As a part of the routine neritic zooplankton collection program in Obhur Creek (central Red Sea, Saudi Arabia), specimens of a pontellid calanoid copepod, Calanopia media Gurney, 1927, were observed and studied. Since the original description was rather brief and drawings limited, especially of mouthparts and legs, which were not illustrated and described, the species is here fully redescribed. Red Sea specimens showed considerable variation in the female genital compound somite, the right caudal ramus and leg 5, as well as in the presence of a medial knob ventrally on the male right prosomal corner. DNA sequences of mtCOI of different specimens did not show any significant differences and supported their identity as one species. Calanopia media exhibited clear diel vertical migration, with high densities of 106 and 150 ind. m^-3 during sunset (6:00 pm; UTC+3) and midnight (12:00 am; UTC+3) collections, respectively. However, this species was not observed in other zooplankton collections from the surface to 20 m depth, at 6:00 am and 12:00 pm (UTC+3).

Calanopia media, copepods, pontellid, redescription, Red Sea

The pontellid (Calanoida) fauna of the Red Sea contains a surprisingly low proportion of the Indo-Pacific fauna, from which it is apparently derived (El-Sherbiny and Ueda 2008; El-Sherbiny 2009). Five genera and 17 pontellid species have been recorded from the Red Sea (Al-Aidaroos et al. 2019; El-Sherbiny and Al-Aidaroos 2017; Razouls et al. 2019), whereas 77 pontellid species are present in the greater Indian Ocean (Razouls et al. 2019). Among the Red Sea pontellids, there are six species of Calanopia (Al-Aidaroos et al. 2016; El-Sherbiny and Al-Aidaroos 2017), namely: C. elliptica Dana, 1849, C. minor A. Scott, 1902, C. media Gurney, 1927, C. kideysi Ünal & Shmeleva, 2002, C. thompsoni A. Scott, 1909 and C. tulina El-Sherbiny & Al-Aidaroos, 2017. The original description of C. media is rather brief, does not include the cephalic and thoracic limbs, and illustrations are confined to habitus drawings of the female and male in dorsal view, the female urosome, the male right antennule ancestral segments 19 and 20 and the male and female leg 5.

During the examination of plankton, samples collected from the Saudi Arabian waters of the central Red Sea, in Obhur Creek, Jeddah, specimens of what we provisionally called C. media were observed. These specimens differ in some respects from Gurney’s (1927) original description. Here, we provide a full description of the species and an account of variability among the Obhur Creek specimens. Comparison with the type specimens held in the Natural History Museum, London (BMNH 1926.2.16.69-88) was carried out, and information on the vertical migration in the water column was provided. In addition, the mitochondrial COI gene of some species from the Red Sea was sequenced and compared with the sequences available in GenBank (NCBI) to determine their affinity.

Zooplankton samples were collected from Obhur Creek (21°42'32.23"N, 39°5'41.56"E) using a 50 cm diameter plankton net of 150 µm mesh size, towed near the surface for 10 minutes at a speed of about 1–1.5 knots and vertically from 20m depth to the surface on 21 January 2016 at 7:00am, 12:00pm, 6:00pm and 12:00am (UTC+3) local time (sunrise at 7:04am and sunset at 6:05pm; UTC+3). A flowmeter (Hydrobios) was attached to the net mouth for estimating the volume of water filtered. Samples were fixed immediately in 95% alcohol. Subsequently, Calanopia media specimens were picked from zooplankton samples collected at midnight. For microscopic examination, specimens were dissected in lactic acid and were observed using bright-field and differential interference microscopy (Nikon DM 6000). Drawings and measurements were made with a camera lucida attached to the microscope and an ocular micrometer. Morphological terminology follows Huys and Boxshall (1991), except for maxillary and maxillipedal appendages, which follow Ferrari and Ivanenko (2008). For scanning electron microscopy, Calanopia specimens were washed three times in filtered seawater and clean distilled water, then dehydrated through a 30–100% ethanol series and dried with hexamethyldisilazane. The specimens were mounted on a stub, coated with gold palladium, and observed with a SEM FEI-QUANTA 250.

For genetic analysis, four intact female specimens of C. media, three of C. minor, one of C. elliptica and two of C. thompsoni (after accurate morphological identification) were sorted out and the genomic DNA was extracted from individual specimens. A portion of the mitochondrial gene cytochrome oxidase subunit I (mtCOI) was amplified using the universal primers of Folmer et al. (1994). Individual copepods were digested in 400 μl ATL buffer (Qiagen) and 20 μl Proteinase K overnight, in a capped 0.2ml microcentrifuge tube. After digestion, 400 μl of AL buffer was added and DNA extraction continued using Qiagen’s Blood and Tissue kit as per the manufacturer’s instructions. DNA was precipitated in 30 μl AE buffer and mtCOI amplicons were amplified using the PCR primers LCO1490 and HCO2198 (Folmer et al. 1994). The reaction conditions were initial denaturation for 5 min at 95 °C followed by 40 cycles of 94 °C (1 min); 47 °C (2 min); 72 °C (3 min). A final extension at 72 °C for 10 min was undertaken. PCR products were purified using ExoStar (Illustra) and sequencing was carried out in an ABI 3730×l Capillary Sequencer. The machine-read sequences were compiled using Sequencing Analysis (Ver. 3.3, ABI prism) and manually checked for accuracy. Available sequences of C. thompsoni were obtained from the NCBI database for comparison. Pairwise distance measures and phylogenetic analyses were conducted using the MEGA X software (Kumar et al. 2018). Ambiguous sites were eliminated from the dataset.

In his original description of C. media, Gurney (1927) provided drawings only for the whole body and leg 5 of both sexes, as well as of the female urosome and the geniculate part of the male right antennule (segments XX and XXI), and compared it with C. elliptica. His description was brief and focused mainly on the female genital compound somite and female leg 5. For the male, he only mentioned that the major differences between C. media and C. elliptica lay in a lack of serrations on the second exopodal segment of right leg 5. In his words, “Fifth legs closely resembling those of C. elliptica, but without the scalloped edge to the broad subterminal joint of the right leg”. Re-examination of C. media from the study area allowed us to more accurately describe this species. The original characters reported by Gurney (1927) are given in brackets: 1) the female genital compound carries one or two spinules on the right side (2 spinules); 2) the female caudal rami are asymmetrical, the right one is expanding anteromedially (symmetrical); 3) the female leg 5 is slightly asymmetrical, with the left leg slightly shorter than the right due to shortness of both the basis and first exopodal segment (mentioned as symmetrical by Gurney, but from his drawings the first exopodal segment is longer on the right side and also the right second exopod segment is longer than the left one); 4) the male prosomal corners project posterolaterally into asymmetrical pointed processes, the right one of which is slightly longer than the left, with a distinct ventral knob on its medial margin, invisible in dorsal view (symmetrical); and 5) the second exopod segment of the male right leg 5 carries one seta medially, which is lacking in Gurney’s description. Moreover, Pesta (1941), in his description from the southern Red Sea, mentioned the lack of spinules on the genital compound somite, while the female leg 5 exopod proportions differed slightly from Gurney’s description (in Pesta’s description, the second exopod segment is 1.1 times as long as the first exopod segment) and the second distal spine on the second exopod segment is shorter than in Gurney’s description.

In addition to the variability noted in the caudal rami of C. media, variation has been reported in C. sewelli Jones & Park, 1967, collected from Marquesas, central Pacific, in which the right ramus was sometimes longer and with a concave medial margin. Moreover, asymmetry of the female caudal rami has been noted in C. asymmetrica Mulyadi & Ueda, 1996 collected from Indonesian waters, where the right ramus was much longer than the left one and expanded posteriorly. This asymmetry was found also in C. australica Bayly & Greenwood, 1966 collected from Moreton Bay, Australia, in which the left ramus was longer than the right. Asymmetry of the male prosomal points in C. media is similar to that in C. sarsi C.B. Wilson, 1950 collected from Fiji waters and C. tulina (El-Sherbiny and Al-Aidaroos 2017) described from the Red Sea.

Calanopia media is closely related to C. tulina from the central Red Sea, but they can be distinguished from each other by the characters listed in Table 2. The most distinctive characters of C. media are: 1) the presence of 2 ventral spinules on the right side of the female genital compound somite, 2) the extreme asymmetry of the caudal rami of the female, 3) the presence of 2 long articulated and 1 short fused terminal spines on the second exopodal segment of the male left leg 5, 4) the first exopodal segment of male right leg 5 with a small thumb located approximately at one-third of the segment length and with the semicircular processes on a flap on its lateral part, and 5) the second exopodal segment of the male right leg 5 shorter than the first exopodal segment, which is curved at mid-length with a blunt apex and 2 medial setae within a shallow medial depression.

In our study, C. media exhibited a clear diel vertical migration (DVM). Sunset ascend and sunrise descent were performed at very low light intensities. This pattern is known as the nocturnal or normal DVM (Forward 1988). Similar observations were made in previous studies in the southern part of the Suez Canal and the southern Red Sea by Gurney (1927) and Pesta (1941), respectively, who reported that C. media was found in high abundance in the coastal waters mainly in the night samples. This pattern was recorded also for C. americana F. Dahl, 1894 in the eastern Gulf of Mexico (Turner et al. 1979) and in the Brazilian waters (Pessoa et al. 2014). This DVM may be performed to avoid UV radiation, light and predators, as well as for availability of food and favorable temperature (Bollens and Frost 1989; Lampert 1989; Andersen and Nival 1991; Pearre 2003; Cohen and Forward 2005). Generally, the light intensity in Obhur Creek drops steeply below 5 m depth, and very little light penetrates into waters below 15 m (Al-Aidaroos et al., unpublished data). Following the same trend, UVB radiation decreased by 74.15% at 5 m depth, while only 2% reached 15 m depth (Duarte et al., unpublished data). This suggests that C. media is avoiding high temperature, high light intensity and/or visual predation during the day and coming up to the surface layers for feeding during night-time. This species was observed to be abundant in night collections by light trap from a shallow coral reef in the central Red Sea (unpublished data). This might be attributed to its benthopelagic behavior, which is clearly visible in terms of the stout, large outer spines on its legs (1–4), as reported in the benthopelagic copepod genus Platycopia (Ohtsuka et al. 1998).

Morphology can be considered as a fundamental method for copepod species identification. However, some pontellid species display considerable intraspecific morphological variations in the female genital compound somite, caudal rami and fifth leg (Jeong et al. 2014). Such differences might be sufficient to prove the presence of a new species, as it is the case for some centropagoid species (e.g. Sakaguchi and Ueda 2010; Soh et al. 2012). The mtCOI gene was proposed as a unique tool for copepod identification with ‘barcoding’ (Bradford et al. 2010; Blanco-Bercial et al. 2014), which can verify the species identity within morphologically varying pontellid specimens. Within Crustacea, the level of genetic variation between congeneric species is 17.16%, while the level of intraspecific variability is 0.46% (Costa et al. 2007). Moreover, variation between calanoid copepod species varied between 13–22%, 17.6–26.7% and 21–23% in previous studies by Bucklin et al. (1999), Eyun et al. (2007) and Soh et al. (2013), respectively. In this study, the intraspecific variation in the COI sequences from the Red Sea specimens of C. media, C. minor and C. thompsoni would confirm the hypothesis of Costa et al. (2007). The Calanopia species nucleotide sequences collected from GenBank indicate no genetic differentiation between the specimens of C. thompsoni from the Red Sea and the Indian Ocean. In contrast, the nucleotide data of C. thompsoni from the Red Sea revealed considerable variations with specimens collected from the China seas, indicating that the two populations are genetically different. This high level of deviation between both populations can by supported by the allopatric speciation hypothesis of Carpenter and Springer (2005), which states that in the Pleistocene the migration of marine organisms from the West Pacific Ocean to the Indian Ocean seems to be blocked by an ecological vicariant. Moreover, sequences included herein from the Red Sea are the first barcodes for these species, and it will be useful in future pontellid barcode studies.

During the last two decades, six species of pontellids have been originally described as new species or were first recorded from the Red Sea: Calanopia kideysi by Ünal and Shmeleva (2002), Labidocera boxshalli El-Sherbiny & Ueda, 2010 by El-Sherbiny and Ueda (2010), Calanopia tulina by El-Sherbiny and Al-Aidaroos (2017), Pontella princeps Dana, 1849 by El-Sherbiny (2009), Labidocera karachiensis Fazal-Ur-Rehman, 1973 by El-Sherbiny and Ueda (2008), and Calanopia thompsoni by Al-Aidaroos et al. (2016). However, the diversity of this family in the Red Sea is rather low (17 species) compared with that of the Indian Ocean (77 species as reported by Razouls et al. 2019), from which the Red Sea fauna has originated (Al-Aidaroos et al. 2019). This low number can be attributed to the characteristic euneustonic or facultative neustonic nature of this group (Mauchline 1998). Thus, inappropriate sampling methods and sampling time and/or limited sampling effort might have resulted in an underestimation of the fauna of Pontellidae. Therefore, this study emphasizes the necessity of understanding the diversity and distribution of pontellid copepods in the Red Sea and their mode of life.

This project was funded by the Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah, under grant No. (DF-400-150-1441). The authors, therefore, gratefully acknowledge DSR technical and financial support. The authors also thank Prof. Janet Bradford-Grieve for her critical comments on an earlier draft. Thanks to Miranda Lowe, the curator of crustaceans, Natural History Museum, London, UK, for loaning us the paratype material. Many thanks also to Dr. Stamatina Isari for assistance and help with sequencing and to Dr. Benjamin Kurten for the translation of Pesta’s paper.