Numerous new records of tropical non-indigenous species in the Eastern Mediterranean highlight the challenges of their recognition and identification

Abstract New data on 52 non-indigenous mollusks in the Eastern Mediterranean Sea is reported. Fossarus sp. (aff. aptus sensu Blatterer 2019), Coriophora lessepsiana Albano, Bakker & Sabelli, sp. nov., Cerithiopsis sp. aff. pulvis, Joculator problematicus Albano & Steger, sp. nov., Cerithiopsis sp., Elachisina sp., Iravadia aff. elongata, Vitrinella aff. Vitrinella sp. 1 (sensu Blatterer 2019), Melanella orientalis, Parvioris aff. dilecta, Odostomia cf. dalli, Oscilla virginiae, Parthenina cossmanni, Parthenina typica, Pyrgulina craticulata, Turbonilla funiculata, Cylichna collyra, Musculus coenobitus, Musculus aff. viridulus, Chavania erythraea, Scintilla cf. violescens, Iacra seychellarum and Corbula erythraeensis are new records for the Mediterranean. An unidentified gastropod, Skeneidae indet., Triphora sp., Hypermastus sp., Sticteulima sp., Vitreolina cf. philippi, Odostomia (s.l.) sp. 1, Henrya (?) sp., and Semelidae sp. are further potential new non-indigenous species although their status should be confirmed upon final taxonomic assessment. Additionally, the status of Dikoleps micalii, Hemiliostraca clandestinacomb. nov. and H. athenamariaecomb. nov. is changed to non-indigenous, range extensions for nine species and the occurrence of living individuals for species previously recorded from empty shells only are reported. Opimaphora blattereri Albano, Bakker & Sabelli, sp. nov. is described from the Red Sea for comparison with the morphologically similar C. lessepsiana Albano, Bakker & Sabelli, sp. nov. The taxonomic part is followed by a discussion on how intensive fieldwork and cooperation among institutions and individuals enabled such a massive report, and how the poor taxonomic knowledge of the Indo-Pacific fauna hampers non-indigenous species detection and identification. Finally, the hypothesis that the simultaneous analysis of quantitative benthic death assemblages can support the assignment of non-indigenous status to taxonomically undetermined species is discussed.


Introduction
The Eastern Mediterranean Sea is a hotspot of non-indigenous species introductions. The opening of the Suez Canal in 1869 broke a long-standing biogeographic barrier and enabled hundreds of Red Sea species to enter the basin and establish populations (Por 1978;Galil 2009;Zenetos et al. 2010Zenetos et al. , 2017Zenetos and Galanidi 2020). These so-called Lessepsian species are now recorded from all countries bordering this basin west to Greece (Katsanevakis et al. 2009;Çinar et al. 2011;Ammar 2018;Zenetos et al. 2018; Bariche and Fricke 2020;Crocetta et al. 2020) and some have already reached the central Mediterranean, e.g., Tunisia (Ounifi-Ben Amor et al. 2015), Italy (Occhipinti-Ambrogi et al. 2011), and even France (Daniel et al. 2009;Bodilis et al. 2011).
The introduction rate is an important metric to describe the invasion process. Genuine variation in this rate can result from changes in vector efficacy, connectivity between the native and introduced range, and environmental conditions in the recipient ecosystem. The introduction rate is often estimated from the discovery record (Solow and Costello 2004). However, even in well sampled and taxonomically well-known groups like mollusks, multi-decadal time lags between introduction and first detection have been quantified (Oliver 2015;Guy-Haim et al. 2017;, suggesting that the detection rate is a poor proxy of the introduction rate. Indeed, although the discovery rate is increasing (Galil 2009;Raitsos et al. 2010), estimates of the introduction rate corrected for temporal variation in sampling effort for Lessepsian fishes showed that it was constant over ~ 1930~ -2010~ (Belmaker et al. 2009). Still, the most recent enlargement of the Suez Canal has raised concerns that the improved connectivity could increase the introduction rate of Lessepsian species . Additionally, rapid climate warming is particularly affecting the Eastern Mediterranean (Ozer et al. 2017), causing, on the one hand, the decline of native species and, on the other hand, more favourable conditions for the establishment of tropical species (Rilov 2016;Albano et al. 2021).
To monitor a dynamic process such as the Lessepsian invasion, intensive fieldwork is mandatory. We indeed show that an intensive sampling effort coupled with identification at high taxonomic resolution and collaborative research among individuals and institutions enabled the detection of 23 new Lessepsian mollusks, another nine species which, upon further inspection, may prove to be new Lessepsian species, nine new records for Eastern Mediterranean countries, and new data for eleven already recognized non-indigenous species. We here describe these new findings, providing detailed collecting data, taxonomic comments, and comparisons with similar species.

Origin of samples
The studied material comes from three main sources. First, sampling on the Israeli Mediterranean shelf performed in the context of the project "Historical ecology of Lessepsian migration" (HELM), in progress at the University of Vienna. Second, benthic assemblage monitoring by the Israel Oceanographic and Limnological Research (IOLR). Third, smaller scale sampling by some of us, further detailed in the Results section.
Sampling in the framework of the HELM project was conducted on soft substrates between 10 and 40 m depth with a van Veen grab, and on hard substrates between 5 and 30 m by diver-operated airlift suction sampling, using 0.5 mm mesh-size net bags. Samples were sieved with a 0.5 mm mesh and the retained material fixed in 95% ethanol. Both living individuals and empty shells were identified and counted.
IOLR conducts regular monitoring of Israeli soft bottom benthic assemblages in the framework of the National Monitoring (NM) and focused sampling for environmental assessment (APM DAN, Shafdan, Via Maris). The NM, APM DAN and Via Maris projects sampled soft substrates with a 0.11 m 2 van Veen grab at depths between 6 and 12.5 m (NM), 22 and 26.5 m (APM DAN) and 18 and 26 m (Via Maris). Samples were sieved with a 250 μm mesh. During the Shafdan project, three replicate sediment samples were taken at each station from a different 0.062 m² box-corer launch (Ocean Instruments model 700 AL) twice a year in spring (May) and fall (October). The samples were sieved on board with a 0.5 mm mesh. All samples were preserved in 99% ethanol, stained with eosin solution (hence the pink hue that some specimens bear) and picked for living individuals.
Finally, we included serendipitous findings by some of us or by colleagues within our extended network, from multiple localities. For each species, we provide detailed collecting data following the guidelines by Chester et al. (2019).

Taxonomic assignment and non-indigenous status attribution
The depth of taxonomic assignment varies across taxa, mostly reflecting the available knowledge on these groups in the Indo-Pacific province (the source pool of most nonindigenous species in the Eastern Mediterranean). For families like the Triphoridae, some of us (PGA, PAJB, and BS) have been conducting taxonomic research for a long time and we have thus been able to describe new species as we have robust knowledge of inter-and intraspecific variability and of type specimens (Albano et al. 2011(Albano et al. , 2017Albano and Bakker 2016). For other families, like the Eulimidae, we focused our attention on highlighting differences from native species and similarities with Indo-Pacific species, because a more thorough coverage would have required revising the taxonomy of entire Indo-Pacific species-groups, a task well beyond our objectives. In all cases, we strove to provide detailed and high-quality images as a basis to foster further research and enable the scientific community to refine our identifications. The use of qualifiers for species left in open nomenclature follows the recommendations of Sigovini et al. (2016).
Acknowledging that an unsettled taxonomic status implies uncertainty in the assignment of non-indigenous status , we here tagged as nonindigenous only the species which: i) unequivocally belong to Indo-Pacific species; ii) belong to clades (genera or families) that do not occur in the Mediterranean Sea, even if left in open nomenclature; iii) belong to species whose diagnostic characters did not enable a clear attribution to a non-Mediterranean clade but that were found alive while not, or only very rarely, in the death assemblage (see Discussion). In contrast, we tagged as "potential" non-indigenous species those whose morphological characters did not allow for an unambiguous attribution to a non-Mediterranean clade and that were found mostly, or exclusively, as empty shells.

Imaging and reporting
Small specimens were photographed with a Zeiss SteREO Discovery.V20 stereomicroscope, larger ones with a Nikon D7200 camera mounted on a stand, using a Nikon Micro-Nikkor 60 mm lens. Photographs were stacked with Helicon Focus 6. Scanning electron microscope (SEM) images were shot with a Fei Inspect S50 at low-vacuum mode without coating. The internal shell morphology of Odostomia (s.l.) sp. 1, with a particular focus on the intorted protoconch, was visualized using a Phoenix v|tome|x s research edition computer tomographic (CT) scanner. The 3D-reconstruction and virtual sections through the shell were produced with VGSTUDIO MAX 2.1 software. The X-ray image stack, mesh files, virtual sections, and a video showing the interior of the shell are available from the Figshare repository (https://doi.org/10.6084/ m9.figshare.c.5215226). Plates were mounted with the image manipulation software GIMP 2.
For each new non-indigenous species record, we report the size of at least one specimen (usually the one figured, unless otherwise stated). The systematic arrangement follows Bouchet et al. (2010Bouchet et al. ( , 2017. Table 1 summarizes the species treated in this work. Table 1. List of the taxa treated in this paper, with indication of the novelty of the records. New NIS for the Mediterranean Sea 12 Figure 6 Opimaphora blattereri Albano, Bakker & Sabelli, sp. nov

Class Gastropoda Cuvier, 1795 Family unassigned (Caenogastropoda)
Unidentified gastropod Figure 1 New records. Israel • 1 spcm; Haifa Bay; 32.8211°N, 35.0196°E; depth 11 m; 2 Aug. 2015; soft substrate; grab; NM project (sample HM27(c)); size: H 2.5 mm, W 1.6 mm. Remarks. We were not able to confidently assign this specimen to any family. The general characters suggest that it is a caenogastropod. This specimen has apparently traces of the animal inside and has thus been considered live collected. However, as it was found in Haifa Bay, we cannot exclude that it comes from freshwater or transitional ecosystems (the adjacent Kishon River and estuary) whose waters flow into the bay. An anatomical study of the soft parts, should another living specimen become available, will clarify the taxonomic placement of this intriguing species.
Remarks. The species has already been reported from Israel and Turkey (Bogi and Galil 1999;Buzzurro and Cecalupo 2006), but only from empty shells. To the best of our knowledge, this is the first record of living individuals from the Mediterranean Sea. Based on our observations, it occurs rather frequently on shallow subtidal rocky substrates. The samples from Turkey led to the description of Parviturbo dibellai Buzzurro & Cecalupo, 2006, at that time supposed to be a native species, but this name was later recognized to be a synonym of the Indo-Pacific Fossarus eutorniscus (Rubio et al. 2015), attributed to Conradia by Janssen et al. (2011). Janssen et al. (2011) highlighted, however, that the Red Sea specimens have seven spiral cords instead of the five cited in the original description based on material from Karachi (Pakistan). Specimens with five spiral cords occur also in the Persian (Arabian) Gulf and rarely in the Red Sea (H. Dekker, pers. comm., November 2020). Further research is required to ascertain if these two morphologies belong to two different taxa.
Remarks. This species has been recently described from sediment collected in 2016 at 33-45 m depth at Karpathos and Samos islands in the eastern Aegean Sea (Agamennone et al. 2020b). The authors discussed but declined the possibility that this is a Lessepsian species, but one of us (BS) observed non-distinguishable specimens from the Red Sea and we received reports of further indistinguishable specimens from the Persian (Arabian) Gulf (H. Dekker, pers. comm., November 2020). This is the first record for Israel; all specimens were live collected.  Figure 3 New records. Israel • 1 sh;Akko;32.92°N, 35.07°E;depth 4 m;22 Oct. 1998; shell grit sample; size: H 0.6 mm, W 1.0 mm.

Skeneidae indet.
Remarks. This tiny gastropod (largest diameter 1 mm) is characterized by a small but solid shell, ~ 0.75 whorls of protoconch with axial costae visible near the prototeleoconch transition (more costae closer to the nucleus may be abraded), and two teleoconch whorls with numerous regular spiral cords. The shoulder is slightly angulated near the lip. Umbilicus open, large. Shell white, slightly translucent. No native Mediterranean species shares these features. Only Skenea catenoides (Monterosato, 1877) has a similarly solid shell with numerous regular spiral cords, but it can be distinguished easily by the three nodulose thicker spiral cords on the base and the lack of angulation at the shoulder. Both Mediterranean (e.g., Circulus striatus (Philippi, 1836) and Red Sea Circulus (e.g., C. novemcarinatus (Melvill, 1906a)) and C. octoliratus (Carpenter, 1856)) can be distinguished by the multispiral protoconch and the much more prominent spiral cords. It is most likely a new, probably still unnamed, Indo-Pacific species in the Mediterranean. New records. Israel

Family Naticidae Guilding, 1834
Eunaticina papilla (Gmelin, 1791) Figure 5 New records. Israel  Remarks. We here report the finding of a living individual of Eunaticina papilla from the Israeli Mediterranean shelf. This juvenile specimen can be assigned to E. papilla because of its overall shape, the sculpture of fine spiral cords, the large umbilicus and the morphology of the thin corneus operculum ( Figure 5D).The species has already been reported in the Mediterranean Sea from Iskenderun in eastern Turkey with a living individual (Öztürk and Bitlis Bakir 2013). An empty shell was collected near Shiqmona, Israel, in November 2019 and reported as Eunaticina linneana (Récluz, 1843) (Schechter and Mienis 2020), a name considered a junior synonym of E. papilla by Beu et al. (2004).
Description. Color: protoconch light brown; first teleoconch whorls whitish, with the first spiral cord becoming brown after one to three whorls. The second spiral cord acquires this brown color only on the dorsal part of the last whorl. The fourth cord, visible only on the last whorl, is brown. The base is light brown.
Teleoconch: 6 (holotype, paratype 1 and 2), 7.5 (Mediterranean specimen) whorls, height: 2.04 mm (holotype), 1.83 mm (paratype 1), 2.43 mm (paratype 2) and 2.95 mm (Mediterranean specimen). The tuberculate first and third spiral cords start simultaneously after the protoconch with the same size, the third later becomes progressively larger and more acute. The second spiral cord appears only on the last whorl and is smaller than the others in front view, becoming of similar size to the first dorsally. In the second half of the last whorl, a very thin smooth suprasutural cord is visible. The base shows a fourth rather smooth cord of the same color as the first, followed by a fifth and sixth cord that are smooth and very pale in color. Anterior siphonal canal short, tubular, and oblique; posterior siphonal canal a simple notch. Peristome without microsculpture and apparently without bifurcating spiral cords.
The Mediterranean specimen is larger, has three white whorls after the protoconch and the second spiral cord appears on the seventh whorl, remaining still smaller than the others.
Etymology. Named after the Lessepsian invasion (Por 1978), because we first found this Red Sea species on the Mediterranean Israeli shelf. The species epithet is an adjective in nominative singular feminine.
Remarks. The Mediterranean specimen is larger and broader than the Red Sea ones. Triphorids do show a morphological dimorphism characterized by smaller and larger morphs and we think that we captured this dimorphism in our samples. See under Opimaphora blattereri Albano, Bakker & Sabelli, sp. nov Diagnosis. Shell cyrtoconoid of less than 3 mm with 11 (holotype) or 12 (paratype 2) whorls and multispiral protoconch. Nucleus with hemispherical granules. Sculpture of three spiral cords with round tubercles larger than their interspaces; the second cord appears only on the fourth whorl, initially as a thin smooth thread. Microsculpture absent on the teleoconch whorls, present on the peristome, which bears bifurcating spiral cords.
Description. Color: protoconch brown; whitish first teleoconch whorls with the first spiral cord becoming brown after two whorls. Light brown irregular patches are randomly distributed on the teleoconch, usually covering one or, more frequently, two tubercles. The base background is white, with the color patches of the last whorl extending onto it. The tip of the anterior siphon is brown.
The tuberculate first and third spiral cords start simultaneously after the protoconch with the same size, whereas the second cord appears from the fourth to the seventh teleoconch whorl, depending on shell size. This cord is initially thin and closer to the first one, it progressively increases its size until reaching that of the other two cords on the last whorl. On the second half of the shell, a very thin smooth suprasutural cord is visible. The second cord bifurcates on the peristome. The base shows a fourth rather smooth cord, and a fifth and sixth smooth ones; these cords become towards the peristome more granulated. On the peristome, below the third spiral thread, microsculpture is visible as fine spiral lines. Anterior siphonal canal tubular, short and oblique; posterior siphonal canal a simple notch.
Etymology. This species is named after Hubert Blatterer, Austrian conchologist, in recognition of his work on Red Sea mollusks. Moreover, he contributed to our work on Lessepsian species by granting us access to the material he collected in the Red Sea and by donating the type series of O. blattereri and Coriophora lessepsiana. The species epithet is a noun in the genitive case.
Remarks. We describe O. blattereri as new because of the similar color pattern to C. lessepsiana Albano, Bakker & Sabelli, sp. nov., even if it has not been reported from the Mediterranean Sea. The two species can be easily distinguished because C. lessepsiana has an monocarinated protoconch while O. blattereri has a bicarinated one; the second spiral cord of O. blattereri never becomes brownish as in C. lessepsiana; O. blattereri has a white background on the base and a distinct brown end of the anterior siphonal canal, whereas C. lessepsiana has a light brown base and the anterior siphon has not a colored end; the teleoconch of O. blattereri has irregular light brown patches, particularly evident on fresh specimens; this feature is totally absent in C. lessepsiana.
We have seen specimens very similar to O. blattereri collected in Madagascar, New Caledonia, and French Polynesia. A revision of the group in the Indo-Pacific province is beyond the scope of this paper; however, this species likely has a broad distribution.
Opimaphora blattereri and C. lessepsiana share their color pattern of brown to orange spiral cords on a white background with other Indo-Pacific species. Litharium bilineatum (Kosuge, 1962) (holotype illustrated by Higo et al. (2001)), Costatophora iniqua (Jousseaume 1898) (= Notosinister kawamurai Kosuge, 1962, type material illustrated by Higo et al. (2001) and ) and Aclophora albozonata Laseron, 1958 can be easily distinguished by having three fully developed spiral cords since the early teleoconch. Iniforis formosula (Hervier, 1898) and Mastonia peanites Jousseaume, 1898 (= Mastonia squamosa Kosuge, 1962, type material again illustrated by Higo et al. (2001) and ) have only two spiral cords, but the former has three or four dark brown lines on the last whorl, whereas the latter has a dark brown last whorl with lighter tubercles. Triphora fulvescens Hervier, 1898 also has a similar color pattern, but the second spiral cord remains a very fine thread even on the last whorl and the tubercles are whitish even on the first cord (on an orange background). Some species show a delayed appearance of the second spiral cord: Nototriphora regina (Hedley, 1903) has a brown tip of the anterior siphonal canal similarly to O. blattereri, but lacks the patches on the whorls and has an orange line on the third spiral cord on the last whorl; Coriophora tigris Laseron, 1958 has a paucispiral protoconch; Cautor similis (Pease, 1871) has larger and more densely arranged tubercles, a brown fourth spiral cord and white base. Last, a few species have a similar color pattern, but with an inverted pattern: the first spiral cord is white and the third orange to brown, like Mastonia cingulifera (Pease, 1861), which also has a dark yellow teleoconch, Mastonia funebris Jousseaume, 1884 and Mastonia tulipa Jousseaume, 1898 with a brown and white base, respectively. Triphora sp.  Remarks. We found a single, adult, empty shell. It likely possesses a large paucispiral protoconch, but it is incomplete in our shell. The second spiral cord starts at midshell height, the fourth and fifth spiral cords are smooth, and the posterior siphonal canal is shallow. It is brown in color with darker spiral cords. We have not been able to assign it to a species so far, but it is distinctly different from all known Mediterranean species and most likely belongs to the Indo-Pacific fauna.  Remarks. This species was first recorded from Israel by Steger et al. (2018) based on three living individuals from Palmachim, southern Israel. We here report multiple living individuals all along the Israeli coast, confirming its establishment. The species shows a broad distribution in the Eastern Mediterranean ranging from Greece to Turkey and Cyprus (Micali et al. 2017;Stamouli et al. 2017;Angelidis and Polyzoulis 2018;Chartosia et al. 2018). Its final taxonomic assignment requires the clarification of the relation between several other Viriola such as V. corrugata (Hinds, 1843), V. senafirensis (Sturany, 1903), and V. tricincta (Dunker, 1882) (Albano and Bakker 2016;Albano et al. 2017.

Family Cerithiopsidae H. Adams & A. Adams, 1853
Cerithiopsis sp. aff. pulvis (Issel, 1869) Figure 9D-F New records. Israel  Remarks. This species superficially resembles the Lessepsian Cerithiopsis pulvis but has a more cyrtoconoid shape and a greater ratio between the height of the last whorl and that of the shell. The base is not concave as in C. pulvis, bears a fourth spiral cord which is more prominently tuberculate, and an additional fifth tuberculate cord that is not present in typical C. pulvis. Additionally, the siphonal canal bears numerous fine cords. The color pattern is similar to C. pulvis which has orange bands on white background; in contrast, in C. aff. pulvis these are brown and yellowish, respectively. It is distinct from any native Mediterranean species and clearly belongs to an Indo-Pacific clade. It is here considered a new non-indigenous species.
Cerithiopsis sp. Figure 10 New records. Israel   Remarks. This beautiful species has almost eight teleoconch whorls bearing two strong spiral cords with oblong tubercles at the intersection with prosocline axial ribs. Interspaces between spiral cords are approximately as large as the cords themselves, and interspaces between the axial ribs are double the size of the ribs. A third smooth thick cord delimits the rather flat base and is visible above the suture throughout most of the teleoconch. The protoconch is smooth with very fine and extremely short axial riblets just below the suture; it is multispiral but broken in our specimen in which only the last two whorls are preserved. The slender shape, the two strong spiral cords and the smooth flat base distinguish it at once from all native Mediterranean species suggesting it is a new non-indigenous species in the basin.
Among Indo-Pacific cerithiopsids, Synthopsis lauta Cecalupo & Perugia, 2013, described from Vanuatu, is among the few similar species we were able to trace. However, the interspace between the spiral cords is broader, the tubercles on the first spiral cord of the last whorl are larger than those on the second cord, and the teleoconch is shorter with just six whorls. Additionally, the color pattern with white tubercles, yellowish interspaces, deep brown suture and violet protoconch is strikingly different from the one of our shell. We have some reservations that S. lauta, as well as our specimen, belong to the genus Synthopsis Laseron, 1956 that was described as bearing three tuberculate spiral cords on the whole teleoconch (Laseron 1956). Pending a molecular phylogeny of the family, we consider this feature important at the genus level. Therefore, we assign our specimen to the nominotypical genus Cerithiopsis, in the wait of a better understanding of cerithiopsid systematics. The specimen identified as Horologica gregaria Cecalupo & Perugia, 2012 and illustrated in the recent revision of Cerithiopsidae from South Madagascar (Cecalupo and Perugia 2014b: fig. 8G) is also similar to ours; that specimen, however, has a distinct basal spiral cord which is absent in our specimen. The latter character, the prominence of the tuberculate spiral cords and the evident but rather flat third cord also raise some doubts that the specimen from South Madagascar is conspecific with the H. gregaria originally described from the Central Philippines (Cecalupo and Perugia 2011). Last, the Sudanese specimen of Horologica cf. taeniata Cecalupo & Perugia, 2013 illustrated by Cecalupo and Perugia (2016: fig. 1P-S) shares the general features of our shell but can be distinguished by the first spiral cord that tends to split into two separate cords, and by the color pattern of white teleoconch and orange base.
Protoconch: composed of 3.5 whorls with no clear demarcation between protoconchs I and II, height: ~ 300 μm, width ~ 200 μm (holotype), but accurate measurement hampered by the last protoconch whorl being covered by the first teleoconch whorl. It appears smooth except for growth lines and fine pustules covering the lower half of the first whorl and sparsely present apically and abapically on the following whorls (only visible with scanning electron microscopy at high magnification).
Teleoconch: 4 whorls (holotype), height: 1.4 mm (holotype). It bears three spiral cords of equal size, with tubercles at the intersection of orthocline axial ribs. The base is contracted and has two additional tuberculate spiral cords. Tubercles become oblong near the lip. Anterior siphonal canal short, reverted upwards, formed by a prong-like protrusion of the anterior outer lip ( Figure 11A); posterior siphonal canal notch-like.
Etymology. The name problematicus refers to the difficult task of recognizing and identifying non-indigenous species belonging to groups whose taxonomy in the tropical seas is poorly known (see Discussion). The species epithet is an adjective in nominative singular masculine.
Remarks. This species is characterized by its bulbous contour and constricted last whorl which justify its inclusion in the genus Joculator Hedley, 1909(Hedley 1909Marshall 1978).
The Cerithiopsidae of the Indo-Pacific have been subject to numerous in-depth studies (Cecalupo and Perugia 2011, 2013, 2014a, b, 2016, 2017a, 2019a. Still, this species does not fit any of the known species. Among the most similar species in terms of shell shape and ornamentation, Joculator itiensis Cecalupo & Perugia, 2014 has one teleoconch whorl more and a different color pattern characterized by light brown first whorl and base, J. olivoideus Cecalupo & Perugia, 2018 can be distinguished by its clearly prosocline axial ribs and greyish tubercles, and J. sekensis Cecalupo & Perugia, 2018 has only two spiral cords and blunter axial ribs on the first teleoconch whorl, in addition to a blunter siphonal canal.
There are several more species of small brown bulbous Joculator often distinguishable only by subtle character differences. Joculator priorai Cecalupo & Perugia, 2012 is corneous in color and has a pointed protoconch with one additional whorl; moreover, in our specimens the interspaces between the spiral cords are smaller. Joculator pupiformis Cecalupo & Perugia, 2012 has one protoconch and one teleoconch whorl more, the tubercles are oblong, and the base lacks a clearly visible fifth tuberculate spiral cord. Joculator fuscus Cecalupo & Perugia, 2012 has much broader interspaces between cords and a wide subquadrangular aperture which is, in contrast, quite small in our specimens. Joculator furvus Cecalupo & Perugia, 2012 has a neat abapical smooth cord on the protoconch, one teleoconch whorl less and a broader aperture. Joculator carpatinus Cecalupo & Perugia, 2012 has one protoconch whorl more, one teleoconch whorl less, a broader aperture and a fine abapical thread on the protoconch. Joculator caliginosus Cecalupo & Perugia, 2012 has one protoconch whorl more and one teleoconch whorl less, the basal fourth and fifth cords are only weakly tuberculate whereas they are neatly tuberculate in our specimens. Joculator coffeus and J. subglobosus, both Cecalupo & Perugia, 2013, have one clear abapical thread on the protoconch, one teleoconch whorl less, the shell has a more roundish shape and the lip does not reach anteriorly the siphonal canal, almost covering it, like in our specimens. The other representatives of Joculator include also other more elongated species that can be easily distinguished from our specimens.
This species is superficially similar to the native Mediterranean Cerithiopsis ladae Prkić & Buzzurro, 2007, which, however, can be distinguished at once for not having the last protoconch whorl partially covered by the first teleoconch whorl and lacking the prong-like process of the anterior outer lip. Additionally, tubercles in C. ladae on the last whorl are more elongated, subrectangular, and the shell profile is less bulbous. Cerithiopsis greppii Buzzurro and Cecalupo, 2005, described from Turkey, has a rather oval profile, but not as bulbous as in our species; additionally, it has a paucispiral protoconch. Cerithiopsis micalii (Cecalupo and Villari, 1997), which also has a somewhat oval shell profile, can be quickly distinguished by its protoconch whose last two whorls bear strong axial ribs.
Unfortunately, a revision of Red Sea Cerithiopsidae is lacking, but given that Joculator is a broadly distributed genus in the Indo-Pacific province, we consider J. problematicus another previously undescribed Indo-Pacific species recently introduced to the Mediterranean Sea.
Remarks. The morphology of this species is unique among the native mollusks of the Mediterranean, which does not host any shallow water Elachisinidae. Therefore, we consider it a new non-indigenous species in the basin.
The only Indo-Pacific Elachisina we are aware of is E. robertsoni Kay, 1979, which indeed shares the general characters of our species. However, it can be readily distinguished by the thicker and fewer spiral cords, less rounded whorls and sigmoid, rather than strongly prosocline, aperture profile. Elachisina sp. is more similar to the West-African E. tenuisculpta (Rolán and Gofas 2003), but the Israeli shells have more rounded whorls, a greater height/width ratio and smaller ratio between aperture and shell height.
The closest match to our specimens is Iravadia elongata (Hornung & Mermod, 1928) which was described from material collected by Arturo Issel in the Red Sea off Massawa, Eritrea, at 30 m depth (Hornung and Mermod 1928). Compared to our material, however, the syntype of I. elongata is larger (height 3.9 mm vs. 2.8 mm in our largest shell) and has seven less convex whorls. Further, the apical part of its spire has a slightly concave profile and thus appears more tapered. According to Issel's description, the sculpture of I. elongata consists of spiral ridges (12 on the penultimate and 22 on the last whorl) as well as growth lines, although the latter are not indicated in the accompanying line drawing. This suggests that the axial component might be less evident in I. elongata than in our specimens, however, the poor preservation of the shell surface of the syntype of I. elongata did not allow a reliable comparison with our material. Slightly eroded shells very similar to our specimens have been collected from the Sudanese Red Sea ( Figure 13G, H), confirming that the material from Israel indeed represents an Indo-Pacific species rather than an undescribed Mediterranean taxon.
Among Mediterranean iravadiids, our specimens superficially resemble only Ceratia proxima (Forbes and Hanley, 1850). This species, however, lacks axial sculpture. Interestingly, Hornung and Mermod (1928) also mention the presence of this latter species at Assab (Eritrea) and "île Saldadin" (Zeila, northern Somalia). While obviously based on a misidentification -C. proxima has an Eastern Atlantic-Mediterranean distribution (Bouchet and Warén 1993;Høisaeter 2009) -one might speculate that this record could be the result of a confusion of C. proxima with the Iravadia presented here.  Remarks. This tiny gastropod defeated all our attempts to identify it. It consists of a protoconch and a teleoconch of ~ 1.5 whorls each. Sculpture is absent, except for two spiral ridges that run on the shoulder and on the base. A third ridge runs periumbilically ( Figure 14E). Broad umbilicus, roundish aperture. Our shell closely resembles the Vitrinella sp. 1 illustrated by Blatterer (2019: plate 127, fig. 12a-j) from the Dahab region in the northern Red Sea, which, however, apparently bears fine spiral threads in the umbilicus ( fig. 12e, and unpublished figures). SEM images of our shell show that its surface is taphonomically altered; additionally, Blatterer's specimens look slightly more mature, reaching 2 teleoconch whorls. The significance of these features should be re-assessed upon a satisfying revision of these tiny gastropods from the Indo-Pacific province. Another similar shell is illustrated by Janssen et al. (2011, plate 19, figs. 3a-b), which apparently has less conspicuous or absent spiral ridges as long as can be judged from the optical illustrations provided. It is worth mentioning that gastropods belonging to the family Clenchiellidae D.W. Taylor, 1966 share the small size, low spire, wide umbilicus and presence of strong spiral keels we observed in our specimen (Ponder et al. 2014); the latter, however, lacks the numerous finer spiral cords that characterize clenchiellids. Additionally, these gastropods occur in mangrove swamps or adjacent habitats in tropical estuaries, a kind of habitat that does not occur in Israel. The shell shape and sculpture (in particular the strong spiral keels) distinguish it at once from native Mediterranean species. The extreme similarity with the shell illustrated in Blatterer's book suggests that the species belongs to a Red Sea clade and is here considered a new non-indigenous species in the Mediterranean Sea.

Hemiliostraca clandestina (Mifsud & Ovalis, 2019), comb. nov.
Remarks. Sticteulima clandestina and S. athenamariae, both Mifsud & Ovalis, 2019, were described on specimens collected in Turkey (Mifsud and Ovalis 2019). However, both belong to species present in the Red Sea and were illustrated by Blatterer (2019)   and Hemiliostraca athenamariae. This is the first record of H. clandestina in Israel, but the species has been recorded for Lebanon based on empty shells collected in 1999 (Crocetta et al. 2020). Consequently, it is likely present here since at least 1999, with a ~ 20 year time-lag in first detection as quantified also for other non-indigenous species in the Mediterranean Sea (Crooks 2005;Albano et al. 2018). This is also the first record of living individuals from the Mediterranean Sea. Despite the relatively large number of living individuals, we did not find any attached to an echinoderm host; this is consistent with the fact that some eulimids actively leave the host if disturbed (Warén 1984). Figure 17 New records. Israel • 5 spcms; Ashqelon; 31.6868°N, 34.5516°E; depth 12 m; 30 Apr. 2018; offshore rocky reef; suction sampler; HELM project (samples S12_1F, S12_1M, S12_3F); size: H 2.7 mm, W 1.0 (illustrated specimen) • 1 spcm; same collecting data as for preceding; depth 11 m; 31 Oct. 2018; HELM project (sample S58_2F) • 3 spcms; Ashqelon; 31.6891°N, 34.5257°E; depth 25 m; 2 May 2018; offshore rocky reef; suction sampler; HELM project (samples S16_1F, S16_2F) • 1 spcm; same collecting data as for preceding; depth 28 m; 31 Oct. 2018; HELM project (sample S59_1F).

Melanella orientalis Agamennone, Micali & Siragusa, 2020
Remarks. This species can be distinguished from Mediterranean Melanella by its gently curved whorls, straight spire with fewer whorls and thinner shell than most species. It superficially resembles the Red Sea "Eulima" orthophyes Sturany, 1903 (type illustrated by Albano et al. (2017)), which can be distinguished because of its slightly bent apical whorls and the unusual pustulous sculpture of the protoconch. The species presented here is apparently already widespread in the Eastern Mediterranean (Agamennone et al. 2020a). We found only living individuals and no empty shells. Because of this, and the low likelihood that a so widespread species in shallow depths in the Remarks. The genus Parvioris Warén, 1981 was erected for a group of numerous conchologically very similar species of which many are still undescribed (Warén 1981). The species here reported is very similar in general shape and size to P. dilecta (Warén 1981: 146), especially the morphs illustrated here in Figure 18I-M. However, it has a multispiral protoconch of ~ 4.5 whorls ( Figure 18H), whereas P. dilecta has a paucispiral protoconch of ~ 1.5 whorls. The type of protoconch is considered to be related to the developmental mode, which was regarded a diagnostic character at the species level for most molluscan lineages (Hoagland and Robertson 1988;Bouchet 1989). Because Warén (1984) suggested that the number of protoconch whorls is rather constant within species in Eulimidae, we currently do not consider our material conspecific with P. dilecta, but only closely related (thus the "aff." notation). However, there is increasing evidence that poecilogony, the intraspecific variation in developmental mode, occurs in Caenogastropoda (McDonald et al. 2014), Neogastropoda (Russini et al. 2020), and Sacoglossa (Krug 1998;Ellingson and Krug 2006;Vendetti et al. 2012).
Parvioris aff. dilecta can be easily distinguished from the native P. ibizenca because of a more arched apical part and because of the protoconch morphology: both have multispiral protoconchs, but P. ibizenca has shorter whorls and a distinct profile which inflates at the third whorl, in contrast with the more slender and regular profile of P. aff. dilecta. Our specimens are likely conspecific with those identified as Melanella sp. 1 by Blatterer (2019) from Dahab, Red Sea, suggesting that it is indeed a new Lessepsian species. An additional issue is whether the animal color is diagnostic at the species level like in other groups whose shells offer few diagnostic morphological characters, e.g., Mediterranean Granulina (Neogastropoda: Granulinidae) and Gibberula (Neogastropoda: Cystiscidae) (Gofas 1990(Gofas , 1992. Some of our live collected specimens show a light yellow-white color (e.g., Figure 18I-K) whereas others have a brownish animal (e.g., Figure 18K-M). The final attribution of our findings to a species requires a thorough revision of Parvioris, which is beyond the scope of this paper. The specimens reported as Parvioris sp. by Albano et al. (2020) from mesophotic reefs off northern Israel belong to this species. Sticteulima sp. Figure 19 New records. Israel • 1 sh; north of Atlit; 32.7820° N, 34.9466° E; depth 10 m; 21 Sep. 2016; sand; grab; HELM project (sample NG10_1F); size: H 1.4 mm, W 0.6 mm (illustrated shell, Figure 19A-F) • 1 spcm; Ashqelon; 31.6891°N, 34.5257°E; depth 28 m; 31 Oct. 2018; offshore rocky reef; suction sampler; HELM project (sample S59_1F).
Remarks. We place this species in Sticteulima due to its small size, slender profile with high and rather flat whorls (Warén 1984). In contrast to the native S. jeffreysiana (Brusina, 1869) and the Lessepsian S. lentiginosa (A. Adams, 1861), it is colorless, also in live-collected specimens, and stouter. Further, this species does not match any of the known small-sized Mediterranean eulimids. It can be readily distinguished from Vitreolina curva (Monterosato, 1874) and Melanella levantina (Oliverio, Buzzurro & Villa, 1994) by the lack of the strongly arched apical whorls. This feature differentiates it at once also form other Red Sea small-sized eulimids (Blatterer 2019). Melanella petitiana (Brusina, 1869) is larger, has more numerous whorls (our Sticteulima has a fully thickened lip suggesting that it is an adult) and has a less prominent lip profile. Nanobalcis nana (Monterosato, 1878) (type illustrated by Appolloni et al. (2018)) has shorter whorls, especially the last one, which is also much broader than in this species. It can also be easily distinguished from Hemiliostraca athenamariae (Mifsud & Ovalis, 2019) by the lack of any color pattern, and the more inflated lip profile with a deeper posterior sinus. Sticteulima sp. may be a new non-indigenous species in the Mediterranean Sea. Remarks. This Vitreolina is extremely similar to the native V. philippi, but the animal is whitish with a yellowish digestive gland ( Figure 16E, F), in contrast to the peculiar color pattern of typical V. philippi with a white background and red dots ( Figure 16D). Vitreolina is known to be gonochorous (Warén 1984) but it is unclear if this different color pattern, never reported from the Mediterranean, can be related to sex. We suspect that this could be another new Lessepsian species for the Mediterranean Sea, because we observed several Mediterranean-Red Sea species pairs that are morphologically extremely similar. If we are correct, the occurrence

Remarks.
Conus fumigatus was first recorded from the Mediterranean Sea in Libya (Röckel 1986) but not recorded again for three decades until a recent report from Syria (Ammar 2018). This is the first finding in Israel, filling the distributional gap from the Suez Canal northward; only shells of juveniles have been found so far.

Family Murchisonellidae T.L. Casey, 1904
Henrya (?) sp. Figure 21 New records. Israel  Remarks. We were unable to assign this species to any Mediterranean or Indo-Pacific species, despite its conspicuous combination of shell characters. Our single specimen has an elongated, pupoid shell with convex whorls, a narrow but deeply incised suture, and a heterostrophic protoconch of type B (diameter: 250 μm). The surface is glossy and smooth except for densely spaced, very fine growth lines. The latter are straight, slightly prosocline on the spire, becoming orthocline near the aperture. The aperture is drop-shaped with a simple, thin lip that is slightly reflected at the columella. An umbilical chink is present. The shell is translucid-white, ornamented with a single, broad, light brown spiral color band. The shell morphology is similar to species of the murchisonellid genus Henrya Bartsch, 1947. However, the three currently known species of that genus were described form the tropical West Atlantic (Florida, Bahamas, and Yucatan) (Bartsch 1947), and none of them has a brown color band. For these reasons, the lack of anatomical and molecular data, and the fact that only a single specimen was available for study, we refrained from a definitive generic assignment. This species is potentially another non-indigenous one originating from the Indo-Pacific.

Family Pyramidellidae Gray, 1840
Odostomia cf. dalli (Hornung & Mermod, 1925) Figure 22 New records. Israel  Remarks. The shell of this species is white and rather solid, with convex, unkeeled whorls and a deep, narrow suture. The columellar tooth is visible in frontal view; there are no lirae inside the aperture. The outer surface appears smooth at first sight but bears numerous very fine spiral lines. The protoconch is of type A2, tending to type B. In ethanol-preserved specimens, the soft body is yellowish-white, with the eyes well visible through the shell ( Figure 22A). Odostomia cf. dalli differs in its shell morphology from all known Mediterranean Odostomiinae, but bears close resemblance to the illustration of the type specimen of Odostomia dalli from Sarad Island ("Ile de Sarato"), Dahlak Archipelago, Eritrean Red Sea (Hornung and Mermod 1925). In contrast to our material, however, O. dalli was described as lacking both, spiral sculpture and a visible columellar fold, although a columellar tooth seems to be indicated in the line drawing accompanying the original description. A rigorous assessment of potential conspecificity between O. dalli and our material therefore awaits a thorough study of the type material of the former, but the close similarity suggests that this is a new non-indigenous species in the Mediterranean Sea. This interpretation is also supported by the lack of empty shells in death assemblages (see Discussion).
Odostomia (s.l.) sp. 1 Figure 23 New records. Israel • 1 spcm; Haifa Bay; 32.8211°N, 35.0196°E; depth 11 m; 2 Aug. 2015; soft substrate; grab; NM project (sample HM27(c)); size: H 1.4 mm, W 0.7 mm Remarks. This species is characterized by a translucid-white, cylindrical shell with ~ 3 whorls, and an intorted protoconch of type C ( Figure 23I) whose columella is oriented at an angle of ~ 160° relative to the teleoconch axis (revealed by μCT-imaging, Figure  23H and additional scans available at https://doi.org/10.6084/m9.figshare.c.5215226). The growth lines are slightly prosocline on the spire while becoming almost orthocline on the body whorl; an extremely faint spiral microsculpture is present on the apical part of the whorls, but only visible in high-magnification SEM images ( Figure 23G). This species differs from Odostomia cf. dalli by its smaller size (height up to 1.4 mm), the more cylindrical shape, shallower suture, and the absence of a visible columellar tooth. Although this species, in terms of size and overall shape, somewhat resembles representatives of the fresh-and brackish water-dwelling family Hydrobiidae, the fact that numerous living specimens were found in a fully marine environment and its heterostrophic protoconch unambiguously identify it as member of the family Pyramidellidae.
Odostomia sp. 1 does not resemble any known Mediterranean pyramidellid; considering the great number of confirmed introductions of Indo-Pacific microgastropods to the eastern Mediterranean Sea, we therefore suspect that also this taxon might be a Lessepsian species. Among Indo-Pacific Odostomiinae, O. bullula Gould, 1861(e.g., Johnson 1964Robba et al. 2004) is similar to our specimens, but differs by its more conical shape and larger size (height to 2 mm, width to 1 mm). Another similar species, O. decouxi Saurin, 1959, was suggested to be a junior synonym of O. bullula (Robba et al. 2004). Odostomia (s.l.) sp. 2 Figure 24 New records. Israel Bogi and Galil (2006)).
Remarks. The first record of this species is based on five well-preserved shells found in a shell grit sample taken in 1995 on the beach of Yumurtalik, Adana, Turkey (Giunchi et al. 2001). In 2006, another beached shell was found at Neve Yam, northern Israel (Bogi and Galil 2006, re-illustrated in Figure 24 herein) and, according to these authors, the species was also found in Israel by J.J. van Aartsen. A specimen of Odostomia sp. 2, from the original lot from Yumurtalik, was recently figured by Giannuzzi-Savelli et al. (2014). Here, we report the first finding of a living individual of Odostomia sp. 2 which was recovered from a sediment sample taken at the Soreq desalination plant, southern Israel. Since the first finding in Turkey 25 years ago, the identity of this most likely nonindigenous species has remained unresolved. It differs from all known Mediterranean Odostomiinae at first glance by the presence of two brown spiral bands. We are unaware of any Indo-Pacific pyramidellid resembling this taxon, and it may well represent an undescribed species. To aid the further study of this taxon and raise awareness of its presence and apparent spread in the Mediterranean, we here re-illustrate the wellpreserved shell from Neve Yam using light and scanning electron microscopy. Figure 25A, B New records. Israel  Remarks. Oscilla virginiae is characterized by a small-sized, white, conical shell with a type A protoconch. The sculpture consists of thick, smooth spiral cords: the first and second whorl bear two cords; the upper cord is broadest and bifurcates on the third whorl, forming three cords on the last whorl, with the newly formed pair remaining positioned very close one to each other ( Figure 25A, B; Peñas et al., 2020). This species has just been described from the infralittoral of Jordan and also occurs in the Egyptian Red Sea (Peñas et al. 2020). It superficially resembles the Indo-Pacific O. appeliusi (Hornung & Mermod, 1925), and indeed, a juvenile shell from Dahab (Egypt) was  fig. 17c, d) under this name. In contrast to O. virginiae, however, O. appeliusi bears spiral cords more similar in thickness which are spaced more equidistantly and closer to each other. Already on the second whorl, three cords are present, and the uppermost cord does not evidently bifurcate (Peñas et al. 2020). Lastly, the illustration of Hornung and Mermod (1925) suggests a greater number of spiral cords on the last whorl. To date, O. appeliusi has not been recorded from the Mediterranean Sea.

Oscilla virginiae Peñas, Rolán & Sabelli, 2020
Within the Mediterranean, O. virginiae is superficially similar only to two other nonindigenous pyramidellids, Cingulina isseli (Tryon, 1886) and Miralda sp. (Figure 25C, D). The latter taxon has previously been reported under the name Oscilla jocosa Melvill, 1904, despite recent evidence by Peñas & Rolán (2017) that it is not conspecific with Melvill's (1904) type material. Oscilla virginiae differs from C. isseli by its broader, more conical shell, fewer whorls, smaller size (C. isseli reaches a height of ~ 3 mm), and the much less pronounced axial sculpture between the spiral cords. Compared to Miralda sp., O. virginiae differs by its smaller size (up to ~ 3 mm in Miralda sp.), the absence of beads on the two upper spiral cords ( Figure 25C Remarks. Parthenina cossmanni has an elongated-conical shell with flat-convex whorls and a protoconch of type C. The whorls of the spire have a subangular profile, whereas the body whorl in adult specimens is more convex and evenly rounded. The axial sculpture is made of strong orthocline ribs that become slightly flexuous on the body whorl in some specimens ( Figure 26A-E). The spiral sculpture on the spire consists of a single, thin, suprasutural cord; a second cord emerges on the penultimate whorl, and three cords are present on the last whorl. The columellar tooth is weak and deeply inset, and in some of the studied specimens hardly visible inside the aperture. The soft body of ethanol-preserved specimens is yellowish, with the eyes visible through the shell ( Figure 26G, H).
The type material of P. cossmanni was collected from the Red Sea of Massawa (Eritrea) at a depth of 30 m (Hornung and Mermod 1924); the species was recently recorded from Dahab (Gulf of Aqaba, northern Egypt) by Blatterer (2019), and from Jordan by Peñas et al. (2020). Outside the Red Sea, it is known from Vietnam (Saurin 1959) and Thailand (Robba et al. 2004).
Among native Mediterranean species, P. cossmanni superficially resembles Parthenina interstincta (J. Adams, 1797). The latter species, however, has only two spiral cords on the last whorl and a more developed columellar tooth. P. cossmanni is further similar to P. indistincta (Montagu 1808) (Figure 26I-L) which has a very weak, internal columellar fold (Warén 1991a: 96, fig. 29f ) and three (rarely four) spiral cords on the last whorl. Compared to P. cossmanni, however, the shell of P. indistincta is more elongated and has two spiral cords on the spire whorls.
We suspect the two shells illustrated as P. indistincta in Öztürk et al. (2011: fig. 10A, B) might also be P. cossmanni, considering their broad shape and overall morphology. Öztürk et al.'s material was collected in 2009 from a mud bottom at 9 m depth in Mersin Bay (stn. 46, 36.7167°N, 34.8667°E), south-eastern Turkey; should our hypothesis be confirmed upon re-examination of these shells, this would suggest that P. cossmanni likely has a wider distribution in the southeastern Mediterranean Sea. Our finding of several living specimens on the Israeli shelf, together with the relative rarity of empty shells in the samples, suggests that this species might have established locally only rather recently. Figure 27 New records. Israel  Remarks. This species is characterized by a straight, conical profile with flat whorls, separated by a deep, canaliculate suture; the abapical part of the whorl is angulated. The sculpture consists of straight axial ribs and a prominent suprasutural spiral cord; the base is smooth except for faint continuations of the axial ribs; an internal columellar fold is present and visible inside the aperture when slightly turning the shell to the left side. The protoconch is of type C and in the illustrated specimen it has a diameter of ~ 240 μm, which is slightly smaller than 270-290 μm stated by Peñas and Rolán (2017) for this species.

Parthenina typica (Laseron, 1959)
Parthenina typica (Laseron, 1959) was described from eastern Australia (Laseron 1959) and subsequently recorded from the Solomon Islands, Fiji and the Philippines at infralittoral to bathyal depths (Peñas and Rolán 2017). To our knowledge, it has not been reported from the Indian Ocean nor from the Red Sea, however, its absence could well represent an artifact of the limited knowledge of the micromollusk fauna of these regions. Among native species, the conchologically highly variable Parthenina interstincta (J. Adams, 1797) andP. monozona (Brusina, 1869) are most similar, however, they differ by having more rounded whorls and a greater number of axial ribs.     7917°N, 98.5500°E;depth 15 m;15 Feb. 1985;sand;A.J. Ferreira leg.;LACM 1985-14.2.
Pyrgulina craticulata has been reported only from the Red Sea so far, however, a search of previously uncatalogued lots of pyramidellid shells in the LACM collection by one of us (PILF) yielded specimens from Madagascar, the Maldives and Thailand, confirming that this species has a much wider distribution in the Indian Ocean. Shells of P. craticulata seem indistinguishable from the illustration of Chrysallida tribulationis (Hedley, 1909), a taxon recorded from Australia and the western Japan Sea (Hedley 1909;Higo et al. 2001). In the light of a possible distribution of P. craticulata also in the West Pacific, we recommend an assessment of potential synonymy between C. tribulationis and P. craticulata.
Here, we provide the first records of P. craticulata for the Mediterranean Sea. Several living specimens were found on hard substrates off southern and northern Israel, suggesting it is established in the region. In terms of shell size, shape and type of ornamentation, this species closely resembles the native Spiralinella incerta ( Figure 28J-N), but has pronounced spiral cords in the interspaces of the axial ribs ( Figure 28I vs. 28N) which enable a reliable segregation of these two species. In addition, the axial ribs are spaced more closely in S. incerta than in P. craticulata.
Remarks. Until now, the first record of this species from the Mediterranean was considered to be by Öztürk and van Aartsen (2006), who reported on material obtained from shallow-water sediment samples collected along the Turkish Levantine (Viransehir, Mersin Bay) and Aegean coasts (Güllük Bay) in 1997 and 2000, respectively. However, already van der Linden and Eikenboom (1992) described and illustrated P. nana from the Levantine Sea (page 60, Figure 41), referring to it as Chrysallida spec. C in the lack of a species-level identification. Their material consisted of a single individual, likely an empty shell, but not specified by the authors, from Mersin (south-eastern Turkey) with unknown collecting date, housed in the collection of J. van der Linden (The Hague, The Netherlands). Although we were unable to examine this material, the excellent and detailed line drawing provided enabled an unambiguous assignment of Chrysallida spec. C to P. nana; thus, van der Linden and Eikenboom (1992) should be regarded the first Mediterranean record. Today, the known Mediterranean distribution of P. nana includes Turkey, Lebanon, and Israel (Bogi and Galil 2006;Giannuzzi-Savelli et al. 2014). To our knowledge, this is the first record of living individuals of P. nana from Israel; here, the species occurs along both the southern (Ashqelon) and northern coasts (west of Rosh HaNikra Islands) on rocky bottoms at 12-28 m depth. Figure 29 New records. Israel Saurin, 1959).

Turbonilla funiculata de Folin, 1868
Remarks. Shells of Turbonilla funiculata are polymorphic with respect to their shape (Peñas and Rolán 2010 and our own observations) but the species can be readily distinguished from all pyramidellids in the Mediterranean -the most similar being the non-indigenous Turbonilla edgarii (Melvill, 1896), Turbonilla flaianoi Mazziotti, Agamennone, Tisselli, 2006 andPyrgulina fischeri Hornung &Mermod, 1925 -by the presence of a very marked subsutural constriction running along the whorls. This constriction separates the pronounced, almost orthocline axial ribs into a larger lower and a narrow upper, crown-like, portion. The ribs extend adapically beyond the suture of the preceding whorl, giving the transition zone between subsequent whorls a wavy appearance. The interspaces of the lower portion of the ribs bear several thin spiral lines, while those between the narrow upper parts of the ribs are smooth. The protoconch is helicoid and of type A.
Turbonilla funiculata has been previously reported from Fiji, Hong Kong, Indonesia, New Caledonia, Thailand, the Solomon Islands and Vietnam from shore to 396 m depth (Robba et al. 2004;Peñas and Rolán 2010). As Robba et al. (2004) already pointed out, the shell figured as Pyrgiscus microscopica (Laseron, 1959) by Okutani (2000: 712, fig. 68) is most likely T. funiculata, confirming that the species also occurs in Japan. This interpretation is re-affirmed by another illustration of the very same Japanese specimen in Mazziotti et al. (2005: 81, fig. 1m, n), showing the subsutural constriction characteristic of T. funiculata. Shells of T. funiculata were found by one of us (PILF) also among hitherto unidentified lots of shells from Pakistan and Sri Lanka housed in the LACM collection, demonstrating that this species also lives in the Indian Ocean. Here, we report the first records of T. funiculata for the Mediterranean, where several living specimens were collected in northern Israel on hard substrates at 12 m depth.

Family Cylichnidae H. Adams & A. Adams, 1854
Cylichna collyra Melvill, 1906    Remarks. We record here for the first time in the Mediterranean 13 living individuals of Cylichna collyra, a cephalaspidean originally described from the Gulf of Oman (Melvill 1906b). Cylichna collyra can be distinguished from the native Mediterranean C. cylindracea (Pennant, 1777) by its more elongated and slender shell, the more tapering apical part, the color pattern characterized by fine brown spiral lines apically and abapically, and the smaller size (C. cylindracea commonly reaches 1 cm in height whereas C. collyra attains approximately half that size). Cylichna villersii (Audouin, 1826), another non-indigenous species of Red Sea origin recorded from the Mediterranean coast of Israel (Bogi and Galil 2013a), is smaller (less than 2 mm), less slender, has a more rounded base and stronger growth marks (not visible in C. collyra), and bears two brown bands apically and abapically instead of the fine brown lines. Cylichna biplicata (A. Adams in Sowerby, 1850), a species occurring on the continental platform in the Indo-West Pacific, shares with our specimens the cylindrical shape and the color pattern of reddish-brown spiral bands apically and abapically (Valdés 2008), but is larger, more elongated anteriorly, with a stronger columellar tooth, and the colored spiral bands become a compact larger band apically. Cylichna collyra has not been recorded from the Red Sea yet (Dekker and Orlin 2000).
Remarks. This species was first recorded in the Mediterranean Sea in 1974 with the finding of few empty shells in the Bardawil Lagoons in Egypt (Mienis 1976). The species has since been recorded also in Greece , Turkey (Çinar et al. 2011), Cyprus (Katsanevakis et al. 2009), and Lebanon (Crocetta et al. 2020). A single record is available from Israel based on shells collected in 2004 off Palmachim (Bogi and Galil 2006). We here report living individuals from Israel for the first time. This is also the first finding of living individuals in the Mediterranean Sea.
Remarks. Atys angustatus was first recorded in the Mediterranean Sea in 1974, based on specimens collected at Haifa, Israel (Aartsen and Goud 2006). It has been reported from Mersin, Turkey, since 1986 (Aartsen and Goud 2006) and multiple records from Israel followed from several locations along its coast (Micali et al. 2016). To the best of our knowledge, this is the first record from Greece.  Remarks. This species was first recorded in Israel in 1960 from Bat Yam with living individuals (Barash and Danin 1973). Further living individuals were recorded in the 1960s and early 1970s from Israel and Egypt (Bardawil Lagoons) (Barash and Danin 1971, 1973, 1977. Since then, however, no further living individuals were found, questioning the persistence of its populations in the Mediterranean. We report here the recent finding of two living individuals, suggesting that the species indeed still occurs in Israel. Figure  Remarks. We here report Lioberus ligneus for the first time from Israel and Cyprus. The Israeli live-collected specimen comes from a patch of Galaxaura rugosa and Cystoseira sp. whereas the two empty but fresh shells from Cyprus were found attached vertically to Cystoseira shoots. These findings suggest that this species indeed prefers vegetated habitats in the shallow subtidal. This species has been previously reported from Lebanon based on shells collected in 1999-2000 (Crocetta et al. 2013), suggesting that it has likely occurred undetected throughout the Levantine Basin for long. The distinction of L. ligneus from the native Mediterranean Lioberus agglutinans (Cantraine, 1835) is not straightforward because both species share the elongated appearance, sculpture limited to concentric striae, and brown color. Moreover, they are morphologically variable. A fairly consistent character of L. ligneus in the samples we inspected from the Red Sea is the darker internal color, often shading into violet. The group would deserve a taxonomic revision to unambiguously distinguish its species.   Musculus costulatus (Risso, 1826): Israel • 9 spcms, 4 vv;Ashqelon;31.6868°N, 34.5516°E;depth 12 m;30 Apr. 2018; offshore rocky reef; suction sampler; HELM project (samples S12_1F, S12_1M, S12_1L, S12_2F, S12_2M) • 3 spcms; Ashqelon; 31.6891°N, 34.5257°E; depth 25 m; 2 May 2018; offshore rocky reef; suction sampler; HELM project (samples S16_1F, S16_2F) • 1 spcm; same collecting data as for preceding; depth 28 m; 31 Oct. 2018; HELM project (sample S59_1M) • 5 spcms; west of Rosh HaNikra Islands; 33.0704°N, 35.0926°E; depth 12 m; 1 May 2018; rocky substrate; suction sampler; HELM project (samples S14_1L, S14_3F, S14_3L, S14_4F) • 56 spcms; same collecting data as for preceding; 29 Oct. 2018; HELM project (samples S52_1F, S52_1M, S52_1L, S52_2F, S52_2M, S52_2L, S52_3F, S52_3M, S52_3L) • 1 v; west of Rosh HaNikra Islands; 33.0725°N, 35.0923°E; depth 20 m; 1 May 2018; rocky substrate; suction sampler; HELM project (sample S13_3F) • 20 spcms; same collecting data as for preceding; depth 19 m; 29 Oct. 2018; HELM project (sample S53_2L).

Lioberus ligneus (Reeve, 1858)
Remarks. We report numerous living individuals of Musculus aff. viridulus from the Mediterranean Israeli coastline. It is the first record of this Indo-Pacific species in the Mediterranean Sea. This species can be readily distinguished from the native M. costulatus because the latter has a more oval outline and a much smaller number of riblets at the same overall shell size ( Figure 36). These riblets are also much larger compared to M. aff. viridulus. The Atlanto-Mediterranean M. discors (Linnaeus, 1767) has a similarly fine posterior sculpture but at the same size is much higher and at all sizes bears much more prominent riblets anteriorly. The taxonomy of Musculus in the Indo-Pacific province is not settled and the available images of Red Sea M. viridulus (Oliver 1992;Zuschin and Oliver 2003) show a more oval species, hence our dubitative identification. Still, we are confident that this is a Red Sea species because we examined indistinguishable specimens from the northern Red Sea ( Figure 36F-H). Blatterer (2019) illustrated these and other similar specimens (plate 10, fig. 18a, b) as Gregariella ehrenbergi (Issel, 1869). We recorded a morphologically distinct species as G. ehrenbergi from a buoy stranded on the Israeli coastline (Steger et al. 2018;Ivkić et al. 2019). Gregariella ehrenbergi type material is corroded by Byne's disease and the original description likely refers to a juvenile specimen; the identity of this species deserves further scrutiny.  Remarks. The discrimination and identification of the species of Isognomon Lightfoot, 1786 is difficult due to their morphological plasticity that is related to their cryptic way of life. Still, the specimens recently reported from Astypalaia, in the Eastern Aegean Sea (Lipej et al. 2017;Angelidis and Polyzoulis 2018), show clear morphological differences from I. legumen (Gmelin, 1791), the established non-indigenous species in the Mediterranean Sea: the main features of the sculpture are radial rather than concentric and the shape of the shell can be very elongated rather than subquadrate. We here report a juvenile living individual and an empty shell from Greece and Cyprus (for which this is a new record), respectively, which are not distinguishable from those previously reported from Astypalaia. Juvenile shells (up to a size of ~ 5 mm) bear a sculpture of radial ribs adorned by tubular spines. We do not comment upon the choice of the australica name for this taxonomic entity by previous authors, lamenting the lack of a thorough revision of this genus in the Indo-Pacific province.  Remarks. Pegophysema philippiana was first found in the Mediterranean Sea in 2018 as a single valve from south of Tel Aviv (Mienis 2019). The specimen here reported is much smaller, and we acknowledge that the identification of juvenile individuals of this genus can only be tentative. We do not consider it conspecific to native species such as Loripinus fragilis (Philippi, 1836) because this latter species is much more inflated at this size, nor Loripes orbiculatus Poli, 1795 which has a different valve profile. If the identification is confirmed, this would be the first live collected specimen of P. philippiana in the Mediterranean Sea. New records. Israel • 1 spcm;Palmachim;31.9292°N, 34.6405°E;depth 36.9 m;29 May 2004;soft substrate;grab;NM project (station 19); size: L 3.2 mm, H 3.2 mm.
Remarks. We report a juvenile but living individual of Chavania erythraea, a lucinid occurring in the Red Sea, the Persian (Arabian) Gulf and the Arabian Sea (Glover and Taylor 2001). This is the first record of this species from the Mediterranean Sea. Adults develop a commarginal lamellar sculpture.   Zorina, 1978 (now in Rugalucina too) from the Mediterranean coast of Israel by Steger et al. (2018), who also illustrated it. A recent molecular phylogeny showed that R. vietnamica is distinct from R. angela and that the non-indigenous species in the Mediterranean belongs to this latter species, which occurs in the Red Sea and northwest Indian Ocean (Taylor and Glover 2019).  Remarks. This non-indigenous species has first been recorded in the Mediterranean Sea by Mifsud and Ovalis (2012) as Nudiscintilla cf. glabra Lützen and Nielsen, 2005, based on five living specimens collected at Yumurtalik, Adana (Turkey) in shallow water. Their tentative identification was primarily guided by the external morphology of the living animals, which had a smooth mantle surface. This feature is characteristic for the monotypic genus Nudiscintilla (hence the genus name), but unusual among scintilloid galeommatids in general. Although no observations on living individuals could be made by us, the shell morphology of our material well matches that of the specimen illustrated in Mifsud and Ovalis (2012: fig. 1), suggesting conspecificity. Our findings represent the first records of this species from Israel. However, the dentition of the right valve as seen in SEM images ( Figure 40C-E) clearly differs from that described by Lützen and Nielsen (2005) for Nudiscintilla: the latter has a single cardinal tooth in each valve and no lateral teeth. However, the studied right valve -the hinge of the single live-collected specimen was not examined to avoid damage -bears what appears to be two cardinal teeth ( Figure 40C) that are fused at their base ( Figure 40D, E), as well as a ridge posterior to the internal ligament which most likely is a lateral tooth. This ridge seems to correspond to the left of the two swellings indicated by a pair of arrows on the right hand side of Mifsud and Ovalis (2012: fig. 1e), while the right swelling might correspond to a narrow ridge visible also on the dorsal margin of our valve. Mifsud and Ovalis (2012) interpreted these features as aberrant shell growth, however, the presence of such ridges also in our right valve ( Figure 40D) speaks against this hypothesis. Furthermore, their living individuals had a small tentacle situated above the widely gaping anterior inhalant region (cf. Mifsud and Ovalis (2012: 8, fig. 2a), however, the illustration of N. glabra in Lützen and Nielsen (2005: 292, fig. 38a) shows a small tentacle in the posterior exhalant region of the reflected mantle. In the light of the poorly developed taxonomy and great species diversity of galeommatid bivalves in the Indo-Pacific, further observations on living specimens, thorough comparisons with the type material from Thailand and molecular analyses are required to definitely clarify the relationship of Mediterranean specimens with N. glabra. Figure 41 New records. Israel • 1 spcm; Haifa; depth 15 m; May 1999; biogenic sediment; B.S. Galil leg.; size: L 6.9 mm, H 5.2 mm.

Scintilla cf. violescens Kuroda & Taki, 1961
Remarks. The single shell found is trapezoid-oval (L:H ratio = 1.33), slightly higher posteriorly than anteriorly, translucid-white, and has a glossy external surface. The valves are narrowly gaping, more widely in their posterior part. The umbones are prosogyrate, pointed and submedian. The commarginal sculpture consists of fine growth lines that are slightly wavy posteriorly, as well as irregular growth marks. Flat radial ribs are present in the posterior part of the shell; they are visible, upon close examination, also on the inside of the valves in the form of shallow markings. The inner surface, particularly of the right valve, is spotted by blister-like markings ( Figure 41C, D). The hinge of the right valve bears a single cardinal tooth, bent towards the anterior, and an elongated posterior lateral tooth. The left valve has two cardinals, but the anterior one is broken off ( Figure 41G); a posterior lateral is present.
Lacking further material and observations on living individuals, which are of great diagnostic importance in galeommatids, we refrained from assigning a definitive specific name to our shell. However, the overall shape, hinge dentition and the presence of radial sculpture match well descriptions of Scintilla violescens Kuroda & Taki, 1961(Arakawa 1961Kuroda and Taki 1961), a species recorded from the intertidal and shallow subtidal of Thailand and Japan (Huber 2015). In contrast to our shell, however, Kuroda and Taki (1961) mention the presence of radial sculpture on the entire surface of the valves of their type material. While our Mediterranean shell is less elongated than the specimens of S. violescens illustrated by Okutani (2000), Lützen and Nielsen (2005), and Huber (2015), it is very similar in outline to the shell shown by Arakawa (1961: fig. 5B) (which was identified by T. Kuroda and I. Taki). Scintilla violescens appears to be variable also with respect to shell size and coloration: specimens from Thailand (maximum length = 10.5 mm, n = 12 spcms) all were considerably smaller than the > 15 mm-long Japanese type (Kuroda and Taki 1961) and had a whitish in-stead of pale violet color (Lützen and Nielsen 2005), like the Israeli shell. Considering this great plasticity in shell characters, and the differences in living animal morphology observed for Thai vs. Japanese specimens of S. violescens by Lützen and Nielsen (2005), the question arises whether more than one biological entity might be involved.
Irrespective of its unresolved specific affinity, the shell presented here clearly differs from all native Mediterranean galeommatids and thus cannot be confused; it can be easily distinguished from the non-indigneous Nudiscintilla cf. glabra (see above) by its less elongated shell, smaller L:H-ratio and, most notably, the presence of radial sculpture on the valves. Apart from the present shell, which was found in 1999 in Haifa Bay, we know of no other material. Remarks. Ervilia scaliola was first recorded from Turkey based on material collected in 2013 by Zenetos and Ovalis (2014) who correctly described the complex taxonomic status of this genus in the Indo-Pacific. We here record the species from Israel for the first time. The complete shell from Ashqelon ( Figure 42) is very fresh, and thus probably originates from an extant population.

Remarks.
We here report the finding of three beached valves of Iacra seychellarum in Kos, Greece. Iacra seychellarum can be readily distinguished from any Mediterranean semelid by its thick valves, large chondrophore and different sculpture in three zones: anterior with fine incised concentric lirations, median to posterior slope of fine incised oblique lines becoming strongly divaricate over the posterior slope, posterior area with closely spaced concentric incised lirations (Oliver 1992;Zuschin and Oliver 2003). It has been recorded from the northern Red Sea by Blatterer (2019) but has a broader distribution in the Indian Ocean (Oliver 1992). Because to our knowledge the finding has not been followed by others, we suggest that living individuals are needed to confirm the introduction and establishment of this species in the Mediterranean Sea. Figure 44 New records. Israel  Remarks. We were unable to identify this peculiar bivalve beyond family level, despite a thorough search in the literature on Mediterranean and Indo-Pacific mollusks. The shell of the single specimen found is roundly subtrigonal, fragile, with a rounded, steeply sloping anterior and a subacute to subtruncate posterior part; the ventral margin is slightly concave, the umbo submedian. The outer surface is smooth, glossy, and only sculptured by very fine growth lines. Although the small size and outline are reminiscent of certain galeommatoid genera such as Bornia Philippi, 1836, which is also represented in the Mediterranean Sea, the presence of both an external and an internal ligament, the latter situated on a well-developed resilifer, is typical for the family Semelidae (Oliver 1992;Beesley and Ross 1998;Huber 2010). The hinge of both valves bears two cardinals; the posterior cardinal of the left valve is becoming obsolete  by encroachment of the internal ligament portion which also extends vertically beyond the hinge line. Two well-developed laterals are present in the right valve, while in the left valve, only a weak tooth-like ridge is present anteriorly, formed by the dorsal shell margin. Due to the extremely smooth and glossy interior of the valves that renders hardly visible even the adductor muscle scars, it remains unclear whether a deep pallial sinus, another feature typical of semelids, is present in the studied specimen.

Semelidae sp.
Lonoa katoi Habe, 1976, a semelid from Japan, shares with our species the small size and irregular outline with an often concave ventral margin (related to its attached lifestyle); however, the outer shell surface of L. katoi bears rough lamellae and fine radial threads (Habe 1976;Okutani 2017), while that of the Israeli specimen is almost smooth. The most similar confamilial species from the Red Sea probably is Abra aegyptiaca Oliver and Zuschin, 2000, however, it differs from the specimen described here in shell shape, sculpture, the prosogyrate umbo, and features of the hinge such as the shape of the anterior lateral tooth of the right valve. Similar-sized juveniles of Mediterranean Abra spp., including A. alba, the most common species on the shallow Israeli shelf, have a hinge morphology comparable to our specimen, but differ in their outline and by having more protruding umbos (Scaperrotta et al. 2013). Abra tenuis is most similar in shape, and a teratological specimen might approach the outline of our shell; however, such a specimen would still differ from Semelidae sp. by the presence of commarginal lines on the early dissoconch (Scaperrotta et al. 2013;Oliver et al. 2020).
Until the finding of further specimens, it remains open whether the present individual is a juvenile or an adult of a small-sized species. Considering that only a single specimen was found so far, the lack of known native Mediterranean species with a similar morphology, and the geographical proximity of the Israeli coast to the Suez Canal, we suspect that this species might be another Indo-Pacific taxon introduced to the southeastern Mediterranean.
Remarks. Clementia papyracea has been recorded in the Mediterranean Sea since 1937, but findings of living individuals are very scarce and limited to samples collected in 1968 in El Arish, Egypt (Barash and Danin 1973), in 1975in Haifa, Israel (Barash and Danin 1977 and in 2012 in Ashqelon, Israel (Crocetta et al. 2016). We here report a further living juvenile individual.
Corbula gibba ( Remarks. Corbula erythraeensis is widespread in the northern Red Sea where it has been recorded from the gulfs of Suez (MacAndrew 1870;Oliver 1992) and Aqaba at Eilat (Edelman-Furstenberg and Faershtein 2010), as well as the northern Bay of Safaga (Egypt) (Zuschin and Oliver 2003); outside the Red Sea, its distribution ranges eastward to Pakistan (Huber 2010). Here we report the first findings of this species from the Mediterranean. Corbula erythraeensis was found sympatrically with the common native Mediterranean Corbula gibba but was always present in very low numbers. No empty shells have been found so far. While being similar in appearance to the morphologically variable C. gibba, C. erythraeensis has a convex anterior dorsal margin of the right valve (usually concave in C. gibba, particularly in smaller specimens), a more inflated umbonal region, a regular concentric sculpture on the left valve (Oliver 1992), and its color always is whitish-yellowish (C. gibba frequently has a rosy pattern). Juvenile individuals are more wedge-shaped than those of C. gibba.

Massive reporting follows intensive fieldwork and broad cooperation
We have covered 52 species, reporting the finding of 23 new Lessepsian mollusks, nine additional species that, upon final identification, may turn out to be further new Lessepsian species, nine new records for Eastern Mediterranean countries and new data for eleven already recognized non-indigenous species. Such a massive report is derived from three characteristics of this study which translate into recommendations for an effective approach to non-indigenous species detection and monitoring. First, the intensive sampling effort and the effective sampling techniques of the "Historical ecology of Lessepsian migration" (HELM) project and of the IOLR monitoring programs. The HELM project in particular targeted also hard substrates, poorly explored at this taxonomic resolution in the Eastern Mediterranean, by suction sampling. This technique has repeatedly proved to be a very effective method on compact (e.g., coral rubble, pebbles (Bouchet et al. 2002;Linnane et al. 2003;Ringvold et al. 2015;Evans et al. 2018)), seagrass (Bonfitto et al. 1998;Albano and Sabelli 2012;Albano and Stockinger 2019) or hard substrates (Templado et al. 2010) and has enabled here the collection of vast amounts of living micromollusks and their shells. Indeed, 26 out of the 52 (50%) species treated here and 17 out of 32 (53%) new and potentially new NIS came from suction samples on hard substrates. The IOLR surveys covered soft substrates along the whole Israeli Mediterranean coastline with a dense station network and multi-seasonal sampling that led to the detection of 16 out of 52 (31%) treated species, notwithstanding several findings had already been published (e.g., Galil 2013b, Aartsen et al. 2015;Lubinevsky et al. 2018). Second, the use of fine mesh sizes. The HELM samples were sieved with a 0.5 mm mesh and those of the IOLR monitoring programs were sieved with either a 0.5 mm or even a 250 μm mesh. Such small sizes enabled retaining the large majority of invertebrates, including mollusks, even those with very small and elongated shells, such as many Pyramidellidae. However, this approach requires an enormous effort when picking and sorting the samples and the availability of high-level taxonomic expertise, since most small-sized species belong to taxonomically challenging groups or represent juvenile individuals. Third, and importantly, cooperation among institutions and individuals. Despite new records of non-indigenous species are often scattered into short papers in the literature and may become difficult to trace in the long term, recent efforts have demonstrated the value of cooperation to build up large datasets (e.g., Katsanevakis et al. 2020). It is also important to highlight the role that citizen scientists had in the detection of the new non-indigenous species treated here by contributing to sample sorting, species identification and their taxonomic study.

The challenge of recognizing and identifying tropical non-indigenous species
Taxonomic uncertainty is recognized as a major impediment to the reliable inventorying of non-indigenous species (McGeoch et al. 2012;Marchini et al. 2015;Katsanevakis and Moustakas 2018). Species whose identification is uncertain or whose taxonomic status is unresolved were suggested to be excluded from inventories . Taxonomic uncertainty may also imply an uncertain non-indigenous status: a species morphologically distinct from the native species pool can either be a yet undescribed native species or a newly introduced species (which may be undescribed too). If the new species belongs to a clade not occurring in the sampled range, then the attribution of the non-indigenous status is well supported. Still, only the finding of clearly conspecific individuals from the source pool would provide final evidence of the non-indigenous status. We exemplified these cases with two new taxa described here: Coriophora lessepsiana Albano, Bakker & Sabelli, sp. nov. clearly belongs to a tropical clade of Triphoridae and the availability of material from the Red Sea enabled the unambiguous attribution of the species to the Red Sea pool (hence the species name lessepsiana). Joculator problematicus Albano & Steger, sp. nov. belongs to a genus absent from the Mediterranean Sea. Despite multiple similar species have been recorded in the Indo-Pacific province, we have not been able to find conspecific Red Sea material (hence the species name problematicus, to highlight the uncertainty in attributing the non-indigenous status when solid taxonomic and faunistic knowledge is lacking).
A more complex case is represented by species that have few diagnostic characters, hampering the unequivocal attribution to a tropical clade (e.g., within Eulimidae and Pyramidellidae). We here propose to exploit the properties of death assemblages to deliver a solid hypothesis of non-indigenous status. Death assemblages, the taxonomically identifiable, dead or discarded organic remains encountered in a landscape or seabed such as molluscan shells, accumulate species richness over time (Kidwell 2013). This property implies that death assemblages are good archives of local species diversity and that species which have long occurred in a study site (like native ones) are likely to be found in a death assemblage even if they are too rare to be regularly detected in living censuses. Indeed, the inclusion of empty shells in field surveys increased the estimations of occupancy and detectability of native land snails (Albano et al. 2015). In contrast, a species which has recently established a population in a new area would be expected to be poorly detectable in the death assemblage because not enough time has elapsed to contribute a significant number of skeletal parts to it. Consequently, newly reported non-indigenous species found alive but not, or very rarely, in the death assemblage may represent newly introduced species. This approach can be applied when living and death assemblages are sampled simultaneously with quantitative methods. The main limitation of this approach is that it is applicable only to organisms with hard skeletal parts like foraminiferans, ostracods and fishes (e.g. Agiadi and Albano 2020).
With biological invasions constituting a major element of global change, there is understandable concern on the accuracy of non-indigenous species inventories Katsanevakis and Moustakas 2018). However, also the risk of underestimating the magnitude of biological invasions must be considered. In this respect, the Lessepsian invasion is a special case. The salinity reduction of the Bitter Lakes (Galil 2006), the containment by dams of the annual Nile flood which used to modulate salinity off its delta (Rilov and Galil 2009), and the multiple enlargements of the Suez Canal  have certainly enhanced connectivity between the Mediterranean Sea and the Indo-Pacific province during the 20 th century. Additionally, in the most recent decades, rising seawater temperatures in the Mediterranean Sea (Ozer et al. 2017) make it increasingly suitable for the establishment of inherently thermophilic Red Sea species. These factors combined render very likely that an increasing number of Red Sea species manages to cross the Canal and to settle in the Mediterranean Sea. The proper understanding of this phenomenon and of its drivers requires a timely and detailed census of assemblages as well as a broad toolkit to overcome the limitations of the taxonomic challenges associated with tropical species.

Authors' contributions
PGA and JS conceived the study. PGA, JS, CB, MB, TGH, MFH, HL, MM, and MS contributed to fieldwork, sample sorting and data acquisition. PGA, JS, PAJB, CB, PILF, and BS identified the specimens and contributed to the taxonomic discussion. PGA, JS, and TGH contributed to the discussion. PGA, JS, PAJB, and MA prepared the figures. PGA and JS wrote the first draft of the manuscript, which then received contributions by all co-authors.

Funding
This research has been conducted in the context of the "Historical ecology of Lessepsian migration" project funded by the Austrian Science Fund (FWF) P28983-B29 (PI: P.G. Albano). Sampling in Crete was supported by the "Kurzfristige wissenschaftliche Auslandsstipendien" program of the University of Vienna to M. Stockinger, the non-profit organization Mare Mundi, and the diving school Dive2gether. Sampling in northeastern Cyprus was conducted in the framework of the project "Classification of coastal habitats of Northern Cyprus according to EUNIS protocol" (PI: F. Huseynoglu). The Faculty of Earth Sciences of the University of Vienna funded a citizen science project conducted in October 2019. MB was supported by an Ernst Mach fellowship of the OeAD (Österreichischer Austauschdienst).