Limnodrilus sulphurensis n. sp., from a sulfur cave in Colorado, USA, with notes on the morphologically similar L. profundicola (Clitellata, Naididae, Tubificinae) Author Fend, Steven V. Author Liu, Yingkui Author Steinmann, David Author Giere, Olav Author Barton, Hazel A. Author Luiszer, Fred Author Erséus, Christer text Zootaxa 2016 4066 4 451 468 journal article 51173 10.11646/zootaxa.4066.4.6 644db33c-001e-4067-92e0-a5fbf7b384cd 1175-5326 261849 87790552-1DFF-41DF-B6C4-5BD43D19E998 Limnodrilus sulphurensis Fend, Liu & Erséus n. sp. ( Figures 1–3 ) Holotype . DMNS ZE .46275, a whole mounted, sexually mature and DNA barcoded specimen; i.e., a small posterior part of the worm was used for DNA extraction. COI barcode = haplotype 1 (CE10482, GenBank KT692957 ; see also "Barcodes" below). Type locality. Colorado, Routt Co., City of Steamboat Springs, Sulphur Cave, high H2S stream in dark zone, W106.841050° , N40.84759° . 11-Apr-2010 . Collected by David Steinmann and Fred Luiszer. Paratypes . All from the type locality. DMNS ZE .46276–46280: 11-Apr-2010 , 2 whole-mounts, 3 dissected. SMNH Type 8779, 20-Sep-2008 , 1 whole-mounted, mature specimen with COI barcode (CE7491, haplotype 2, GenBank KT692956 ). USNM 1283518, 18-Aug-2007 , 1 whole-mount, mature, with COI barcode (CE7489, haplotype 1); USNM 1283512-1283516, 11-Apr-2010 , 2 whole-mounts, 3 dissected; USNM 1283517, 1-Aug- 2009 , 1 sagittally sectioned. CASIZ 197459, 1-Aug-2009 , 1 whole-mount; CASIZ 197460, 11-Apr-2010 , 1 wholemount, 3 dissected. ZMH Ol 15446 -15450, 1-Aug-2009 , 2 whole-mounts; 11-Apr-2010 , 3 dissected. All collected by David Steinmann and Fred Luiszer. Additional material. The type locality, 18-Aug-2008 , SMNH 150175 (CE7486); 20-Sep-2008 , SMNH 150176 (CE7490); both whole-mounts with COI barcodes (haplotype 2). 9-Mar-2009 , 10 whole-mounts (6 immature), 1 sagittally sectioned. 1-Aug-2009 , 5 whole mounts. 11-Apr-2010 , 5 whole mounts, 12 dissected. Colorado, Routt Co., Surface spring near Sulphur Cave, 11-Apr-2010 , 4 whole mounts (immature). SMNH 150177 , 11-Apr-2010 , 1 whole-mount, immature, with COI barcode (CE10479, haplotype 1). Colorado, Routt Co., Black Sulphur Springs, 11-Apr-2010 , 1 whole mount, 1 whole mount with COI barcode ( SMNH 150178 , CE10477, haplotype 1). All collected by David Steinmann and Fred Luiszer. Etymology. Named for the type locality. Description. Specimens relaxed in dilute alcohol and fixed in FAA 18–25 mm long; formalin-preserved worms more contracted, 12–19 mm . Width 0.52–0.99 mm in VIII, to 1.05 mm in XI, tapering in posterior segments. 51–78 (median 64) segments. Prostomium rounded, slightly wider than long in FAA-preserved worms; broadly conical in formalin-fixed specimens ( Fig. 1 A). Clitellum weak in all available material, X–XII. Segments II–VII or VIII with 5–11 fully developed chaetae per bundle (median 7 dorsal, 8 ventral), and a similar but indeterminate number of partially developed replacement chaetae, all in a descending, fan-like series ( Fig. 1 B). Number of chaetae per bundle decreasing posteriorly; by XX 2–3, posterior segments at most 1–2; most specimens with few or no chaetae from about the posterior 1/4 of the worm. All chaetae bifid, sigmoid, with variably developed nodulus at about the distal 1/3 ( Fig. 1 B–E). Chaetae in anterior part of body (to about X) with teeth approximately equal in length and thickness; teeth long (12–20 µm), nearly straight (never strongly hooked), and nearly perpendicular to the distal part of the shaft ( Fig. 3 A–C). Chaetae gradually smaller, and teeth shorter and less strongly angled relative to shaft in posterior segments ( Figs. 1 C–E, 3D). Ventral chaetae absent in XI in mature worms; no modified spermathecal chaetae in X. Chaetal length 84–185 µm in II–X (maximum length in IV–VI); 90–125 in XI–XX; 50–100 µm in posterior segments. In anterior (preclitellar) segments, the ventral chaetae about 30% longer and slightly thicker than the corresponding dorsal ones ( Fig. 1 C vs. 1E). Pharynx in II–III, weakly differentiated from esophagus; about equally developed dorsally and ventrally. Chloragogen cells dense on gut beginning in VI, with dark inclusions in worms fixed in the field ( Fig. 3 E,F). Branches of dorsal blood vessel join to form the ventral vessel at about 2/3. A well-developed supra-intestinal vessel from about 5/6, joining the dorsal vessel in IX; the conjoined "dorsal" vessel curving ventrad along the left side of the gut, and remaining in a lateral or even ventrolateral position until the posterior segments. Dorsal and ventral vessels connected by long, convoluted, commissural vessels in II–VII and in IX–XI. In VIII, commissures shorter and modified as "hearts"; these greatly dilated (to 150 µm), but not obviously muscular, and joining the supra-intestinal vessel instead of the dorsal. A network of fine peripheral, capillary blood vessels visible in posterior segments, extending into the epidermal layer ( Fig. 3 F); relatively sparse beginning in about XXV, forming a dense network beginning in about XXXV. Branches of capillary vessels coalesce to a short ventrolateral vessel, which joins the ventral vessel near the posterior septum. A second main branch from the capillary blood vessels joins a pair of dorsolateral vessels extending from the dorsal vessel to both sides of the ventral body wall. These dorsolateral vessels join the dorsal vessel near the posterior septum, the one going to the right side passing over the gut; apparently no direct connection to the ventral vessel ( Fig. 1 F,G). FIGURE 1. Limnodrilus sulphurensis n. sp. from Sulphur Cave, exterior and somatic characters. A. Anterior ends of two whole, unmounted worms. B. Ventral chaetal bundle in IV, from a mature worm (n=19, including 10 partially developed). C. Left to right: two ventral chaetae in V, one from XX, from a mature worm. D. Ventral chaetae (n=13, including 6 partially developed) in V, from a very small, immature worm (diameter 0.3 mm). E. Dorsal chaetae from a mature worm, two each from V, X, and XXX. F. Ventral view of blood vessels in a posterior segment, showing junction of capillary vessels with ventral blood vessel, lateral vessels and (laterally displaced) dorsal blood vessel. G. Lateral view of dorsal and ventral blood vessels, as in Figure 1F. FIGURE 2. Limnodrilus sulphurensis n. sp. from Sulphur Cave (A–D, F, G) and from Black Sulphur Springs (E), reproductive characters. A. Atrium and penis sheath of an unmated, nearly mature worm. B. Atrium and penis sheath from a mature, mated worm. C. Male duct from a mated worm. D, E . penis sheaths. F. Spermatozeugmata. G. Spermatheca. FIGURE 3 . Limnodrilus sulphurensis n. sp. from Sulphur Cave. A. Dorsal and ventral chaetae from anterior segments, worm mounted in Canada balsam. B. Ventral chaetae in V, worm mounted in CMC-10. C. Ventral chaetae in VI, from the holotype. D. Dorsal chaetae in XLVII. E. Lateral view of segments V and VI, in a cleared whole mount, showing beginning of chloragogen in VI. F. Sagittal section from a posterior segment, showing dark chloragogen, a transverse blood vessel, and capillary vessels in epidermal layer. G, H. Sections through atria, showing junction with prostate gland. I. Penis sheaths, from a whole mount. J. Penis sheath, from the holotype. K. Ectal end of penis and sheath, from a sectioned, mature worm fixed in formalin. L. Ectal end of penis sheath, from a sectioned, nearly mature, unmated worm fixed in FAA. M. Spermatheca, from a transverse section. N. Spermatozeugma, from the holotype. O. Detail of spermathecal duct. Scale bars: 50 µm (A–D, F–H, K, L, N, O), 100 µm (E, I, J, M). Nephridia in XIII and many posterior segments; a convoluted duct covered with indistinct, granular cells forming a dense mass, extending ventrally through most of segment, terminating in a short ectal duct, which widens into a small vesicle at the nephridiopore, anterior to the ventral chaetae. Male ducts and spermathecae paired. Male pores narrow, transverse openings, in line with ventral chaetae, at about the posterior 1/4 of segment XI ( Fig. 1 A). Spermathecal pores prominent and transverse, in line with ventral chaetae, midway between chaetae in X and anterior septum (9/10). Male funnel 100–120 µm long, either narrowly conical with upper and lower lips about the same length, or with lips laterally spread; projecting into X from 10/11. Vas deferens very long, about 4000–6400 µm, diameter 22–48 µm, generally thinnest in formalin-preserved worms; of nearly uniform diameter and histology ( Fig. 2 C). Vas deferens composed of a single layer of ciliated epithelium 7–10 µm thick, without obvious muscle layer. Atrium narrowly fusiform, narrowing gradually to vas deferens and ejaculatory ducts; length 240–490 µm, diameter to 65–94 µm ental to prostate attachment, 42–85 µm in ectal part ( Fig. 2 A–C). Atrial epithelium ciliated, cells granular, irregular, indistinct in ental part; often more regular and less granular in ectal part ( Fig. 3 H). Atrium joined near midpoint by a single, stalked prostate gland ( Figs. 2 B, 3G,H); prostate a fan-like cluster of irregular lobes; entire mass 240–380 µm wide; prostatic cells granular and indistinct. Ectal (ejaculatory) duct of atrium 250–480 µm long by 26–38 µm wide, with a thin (5–7 µm), non-ciliated epithelium and indistinct muscle layer about 2–3 µm thick; appearing slightly wrinkled and less evenly cylindrical than vas deferens. Penis without obvious spiral muscles; penis sheath surrounded by a closefitting sac to 120–240 µm deep, consisting of columnar cells and a thin, indistinct muscle layer. Shaft of cuticular penis sheath nearly tubular, expanded slightly at ends; straight or somewhat curved in slide-mounted specimens ( Figs. 2 D, 3I ,J); wall thickness up to 3–4 µm. Sheath length 240–410 (median 333) µm; diameter in middle 34–49 µm, 43–82 µm at ental end, 40–60 µm at ectal end (below head); ratio of length to basal width 4.1–5.7 (median 4.9). Ectal end ("head") broad and plate-like, appearing flat or recurved ( Fig. 3 I,K,L); plate circular and symmetrical, diameter 120–170 µm in fully mature worms. Plate may be indistinct, with very thin cuticle, particularly in unmated specimens ( Fig. 2 A). Ectal part of spermathecal duct about 200–300 µm long, thick and pyriform; widest entally, to 121–145 µm thick; ectally narrowing to 80–95 µm ( Fig. 2 G). Duct composed of an outer, transverse-circular muscle layer, and an irregular epithelium ( Fig. 3 M,O); lumen narrow ectally, wider entally. Duct surrounded by an irregular layer of cells. Remaining (ental) part of duct narrower ( Fig. 3 O), but often medially inflated ( Fig. 2 G); about 60–100 µm long. Spermathecal ampulla nearly spherical to irregular, sacciform. Spermatozeugmata 200–310 µm long, narrow, slightly widened at one end, to 37–55 µm ( Figs. 2 F, 3N). Barcodes. Data on the COI barcodes of seven specimens are summarized in Table 2 . Six of them were successfully sequenced to the standard length of 658 base pairs (bp), all barcodes covered positions 2–658 in the alignment. Two different haplotypes (Ht 1 and 2 in Table 2 ) within this interval were recognized, with variation (uncorrected p-distance 0.15%) only in one position (181). When blasting the haplotypes against the BOLD and NCBI/Genbank databases,>80% resemblance was found with several other North American clitellates, including a barcoded worm (about 82 % similar) identified as Limnodrilus hoffmeisteri (Genbank EF089358 ; http:// www.ncbi.nlm.nih.gov/nucleotide/). The closest match (90.21%) was between haplotype 1 and an unidentified " Tubificidae " collected in Manitoba, Canada , and in Early-Release status in BOLD (http://www.boldsystems.org/). Remarks. The new species clearly belongs to Limnodrilus as defined by Brinkhurst (1971) : vas deferens long and ciliated, atrium short with a single large prostate, ejaculatory duct long, leading to a thick, cylindrical penis sheath; genital chaetae absent, somatic chaetae all bifids; spermatozeugmata present in spermathecae; prominent coelomocytes absent. Limnodrilus sulphurensis is most easily differentiated from congeners by the very long, sharply angled teeth of both ventral and dorsal chaetae in anterior segments. The distinctive chaetae are not restricted to mature worms, and are consistently seen in very small, immature specimens from Sulphur Cave ( Fig. 1 D), as well as in worms from nearby surface springs. The large number and continuous progression of developing, replacement chaetae in anterior segments of L. sulphurensis are also unusual. Interestingly, Fig. 1 A in Snimshchikova (1998) shows several replacement chaetae in anterior ventral bundles of Limnodrilus dybowskii ( Grube, 1873 ) . Based on proportions and morphology of the penis sheath, the characters most often used to distinguish Limnodrilus species, L. sulphurensis appears closest to the widespread L. profundicola , and will key to that species using Brinkhurst (1971: 463) . It also somewhat resembles the widespread Limnodrilus udekemianus Claparède, 1862 , the Jamaican Limnodrilus variesetosus Brinkhurst, 1979 , and a group of endemic Lake Baikal species associated with L. dybowskii . Taxonomy of L. profundicola is problematic, due to a very limited original description and lack of useable type material; in the following we attempt to clarify some of these issues with observations on penis sheaths and chaetae in museum specimens and in new collections (see below). Based on new and previously published measurements, the penis sheath is similar in general shape and proportions to that of L. sulphurensis , but is smaller ( Table 1 , Fig. 4 ), and the terminal expansion ("head") is smaller, but usually more distinct. The profundicola atrium is also smaller ( Table 1 ). Limnodrilus profundicola has rather ordinary chaetae for the genus, with short, nearly equal teeth. Chaetal bundles have a similar number of well-developed chaetae in the two species ( Table 1 ), although profundicola does not have the large number of developing chaetae seen in L. sulphurensis (see below; Fig. 1 B vs. 4G). Chaetae of Limnodrilus alpestris Eisen, 1879 , and Limnodrilus monticola Eisen, 1879 —both taxa that have since been synonymized with profundicola ( Brinkhurst 1965 ) —are not well described. There seems to be good agreement in the literature on the status and diagnostic characters of L. udekemianus . Penis sheaths of this species may have similar proportions to those of L. sulphurensis (length/width ratio 2–5 [ Kennedy 1969 ], or up to 4 [ Timm 2009 ]), but despite high variability, they are usually illustrated as shorter and broader, gradually expanded, and without a plate-like head ( Claparède 1862 , Pl. I, Fig. 4 ; Brinkhurst 1971 , Fig. 8.4B; Kennedy 1969 , Fig. 9). Nevertheless, the distal end may be variably expanded and reflexed ( Dzwillo 1984 , Fig. 3 ; Piguet 1913 , Fig. 10a). The anterior chaetae of L. udekemianus are considered diagnostic: distal teeth are enlarged, thickened, and sometimes strongly curved, but the proximal teeth are of normal size, differing from those of L. sulphurensis (e.g., Kennedy 1969 , Fig. 1 ; Brinkhurst 1965 , Fig. 4 ). Timm (2009) also considers the long, thin posterior end of the body to be distinctive. Penis sheaths of L. variesetosus appear quite similar to those of L. udekemianus ; the anterior ventral chaetae also have a very long distal tooth, but posterior chaetae have a short, thin distal tooth ( Brinkhurst 1979 , Figs. 1 A,E; Rodriguez 2002 , Fig. 2 B,C). The male duct of L. dybowskii , as illustrated in Fig. 2 of Snimshchikova (1998) , appears quite similar to that of L. sulphurensis ; measurements of dybowskii atria and penis sheaths (400–520 and 350 µm, respectively) are also similar. Two other Lake Baikal endemics, Limnodrilus nitens ( Semernoy, 1982 ) and Limnodrilus tendens ( Semernoy, 1982 ) appear to have similar penial morphology to dybowskii ( Semernoy 2004, Figs. 117,118 ); the latter was synonymized with dybowskii by Snimshchikova (1998) . It should be noted that Semernoy (1982) compared the penis sheath of L. tendens to that of L. helveticus , and gave similar measurements for L. nitens . All three of the Baikal species are distinguished from congeners by enlarged and highly modified chaetae in some anterior ventral (but not dorsal) segments. The enlarged chaetae are fewer in number and simple-pointed (or with small distal tooth), and thus bear little resemblance to those of L. sulphurensis n. sp. ( Snimshchikova 1998 , Figs. 1 , 3 ; Semernoy 2004 , Figs. 116, 117, 118). The material. The morphological description is largely based on material collected on three dates in March 2009 (fixed in FAA ), August 2009 , and April 2010 (fixed in 10% formalin). The March collection included only two mated specimens (only one with developed eggs), a few "nearly mature", and many immature specimens. Penis sheaths in particular were not well developed, and the ectal ends were indistinct. Most measurements used in the description were based on the large series of mature, formalin-fixed worms collected in April 2010 . The formalin-fixed material from August contained several mated worms and many immatures. Some histological differences among collection dates appear related to the fixation method, as formalin-fixed specimens were highly contracted, and tissues appeared shrunken and separated (e.g., Fig. 3 K [formalin] vs. 3L [ FAA ]). Chaetae of specimens mounted in CMC-10 are similar to those in Canada balsam, but teeth appear slightly thicker ( Fig. 3 B vs. 3A). Penis sheaths have similar length and proportions, although the thinner cuticle of the head may be indistinct in CMC-10. TABLE 1. Selected measurements of L. sulphurensis n. sp. and L. profundicola from new material, compared with literature values for profundicola . Values are mean and (range) , where available.

L. sulphurensis n. sp.	Number of chaetae Maximum chaeta in V, ventral bundle length in V	Penis sheath length Penis sheath Penis sheath head Atrium length length/ basal width width
Sulphur Cave (n=26–33)	8.1 (4–11) 165 (146–185)	333 (240–410) 4.9 (4.1–5.7) 148 (120–170) 366 (240–492)
Black Sulphur Spring (n=2)	(8–9) (150–175)	(395–415) (4.1–5.6) (168–192) (385–550)
L profundicola new material
Estonia and European Russia (n=12–16)	6.4 (4–8) 96 (79–110)	191 (156–221) 4.6 (3.8–5.7) 65 (48–77) 171 (150–190)
Kamchatka (Lake Kuril'skoe) (n=3)	5.8 (5–7) 114 (103–122)	241 (226–270) 4.9 (4.3–5.3) 66 (58–72) 236 (220–252)
Japan, Lake Biwa (n=4)	6.3 (6–7) 94 (89–98)	194 (180–221) 4.8 (4.4–5.1) 59 (53–65) 171 (145–190)
Lake Michigan (n=15–22)	6.1 (4–9) 113 (89–139)	222 (178–270) 4.7 (3.8–5.7) 63 (53–79) 186 (145–252)
L. profundicola literature values
Sokol’skaya 1972, Kamchatka (as L . helveticus )	(252–340) (4–6)
Piguet 1913, Switzerland ( L. helveticus )	maximum8	206 ( 148–264) 5 69
Brinkhurst 1965 and Kennedy 1969, Europe and North America (incl. helveticus , monticola , profundicola )	6.4 ( 5–9 )	240, 340, 270 4.5 ( 2–7 )
Ohtaka 1992, Japan	3.2
Cui et al. 2015, Tibet, China	(6–11)	290 5 238
Semernoy 2004, Lake Baikal	(198–234) (4–5)

TABLE 2 . COI barcodes of 7 specimens of Limnodrilus sulphurensis n. sp. , from Sulphur Cave dark zone (SC), Sulphur Cave surface spring (SS) and Black Sulphur Spring (BS). Two haplotypes (Ht #1, #2) are recognized in positions 2-658, i.e. the interval covered by all sequences; 658 base pairs (bp) are the targeted length of this barcode, but first nucleotide is unknown for 1 specimen.

GenBank#	Individual#	Length	Ht#	Site	Museum voucher
-	CE7486	658 bp	2	SC, 18 Aug 2008	SMNH 150175
-	CE7489	658 bp	1	SC, 20 Sep 2008	USNM 1283518 **
-	CE7490	658 bp	2	SC, 20 Sep 2008	SMNH 150176
KT692956	CE7491	658 bp	2	SC, 20 Sep 2008	SMNH Type 8779**
-	CE10477	657 bp	1	BS, 11 Apr 2010	SMNH 150178
-	CE10479	658 bp	1	SS, 11 Apr 2010	SMNH 150177
KT692957	CE10482	658 bp	1	SC, 11 Apr 2010	DMNS ZE.46275*

* Holotype (sexually mature); ** Paratypes (mature) Earlier ( September 2007 and August 2008 ) collections included several mature worms, but they had been preserved in strong alcohol, and thus were brittle and contracted. Chaetae were mostly broken, male ducts invisible against the dark chloragogen, and penis sheaths broken and twisted. Distribution. Other than the cave habitat, L. sulphurensis has only been collected at the adjacent surface spring and at Black Sulphur Spring. The distinctive chaetal morphology allows the attribution of immature specimens, confirmed by reproductive characters in the two mature worms ( Fig. 2 E, Table 1 ). DNA, as well as proximity of the three sites, suggests that all are from the same population.