Description of a common stauromedusa on the Pacific Coast of the United States and Canada, Haliclystus sanjuanensis new species (Cnidaria: Staurozoa)

Introduction

Staurozoa is a cnidarian class with about 50 valid species (Miranda et al., 2018). They are a group of benthic marine animals generally attached to algae or rocks in intertidal pools or in the shallow subtidal (Miranda et al., 2018), generally rare or overlooked but occasionally found in great numbers (Miranda, Morandini & Marques, 2012). Haliclystus James-Clark, 1863 is currently the most diverse genus of the class, with 12 valid species and one additional species assigned as a nomen nudum (i.e., a Latin term referring to a species name that was not described in conformance with the International Commission on Zoological Nomenclature, 1999), Haliclystus “sanjuanensis” (see Miranda et al., 2018).

This little animal has been known by naturalists and scientists since at least the 1920s and has appeared in most guidebooks to the NE Pacific coastal fauna for the last nearly one hundred years. The species synonymy that we present here is both impressively long and indicative of the persistent confusion over what to call this stalked jellyfish.

Haliclystus “sanjuanensis” was first described in the Master’s Thesis by Gellermann (1926), a student at the University of Washington Friday Harbor Laboratories in Friday Harbor, Washington, USA. That species name was subsequently used by other visitors to the Friday Harbor Laboratories including Guberlet (1936, 1949) and Hyman (1940a, 1940b) as if it had been formally described, insinuating itself into the literature. In fact, Hyman (1940b) has sometimes been indicated as the author/descriptor of H. “sanjuanensis” (e.g., Kramp, 1961; Hirano, 1997; Collins & Daly, 2005), although Hyman was apparently unaware of the necessity to formally describe H. “sanjuanensis” and her work did not address the requirements for species descriptions (e.g., a complete description and the designation of a type, see also International Commission on Zoological Nomenclature, 1999, Article 13).

Based on morphology, Gwilliam (1956) synonymized H. “sanjuanensis” (which he had studied in False Bay on San Juan Island, Washington, USA) with Haliclystus auricula James-Clark, 1863 in his Ph.D. dissertation. Gwilliam never published his taxonomic conclusions, but he used the same identification in an article derived from that dissertation on the neuromuscular physiology (Gwilliam, 1960) of H. “sanjuanensis” (as H. auricula) collected on San Juan Island, Washington. Kramp (1961) followed Gwilliam’s conclusions and included H. “sanjuanensis”, mentioned by Hyman (1940b), in the synonymy of H. auricula, and that species name was also selected for the much-used keys for identification of Pacific Northwest invertebrates by Kozloff (1974).

Meanwhile, many authors misidentified specimens of H. “sanjuanensis” collected between British Columbia (Canada) and California (USA) as a similar stauromedusa, Haliclystus stejnegeri Kishinouye, 1899 (e.g., Ricketts, Calvin & Hedgpeth, 1968; Otto, 1976, 1978; and a number of guidebooks written for the Pacific coast of mainland USA from the 1920s into the 1960s, see synonymies below), perhaps based on the record of this latter species by Bigelow (1920) at Port Clarence, Alaska (USA, near the Bering Strait). Contributing to this chaotic scenario, H. “sanjuanensis” has also been called Haliclystus octoradiatus James-Clark, 1863 (e.g., Eckelbarger & Larson, 1993). In addition, there is at least one other not-formally-described species of Haliclystus in the same region (partially sympatric), also proposed by Gellermann (1926) in her Master’s Thesis (this species later referred to as Haliclystus salpinx James-Clark, 1863 in many publications including Otto, 1976, 1978; Mills, 1987; Strathmann, 1987; Wrobel & Mills, 2003; Mills & Larson, 2007; but this synonymy is yet to be tested and is to be dealt with in a future publication).

On the other hand, there are studies supporting the validity of H. “sanjuanensis”. Hirano (1997) morphologically revalidated H. “sanjuanensis” as different from H. auricula based on the presence of prominent white spots (sacs) composed of masses of nematocysts in H. “sanjuanensis”; these nematocyst-filled structures are absent in H. auricula. Hirano (1997) also distinguished H. “sanjuanensis” from H. stejnegeri and H. octoradiatus based on the distribution of the white nematocyst spots in the subumbrella. Phylogenetic analyses based on ribosomal and mitochondrial genes corroborate that H. “sanjuanensis” is clearly distinct from H. stejnegeri and H. octoradiatus and closely related to the clade composed of Haliclystus antarcticus Pfeffer, 1889 and H. auricula (Miranda, Collins & Marques, 2010; Miranda et al., 2016a; Holst & Laakmann, 2019; Holst, Heins & Laakmann, 2019). The genetic difference (in % based on p-distances) of H. “sanjuanensis” to H. auricula and H. antarcticus is about 3.49–4.31 for 16S and 10.08–11.27 to COI molecular markers (Holst, Heins & Laakmann, 2019).

Based on this evidence, and since “nomina nuda” and other unavailable names can be made available if they are published again in a way that meets the criteria of availability (International Commission on Zoological Nomenclature, 1999; see https://www.iczn.org/outreach/faqs/ “What is a nomen nudum”; and Article 13), our aim in this study is to formally describe Gellerman’s Haliclystus “sanjuanensis” using the species name that she originally applied. We also provide a table comparing Haliclystus sanjuanensis n. sp. with the other twelve described species in the genus Haliclystus.

Material and Methods

Results and Discussion

SYSTEMATIC ACCOUNT

Phylum Cnidaria Verrill, 1865

Subphylum Medusozoa Petersen, 1979

Class Staurozoa Marques & Collins, 2004

Order Stauromedusae Haeckel, 1879

Suborder Myostaurida Miranda et al. (2016a)

Family Haliclystidae Haeckel, 1879

Genus Haliclystus James-Clark, 1863

Haliclystus sanjuanensis new species

Synonymy.

“Haliclystus sanjuanensis n. sp.” in Gellermann 1926 (pp. 80–84, Plate XX figures 20, 20a, Plate XXXVI figures 49, 50, 51) San Juan Island, Washington, USA.

“Haliclystus stejnegeri” in Ricketts 1929 (p. 20).

“Stalked Jelly Fish Haliclystus sanjuanensis” in Guberlet 1936, 1949 (pp. 46, 47, 409).

“Haliclystus” in Ricketts & Calvin 1939 (p. 87).

“Haliclystus (probably stejnegeri)” in Ricketts & Calvin 1939 (p. 195) Port Townsend and Friday Harbor, Washington, USA and from Point Joe, Monterey, and Carmel Bay, California, USA.

“Haliclystus stejnegeri” in Ricketts & Calvin 1939 (Figure 86 and p. 261).

“Haliclystus” in Hyman 1940a (Figure 165E and caption on p. 510, pp. 534, 535) Puget Sound (San Juan Island), Washington, USA.

“Haliclystus” in Hyman 1940b (Figure 8 caption and text on p. 292, pp. 282, 293, 295) San Juan Island, Washington, USA.

“Haliclystus sanjuanensis” in Hyman 1940b (p. 292) San Juan Island, Washington, USA.

“Haliclystus” in Ricketts & Calvin 1948 (p. 87).

“Haliclystus (probably stejnegeri)” in Ricketts & Calvin 1948 (p. 195) Port Townsend and Friday Harbor, Washington, USA and from Point Joe, Monterey, and Carmel Bay, California, USA.

“Haliclystus stejnegeri” in Ricketts & Calvin 1948 (Figure 86 and p. 296).

“Haliclystus stejnegeri” in MacGinitie & MacGinitie 1949 (Figure 19 on p. 120, p. 121) found in great numbers in Puget Sound and occasionally in Monterey Bay, USA.

“Haliclystus sanjuanensis” in Knapp 1950, Neptune State Park, Oregon, USA.

“Haliclystus” in Ricketts, Calvin & Hedgpeth 1952 (p. 85).

“Haliclystus (probably stejnegeri)” in Ricketts, Calvin & Hedgpeth 1952 (p. 249) Port Townsend and Friday Harbor, Washington, USA and from Point Joe, Monterey, and Carmel Bay, California, USA.

“Haliclystus stejnegeri” in Ricketts, Calvin & Hedgpeth 1952 (Figure 86 on p. 284, p. 416).

“Two species of Haliclystus are found in the Friday Harbor region” in Smith et al. 1954 (p. 41) San Juan Islands, Washington, USA.

“Haliclystus auricula” in Smith 1955, Field Notebook #3, June 24, 1955, Davis Bay, Lopez Island, Washington, USA.

“Haliclystus auricula (Rathke)” in Gwilliam 1956 (pp. 2, 8, 17, 20, 21, 43–55, 56, 57, 61–66, 79, 80, 94–97, 110–180) San Juan Island, Washington, USA.

“Haliclystus sanjuanensis” in Gwilliam 1956 (pp. 13, 14, 20) San Juan Island, Washington, USA.

“Haliclystus auricula” in Anonymous 1958 (p. 2) San Juan County, Washington, USA.

“Haliclystus auricula (Rathke)” in Gwilliam 1960 (pp. 454–473) San Juan Island, Washington, USA.

“Haliclystus auricula (Rathke, 1806)” (in part) in Kramp 1961 (pp. 292–293).

“Haliclystus sp.” in McLean 1962 (p. 106) scarce on rock walls along the open coast of the Monterey Peninsula, nine miles south of Carmel near Granite Creek, California, USA.

“Haliclystus” in Ricketts, Calvin & Hedgpeth 1962 (p. 85).

“Haliclystus (probably stejnegeri)” in Ricketts, Calvin & Hedgpeth 1962 (p. 249) Port Townsend and Friday Harbor, Washington, USA and from Point Joe, Monterey, and Carmel Bay, California, USA.

“Haliclystus stejnegeri” in Ricketts, Calvin & Hedgpeth 1962 (Figure 86 on p. 284, p. 416).

“Haliclystus sp.” in MacGinitie & MacGinitie 1968 (Figure 19 on p. 120, p. 123) abundant in Puget Sound waters, USA.

“Haliclystus” in Ricketts, Calvin & Hedgpeth 1968 (p. 130).

“Haliclystus (probably H. stejnegeri)” in Ricketts, Calvin & Hedgpeth 1968 (pp. 299–300) Port Townsend and Friday Harbor, Washington, USA and from Point Joe, Monterey, and Carmel Bay, California, USA.

“Haliclystus stejnegeri” in Ricketts, Calvin & Hedgpeth 1968 (Figure 221 on p. 299, p. 462).

“Haliclystus” in Westfall 1968 (p. 803), San Juan Island, Washington, USA.

“Haliclystus auricula” in Westfall 1968 (p. 804), San Juan Island, Washington, USA.

“Two species of Haliclystus are found in the Friday Harbor region” in Smith 1970 (p. 41) San Juan Islands, Washington, USA.

“Haliclystus auricula” in Westfall, Yamataka & Enos 1970 (p. 545), San Juan Island, Washington, USA.

“Haliclystus auricula” in Westfall 1973 (pp. 239–243), San Juan Island, Washington, USA.

“Haliclystus auricula” in Kozloff 1973 (Figure 206 and text, p. 247).

“Haliclystus auricula” in the Bodega Marine Lab Synoptic Collection, collected in 1973 at Shell Beach, Sonoma County, California, USA.

“Haliclystus auricula (Rathke, 1806)” in Kozloff 1974 (pp. 22, 219).

“Haliclystus auricula” in Smith 1975 (Figure 21 on p. 88, p. 96).

“21, Haliclystus, from False Bay near Friday Harbor, Wash.” in Smith 1975 (caption on p. 89).

“Haliclystus auricula (Rathke, 1806) (= H. sanjuanensis of several authors)” in Smith 1975 (p. 96) Friday Harbor, Washington, USA.

“Haliclystus auricula” in Smith 1976 (Figure 121 on p. 115) Pacific Northwest.

“Haliclystus stejnegeri” in Otto 1976 (pp. 319–322, 328) San Juan Island, Washington, USA.

“Haliclystus stejnegeri” in Singla 1976 (pp. 533–539) on eelgrass, Victoria, British Columbia, Canada.

“Haliclystus auricula” in Flora & Fairbanks 1977 (Figure 48 on p. 56) on kelp and eelgrass in protected bays in Puget Sound, USA.

“Haliclystus stejnegeri” in Otto 1978 (p. 22) San Juan Island, Washington, USA.

“Haliclystus auricula” in Snively 1978 (p. 162).

Not “Haliclystus auricula” = Manania handi in Snively 1978 (Figure 79 on p. 43).

“Haliclystus” in Brusca & Brusca 1978 (pp. 52–54, Figure 24 on p. 53) in Northern California, USA attached to seaweeds in quiet tidepools, or to eelgrass in Humboldt Bay, USA.

“Haliclystus auricula” in Caine 1980 (p. 162) San Juan Island, Washington, USA.

“Haliclystus sp.” in Barr & Barr 1983 (Figure 32 on p. 9, pp. 81–82) Yakobi Island, Southeast Alaska, USA.

“Haliclystus stejnegeri and H. auricula” in Phillips 1984 (p. 40) Pacific Northwest.

“Haliclystus auricula (Rathke, 1806)”, in part, in Austin 1985 (p. 70).

“Haliclystus steinegeri Kishinouye, 1899”, in part, in Austin 1985 (p. 70).

“Haliclystus” in Ricketts et al. 1985 (pp. 154–155) common “in the north” and rare to occasional at Monterey, USA.

“Haliclystus auricula” in Ricketts et al. 1985 (pp. 343–344, Figure 267 on p. 344, p. 513) on eelgrass in Puget Sound and northern California, USA.

“H. stejnegeri” in Ricketts et al. 1985 (p. 513).

“H. sanjuanensis” = Haliclystus auricula (Rathke, 1806) in Hirano 1986 (p. 183).

“Haliclystus” in Westfall 1987 (pp. 9, 11, 12), San Juan Island, Washington, USA.

“Haliclystus stejnegeri Kishinouye, 1899” in Mills 1987 (p. 67).

“Haliclystus stejnegeri Kishinouye (= H. sanjuanensis and H. auricula, E. Kozloff personal communication to C.E.M. c. 1984)” in Strathmann 1987 (p. 79) San Juan Island, Washington, USA.

“Manania (?)” DP Abbott in Hilgard 1987 (p. 22) July 2, 1974, 15–20 feet deep on Plocamium, off Malpaso Creek, Monterey County, California, USA.

“Haliclystus octoradiatus (Lamarck, 1816)” in Larson 1990 (pp. 546–550, Figure 1A on p. 551) Alaska to central California, USA.

“Haliclystus stejnegeri Kishinouye, 1899” in Cairns et al. 1991 (p. 11 (in part), Figure 1 on p. 77).

“Haliclystus octoradiatus” in Eckelbarger & Larson 1993 (pp. 225–236) Victoria, British Columbia, Canada.

“Haliclystus salpinx” in Eckelbarger 1994 (p. 16, Figure 2.3.d on p. 19, pp. 20, 26) Victoria, British Columbia, Canada.

“Haliclystus” in Kozloff 1996 (pp. 322, 323).

“Haliclystus stejnegeri” in Kozloff 1996 (p. 323, Figure 345 on p. 324).

“Haliclystus stejnegeri Kishinouye, 1899” in Mills 1996 (pp. 67, 489).

“H. sanjuanensis Hyman, 1940” in Hirano 1997 (pp. 247, 249–251).

“Haliclystus stejnegeri” in Heeger 1998 (p. 74).

“Haliclystus stejnegeri” in O’Clair & O’Clair 1998 (Figure 13 on p. 15, pp. 15–16) Bertha Bay, west side of Chichagof Island, Southeast Alaska, USA.

“Haliclystus sanjuanensis” in Wrobel & Mills 2003 (p. 49).

“Haliclystus “sanjuanensis” Gellermann, 1926 or Hyman, 1940 or Hirano, 1997 or undescribed?” in Mills 1999–present accompanied by C.E. Mills photograph of two stauromedusae on one piece of eelgrass (accessed March 2023).

“Haliclystus sanjuanensis” in Collins 2000 (pp. 10, 11, 14–16).

“Haliclystus sp. undescribed “sanjuanensis”” in Cairns et al. 2002 (pp. 9, 47).

“Haliclystus stejnegeri” in Cairns et al. 2002 (Figure 3 on p. 109).

“Haliclystus sanjuanensis” in Collins 2002 (p. 420).

“Haliclystus sanjuanensis Hyman, 1940” in Zagal 2004 (table 2 on p. 339).

“Haliclystus sanjuanensis Hyman, 1940” in Collins & Daly 2005 (pp. 221, 222, 225, 226, 229).

“Haliclystus sanjuanensis” in Collins et al. 2006a (p. 3).

“Haliclystus sanjuanensis” in Collins et al. 2006b (Figure 2 on p. 101, Figure 3 on p. 102, Figure 4 on p. 103, Figure 6 on p. 106, p. 115).

“Oval-anchored stalked jelly Haliclystus stejnegeri” in Lamb & Hanby 2005 & 2015 (p. 111 photo and text) intertidal to 10 m, Sekiu, Olympic Peninsula, Washington, USA.

“Haliclystus sp.” in Mills & Hirano 2007 (Figure 1 on p. 541).

“Haliclystus sp. (undescribed species)” in Mills & Larson 2007 (p. 170).

“Haliclystus sp. (“sanjuanensis”)” in Mills & Larson 2007 (Plate 63A on p. 171).

“Haliclystus sp. (=H. “sanjuanensis”, nomen nudum)” in Mills & Larson 2007 (p. 172).

“Haliclystus “sanjuanensis” (nomen nudum)” in Miranda, Morandini & Marques 2009 (pp. 1507, 1508, 1513–1515).

“Haliclystus “sanjuanensis” (nomen nudum)” in Miranda, Collins & Marques 2010 (pp. 3–5, 7).

“H. sanjuanensis” in Miranda, Collins & Marques 2010 (p. 4).

“Haliclystus “sanjuanensis” nomen nudum” in Kahn et al. 2010 (pp. 49, 52, 54–57).

“Oval-anchored stalked jelly (Haliclystus stejnegeri)” in Brewer et al. 2011 (p. 40 photo and text) Cape Prominence, Unalaska Island in the Aleutians, USA.

“Haliclystus “sanjuanensis”” in Kayal et al. 2012 (pp. 2–4, 11).

“Haliclystus sanjuanensis” in Kayal et al. 2013 (Figure 2 on p. 5, table 3 on p. 13).

“H. “sanjuanensis”” in Miranda, Collins & Marques 2013 (p. 1381).

“Haliclystus sanjuanensis” in Zapata et al. 2015 (pp. 5, 9, Figure 4 on p. 7) San Juan Island, Washington, USA.

“Haliclystus ‘sanjuanensis’ nomen nudum” in Westlake 2015 (185 pp.) Victoria, British Columbia, Canada.

“H. “sanjuanensis” nomen nudum” in Miranda et al. 2016a (pp. 6, 13–15, 17, 19, 37).

“H. “sanjuanensis” (nomen nudum)” in Miranda et al. 2016b (pp. 3, 72, 73).

“Haliclystus sanjuanensis” in Li, Sung & Ho 2016 (pp. 113–114).

“Haliclystus ‘sanjuanensis’ nomen nudum” in Westlake & Page 2017 (pp. 29–49) Victoria, British Columbia, Canada.

“Haliclystus “sanjuanensis”” in Kayal et al. 2018 (pp. 4, 13, 15, Figure 1b on p. 2, Figure 5 on p. 7, Figure 6 on p. 9) from Cattle Point, San Juan Island, Washington, USA.

“H. “sanjuanensis” nomen nudum” in Miranda et al. 2018 (pp. 1697, 1699, 1703, 1706, 1707, 1709).

“Haliclystus sp.” in Jensen, Gotshall & Flores Miller 2018 (p. 28) British Columbia, Canada to central California, USA.

“Haliclystus sanjuanensis” in Nicosia et al. 2018 (Table 2 on p. 5, Figure 6 on p. 13).

“Haliclystus “sanjuanensis”” and “Haliclystus sanjuanensis” in Holst & Laakmann 2019 (Table 1 on p. 1062, Figure 2 on p. 1064).

“Haliclystus “sanjuanensis”” in Holst, Heins & Laakmann 2019 (Figure 6 on p. 1787, Table 4 on p. 1788, pp. 1789, 1793, 1794).

“Haliclystus “sanjuanensis” (nomen nudum)” in Miranda & Collins 2019 (p. 8).

“Haliclystus sanjuanensis” in Koch & Grimmelikhuijzen 2019 (Table 1 on p. 3, Table 3 on p. 5, Figure 3 on p. 9, Figure 4 on p. 10).

“Haliclystus sanjuanensis” in Lewandowska, Hazan & Moran 2020 (p. 11).

“Haliclystus sanjuanensis” in Gornik et al. 2021 (p. 1755).

“Haliclystus sanjuanensis” in Hartigan et al. 2021 (Figure 4 on p. 11).

“Haliclystus “sanjuanensis” (nomen nudum)” in Holst et al. 2021 (p. 249).

“Haliclystus sanjuanensis” in Adriansyah et al. 2022 (Figure 8A on p. 265, Table 2 on p. 269).

“Haliclystus sanjuanensis” in Santander et al. 2022 (Figure 1C on p. 2).

“Haliclystus sanjuanensis” in Vöcking et al. 2022 (p. 11).

“Haliclystus sanjuanensis” in Zhang & Jacobs 2022 (Table 1 on p. 5, Figure 4 on p. 8).

“Haliclystus sanjuanensis” in Steinworth, Martindale & Ryan 2023 (Table 2 and text on p. 3, p. 9).

We remain unsure whether the following records refer to H. sanjuanensis:

“Haliclystus stejnegeri Kishinouye” in Bigelow (1920) (p. 12H and Plate 2 Figure 4 on p. 20H) Port Clarence, Alaska, USA, 2–3 fathoms. There is insufficient information to distinguish if this was H. stejnegeri or H. sanjuanensis.

“Haliclystus” in 1946 field notes by Ed Ricketts (Transcript of the 1946 trip to West Vancouver Island and to the Queen Charlotte Islands), Massett, Queen Charlotte Islands, Canada, June 27, 1946, and July 2, 1946, “lots of Haliclystus” on eelgrass in two locations near Massett in the very low intertidal, in Steinbeck, Ricketts & Hedgpeth (1978) (pp. 115, 118), and Rodger (2006) (pp. 306, 310).

“Haliclystus auricula (Rathke, 1806)” in Nuclear Regulatory Commission (2001), Pacific Gas & Electric Environmental Report, ~2001, table 2.3–5, Sheet 7 of 23, p. 173 of 202, from coastline near Diablo Canyon, California, USA, (confirmed by personal communication from David Behrens, who did much of the collecting and identifying for that report, but since he could not locate a photograph of the specimen for us in 2019, we cannot be positive of this material, which would otherwise represent one of the southernmost collection locations of H. sanjuanensis n. sp.).

“Haliclystus stejnegeri” Royal British Columbia Museum (RBCM) #015-00395-001. Single animal collected from Egeria Bay, Langara Island, Haida Gwaai, British Columbia, Canada, May 22, 2015 by Heidi Gartner, Melissa Frey, Gavin Hanke, Tommy Norgard, Leslie Barton and Vanessa Hodes is too young to determine whether it is H. sanjuanensis n. sp. or H. stejnegeri. Claudia Mills notes upon inspection of the specimen, April 26, 2019.

Holotype. Smithsonian Institution, National Museum of Natural History, USNM 1522349 (Table 1). Northeast Pacific, collected July 15, 2019 at Cattle Point, San Juan Island, San Juan County, Washington, USA (48.452131° N, 122.961750° W) on a blade of algae at approximately 0.5 m deep. Preserved in 10% formalin in seawater. Collector Claudia E. Mills.

Other material examined. Paratypes and additional specimens (Table 1) were collected July 3, July 15 and July 29, 2019, on low tides at Cattle Point, San Juan Island, San Juan County, Washington, USA (48.452131° N, 122.961750° W) on red or green algae or surfgrass Phyllospadix at approximately 0.5 m deep, collector Claudia E. Mills. The animals were preserved in 10% formalin in seawater or preserved directly into 95% ethanol. Both formalin and ethanol preserved paratypes and additional specimens were deposited at the Smithsonian Institution, National Museum of Natural History in Washington, DC; the Florida Museum at the University of Florida in Gainesville; the Royal British Columbia Museum in Victoria, British Columbia; the California Academy of Sciences in San Francisco, California; and the Synoptic Collection of the University of Washington Friday Harbor Laboratories in Friday Harbor, Washington (Table 1).

Further evidence. Photographs (mainly from https://www.inaturalist.org/), specimens in public and particular collections, and personal communications, not otherwise published, South to North, are available in Table S1.

Diagnosis. Haliclystus with broadly conical calyx, wider than high (Figs. 1 and 2). Calyx well demarcated from the stalk, where the profile goes from conical to cylindrical (Figs. 1 and 2). Stalk length variable, from as long as the calyx is tall to about 1/2 length of calyx. Gonads extending from base of the calyx to tips of each of eight arms (Figs. 1 and 2); 200 to 300 tightly-packed gonadal vesicles comprising each gonad, with 10 to 22 abreast at the widest (Figs. 1 and 2). The gonads are broadly leaf-shaped, filling much of the available subumbrellar surface, and slightly folded over upon themselves in larger animals; the proximal half of each gonad may be turned over upon itself in the perradii and the distal half turned over in a similar manner in the interradii (Figs. 1 and 2). Each arm tipped with a cluster of usually up to 130 (but may be more than 200 in larger specimens) capitate secondary tentacles (Figs. 1–3). Eight anchors along the calyx margin on interradial and perradial axes, alternating with the eight adradial arms—the anchors are roughly oval in outline and vary from fairly spherical to elongate, sometimes with a remnant of the primary tentacle in the center, usually disappearing in older animals (Figs. 1, 2, 3A–3E). The eight marginal notches between arms are all approximately equal in size (Figs. 1 and 2). Bright-white nematocyst spots about the same size as the gonadal vesicles (Figs. 4E–4G), closely grouped in a single row along the distal perradial edges of the gonads from the tips of the arms to the widest point of the gonads or a little further down, but also rarely seen along the perradial calyx margin (Figs. 1–4). Often, but not always, with four dark pigment stripes running up the perradii of the stalk and bifurcating (into eight stripes) at the stalk-calyx junction, extending halfway up the calyx on the exumbrella along the interradial edge of each gonad, and sometimes also with lighter stripes running along both upper edges of the gonads near the calyx margin (Fig. 2).

Description. (General description using Gellermann, 1926; Gwilliam, 1956; Westlake, 2015; photographs by the authors and available online, and observation of living and preserved individuals).

Calyx broadly conical, with eight arms; arms sometimes recurved toward the stalk in life; calyx usually up to about 30 mm wide (rarely to 40 mm wide) and clearly demarcated from the stalk, wider than high (Figs. 1 and 2). Stalk narrow and varies from about 1/2 the length of calyx to as long as the calyx is tall, rarely a little longer than the calyx is tall (Figs. 1, 2 and 5). Well-developed basal pedal disc at base of stalk (Figs. 5B and 5C). Stalk quadratic in cross section, with four usually darkly-pigmented perradial gastric chambers (delimited by gastrodermis) and four interradial longitudinal muscle bands, embedded in mesoglea (Figs. 5 and 6A). Perradial chambers fusing into one single chamber at stalk-calyx connection, with four interradial septa associated with the four interradial longitudinal muscle bands (Figs. 6B, 6C and 7). Four infundibula funnel-shaped with blind end, delimited by epidermis, deeply developed down to base of calyx, widening apically, with broad apertures on subumbrella (Fig. 7). Gastrovascular cavity without claustrum. Numerous gastric filaments in gastrovascular cavity (Figs. 3F and 3G). At base of infundibula, the four interradial longitudinal muscle bands of the stalk divide into eight bands (Figs. 6C–6G, 7D and 7E), continuing up through the calyx along interradii, reaching out into the eight adradial arms. Gastrovascular cavity divided into four perradial gastric pockets at the manubrium region (Figs. 6F and 6G), where gonadal vesicles are located in mature individuals. Four perradial gastric pockets bifurcating distally to form eight adradial branches, each of which is running into each arm. The manubrium is short and quadrangular, with dark-pigmented perradial corners, and with the four oral lobes thin and folded (Figs. 4A–4D, 6F and 6G).

Eight adradial arms, all alike, and notches between arms are all approximately equal in size (Figs. 1 and 2). However, arms occasionally exhibit some perradial pairing, with the perradial notches between arms sometimes a little narrower than the interradial notches. Tip of each arm with a cluster of up to 130 (or more than 200 in larger specimens) knobbed (capitate) secondary tentacles; all tentacles are short, from 1 to 3 mm long (Figs. 1–3 and 8A–8H).

Intertentacular lobules at the internal tip of the arms (at the base of tentacular clusters) (Figs. 8A–8C). Each secondary tentacle in the arms is ensheathed with longitudinal muscle fibers (Figs. 8D–8F). Four groups of perradial muscular fibers extend from the manubrium lips (Figs. 4C and 4D) down to the four perradial corners of the manubrium. The coronal muscle lies beneath the subumbrellar surface at the periphery of the calyx and is comprised of four interradial and four perradial segments.

Eight anchors on the exumbrella, near the margin, one between each pair of arms in the interradii and perradii outside of the low-point of each notch (Figs. 1, 2 and 3A–3E). The anchors are irregularly oval in outline and about half as long as the diameter of the stalk, but quite variable in shape and size between animals. The primary tentacle (the developmental origin of the anchor) capitulum usually disappears in older specimens, but is present in developing animals to a greater or lesser extent depending on animal size (Figs. 3A–3E and 6H). The anchors may be fairly flat in profile or rather elongate, but they are never flaring or trumpet-shaped with thickened or rolled edges (Figs. 1, 2 and 3A–3E); the cavity of each anchor communicates with the general body cavity of the animal by means of a very short canal through the anchor peduncle (Fig. 3E).

The eight adradial gonads extend the full length of the eight arms, are very broad in the middle, filling much of the available subumbrellar surface, and in large specimens slightly folding over on themselves, and narrower at both ends (Figs. 1, 2 and 4E–4G). The gonads in the largest animals are composed of 200 to 300 tightly-packed gonadal vesicles that are roughly spherical, but may be so tightly packed together that their shape may be modified (Figs. 1, 2 and 4E–4G). The gonadal vesicles are not arranged in rows; there are 10 to 22 vesicles abreast at the broadest part of each gonad (usually 12–16) (Figs. 1 and 2). In the largest animals, the proximal half of each gonad is turned over upon itself in the perradii and the distal half is turned over in a similar manner in the interradii (Figs. 1A and 1E). The gonads are separated in the interradii, but are closer at their bases in the perradii, making four pairs of gonadal bands (Figs. 1 and 2).

The exumbrellar surface is finely granulated with numerous, scattered nematocyst clusters (Fig. 9L). The subumbrella is marked with distinctive bright-white nematocyst spots that are about the same size as the gonadal vesicles (Figs. 4E–4G)—these white spots are arranged in the perradii, mostly closely grouped in a single row along the distal perradial edges of the gonads from the tips of the arms to the widest point of the gonads or a little further down, but also occasionally located along the calyx margin (perradial center) in large specimens (Figs. 1, 4E–4H and 8I). The subumbrellar epidermis is raised in slight bulges over each white nematocyst spot (Figs. 4H and 8I).

Overall color in life is highly variable, but each individual is of a single color, from brown or beige to orangey red to peach, rusty or wine red, or greenish, usually cryptically matching the color of its eelgrass or algal substrate (Figs. 1 and 2); young (small) animals are less pigmented. Gonads and anchors golden brownish to green to orange to wine red, bright-white nematocyst spots along perradial edges of gonads and (rarely) subumbrellar margin; tentacles greenish to golden yellow to reddish or pink (Figs. 1 and 2). Often, but not always, with four dark brown to orange longitudinal stripes of varying intensity on the perradial corners of the manubrium (Figs. 4A–4D) and on the perradii of the stalk, bifurcating at the stalk-calyx junction (Figs. 1, 2 and 5F), extending halfway up the calyx on the exumbrella along the interradial edge of each gonad, and sometimes also with lighter stripes running along both upper edges of the gonads near the calyx margin (Figs. 2B and 2C). Muscles whitish to colorless in the stalk and calyx (Figs. 5D and 5E).

Cnidome. Secondary tentacles with two types of nematocysts: isorhiza (abundant), length 9.54–13.50 μm (mean 12.11 μm, number of capsules measured n = 62), diameter 2.26–4.18 μm (mean 3.51 μm, n = 62); and microbasic eurytele (scarce), length 15.69–19.85 μm (mean 17.73 μm, n = 24), diameter 7.39–9.36 μm (mean 8.52 μm, n = 24) (Figs. 9A–9F). White nematocyst spots with one type of nematocysts: birhopaloid type II (two dilatations close together) (abundant), length 11.63–17.13 μm (mean 14.60 μm, n = 50), diameter 7.95–11.39 μm (mean 9.96 μm, n = 50) (Figs. 9G–9I). Gastric filaments with one type of nematocysts: rhopaloid (not enough information to identify them as eurytele, stenotele, or birhopaloid) (abundant), length 7.10–10.70 μm (mean 8.90 μm, n = 11), diameter 4.13–6.80 μm (mean 5.47 μm, n = 11) (Figs. 9J and 9K). Exumbrella with one type of nematocysts: rhopaloid (not enough information to identify them as eurytele, stenotele, or birhopaloid), mean length 11.04 μm (n = 13), mean diameter 4.46 μm (n = 13) (Figs. 9L and 9M).

Behavior. This animal is almost always attached to the substrate by its pedal disc; some use of the anchors and secondary tentacles for anchorage has also been observed (Figs. 1 and 2). Only once or twice in many visits to the seashore have we seen a specimen free in the water. This stauromedusa produces a periderm in the form of a thin transparent cup that encases the pedal disc and is left behind when the staurozoan moves (Fig. 10); a new periderm is formed when the staurozoan reattaches. Small (usually arthropod) prey was captured by flexing (not contracting) one or more secondary tentacles on one or more adjacent arms, which immediately then bent in to the manubrium to transfer the captured prey to the mouth. Bending of an arm or arms was often accompanied by bending or twisting of the stalk. Spontaneous bending of single and multiple arms was typically observed. The anchors did not participate in feeding behaviors; their sole purpose appears to be as adhesive organs. One animal spawned spontaneously in the laboratory in May 2015, releasing many 35 µm diameter eggs in sticky mucus strands through its mouth opening of the manubrium (Fig. 11).

Etymology. The specific name, originally assigned by Gellermann (1926) in her M.S. Thesis, refers to the San Juan Archipelago, San Juan County, Washington State, USA, where this animal is abundant in some more-exposed areas, and where she studied it in detail. This location in NW Washington is approximately the epicenter of its presently-known distribution along NE Pacific shorelines and the location of some of the most-dense populations.

Habitat. On living eelgrass and surfgrass, Zostera marina and Phyllospadix scouleri, or on various species of red, green, and brown algae in the low intertidal to shallow subtidal, to at least 10 m deep in areas of moderate wave activity.

Distribution. Along the Pacific coast of North America at least from the eastern Aleutian Island to just north of Point Conception in central California, USA (Fig. 12; Table S1). It does not seem to occur south of Point Conception, as this species is not listed in the Southern California Association of Invertebrate Taxonomists species list for the Southern California Bight (Cadien et al., 2018), nor has it been seen by naturalist diver-photographers Kevin Lee, Merry Passage and Phil Garner in about 4,500 dives near Los Angeles (K. Lee & M. Passage, 2019, personal communications to C.E.M.).

On San Juan Island, Washington, USA, the type locality (Fig. 12), very small H. sanjuanensis n. sp. stauromedusae can be found beginning in approximately mid-April, and the last specimens persist into late November. No recognizable life history stages have been found from late fall until mid-spring.

Genetic sequence data. GenBank accession numbers: LSU (28S): KU308593, AY920782; SSU (18S): AF358102, KU308562, MN329739; SSU-ITS1-5.8S-ITS2-LSU: FJ874776, HM022143, HM022144, HM022145, MN329741; 16S: AY845339, HM022149, HM022150, HM022151, MN329738; COI: KU257477, MH242772, MN329740 (Collins, 2002; Collins & Daly, 2005; Collins et al., 2006b; Miranda, Collins & Marques, 2010; Miranda et al., 2016a; Westlake & Page, 2017); Transcribed RNA sequences: HAHB01000001–HAHB01134305 (Kayal et al., 2018); Partial genome: JN700944 (Kayal et al., 2012).

Remarks. The genus Haliclystus is comprised of species with four interradial longitudinal muscles in the stalk (=peduncle) (Figs. 5, 6 and 7), a shared character of the suborder Myostaurida (see Miranda et al., 2016a). Within this suborder, Haliclystus is placed in the family Haliclystidae, characterized by the presence of perradial and interradial anchors (=rhopalioids, adhesive and sensorial structures derived from primary tentacles, located between adjacent pair of arms) (Figs. 1–3) (Miranda et al., 2016a; Miranda & Collins, 2019). The family Haliclystidae is also represented by species of the genera Depastromorpha Carlgren, 1935, Depastrum Gosse, 1858, Halimocyathus James-Clark, 1863, and Manania James-Clark, 1863. However, Haliclystus differs from these genera in lacking a claustrum, a tissue composed of a thin layer of mesoglea surrounded by a double layer of gastrodermis (Miranda et al., 2016a, 2016b, 2017). In addition, species of Haliclystus generally have four perradial chambers in the stalk (although this character is variable during development) (Figs. 5 and 6) (Miranda et al., 2016a, 2016b) and the coronal muscle at the subumbrellar margin can be divided or entire (Miranda et al., 2016a). A monophyletic Haliclystus has been recovered in all phylogenetic analyses for the class (Collins & Daly, 2005; Miranda, Collins & Marques, 2010; Miranda et al., 2016a; Holst & Laakmann, 2019; Holst, Heins & Laakmann, 2019). However, in spite of recent efforts (e.g., Hirano, 1997; Miranda, Morandini & Marques, 2009; Miranda et al., 2016a; Kahn et al., 2010), the taxonomy of the genus is still a challenge, as characters generally mentioned in the description of the species, such as number of tentacles and number of chambers in the stalk are known to show intraspecific variation (Uchida, 1929; Hirano, 1986; Miranda, Morandini & Marques, 2009; Kahn et al., 2010).

Haliclystus sanjuanensis n. sp., discussed as a nomen nudum in different articles (see above), is herein properly described. Haliclystus sanjuanensis n. sp. is unique among other species of Haliclystus because of the arrangement of the bright-white nematocyst spots in perradii only, forming a conspicuous line along the perradial edges of gonads and rarely also a few near the perradial calyx margin (Figs. 1 and 4; see also Table S2; Hirano, 1997). Haliclystus antarcticus and H. auricula, the species phylogenetically most closely related to H. sanjuanensis n. sp. (Miranda, Collins & Marques, 2010; Miranda et al., 2016a; Holst, Heins & Laakmann, 2019), do not have these white nematocyst spots (see Hirano, 1997; Miranda, Morandini & Marques, 2009; Miranda, Collins & Marques, 2013). Haliclystus borealis Uchida, 1933, Haliclystus californiensis Kahn et al., 2010, Haliclystus inabai (Kishinouye, 1893), Haliclystus monstrosus (Naumov, 1961), Haliclystus sinensis Ling, 1937, Haliclystus tenuis Kishinouye, 1910, H. salpinx, and H. octoradiatus have white nematocyst spots in both perradii and interradii (see also additional differences in Table S2). There is no information concerning white spots for Haliclystus kerguelensis Vanhöffen, 1908. However, H. sanjuanensis n. sp. can be distinguished from H. kerguelensis based on the proportion between calyx and stalk and on the dark stripes (Table S2). Nevertheless, it is important to highlight that H. kerguelensis was described in 1908, based on a drawing and not specimens; Kramp (1957) reports three additional specimens collected in 1929–1930, but he does not include figures, so the validity of this species has never been tested.

The only Haliclystus species left for comparison is H. stejnegeri, which is morphologically very similar to H. sanjuanensis n. sp. (Table S2). However, although both species have white nematocyst spots found in perradii only, in H. sanjuanensis n. sp. the white spots are relatively abundant along the perradial edges of the gonads, running from near the margin to considerably deep into the subumbrella (Figs. 1 and 4), whereas in H. stejnegeri the white spots are relatively sparse and mainly found along the edge of gonads near the calyx margin (Fig. 13; Table S2). One additional difference is the pattern of pigmentation in the calyx (exumbrella). In H. sanjuanensis n. sp., eight dark stripes may be present, running halfway up the calyx along the interradial edges of gonads, which join as four stripes in the perradii running the length of the stalk; dark stripes may also be present near the calyx margin running along both edges of gonads (Fig. 2; Table S2). On the other hand, H. stejnegeri is characterized by its exclusive pattern with eight adradial pairs of dark stripes running the length of the calyx, and each two pairs united at the perradii of calyx base continuing as four dark stripes running down the stalk; the calyx margin is also tinted in the same color (Kishinouye, 1899; Fig. 13; Table S2). Unfortunately, preserved stauromedusae frequently lose their live coloration, so this character can be difficult to assess in preserved specimens.

Although some morphological characters used to differentiate species of Haliclystus vary with development (Uchida, 1929; Hirano, 1986; Miranda et al., 2016b), the current valid species have good support based on molecular evidence. Nine out of 13 described species of Haliclystus have been phylogenetically placed (Miranda et al., 2016a; Holst & Laakmann, 2019; Holst, Heins & Laakmann, 2019), including H. sanjuanensis n. sp.. Interestingly, although morphologically similar, H. sanjuanensis n. sp. and H. stejnegeri did not come out as closely related species (Miranda et al., 2016a; Holst & Laakmann, 2019; Holst, Heins & Laakmann, 2019). Haliclystus sanjuanensis n. sp. proved to be more closely related to H. antarcticus and H. auricula (Miranda, Collins & Marques, 2010; Kayal et al., 2018; Holst, Heins & Laakmann, 2019), despite the huge geographical gap between the known distributions of these species: Western United States and Canada for H. sanjuanensis n. sp. (Fig. 12); Southern Chile and Argentina and the Antarctic Peninsula for H. antarcticus (Miranda et al., 2018); and Eastern United States and Europe for H. auricula (Miranda et al., 2018).

The geographic distribution of H. sanjuanensis n. sp. in the North Pacific Ocean begins at about 34.5° N in central California and extends up at least to Southeast Alaska (about 58.2° N) and also west into the Aleutian Islands (to at least 166.8° W) (Fig. 12). It may or may not overlap H. stejnegeri at the upper end of its range in the NE Pacific, but then H. stejnegeri appears to swing across and down the NW Pacific to the Kurile Islands and to northern Japan (Miranda et al., 2018). Haliclystus californiensis is a very distinctive orangey-red species that is sympatric with H. sanjuanensis n. sp., occurring from southern California to at least central British Columbia (Kahn et al., 2010; our studies, see also Table S2). Four more described species of Haliclystus, easily separated from H. sanjuanensis n. sp. because of the pattern of white nematocyst spots and the number of secondary tentacles on each arm in adult stauromedusae (see Table S2), are found in NW Pacific waters of China, Far East Russia, Japan, and Korea (H. sinensis, H. borealis, H. tenuis, and H. inabai) (see Miranda et al., 2018).

At least one more possibly undescribed species of Haliclystus (Gellermann, 1926) occurs at about the midpoint of the south to north distribution of H. sanjuanensis n. sp., from the San Juan Islands and southern Vancouver Island to at least SE Alaska and possibly all the way to the Kuriles, northern Japan, and mainland far-east Russia, where a look-alike has been described as H. monstrosus (Naumov, 1961; Sheiko & Stepanyants, 1990). This possibly undescribed species of Haliclystus is also similar to H. salpinx, a species found in the North Atlantic (James-Clark, 1863; Berrill, 1962; but see also Sheiko & Stepanyants, 1990), raising the hypothesis that all records of H. salpinx in Washington State (Otto, 1976, 1978; Mills, 1987; Strathmann, 1987; Wrobel & Mills, 2003; Mills & Larson, 2007) are misidentifications, but we need to complete further work to be sure. We are in the process of collecting more material for DNA and morphological comparisons to determine how many other Haliclystus species are present in the boreal North Pacific.

Supplemental Information