taxonID	type	description	language	source
03821554FFCD310FC9239A32FC06FC71.taxon	description	Henricia hedingi Henricia perforata Dense coverage of spines on dorsal pseudopaxillae Dorsal skeleton open reticulum of thin plates Dorsal pseudopaxillae crescent-shaped Dorsal spines single or few, in one to two irregular rows Dorsal spines with short distal thorns Spines large with blunt, rough, distal end Henricia lisa ingolfi Henricia eschrichti Few spines (<10) on dorsal pseudopaxillae Dorsal spines usually cylindrical closely set in irregu- Dorsal spines with three to five unequally long distal thorns lar double row Adambulacral spines in irregular double row Ventral skeleton not in distinct rows Henricia sanguinolenta Henricia oculata Sparse coverage of spines (10 – 15) on dorsal Dorsal plates crowded with up to 25 spines in multiple pseudopaxillae rows Dorsal spines with three to seven short distal thorns Dorsal spines short and cylindrical Adambulacral spines in irregular double row Ventral skeleton in distinct transverse and longitudinal rows Henricia spongiosa Dense coverage of spines (30 – 40) on dorsal pseudopaxillae Dorsal pseudopaxillae oval or circular Adambulacral spines in irregular triple row Henricia pertusa Sparse coverage of spines (10 – 15) on dorsal pseudopaxillae Dorsal spines with three to five unequally long distal thorns Adambulacral spines in three to four irregular rows Henricia cylindrella Few spines (<10) on dorsal pseudopaxillae Dorsal spines with a single long distal thorn Two furrow spines within each group overlap to some degree, and species identification based on morphology is still difficult, not only for those unfamiliar with the morphology of asteroid echinoderms, but also for specialists. Among the taxa revised by Madsen (1987), H. eschrichti, H. hedingi, H. lisa ingolfi and H. spongiosa are omitted from the later work by Clark & Downey (1992), who only focused on fauna collected as far north as Belle Isle, Canada in the west and Trondheim, Norway in the east. Moreover, Clark & Downey (1992) only mentioned H. perforata in their identification key. One reason for the omission of these taxa is that the two works were completed roughly simultaneously (despite the difference in publication date; see Clark & Downey, 1992). Consequently, Madsen (1987) is the most recent authority on species of Henricia in the north Atlantic. An alternative approach used to simplify species diagnosis has been to use molecular data instead of, or in addition to, morphological characters. DNA barcoding with an appropriate species-specific genetic marker (Hebert et al., 2003; Pečnikar & Boznan, 2014) has become a commonplace alternative to morphological species identification when correct species designation is difficult given variable or few, easily assessed diagnostic morphological characteristics. Typically for Metazoa, the gene coding for cytochrome c oxidase subunit I (COI) has been used as a species-specific marker, and, previously, COI has been shown to be effective in delineating echinoderm species (Ward, Holmes & O’Hara, 2008). Even though it has been challenging to design primers that amplify COI consistently in asteroids, especially Henricia (KEK, personal observations; Hoareau & Boissin, 2010; Zulliger & Lessios, 2010), the DNA barcoding approach has been used successfully to assess species identification and phylogenetic relationships in some asteroid genera (Linckia: Williams, 2000; Astropecten: Zulliger & Lessios, 2010; Luidia: Xiao et al., 2013; Echinaster: Lopes et al., 2016). Alongside the use of COI for DNA barcoding, some studies have also employed sequences of the 16 S ribosomal subunit as species-specific markers (Naughton & O’Hara, 2009; Mah & Foltz, 2011; Janosik & Halanych, 2013; Lopes et al., 2016). Molecular data (allele frequencies at the glucose- 6 - phosphate isomerase locus) have previously been shown to support the distinction of the ‘ pertusa group’ and the ‘ perforata group’ (Ringvold & Stien, 2001), but the use of DNA barcoding for identification of Henricia species in the North Atlantic Ocean has not been explored. The difficulty of identifying species in Henricia using only morphological characteristics also creates difficulty when attempting to use a DNA barcoding approach. This is because DNA barcoding requires a reference database of barcode sequences from known species for comparison and identification of unknowns (Hebert et al., 2003; Hajibabaei et al., 2007). When known species cannot be identified with certainty, it is difficult to develop a reference database. Moreover, when there is no reference database from known species, the DNA sequence chosen as a barcode cannot be evaluated for its specificity and utility in species identification. Indeed, errors in public databases, including species misidentification, are not uncommon for DNA barcoding data sets and create problems for researchers hoping to use this method (Bucklin, Steinke & Blanco-Bercial, 2011). In this study, we evaluated whether (1) traditional classification was possible for Henricia collected from the North Atlantic Ocean, following Madsen’s (1987) key and morphological diagnostics; (2) our classifications based on morphology could be supported by phylogenetic analysis of DNA sequence data from COI and 16 S genes and (3) clades identified in the molecular analysis were also supported by a re-evaluation of morphological features. Our approach is cautious, but realistic, considering that both morphological and molecular data could be subject to misinterpretation. Importantly, following this approach allows us to evaluate how well the different data sets can be used to distinguish species and allows us to identify those cases for which additional study is needed.	en	Knott, K Emily, Ringvold, Halldis, Blicher, Martin E (2018): Morphological and molecular analysis of Henricia Gray, 1840 (Asteroidea: Echinodermata) from the Northern Atlantic Ocean. Zoological Journal of the Linnean Society 182 (4): 791-807, DOI: 10.1093/zoolinnean/zlx066, URL: https://academic.oup.com/zoolinnean/article/182/4/791/4558517
03821554FFCB310ACABF9821FB7DF997.taxon	description	61.71193 61.97528 65.81902 58.87694 60.29781 information analyses of clade. SMNH to. Continued 49364 show: we collection in phylogenetic in each genetic according Table 2 PA 04016 PA 04021 04116 PA SMNH Tussøyna, For each revealed specimens of * Identification the amplification reactions was checked on a 1 % agarose gel and then, amplification products were treated enzymatically with Exonuclease-I and Fast-SAP (shrimp alkaline phosphatase; both from Fermentas, Thermo Scientific). These products were then sequenced along both strands in 10 µL reactions with 0.165 µM primer, 1 × Sequencing Buffer and 0.5 µL Big Dye v. 3.1 PreMix (Applied Biosystems, Thermo Scientific) following a standard thermocycling protocol. Sequencing products were separated on an ABI 3130 xl sequencer and bases were called using Sequence Analysis 6 software (both from Applied Biosystems, Thermo Scientific). PHYLOGENETIC ANALYSES Data from the different genes were aligned and analyzed as separate data sets. DNA sequence quality was checked from the sequence electropherograms and the data were input to MEGA 7 software (Kumar, Stecher & Tamura, 2016). Both sequenced strands were aligned using the clustal W algorithm in MEGA 7 and any ambiguous bases were checked with sequence from the complementary strand when creating a consensus sequence for each individual. Best models for use in phylogenetic analyses were chosen according to the Bayesian Information Criterion (BIC) using jModelTest 2 (Darriba et al., 2012; http: // jmodeltest. org /). Phylogenetic analyses were performed using the maximum likelihood algorithms implemented in MEGA 7 and PhyML 3.0 (Guindon et al., 2010; http: // www. atgc-montpellier. fr / phyml /) using the HKY + G model, for both data sets. Five hundred replicates were performed to estimate bootstrap values. In addition, we also analyzed the data sets using the algorithm implemented in RAxML 8.2.7 (Stamatakis, 2014) with the plugin available in Geneious v. 10.2 (Biomatters Limited; www. geneious. com). In this case, the GTR + G model was used and 1000 bootstraps were performed. For outgroups, we chose sequences from distantly related asteroids in the genus Echinaster available in GenBank [16 S: Echinaster sentus (Say, 1825) DQ 297088.1; COI: Echinaster spinulosus Verrill, 1829 GAVE 01107285.1]. MEGA 7 was also used to calculate genetic distances between the clades identified in the phylogenetic analyses. In these calculations, the Kimura 2 parameter evolutionary model was used. The sequences obtained in this study are available in GenBank with accession numbers KY 853246 – KY 853395 (Table 2).	en	Knott, K Emily, Ringvold, Halldis, Blicher, Martin E (2018): Morphological and molecular analysis of Henricia Gray, 1840 (Asteroidea: Echinodermata) from the Northern Atlantic Ocean. Zoological Journal of the Linnean Society 182 (4): 791-807, DOI: 10.1093/zoolinnean/zlx066, URL: https://academic.oup.com/zoolinnean/article/182/4/791/4558517
