Zoologica Scripta Phylogenetics of Acrocirridae and Flabelligeridae (Cirratuliformia, Annelida) KAREN J. OSBORN &GREG W. ROUSE Submitted: 26 July 2010 Osborn, K. J. & Rouse, G. W. (2010). Phylogenetics of Acrocirridae and Flabelligeridae Accepted: 5 October 2010 (Cirratuliformia, Annelida). — Zoologica Scripta, 40, 204–219. doi:10.1111/j.1463-6409.2010.00460.x When seven deep-sea, swimming cirratuliforms were recently discovered, the need for a thorough phylogenetic hypothesis for Cirratuliformia was clear. Here, we provide a robust phylogenetic hypothesis for the relationships within Acrocirridae and increase the taxon sampling and resolution within Flabelligeridae based on both molecular (18S, 28S, 16S, COI and CytB) and morphological data. Data partitions were analyzed separately and in combination. Acrocirridae and Flabelligeridae were reciprocally monophyletic sister groups when rooted by cirratulids. The seven recently discovered species form a clade within Acrocirridae and will be designated as four genera based on phylogenetic relationships and apomorphies. A revised diagnosis is provided for Swima, restricting the genus to three spe- cies distinguished by a thick gelatinous sheath, transparent body, simple nuchal organs, a single medial subulate branchia, and four pair of small segmental branchiae modified as elliptical, bioluminescent sacs. Helmetophorus and Chauvinelia are maintained as separate genera based on morphological differences. Evidence for flabelligerid branchiae being seg- mental is provided, the papillae on segment two of most acrocirrids is confirmed to be the nephridiopores, and scanning electron microscopy is used to examine acrocirrid spinous chaetae in comparison with flabelligerid segmented chaetae. Corresponding author: Karen J. Osborn, Institute of Marine Science, University of California, Santa Cruz, Earth & Marine Sciences Building, 1156 High Street, Santa Cruz, CA 95064, USA. E-mail: [email protected] Greg W. Rouse, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093-0202, USA. E-mail: [email protected] Introduction abyssorum McIntosh 1885 (Mesnil 1899; Fauvel 1916), The recent discovery of seven new members of Acrocirri- Flota spp. and Poeobius meseres Heath 1930, and these like- dae (Fig. 1; Table 1; Osborn et al. 2009) emphasized the wise differ dramatically from their benthic relatives. These need for a thorough phylogenetic treatment of Cirratuli- swimming flabelligerids and acrocirrids possess features, formia. Addition of these new species to Acrocirridae, one e.g. thick gelatinous sheathes and long, clavate papillae, of which, Swima bombiviridis, was formally described in numerous elongate chaetae (P. meseres excepted), broad, Osborn et al. (2009), markedly increased the morpho- flat chaetae (two of the recently discovered species), trans- logical and size diversity within the group. The striking parent bodies, and bioluminescence that may be homolo- differences observed between the recently discovered gous or convergent. Yet to understand the origin of the swimming species and other known acrocirrids were not unusual structures found in the various swimming species surprising considering the pelagic (living in open water, and to understand the evolutionary transformations that not associated with the seafloor) and benthopelagic (living took place in the transition from benthic to pelagic, a both in the water column and associated with the seafloor) well-resolved evolutionary history of Cirratuliformia is lifestyles of the new species. Interestingly, some acrocirrids needed. Helmetophorus rankini Hartman 1978; Chauvinelia spp. and Rouse & Fauchald (1997) found Acrocirridae, Cirratuli- possibly even some Acrocirrus were previously suspected of dae, Ctenodrilidae, Fauveliopsidae, Flabelligeridae, Floti- being able to swim (Banse 1969; Averincev 1980; Kirkeg- dae, Poeobiidae and Sternaspidae formed a clade later aard 1982). Also, several flabelligerids are pelagic, Buskiella named Cirratuliformia (Rouse & Pleijel 2001). Poeobiidae 204 ª 2010 The Authors d Zoologica Scripta ª 2010 The Norwegian Academy of Science and Letters, 40, 2, March 2011, pp204–219 K. J. Osborn & G. W. Rouse d Acrocirrid phylogeny A B C D E F G H I Fig. 1 Swimming acrocirrids. —A. Swima bombiviridis, photo C. W. Dunn. —B. ‘Tawi-tawi Bomber’, photo L. P. Madin. —C. Chauvinelia arctica ZMUC POL-0159. —D. ‘Shining Bomber’ missing all elliptical branchiae. —E. ‘Squidworm’ missing a branchia, photo M. Aw. —F. ‘Horned Bomber’ missing three elliptical branchiae. —G. ‘Tiburon Bomber’ missing six elliptical branchiae, photo F. Pleijel. —H. ‘Juanita worm’ missing seven branchiae, photo F. Pleijel. —I. Helmetophorus rankini type USNM-46749. Scale bars: A–B and D–H = 10 mm, C and I = 1 mm. ª 2010 The Authors d Zoologica Scripta ª 2010 The Norwegian Academy of Science and Letters, 40, 2, March 2011, pp204–219 205 206 Acrocirrid phylogeny d K. J. Osborn & G. W. Rouse Table 1 Characteristics of swimming acrocirrids Sizea Taxa Authority Type locality Depth (m) (mm) Subulate branchiae Main branchiae Nuchal organs Neurochaetae Apomorphy Swima bombiviridis Osborn et al. 2009 Monterey Canyon, NE 2732–3744 28 Single medial Elliptical, smallb Straight ridge Capillary Transparent anterior gut Pacific b ª Shining Bomber This report Monterey Canyon, NE 3267–3625 30 Single medial Elliptical, small Straight ridge Capillary Dark pigmented anterior gut 2010 The Authors Pacific Tawi-tawi Bomber This report Celebes Sea, 2836 45 3 in 1 row Elliptical, smallb Straight ridge with Compound 3 subulate branchiae, broad, flat compound Philippines single 180° bend neurochaetae Horned Bomber This report Astoria Canyon, NE 1863 42 6 pairs in 3 rows Elliptical, largeb Convoluted ridgec Compound Number and arrangement of subulate Pacific branchiae d b c Zoologica Scripta Tiburon Bomber This report Juan de Fuca Ridge, 2367 59 8 pairs in 4 rows Elliptical, large Convoluted ridge Compound Notochaetae broad and flat with sharp NE Pacific point and 90 degree elbow at tip Helmetophorus Hartman 1978 Weddell Sea, 3111 6 3 pairs in 2 rows Elliptical, smallb Convoluted ridgec Capillary Helmet-like cephalic hood rankini Antarctica Chauvinelia Laubier 1974 Bay of Biscay, NE 4455 16 3 pairs in 2 rows Unknown Convoluted ridgec Compound Unknown ª 2010 The Norwegian Academy of Science and Letters, biscayensis Atlantic Chauvinelia arctica Averincev 1980 Polar Sea, Canada 3290–3380 20 0–3 pairs in 3 rows Unknown Convoluted ridgec Compound Foliose palps Squidworm This report Celebes Sea, 2028–1912 90 None Elongate Convoluted ridge Capillary Broad, flat notochaetae abruptly tapered to Philippines extending onto minute tip oppositely branched, free- standing structures Juanita worm This report Juan de Fuca Ridge, 2257–2291 41 None Elongate Convoluted ridge Pseudocompound 3 elongate featureless anterior segments, NE Pacific extending onto nuchal organs extending onto spiral, free- oppositely branched standing structures, pseudocompound and spiral, free- neurochaetae standing structures aMaximum length of types; bSmall is less than half width body at widest, large is more than half width body at widest; cWith multiple 180° bends. 40 , 2, March 2011, pp204–219 K. J. Osborn & G. W. Rouse d Acrocirrid phylogeny and Flotidae were recently synonymized with Flabelligeri- (Burnette et al. 2005; Osborn & Rouse 2008; Osborn et al. dae (Poeobius by Burnette et al. 2005; Flota by Osborn & 2009), but many questions remain, and a combined mole- Rouse 2008) and Ctenodrilidae placed within Cirratulidae cular and morphological approach has not yet been (Bleidorn et al. 2003a; Rouse & Pleijel 2003). The question attempted. Here, we use morphological (29 characters) of the affinities of Sternaspidae and Fauveliopsidae remain and molecular [18S, 28S, 16S, COI and Cytochrome B unresolved, but Struck et al. (2007) and Rousset et al. gene (CytB) sequences] evidence to reconstruct the evolu- (2007) both found these taxa to be sistergroups and not clo- tionary history of 17 flabelligerids and 11 acrocirrids, with sely related to other Cirratuliformia. Thus, Cirratuliformia seven cirratulids as nominal outgroups. Our objectives could be considered to consist of three families, Cirratuli- were to: (i) place the seven recently discovered acrocirrids dae, Acrocirridae and Flabelligeridae. These three families in a stable phylogenetic framework in preparation for the share several distinctive features including: a pair of large description of the six remaining undescribed species (Os- anterior nephridia, chlorocruorin as a blood pigment, and a born, Haddock & Rouse in review; Osborn, Madin & heart body (Banse 1969), as well as paired palps (Rouse & Rouse in review; Osborn & Pleijel in prep.), (ii) use com- Fauchald 1997), though whether any of these are apomor- parative methods to determine structure, function and phic is not yet resolved as all of these features are found in homology of morphological features of acrocirrids and fla- other polychaete groups (Rouse & Fauchald 1997). belligerids, (iii) elucidate historical relationships within Cirratulidae and, to an extent, Flabelligeridae were for Flabelligeridae and Acrocirridae and (iv) generate an evo- many years ‘dumping grounds’ for taxa of uncertain affini- lutionary framework on which to assess questions about ties and both suffer from poorly understood phylogenies. adaptations for life in the water column. Acrocirridae was erected from two genera previously included in Cirratulidae, Acrocirrus and Macrochaeta, by Materials and