ", Christoffersen (1979). Further details in the remarks section.)

Living specimens 4–7 mm long (COD: 3–4 mm, CHR: 3.5–8.5 mm) and 0.18–0.2 mm wide, preserved specimens 3.5–5.5 mm long and 0.1–0.15 (-0.19) mm wide ( CHR: 0.09-0.29 mm wide, up to 0.32 mm in clitellar region). Segment number (27)-30-33-(38) ( CHR: 30-46).

Pyriform glands absent. Hyaline epidermal glands cells numerous ( Fig. 4 A,B,D,I), in 5–6 transverse rows per segment, 2–10 cells per row, more cells in dorsal body half; some specimens with 8 cells in a longitudinal row dorsally. Cells inconspicuous in some specimens. Cells hyaline, no granular framing distinguished. Cells also present in mid-dorsal clitellar region.

Body wall thin ( Fig. 4 A–K), cuticle variable; about as thick as body wall in anterior segments (up to 6 µm), posteriorly thinner; or anteriorly thin (1–2 µm), posteriorly up to 13 µm thick (N=1); thickening of cuticle always stronger dorsally than ventrally. Septa 4/5 and 5/6 thickened compared to the rest; septa 6/7 and 7/ 8 extremely thin or absent, not always distinguished ( Fig. 4 A,C); following septa variable; postclitellar septa often well-developed, often as thick as septa at 4/5 and 5/6, especially laterally and ventrally, dorsally often thin.

Head pore on prostomium, dorso-frontal. Frontal prostomial epithelium thickened between head pore and mouth opening, with a small recess immediately below head pore. A pair of elongate prostomial interior papillae present dorso-laterally (prostomial ganglia?), dorsally of and beneath prostomial nerve, connecting with frontal prostomial surface. Prostomial musculature weakly developed. Brain ( Fig. 4 C) twice as long as wide, sides parallel or slightly indented, posteriorly truncate, anteriorly convex. Suboesophageal ganglion in II–IV, wider than rest of ganglia. From V on one ganglion per segment, ganglia separate from each other, a few perikarya often present on anterior connectives of 6/7–8/9 (in 24 out of 34 specimens, comp. Fig. 4 A). Ganglion in V shorter and wider than the rest. Two pairs of post-pharyngeal bulbs.

Oesophageal appendages absent. Pharyngeal glands ( Fig. 4 A,C,F,G) in IV and V each with one unpaired dorsal lobe and a pair of ventral lobes in wide connection with dorsal lobe; ventral lobes in IV sometimes absent ( Fig. 4 F). Glands in VI dorsally separate, shape and size variable, from small, knob-like in anterior of VI to large, extending over VI and VII, occasionally bipartite ( Fig. 4 F). Gut dilatation somewhere in VI: at beginning, in the middle, or—rarely—in posterior segment region ( CHR: anywhere between the anterior portions of VI and VII, respectively), abrupt or gradual; when abrupt, occasionally with small circumferal diverticulum ( Fig. 4 G); epithelium more than 2x as thick here as anteriorly (15–20 µm), cells larger, finely granulated, often with a slight brownish tint; this specialized region extending over 1-1.5 segments (e.g., VI-1 / 2 VII or 1/ 2VI-VII, 3/4 VI-1 / 4 VIII); posteriorly of this region, intestine narrower, epithelium flatter, no tint, no granulation, often strongly meandering, blood sinus often wide (comp. Fig. 4 A). Chloragocytes from V, dense layer from VI-VII. Cells filled with refractile brownish granules. Dorsal blood vessel ( Fig. 4 A) from XIII–XV in fully mature specimens, not attached to intestine in preclitellar segments, often meandering here (COD: dorsal blood vessel from XI, CHR: from XI –XIV). Pars tumida of posterior midgut ( Fig. 4 H) in XVI–XXVIII, extending over (3.5)-5-6-(10) segments in mature specimens, developed ventrally only; cell height variable, epithelium often cuboidal, not cylindrical. Region not distinguished in one specimen.

Preclitellar nephridia ( Fig. 4 A,C,G) in two pairs, at 6/7 and 8/9, absent at 7/8, most often 3–4 nephridia altogether; nephridia elongate, usually bent, anteseptale distinctly longer than wide, nephrostome embedded obliquely in nephridial body; postseptale slightly narrowed towards ectal pore, no efferent duct; terminal vesicle large. ( CHR: some specimens with preclitellar nephridia only at 8/9). Postclitellar nephridia ( Fig. 4 H): first at 13/14 or more posteriorly (e.g., 16/17), as long as or longer than one segment, often bent; anteseptale up to 3x as long as wide, shape variable, same as preclitellar nephridia, occasionally with constriction at septum and/or thinning of ectal third of postseptale; terminal vesicle always large. Proximal region of pre- and postclitellar nephridia often granulated.

Coelomocytes ( Fig. 4 L) not abundant, cell outline irregular, length: width ratio 1–1.5: 1; cytoplasma filled with vesicles, either conspicuous or inconspicuous, aspect inhomogeneous, some vesicles dark; cell aggregations dark and opaque but usually not coloured, occasionally distinctly brown in aggregations; cells ca. 20-30 µm long (viv), 15–25 µm long in fixed specimens. Pygidium ( Fig. 4 K) not longer than wide.

Reproductive organs in usual position: Testis and sperm funnels in XI, ovary and male pores in XII.

Clitellum ( Fig. 4 D,E) from XII-1 /2 XIII, not developed dorsally and ventrally, cell-free area narrower dorsally than ventrally; gland cells not projecting proximally into coelom. Dorsally hyalocytes and granulocytes alternating, hyalocytes larger (11–13 µm long and 16–28 µm wide), in two irregular longitudinal rows, embedded in granulocytes, i.e., 1 or 2 hyalocytes per transverse row on each side of the body ( MZUSP #350 and ZMH OL #14518: three irregular longitudinal rows, i.e. up to 3 hyalocytes per transverse row on each side). At dorsal margin of clitellum only granulocytes. Clitellum-free mid-dorsal field with epidermal gland cells ( Fig. 4 D). Ventro-laterally only granulocytes, in transverse rows.

Seminal vesicle absent, cysts free in XI ( CHR: present, inconspicuous, may reach septum 8/9; seen here only in 3 specimens of MZUSP # 359, absent in rest of reinvestigated CHR material). Spermatozoa 45–55 µm long, heads 20-23 µm long (viv). Sperm funnel ( Fig. 4 E) variable in shape and size; cylindrical or pearshaped, 1/4–1/2 as long as body diameter (45–80 µm long, 25–60 µm wide, fix), 1–1.5x as long as wide, collar distinct, as wide as funnel body, hyaline ( CHR: from 60: 40 µm up to 200: 100 µm, up to 2.7x as long as wide); funnel body distinctly vesicular, slightly flattened; hyaline collar and vesicular funnel body optically contrasting (viv). Vas deferens ( Fig. 4 E) comparatively short but not straight, usually with dense irregular coils in XII ventrally, diameter ca. 5 µm throughout (fix). Male pores ( Fig. 4 E) in XII, apparently on body surface, no bursa (= epidermal invagination) distinguished, widely separate, surrounded by a simple flat and compact glandular body; size of glandular body variable: 20–45 (-90) µm long and 12–20 (-60) µm wide, 1.5– 2 x as long as wide ( CHR: 50–100 µm long, 25–50 µm wide); shape roughly oval or heart-shaped, with middorsal indentation. Copulatory body muscles present. No accessory glands near male pore.

Spermathecae simple, blind-ending, tube-like, without ectal asymmetry, varying in size: confined to V or extending into VI ( CHR: ending in VI, VII or VIII); ectal pores lateral, right below lateral line; cuticularized ectal duct short, widened distally at ectal pore; ampulla simple, often not set off against ectal duct, with or without ectal sperm-containing dilatation, proximal end most often widened, spherical or club-shaped, usually narrower than pharyngeal gland lobes in VI; sperm may be present in all parts of ampulla. One mature egg at a time, extending over 2–3 segments.

Two specimens—one submature spcm collected by us ( Fig. 4 M) and another one collected by Christoffersen (1979, Fig. 3, "variety D", MZUSP #359, see Table 3 ) —with greatly enlarged male reproductive apparatus: sperm funnel half as wide as body diameter, 2.5- 3 x as long as wide; collar with two long extensions, these extensions about as long as the funnel body itself, conspicuous by spermatozoa attached to the collar, giving the aspect of a spruce twig or feather boa. Gland surrounding male pore ca. 80 µm long and 55 µm wide in the collected submature spcm (viv; fixed 70 µm long, 30 µm wide, 40 µm and high), 130: 70 µm in mature MZUSP specimen; gland broadly heart-shaped in lateral view, i.e. with mid-dorsal indentation; cells inflated by masses of secretory products. Spermathecal ampullar ental sac club-shaped, extending over 2–3 segments, down to VIII or IX, half as wide as body diameter, thick-walled, densely packed with sperm. No further differences between these specimens and the rest.

Remarks. Species diagnosis: Our material differs in two important traits from the original description of A. neotropica : (1) The most anterior nephridium is situated at 6/7; (2) spermathecal ectal pores are in lateral position. According to Černosvitov (1937: 156), the most anterior nephridia are behind septum 6/7, i.e. at 7/8, and the spermathecal ectal pores are on the ventral surface of the animal. These differences were also recorded in the redescriptions of A. neotropica in Righi (1974) and Christoffersen (1979), based on new material from Brazil, but they were not recognized or discussed as differences there. The mentioned two traits are, from all available knowledge, intra-specifically constant. So when we found specimens from the Manaus region with the same differences, we ascribed them to a new species, Achaeta becki (Schmelz & Collado 2005) , and we questioned the identification of the specimens in Righi (1974) and Christoffersen (1979) as A. neotropica . However, our reinvestigation of the types of A. neotropica revealed that the differences are - once again, see A. piti - due to faults in the original description, since the type specimens have lateral spermathecal pores and nephridia at 6/7. As a result, Righi and Christoffersen were right, 'by default' as it were, the Mata Atlântica specimens can be identified as A. neotropica , and the status of A. becki as a species different from A. neotropica has become questionable.

A. neotropica exhibits a peculiar mosaic of variable and invariable characters. Christoffersen (1979) was the first to note considerable variation in A. neotropica . Most striking here is the development of spermrelated sexual organs. The extremes as outlined in Table 3 would justify the recognition of two different species, were it not for intermediate character states present in the series of specimens investigated here, especially if A. becki is included, where the spermathecae extend into VII–VIII but a seminal vesicle is lacking (comp. Schmelz & Collado 2005, Fig. 1 A). Further variations concern (1) body size and (2) segment number, (3) thickness of cuticle, (4) shape, size and extension of pharyneal glands in VI, (5) thickening of septa, (6) the transition oesophagus/intestine in VI, (7) texture of coelomocytes, and (8) number and distribution of clitellar hyaline gland cells. Some of these (Nos. 1, 2, 4) were already noted by Christoffersen (1979), who created four informal varieties ("A, B, C, D"), without achieving clearcut distinctions among specimens (see below). The variation range in our specimens is narrower than in Christoffersen's material and less divergent from the original description and the type series, the single specimen with huge reproductive organs excepted.

Variation is contrasted by ca. 20 invariant characters that in combination distinguish A. neotropica from all congeners, these are: (1) Body length below 1 cm; (2) below 50 segments; (3) no pyriform glands; (4) 20– 30 lentiform epidermal gland cells per segment, (5) all of one type, hyaline; (6) cuticle distinct; (7) brain twice as long as wide, (8) ganglia on prostomial nerves; (9) ganglia of II–IV fused into suboesophageal ganglion, from V on perikarya in segmental ganglia; (10) pharyngeal glands dorsally connected in IV and V, dorsally separate in VI(/VII); no secondary ventral lobes; (11) oesophageal appendages absent; (12) gut epithelium thickened in VI and VII; (13) dorsal blood vessel originating in or near clitellar region; (14) preclitellar nephridia at 6/7 (rarely absent here) and 8/9, never at 7/8; (15) nephridia elongate, anteseptale distinctly longer than wide, narrowed at septum, (16) with large terminal vesicle; (17) pars tumida of posterior midgut ventral, extending over 4–10 segments; (18) clitellum interrupted dorsally and ventrally, (19) with hyalocytes latero-dorsally, not compacted, in 2–3 irregular longitudinal rows, not extending beneath body muscles; (20) spermathecal ectal pores lateral. The diagnosis applies to the entire series investigated for this study, i.e. specimens collected in the Mata Atlântica, the type series of A. neotropica , the reference material of A. neotropica in Righi (1981, those of 1974 were not investigated here) and Christoffersen (1979, not all specimens scrutinized here), the type series of A. becki (character 12 excepted, which was not investigated), and the single specimen from Mexico. Other species of Achaeta differ in at least four characters from this list (exceptions: A. iridescens , see below, and the ill-described A. silvatica , see above, A. hanagarthi ).

As a result, with the outer limits of A. neotropica being clearcut and inner limits not being evident on morphological grounds, we follow Christoffersen (1979) and ascribe all material to one nominal species. We hypothesize, however, that A. neotropica is genetically heterogeneous, a complex of incipient or well-established species, that may be distinguishable with more refined methods than the ones applicable here. As an alternative explanation, one may hypothesize an extraordinary morphological plasticity of this species, partly fixed genetically, partly induced by environmental conditions.

Type series and original description: Part of the variations are already present in the type series. Černosvitov (1937: 156; Figs 19, 20) describes smaller spermathecae, confined to V, and larger ones, extending into VI (not into VII as put erroneously in Schmelz & Collado 2005, Table 1 ). The small ones are present in the two sectioned specimens, the larger ones in the ethanol-preserved material. Černosvitov ascribed the variation to different developmental stages, but the sectioned specimens have a fully developed clitellum and a mature egg, i.e. they are mature just like the other specimens. Spermathecae do not extend beyond VI in the type series, and the sperm funnels are shaped as originally described, i.e. conical. Seminal vesicles are absent. The male glands are not as large as described and illustrated in Černosvitov (1937: 156, Fig. 16), but small and inconspicuous, ca. 35 µm long.

There are a few minor incongruences between original account and type series: In some of the specimens the posterior pharyngeal gland lobes do not extend into VII (comp. Černosvitov 1937, Fig. 21). According to Černosvitov (1937, 156; Fig.: 22), there are 1–2 or 3 rows of epidermal gland cells per segment, resulting in only 6–14 cells per segment; the types, however, have more glands and more than 3 rows per segment, i.e., they agree more with our account. The ganglion in V is not larger than the following ones, contrary to the original description (Černosvitov 1937: 156; Fig. 24). Being the first ganglion behind the suboesophageal ganglion, it is often a little wider and a little shorter than the following ganglia, which may create the impression that it is larger. The clitellum does not extend over XII–XIII as originally described, i.e. over two full segment lengths. Apart from these minor incongruences, the original description is admirably concise.

Černosvitov (1937) describes and illustrates exterior papillae on the prostomium, but the prostomium of the types is smooth, as in the rest of the investigated specimens. Prostomial papillae—exterior sense organsare common in enchytraeids, Achaeta included ( Römbke & Schmidt 1999), and in other microdrile oligochaetes as well ( Yañez et al. 2005), so they may be just more or less conspicuous, depending on some physiological state of the animal, as Christoffersen (1979) already assumed.

Comparison with A. becki: About half of the differences between A. becki and A. neotropica as listed in Schmelz and Collado (2005) have become obsolete after the reinvestigation of the types series of A. neotropica: Location of spermathecal ectal pores, location of most anterior nephridia, size of ganglion in V, size of the male gland. Further possible peculiarities of A. becki are shared by one or another specimen of A. neotropica investigated here, Christoffersen's series included, or they need reinvestigation: (1) A subdivision of the male gland into two separate glandular bodies (Schmelz & Collado 2005: Fig. 1 A,C) is present in one of the specimens of variety "D" in Christoffersen's series of A. neotropica . Such separation of the male gland into two bodies is foreshadowed by the median incision present in the male gland of several specimens investigated here (see also Černosvitov 1937: Fig. 16). The same type of variation has been described for Marionina appendiculata Nielsen & Christensen, 1959 ( Healy 1996) . (2) Figure 1 B in Schmelz and Collado (2005) shows three irregular longitudinal rows of clitellar hyalocytes, whereas only two rows are present, respectively, in the types of A. neotropica , in most of Christoffersen's specimens and in the material collected by us. The single specimen from Mexico, however ( ZMH OL #14518), and whole-mounted specimens of Christoffersen's "variety B" ( MZUSP #350) blur the distinction, with an arrangement of clitellar hyalocytes exactly as illustrated for A. becki . (3) The size difference of coelomocytes (15–18 µm in A. becki , 20–28 µm in A. neotropica ) needs reinvestigation. (4) Spermathecae do not extend beyond VI in the type series and in our material; in A. becki they extend into VII–VIII, in the single specimen from Mexico into VII, and in Christoffersen's "variety D" into VII, VIII or IX. (5) The straight and short vas deferens in A. becki is the only clearly distinguishing character. In all specimens of A. neotropica investigated here the vas deferens describes at least a few coils. However, length varies as well, and in some specimens the vas deferens is fairly short (comp. Fig. 4 E). Furthermore, the straight vas deferens may be an intraspecific variation, for example as a stage of resorption after sexual maturity (compare A. singularis , below). For an appraisal of this possibility, more material from the type locality of A. becki is necessary.

Comparison with Achaeta iridescens Christoffersen, 1979 : Our hypothesis that A. neotropica is a species complex rather than a single and enormously variable species finds support in a comparison of A. neotropica with A. iridescens , a species originally described in the same paper that first highlighted the variability in A. neotropica ( Christoffersen 1979) ; both species were collected at the same site and the same date. A. iridescens is characterized as being "most closely related to A. neotropica " ( Christoffersen 1979: 158) . In fact, the only clearcut differences of A. iridescens are body size (13.5–14.5 mm) and segment number (55–60); the other four distinguishing traits mentioned (ibid.) add just a little bit more to the variability already present in A. neotropica , or they are doubtful and need reinvestigation: Most important here is (1) the presence of true intestinal diverticula in A. iridescens . They are described as a pair of lateral pouches in VII, two-lobed each, with the dorsal lobe larger than the ventral one. A fold of the intestine in the same region occurs in several specimens of A. neotropica (comp. Fig. 4 F, see Černosvitov 1937, Fig. 21), but here the fold is less developed, and circumferal, like a secondary effect of an invagination of the oesophagus backwards into the intestine. 'Differentiation of lateral pouches in A. iridescens ' instead of 'one circumferal pouch' would make a true difference, however, "... in some worms these pouches were less developed than in others, suggesting a homology with the condition found in A. neotropica ..." ( Christoffersen 1979: 158); furthermore, a figure in the original description ( Christoffersen 1979, Fig. 17) suggests that the fold is also dorsal and not only lateral. Further differences, as listed in the original description, are as follows: (2) Arrangement of clitellar gland cells. It is not specified however in what they differ. From Fig. 18 in Christoffersen (1979) it might be inferred that there are up to four hyaline cells per transverse row on each side. (3) Shape of cerebral ganglion; (4) shape of posterior pharyngeal gland lobes. The latter two traits are within the range of variation in A. neotropica as established in this study.

We do not contend here that an A. neotropica -worm of 3–4 mm length and appr. 30 segments is likely to interbreed successfully with an A. iridescens -worm of 14 mm length and 60 segments, but we expect future finds of specimens with intermediate body size and segment number. Christoffersen's variety "D" of A. neotropica is already in the middle between A. iridescens and A. neotropica as originally described. The currently clearcut distinction between A. neotropica and A. iridescens may be a random effect of low sampling intensity in an area as vast as Central and South America, where A. neotropica is apparently widely distributed. On the other hand, A. iridescens and several 'varieties' of A. neotropica were found at the same site and the same date ( Christoffersen 1979), which weakens the alternative hypothesis that environmental factors are responsible for the differences encountered in the material. So these remarks are not meant to question the status of A. iridescens as a species different from A. neotropica , but rather to illustrate our hypothesis that A. neotropica as circumscribed here is a species complex not completely resolvable with conventional morphological comparisons.

Further reference material of A. neotropica: Christoffersen (1979) created four informal varieties, A, B, C, D, to cope with the variations found in material collected in rain forest soils in the state of São Paulo, some 300 km north of our sampling sites, but he also remarked that the discriminating characters were unstable and overlapping (comp. Christoffersen 1979: Table 1 ). We reinvestigated parts of the material. Their quality varies among specimens, full redescriptions are not possible. Variety A, originally represented by four specimens, is said to have nephridia "usually only [at] 8/9" (ibid.); this trait is present in the three remaining specimens of the series (the fourth specimen is represented by a posterior end only). This trait is perhaps without exception and a possible distinguishing character of a subtaxon of A. neotropica ; in no other specimen investigated here could the lack of nephridia at 6/7 be ascertained (which means that nephridia were either present at 6/7 or the character state could not be ascertained). The specimens are ill-preserved, but fully mature and with large and conspicuous clitellar hyalocytes, not recognized by Christoffersen (comp. Christoffersen 1979: Table 1 ). Variety D harbours the largest specimens and the single one with huge sperm-related sexual organs, dealt with above. All specimens (6 out of 10 were suitable for investigation) have spermathecae that extend into VII, VIII or IX with an ampullar diameter between 50 and 86 µm, contrasted by spermathecae in V or VI and a maximum ampullar diameter of 35 µm in our Mata Atlântica specimens and in the types. Varieties B and C are more or less within the variation range of our material, except for the arrangement of clitellar hyalocytes (3 transverse rows) in variety B and a more anterior dorsal blood vessel origin ( XI, congruent with the type series) in C; specimens of the latter variety were not reinvestigated here.

Christoffersen (1979) further describes an ectal thickening of the spermathecal ectal duct near the ectal pore in all specimens. This trait is evident in the material, but not in our specimens, and not in the types, either. The reason for this difference is obscure, perhaps it is a fixation or anaesthetization artefact.

Righi (1974: 128, Fig. 4), in a single specimen from Amazonia, describes and illustrates a deep dorsal transverse groove of the brain at about 1/3 of from its anterior tip, dividing the brain into a conical anterior part and a quadrangular posterior part, in lateral view. This trait, evident also in the reference material of Righi (1981) and in some type specimens, is a fixation artefact caused by strong body contractions; it may nonetheless indicate a histological peculiarity of the brain in this species.

The single specimen from Mexico ( ZMH OL #14518) was found together with a single earthworm specimen, identified and redescribed as Zapotecia mecamecae (Eisen, 1990) (Acanthodrilidae) by Zicsi and Csuzdi (1991); this is the first record of an Achaeta species in Mexico.

Figures

FIGURE 1. Location of the ' Reserva Natural Rio Cachoeira', study area and type locality of Achaeta hanagarthi, A. paranensis and A. singularis spp. nov.

FIGURE 2. Achaeta piti Bittencourt, 1974, specimens from the Cachoeira Natural Reserve. B is a freehand sketch from living specimens, the rest is from stained whole mounts. Bar = 100 µm, valid for A, C, D. A. Submature specimen, side view, anterior 9 segments, with thick body wall and dorsally thick cuticle (1). There are dorsal, lateral and ventral pyriform glands (2,3,4, respectively; only glands of left body side shown). The ventral glands insert more posteriorly than dorsal and lateral glands. The dorsal blood vessel (5) begins at 1 / 3 VIII in this specimen. The thickened septa in 4 / 5 – 7 / 8 are a constant feature of this species. 6: Oesophageal appendage; 7: pharyngeal gland; 8: postpharyngeal bulb; 9: brain; 10: suboesophageal ganglion; 11: ganglion of segment V. Perikarya are omitted here, they are as in Figs 3 B, 4 A. 12: nephridia; 13: intestine with chloragocytes; 14: spermatheca with ectal asymmetry; 15: recesses in frontal prostomial epithelium. B. Oesophageal appendage and pharyngeal glands, dorsal view, as seen in a living specimen. The two separately winding canals in IV, one for each glomerulus in V, were already mentioned in the original description. C. Postclitellar nephridia, side view. The organs lie in consecutive segments, one to the left, one to the right. D. Clitellum and male reproductive system, side view, specimen with thick body wall and cuticle. The dorsal pyriform gland of XI is drawn in (1). The clitellar border cells (2) extend farther ventrally than the granulocytes.

Tables

TABLE 1. Comparison of Achaeta species with reproductive organs shifted one segment forward (testis and sperm funnels in X, ovary, vasa deferentia, male pores and citellum in XII, spermathecae in usual position). Common characters: body length 3 - 5 mm, oesophageal appendages present.

Pyriform glands absent	absent	absent	6 per segment 2 per segment 2 per segment
Lentiform epi- 4–8 dermal glands	8–12	6	absent present, present, inconspicuous inconspicuous
Preclitellar 1 pair nephridia at 7/8	3 pairs at 6/7 – 8/9	2 pairs at 6/7, 7/8	2 pairs 1 pair 1 pair at 7/8, 8/9 at 6/7 at 6/7

TABLE 2. Comparison of Achaeta hanagarthi with congeneric species with sexual organs in normal position and without pyriform glands. Five characters are mainly considered: (1) animals much larger, more than 60 segments; (2) oesophageal appendage in one segment only (either IV or V), or absent; (3) pharyngeal glands with secondary ventral lobes; (4) more than one pair of preclitellar nephridia; (5) clitellum dorsally closed. In the case of insufficiently described species, characters other than (1) - (5) are added. In all species except A. camerani the original description is the only currently available source of morphological information. As to A. becki, A. neotropica, A. iridescens, A. paranensis and A. singularis, see below, descriptions, remarks, and Table 4.

Species name and author A. afolliculata Sesma & Dózsa-Farkas, 1993	Traits that differ from A. hanagarthi sp. nov. (2), (4)
A. brevivasa Graefe, 1980	(2), (3), (4), (5)
A. camerani (Cognetti, 1899) A. etrusca Rota, 1995	(2), (3), (4), (5) (3), (4)
A. gigantea Dózsa-Farkas, 2000	(1), (4), (5)
A. hallensis Möller, 1976 A. iberica Graefe, 1989	(3), (4) (3), (4)
A. indica Prabhoo, 1961	(5)?, coelomocytes strongly granulated
A. littoralis Lasserre, 1968 (4) A. macrocyta Christensen & Dózsa-Farkas, 1999	spermathecal ectal pores ventral (2)?, only one segmental pair of epidermal gland blood vessel from VIII	cells, dorsal
A. segmentata Prabhoo, 1966 A. silvatica Nurminen, 1973	(1), (2) (2)?, sperm funnel as wide as long, collar wider than	funnel body

TABLE 3. Contrasting extremes in the development of sperm-related sexual organs in A. neotropica. The maximum refers to two specimens, one in Christoffersen (1979: 154, Fig. 3), reinvestigated here (MZUSP # 359), one in our collection (Fig. 4 M); the minimum is present in the type series (sections), in our collection (Fig. 4 A, E), and in Christoffersen (1979).

	Minimum	Maximum
Seminal vesicle	absent	large, 2 segments
Sperm funnel size	1/4 as long as body diameter	longer than body diameter
	1.5x as long as wide	3x as long as wide
Sperm funnel collar	as wide as funnel body	with 2 lateral extensions, longer than body diameter
Gland around male pore	20 µm long, 15 µm wide	130 µm long, 70 µm wide
Spermatheca extending over	one segment length, confined to V	five segment lengths, extending into IX
Diameter of spermathecal ampulla	1/7 body diameter	1/2 body diameter