Elysia papillosa Verrill, 1901

Krug, Patrick J., Vendetti, Jann E. & Valdés, Ángel, 2016, Molecular and morphological systematics of Elysia Risso, 1818 (Heterobranchia: Sacoglossa) from the Caribbean region, Zootaxa 4148 (1), pp. 1-137 : 39-46

Elysia papillosa Verrill, 1901

( Figs. 6 O, 18–20)

Elysia papillosa Verrill 1901: 31 , pl. 4, fig. 3 (Type locality: Hungry Bay, Bermuda) — Pruvot-Fol 1946: 36; Er. Marcus 1957: 410, 415; Ev. Marcus & Er. Marcus 1967: 27 –28, figs. 22–25;? Er. Marcus & Ev. Marcus 1970: 45; Ev. Marcus 1980: figs. 9, 48; Jensen 1980: fig. 1C, tables 2,5; Clark 1984 figs. 15, 17;? Hess et al. 1994: 161 –163; Redfern 2013: 284, figs. 787A–B.

Elysia patina [non Elysia patina Ev. Marcus, 1980 ] — Ortea, Caballer, Moro & Espinosa 2005:

497–498, 505–512, fig. 5, pl. 1C; Händeler et al. 2009: figs. 6, 7; Curtis, Schwartz & Pierce 2010: 299–302, figs. 1C, 5; Christa et al. 2014: figs. 1C, 3; Curtis et al. 2015: 27, fig. 2

Checholysia patina [non Elysia patina Ev. Marcus, 1980 ] — Espinosa, Ortea, Caballer & Moro 2005: 56; Ortea et al. 2005: 512.

Elysia annedupontae Ortea, Espinosa & Caballer in Ortea, Caballer, Moro & Espinosa 2005: 502 –505, fig. 3, pl. 1B (Type locality: Ensenada de Bolondrón, Guanahacabibes, Cuba) n. syn.

Checholysia annedupontae (Ortea, Espinosa & Caballer in Ortea, Caballer, Moro & Espinosa 2005) — Espinosa et al. 2005: 56; Ortea et al. 2005: 512 n. syn.

Elysia sp. 1 — Valdé s et al. 2006: 72–73.

Type material. Elysia papillosa— untraceable, not found at the (YPMNH); Elysia annedupontae— holotype at IESH (no registration number given); paratype at MCNT (no registration number given).

Material examined. Bocas del Toro, Panama, 19 February 2004, 4 specimens ( LACM 178616–19 ) ; Discovery Bay, Jamaica, 7 March 2006, 1 specimen ( LACM 178620 ) ; Bermuda, 2006, 4 specimens ( LACM 178600–02 , LACM 178621 ) ; Florida, USA: 2009, 1 specimen ( LACM 178624 ), Geiger Beach , 6 October 2006, 4 specimens ( LACM 178612–13 , LACM 178622–23 ) ; Great Stirrup Cay , Bahamas, July 2007, 2 specimens ( LACM 178608 , LACM 178611 ) ; Sweetings Cay , Bahamas, July 2010, 2 specimens ( LACM 178614–15 ) ; Stocking Island , Exumas, Bahamas, 29 January 2009, 2 specimens ( CPIC 0 0 0 70, CPIC 00073 ) .

Additional material examined. Bocas del Toro , Panama, 19 February 2004, 4 specimens (isolate Epap_04Pan01, isolate Epap_ 04Pan04, isolate Epap_04Pan07, isolate Epap_04Pan08); Discovery Bay , Jamaica, 7 March 2006, 1 specimen (isolate Epap_06Jam09) ; Bermuda, 2006, 3 specimens, (isolate Epap_06Ber05-07); Geiger Beach, Florida , USA, 6 October 2006, 1 specimen (isolate Epap_06Gei24) ; Bahamas, July 2010, 1 specimen (isolate Epap_10LSS01).

Live animal. Specimens swim readily when disturbed by undulating their parapodial margin.

External anatomy. External coloration highly variable. Overall body color generally light green, but ranging from white or tan to olive green. Parapodial margin tan to dark brown ( Fig. 18 C–D). Medial band running longitudinally along head between eyes, from front of face to pericardium, color ranging from cream to tan to brown; some specimens with darker brown to black streaks along sides of this band, running through eyes ( Fig. 18 E). Sides of head lighter green to white ( Fig. 18 D, F). Most specimens with one or two large white papillae between eyes, with scattered, smaller white papillae across head on some. Rhinophores elongated, rolled, with white to tan ground color; rounded white papillae of varying sizes dot surface. Rhinophores blunt-ended, sometimes with gently curving edge. Distinctive wide, dark band appearing about ⅓ of the way up rhinophore, not perpendicular to rhinophoral axis but rather at an angle such that band forms a rhomboid shape when viewed from above ( Fig. 18 A–E). Second, fainter band or streak present ⅔ of way up rhinophore on some specimens.

Parapodia relatively low, not covering pericardial complex. Outer parapodial surface bearing rows of white papillae, varying in size ( Fig. 18 C–F). Parapodia dotted with black spots same size as eyes, and with scatted brown specks. Lower portion of parapodia green to tan, grading to tan-brown along distal portion. Margin thickened into row of white-tipped, rounded protrusions ( Fig. 18 F–H). Color of margin either not distinct from upper parapodial surface, or more pronouncedly brown. Parapodial margin with scalloped edge, laterally undulating with threepronged, pointed side flaps regularly spaced along entire length, forming series of siphonal openings ( Fig. 18 C–E).

Inner surface of parapodia and dorsum varying greatly in color among specimens. Ground color generally mottled light to dark green, usually pigmented by patches of white and minute flecks of iridescent blue, green or orange. Inner parapodial surface and dorsum lightly to heavily speckled with brown or black spots, and with scattered white, rounded papillae ( Fig. 18 G–H). Posterior end of body narrowing to form short, triangular, pointed tail, much elongated on some specimens ( Fig. 18 C–F).

Pericardium round, with brown streaks and spots and low white papillae ( Fig. 18 G–H). Renopericardial extension emerges from pericardium as narrower tube, clear with orange-brown spots and faint speckling of white to light blue. Renopericardial extension running less than halfway down body, relatively straight, terminating in pair of wide posterior vessels. Dorsal vessels sometimes clear, otherwise lined with patches of white or light blue speckles. Two to three vessels typically emerging on either side of the pericardium, and a further four to five vessels on each side of the renopericardial extension ( Fig. 19 ). Vessels relatively symmetric in placement, but usually one extra vessel emerging on right side. Vessels wider than in most elysiids, initially straight and unforked, angling back towards posterior end of body. Vessels with extending thin side branches appearing at irregular intervals near parapodial margin, with terminal branches anastomosing on some specimens. Elongated pair of posterior vessels emerging from terminus of renopericardium, and running almost to tail with numerous lateral branches, some anastomosing with side branches of more anterior vessels.

One pair of large sperm-storage vesicles, typically visible on large adult specimens, and not juveniles, as irregularly shaped, greyish protrusions ( Figs. 18 G–H, 19 lower panel). Sperm-storage vesicles shall herein be referred to as “gametic” vesicles, sensu Clark (1984) [not “gametolytic” vesicles sensu Marcus (1983)] to denote that vesicles are filled with allosperm after mating without implying that sperm are lysed or digested there. Gametic vesicles usually between 5th and 6th, or 6th and 7th dorsal vessels, just anterior to end of renopericardial extension.

Internal anatomy. Radula with 10–13 teeth (CPIC 0 0 0 70, CPIC 0 0 0 73, LACM 178608, LACM 178613–18, LACM 178620, LACM 178622–23, isolate Epap_06Ber07, isolate Epap_10LSS01), 4–6 in ascending limb and 6– 9 in descending limb ( Fig. 20 A). Leading tooth elongate and slender, with cusp bearing 18–30 denticles ( Fig. 20 B). Tooth length, width in lateral profile, and degree of mid-tooth angle variable among specimens. Variability in length of leading tooth from 60–200 µm, tooth width in lateral profile from 5–28 µm, with width to length ratios from 8.3–15.5, and mid-tooth angle from 11° to 33° (n=18). Housing depression for interlocking teeth “V”-shaped and extending ¾ total tooth length ( Fig. 20 C). Base of tooth ½ to ⅓ total tooth length. Ascus containing jumbled heap of discarded teeth (not figured).

Penis thin and elongate, with rigid musculature that did not deform after drying ( Fig. 6 O, Fig. 20 D), bearing a spoon-shaped stylet (LACM 178600, LACM 178616, LACM 178618–21) opposite a medial flange ( Fig. 20 D–E). Curved hook on stylet tip considered anomalous as observed in only one specimen ( Fig. 20 E). Deferent duct long, thin, and convoluted ( Fig. 6 O).

Reproduction and development. Development was planktotrophic for specimens from Panama, Curaçao, and the Bahamas. Newly hatched larvae lacked eyespots, fed readily on cultured phytoplankton, and did not metamorphose. Egg masses contained an irregular ribbon of white ECY which contacted most, but not all, egg capsules ( Fig. 18 I). Mean egg diameter was 67.2 µm ± 2.4 SD (n = 17 ova) for one clutch from Curaçao. Mean shell length at hatching was not determined for specimens that could be confirmed as E. papillosa by DNA sequencing or internal morphology.

Development was described as lecithotrophic for “ E. papillosa ” by Clark & Goetzfried (1978), with eggs measuring 91.9 µm in diameter and having a flat watch-spring spiral of ECY. Unfortunately, ECY color was not described. Either E. papillosa is poecilogonous and some specimens from Florida produce lecithotrophic larvae, or Clark & Goetzfried (1978) and Clark & Jensen (1981) misidentified E. patina (which is lecithotrophic) as E. papillosa . Krug (2009) described development for E. patina under the name “ E. papillosa ”, following Ortea et al. (2005).

Host ecology. Field surveys and laboratory observations confirm that E. papillosa specializes on the algal genus Penicillus , as described by Jensen (1980). In field surveys, over 300 specimens were collected from Penicillus spp. at 12 sites over a decade of sampling. In the laboratory, some starved specimens fed on Rhipocephalus brevicaulus , but only a single large specimen was found on R. brevicaulus in the field. In an unreplicated choice experiment, 15 specimens of E. papillosa from Panama were placed in a dish with 200 mL seawater and stipes of two different algae, and left for 24 hr. Choices were: (1) P. capitatus vs. Udotea flabellum ; (2) P. capitatus vs. Halimeda incrassata ; and (3) U. flabellum vs. H. incrassata . After a day, 10 of 15 specimens were physically associated with P. capitatus in dish 1, and 12 of 15 with P. capitatus in dish 2; none were associated with the non-host alga, and the remainder were crawling on the glass or undersurface of the water. In dish 3, two slugs chose U. flabellum , one chose H. incrassata , and 12 of 15 chose neither alga. All available data thus indicate that the only ecologically relevant host for E. papillosa is the algal genus Penicillus .

Numerically, E. papillosa is one of the most abundant species of Elysia in the Caribbean; a collection of P. capitatus in Panama yielded 181 specimens from 364 individual algal stipes, or about one slug per two stipes. A discussion of maintenance of host chloroplasts in E. papillosa is provided by Curtis et al. (2010) under the name E. patina (due to misidentifications based on Ortea et al. (2005); see below). Like most Elysia spp., E. papillosa has short-term retention of functional chloroplasts from Penicillus , with degradation delayed for up to a week after phagocytosis by cells lining the digestive tubules ( Curtis et al. 2010).

Based on the COI barcode used to identify slug species, Christa et al. (2014) also report the diet-derived plastids of E. papillosa under the name “ E. patina ”, and vice-versa report the diet of E. patina as “ E. papillosa ”, reflecting the prior reliance of most authors on the misidentification in Ortea et al. (2005) which swapped the names of these two species. Christa et al. (2014) reported plastids from Udotea and an unidentified alga in Udoteacea as the diet of E. papillosa (their “ E. patina ”) based on barcoding of plastid DNA. However, inspection of the tree used to match plastid sequences to algal reference sequences (Figure S 1 in Christa et al. 2014) reveals that plastid sequences from the true E. papillosa , and also Cyerce antillensis , match perfectly or closely the reference sequence for Penicillus capitatus , which is the primary food of both slug species. Udotea spp. are not monophyletic on the reference tree used to identify food sources by Christa et al. (2014), and plastid sequences could equally well correspond to Penicillus , Rhipocephalus , Udotea , or Chlorodesmis on this tree; why Udotea was reported as the diet source of “ E. patina ” is thus unclear. This inability to match unambiguously plastid barcodes to reference sequences highlights a serious concern with inferring diet via tree-based methods when algal relationships are not resolved by the available sequence data. The species reported as “ E. papillosa ” by Christa et al. (2014) is actually E patina , for which the ecologically relevant host genus Halimeda was indeed recovered.

Phylogenetic relationships. Elysia papillosa belongs to subclade 1, a putatively basal lineage of Caribbean elysiids, and was recovered as sister to E. taino n. sp. (described subsequently) ( Fig. 4 ).

Range. Range data for E. papillosa are compromised by frequent misidentifications in the literature. We confirm the species is present in Bermuda (the type locality), Panama, Jamaica, Florida, USA, Bahamas, U.S. Virgin Islands, Antigua, and Curaçao. Records from Mexico and Cuba ( Ortea et al. 2005) and Florida ( Curtis et al. 2010) as E. patina and E. annedupontae also refer to E. papillosa .

Remarks. Perhaps no Caribbean elysiid has had as complex a taxonomic history as E. papillosa . Confusion stemmed in part from the absence of type material and lack of relevant anatomical data in the initial description, and was compounded by the presence of a cryptic sister species in the central Caribbean. We distinguished E. papillosa from its cryptic sister species based on molecular phylogenetic analyses combined with subtle differences in radular morphology. In a population genetic survey, the barcoding COI gene and the nuclear H3 gene were sequenced from 174 specimens that superficially resembled E. papillosa ( Trathen 2010 as “ E. patina ” sensu Ortea et al. 2005 ; authors’ unpublished data). In phylogenetic analyses, COI haplotypes formed two clades between which TrN distance ranged from 8.8 to 12.8% (mean distance = 10.9%), above the 8% threshold for species-level divergence in Elysia (Krug et al. 2013) . In contrast, maximum pairwise COI divergence was 5.6% within E. papillosa , consistent with moderate phylogeographic structure in each species (as in other Elysia spp.; Krug 2011; Krug et al. 2013). Different alleles at the H3 locus were fixed in the two divergent COI clades, even where both clades were sympatric, indicating the two distinctive COI lineages do not interbreed. When these data were re-analyzed in the present study by ABGD, two distinct E. papillosa -like species were recovered across a wide range of priors on allowable intraspecific divergence ( Fig. 3 A). Branch lengths on the ML tree based on a concatenated four-gene alignment also show a comparable level of genetic divergence between these two taxa as exists for many other pairs of well-recognized sister species of Elysia ( Fig. 4 ). Thus, all available molecular data support the distinction of E. papillosa from E. taino n. sp.

All E. papillosa -like specimens from Bermuda, the type locality of E. papillosa , grouped genetically with all specimens from Florida, Curaçao, and Panama, and with most specimens from three Bahamas islands (Sweetings Cay, Bimini, and Little San Salvador); these specimens formed ‘clade 1’ in Trathen (2010). Conversely, all specimens from Dominica grouped with most specimens from Jamaica and the remaining Bahamas islands (Stirrup Cay, San Salvador, Plana Cays, Compass Cay, Northern Exumas), forming a divergent lineage (‘clade 2’ of Trathen 2010). This lineage is subsequently described as E. taino n. sp. based on genetic divergence and differences in radular morphology from E. papillosa . The two species also had minimally overlapping geographical distributions, with only 1-3 specimens of E. papillosa sampled from each site where E. taino n. sp. was the predominant Elysia feeding on Penicillus .

Despite the existence of the morphologically similar sister species E. taino n. sp., much of the confusion in the literature over the identify of E. papillosa involved even more distantly related species, including E. patina , E. zuleicae , and E. pawliki n. sp. The original description of E. papillosa by Verrill (1901) was brief, but included important details:

“A small, grayish, distinctly papillose species. Body rather elongated in extension; head large; neck long; rhinophores large; strongly folded and wide at the tips. Side-flaps large, thin, usually with the edges deeply undulated. Whole surface of body, head, and outside of flaps thickly covered with small conical papillae. Color of head, neck, and outside of flaps grayish blue, paler anteriorly, and spotted with darker gray on the outside of the flaps, and speckled with flake-white over the whole surface. Inside of flaps darker ash-gray; the edges bordered with white. Rhinophores are like the head, but with two indistinct transverse bands of orange-brown on the posterior side. Length, about 12 mm in extension. Hungry Bay, under stones, at a very low-tide, April 5, 1901. (A.H.V.) Rare. This species can swim freely by means of its ample lateral flaps.”

The transverse brownish bands on the rhinophores and papillose body surface match the first authoritative redescription of E. papillosa by Ev. Marcus & Er. Marcus (1967), who noted that the specimen they examined conformed to the limited details in the original description. The teeth were described as ~200 µm long and serrulated with coarse denticles; the penis as muscular, 700 × 300 µm, with a triangular stylet 60 µm long ( Ev. Marcus & Er. Marcus 1967). The species we recognize as E. papillosa is the only Caribbean species that matches both Verrill (1901) and Ev. Marcus & Er. Marcus (1967) in having the following distinguishing characteristics: (i) common in Bermuda; (ii) external morphology of some specimens fits the original description; (iii) swims readily when disturbed; (iv) has a penial stylet; (v) has a coarsely serrated, straight-edged radula. The morphologically similar species Elysia taino n. sp. was not sampled in Bermuda, and has shorter, wider radular teeth (see remarks of E. taino for more details). Radular characters readily distinguish E. papillosa from all other related species that swim, allowing us to match our material to E. papillosa by Ev. Marcus & Er. Marcus (1967), and the description of radular anatomy and feeding ecology by Jensen (1980). Related swimming species ( E. zuleicae , E. buonoi n. sp., E. patina ) have curved, narrow, pointed radular teeth, which was termed the “ Halimeda spur” ( Clark & DeFreese, 1987). The pointed tip of such curved teeth is used to pierce the narrow utricles accessible on the surface of the heavily calcified, inter-utricular matrix of udotacean algae such as Halimeda and Udotea . The serrated, straight, blade-shaped tooth of E. papillosa and E. taino n. sp. is used to feed on the long, wide filaments of Penicillus , and has diverged rapidly from the tooth shape of all closely related species. Host ecology similarly helps distinguish E. papillosa from most related species, but not from its sister taxon.

Ev. Marcus (1980) described the dorsal vessels of E. papillosa as having only one pair of vessels with a posterior orientation and many lateral side branches (pg. 57: fig. 9); she may have illustrated a specimen with multiple side branches emerging from the elongated posterior vessels, and failed to note smaller, transparent vessels anterior to the last pair ( Fig. 19 ). Otherwise, the dorsal vessels figured by Ev. Marcus (1980) for E. papillosa do not match the pattern on any specimens we have seen, nor do they match any other Caribbean elysiid. Vessels on E. papillosa appear to be slightly wider on average than the vessels of E. taino n. sp., and sperm-storage vesicles form closer to the posterior end of the renopericardium on E. papillosa .

Some similarities exist between E. papillosa and E. zuleicae ; E. zuleicae can swim, occurs in Bermuda, and the planktotrophic egg masses of E. zuleicae are similar to those of E. papillosa . However, the coloration of typical specimens, rhinophores, and extended tail of E. zuleicae are not consistent with the description of E. papillosa . Clark (1984) lumped specimens of E. zuleicae (pg. 89: figs. 16, 18–20) in with E. papillosa (pg. 89: figs. 15, 17); his posthumously obtained notes and drawings indicate he recognized that his 1984 material included species other than E. papillosa , and Clark (1994) acknowledged that his E. papillosa was a species complex.

Another species often confused with E. papillosa is E. patina . The coloration described by Verrill also describes well typical specimens of E. patina , which also swims, and the dorsal vessels of E. papillosa and E. patina are very similar. Indeed, the specimen of E. papillosa illustrated by Ev. Marcus & Er. Marcus (1967) (reproduced here as Fig. 18 B) resembles specimens of both E. papillosa (e.g., Fig. 18 F) and E. patina (e.g., Fig. 42 A–B). Small specimens of E. patina and E. papillosa can be very difficult to distinguish externally, but the species can clearly be distinguished by a number of criteria. Their host algae and radulae are entirely different: E. patina feeds on H. opuntia and its radular teeth have a curved “ Halimeda spur”. Egg mass characters clearly discriminate between the two: E. patina has lecithotrophic development and orange ECY in a flat ribbon ( Fig. 42 G), whereas E. papillosa is planktotrophic in all surveyed populations, with white ECY ( Fig. 18 I). The spermstorage vesicles form posterior to the renopericardium in E. patina , but are more anterior in E. papillosa . Molecular data also clearly differentiate the taxa ( Figs. 3–4 ).

Despite these differences, Ortea et al. (2005) identified specimens of E. papillosa as E. patina based on their dorsal vessel pattern. Ortea et al. (2005) confusingly assert that Ev. Marcus mixed two species in her original description of E. patina , presumably including one with a blade-shaped, serrated tooth, despite the absence of any supporting evidence. While Ev. Marcus (1980) suggested her paratype specimen from the Bahamas was likely a different species from E. patina , inspection of the holotype material of E. patina confirms that the type specimen had the curved, pointed radula illustrated in the description, and not the blade-shaped, heavily serrated tooth of E. papillosa . The species Ortea et al. (2005) called E. patina thus cannot be E. patina Ev. Marcus 1980 . Based on the radular teeth drawn in Ortea et al. (2005: fig. 4D), their E. patina is most likely E. papillosa , but could potentially be E. taino n. sp. Also, Ortea et al. (1998) reported E. papillosa from the Canary Islands, but this is another misidentification.

Verrill (1901) drew the parapodial margin of E. papillosa undulating in a series of scalloped segments (reproduced here as Fig. 18 A); the margin on some specimens bears unbranched conical papillae that create the appearance of points on a crown (e.g., Fig. 18 C–D), whereas on other specimens, the margin has only low papillae and appears relatively smooth ( Fig. 18 E–F). The lateral undulations allow the parapodia of E. papillosa to interlock and cover the dorsum, as in the drawing by Ev. Marcus & Er. Marcus (1967) ( Fig. 18 B). Verrill (1901) drew E. papillosa resting on a stipe of Halimeda incrassata ( Fig. 18 A); the alga was misinterpreted by Ortea et al. (2005) as a row of digitiform, branching papillae along the parapodial margin of the animal. The species called E. papillosa by Thompson (1977) and Ortea et al. (2005) does not swim, is unknown from Bermuda, and has other features that are incompatible with the details provided by Verrill (1901) for E. papillosa , and must therefore be a different species (which we describe subsequently as either E. pawliki n. sp. or E. zemi n. sp.)

Based on external, radular and penial morphology, E. annedupontae (Ortea, Espinosa & Caballer in Ortea, Caballer, Moro & Espinosa, 2005) is a junior synonym of E. papillosa . Ortea et al. (2005) noted that key anatomical features of E. papillosa given by Ev. Marcus & Er. Marcus (1967) were all present in E. annedupontae , including radular characters, penial morphology, and shape and banding pattern of the rhinophores. Ortea et al. (2005) claimed E. annedupontae was a distinct species because they erroneously interpreted Verrill’s drawing as indicating long, branching papillae along the parapodial rim of E. papillosa ; as noted above, no such papillae are indicated on the drawing, nor are the papillae of E. papillosa described as branching by Verrill (1901). By all criteria, E. annedupontae is therefore synonymous with E. papillosa .

Figures

FIGURE 3. Mismatch distributions based on TrN-corrected pairwise distances at the COI locus for three subclades of Elysia. Based on ABGD analysis, genetic distances were classified as intra-specific (white bars) or inter-specific (black bars). Dashed horizontal line indicates the a priori threshold distance used provisionally to delimit candidate species.

FIGURE 4. Phylogenetic hypothesis for family Plakobranchidae rooted on genus Bosellia (not shown) to illustrate relationships within the genera Elysia, Plakobranchus and Thuridilla. Topology and branch lengths are from Maximum Likelihood analysis of concatenated DNA sequences (2,807 bp total) representing portions of two mitochondrial (COI, 16 S) and two nuclear (H 3, 28 S) genes. Significant support values are given as ML bootstrap percentages (below branch), or BI posterior probabilities (above branch); asterisk = 1.0 or 100 % support. Species discussed in this study are bolded. Triangle denotes presence of a penial stylet.

FIGURE 6. Penial morphology of some species examined. A, Elysia ornata (LACM 178583). B – E, Elysia crispata (LACM 178640) (B); (LACM 178641) (C); (LACM 2004.5.1) (D); (isolate Ecri _ 10 LKS 03) (E). F, Elysia chlorotica (LACM 178597). G – H, Elysia patina (LACM 178650) (G); (LACM 178651) (H). I, Elysia flava (LACM 178626). J, Elysia subornata (LACM 178629). K, Elysia canguzua (LACM 178644). L, Elysia serca (CPIC 00027). M, Elysia evelinae (MZUCR INB 0003312779). N, Elysia velutinus (LACM 178642), O, Elysia papillosa (LACM 178607). P, Elysia cornigera (LACM 173227). Q, Elysia marcusi (LACM 178647), R, Elysia pratensis (CPIC 00068). S, Elysia zuleicae (LACM 178656). Abbreviations: bw, body wall; dd, deferent duct; pe, penis; s, stylet.

FIGURE 18. Elysia papillosa, external morphology and egg mass. Live specimens were photographed upon field collection from Bocas del Toro, Panama (C – E, G – H) or Discovery Bay, Jamaica (F). A, Illustration accompanying original description (Verrill 1901: fig. 2), showing lateral undulations of the parapodial margin but no branching papillae; specimen is shown resting on stipe of H. incrassata. B, Illustration accompanying re-description of E. papillosa (Ev. Marcus & Er. Marcus 1967: figs. 22 and 23). C, Side view of isolate Epap _ 04 Pan 01 (8 mm length) showing brown parapodial margin with crown-like lateral undulations. D, Side view of isolate Epap _ 04 Pan 04 (5 mm). E, Top view of isolate Epap _ 04 Pan 08 (6 mm), showing elongated tail. F, Side view of juvenile LACM 178620 (3 mm). G – H, Dorsal surface of relaxed isolates Epap _ 04 Pan 07 (G) and Epap _ 04 Pan 01 (H), showing renopericardial complex, vessel network and greyish-white gametic vesicles. I (bottom right panel), Egg mass from Curaçao specimen showing irregular ribbon of white ECY twisting around uncleaved ova; field of view = 1.75 mm.

FIGURE 19. Elysia papillosa, renopericardial complex and dorsal vessel network traced from digital photograph of live specimens from Panama. Top, isolate Epap _ 04 Pan 01, length = 12 mm. Bottom, Epap _ 04 Pan 07, length = 10 mm. Grey areas represent sperm-storage vesicles.

FIGURE 20. Elysia papillosa, SEM of the radula and penis. A, Radula minus a jumbled ascus (LACM 178613). B, Leading tooth (LACM 178613). C, Leading tooth, (LACM 178600). D, Penis, with stylet (LACM 178600). E, Penial stylet, dorsal and ventral views (LACM 178618).

FIGURE 42. Elysia patina, external morphology and egg mass; all specimens 6 – 8 mm in length from Sweetings Cay, Grand Bahamas Island, Bahamas. A – D, Views of four different live specimens showing brown transverse bands on rhinophores, and color variation from white to tan to brown-green, with white papillae. E, Relaxed specimen (8 mm) showing elongated renopericardial complex, dorsal vessel network, and opaque white gametic vesicles. F, Live specimen (6 mm) with parapodia held open showing dorsal vessels lined in iridescent blue, and gametic vesicles in a semi-transparent state. G, Egg mass spiral showing flat orange ECY ribbon above uncleaved ova. Actual diameter = 2 mm.