A new subfamily classification of the highly diversified Dorippidae H. Milne Edwards, 1837 (Crustacea, Decapoda, Brachyura, Dorippoidea), using morphological, molecular and palaeotonlogical data, with special emphasis on its unique female reproductive system Author Guinot, Danièle Institut de Systématique, Évolution, Biodiversité (ISYEB), Muséum national d’Histoire naturelle, CNRS, Sorbonne Université, EPHE, Université des Antilles, case postale 53, 57 rue Cuvier, F- 75231 Paris cedex 05 (France) daniele. guinot @ mnhn. fr Dedicated to the memory of my colleague and dearest friend Ngan Kee NG (1966 - 2022) guinot@mnhn.fr text Zoosystema 2023 2023-06-05 45 9 225 372 journal article 10.5252/zoosystema2023v45a9 1638-9387 8071253 urn:lsid:zoobank.org:pub:69C34731-8C25-4A1E-B336-B222CD3CBAC3 REMARKS ABOUT THE VALIDITY OF MEDORIPPE CROSNIERI CHEN , 1988 A few years ago around 2010-2012, we had a project with J. C. Y. Lai (ZRC) to revise the status of Medorippe crosnieri Chen, 1988 by re-examining the Malagasy Chen’s type material: holotype : MNHN-IU-2009-1995 (= MNHN-B18269); paratypes : MNHN-IU-2009-1996 (= MNHN-B18269) and MNHN-IU-2009-1997 (= MNHN-B18365). The main issue was to compare it with specimens in the MNHN and ZRC collections assigned to M. lanata but collected outside of its typical occurrence (the Mediterranean Sea and the West African coast), for example in the Mozambique Channel and in South Africa , by also including records in the literature. To test the hypothesis based on morphological characters that M. crosnieri could be a valid species, molecular analyses of COI gene of dorippids from West Africa and the Indian Ocean were undertaken. Unfortunately, this study could not be carried out due to inconclusive gene sequence results, awaiting further molecular analyses. The data summarised below are from an advanced draft in collaboration with J. C. Y Lai but is currently on hold. A second species of Medorippe , M. crosnieri , from the north-west coast of Madagascar , was established by Chen (1988: 681 , fig. 2, pl. 1D, E, table 2; paper often erroneously reported from 1987), based mainly on the smooth dorsal margins of P2 and P3 meri, as opposed to the spinulated margins in M. lanata . But M. crosnieri was immediately synonymised with M. lanata by Holthuis & Manning (1990: 89, 93), who argued that Chen’s description was “based upon juveniles” and who reproduced (1990: fig. 38) all Chen’s (1988) figures as M. lanata . Examination of the Chen’s material, deposited in the MNHN, shows that the male holotype MNHN-IU-2009-1995 of M. crosnieri is a pre-adult with full-grown but slender gonopods and still symmetrical chelipeds; the two young paratype males MNHN-IU-2009-1997 (instead of females as mentioned by Chen [1988] ) both have undifferentiated gonopods; another paratype MNHN-IU-2009-1996 is an immature female without opened vulvae. All these specimens have cylindrical, slender and rather long P2 and P3 meri, and evidently smooth ( Chen 1988 : fig. 2c, pl. 1D, E). While the stout G1 ending in triangular apex in the Chen’s species is similar to that of M. lanata , the main distinguishing feature of M. crosnieri was related to the smooth dorsal margins of the P2 and P3 meri, except for a few low teeth on the basal part of the P2 merus and one or two on the P3 merus ( Fig. 23A, B, D ), instead of the distinctly spinulated meri of M. lanata ( Figs 22A, B ; 23C ). Additional differences were: in M. crosnieri body with short thin hairs (versus with club hairs in M. lanata ); carapace surface with less prominent granules and tubercles in M. crosnieri than in M. lanata ; and the surface of the male pleon not densely covered with short hairs (versus densely covered with rather long hairs in M. lanata ). Two specimens of Medorippe from the Mozambique Channel were collected by the MAINBAZA Expedition in 2009 (MNHN-IU-2016-1335 and MNHN-IU-2016-1336), a male 21.1 × 25.9 mm with dimorphic chelipeds and a young male 12.0 × 15.0 mm with symmetrical chelipeds, both having smooth meri on P2 and P3, which are long and cylindrical in the young ( Fig.23B ), shorter and more robust in the adult ( Fig. 23A ). Both are related to M. crosnieri by their smooth P2 and P3 meri. Our Figure 23C, D seems to show that the merus is proportionally more elongated in M. lanata than in M. crosnieri but this criterion must be used with prudence as the size of the merus varies with age, becoming wider and stouter when the individual grows. It seemed possible that this material from Mozambique belonged to M. crosnieri and that the refutation by Holthuis & Manning (1990) of the characters used by Chen (1988) to distinguish it from M. lanata needs to be re-evaluated. It should be noted that Holthuis & Manning (1990: 89, fig. 37b) assigned to M. lanata a male from Mozambique and several South African specimens, the latter being figured with spinulated meri. Chen (1988) added as possible synonyms of her M. crosnieri some Dorippe lanata of the literature, including that of Barnard (1950: 389 , fig. 73d; 1955: 4) from South Africa (Natal). The material from Spanish Sahara , Sierra Leone , Gabon , Cabinda and Congo reported by Capart (1951b : fig. 6, as Dorippe lanata ) has spines on the P2, P3 meri, a V-shaped cardiac region, and a G1 distally curved at right angle. As the southern limit of Medorippe lanata along the West African coast includes Angola with a reasonable certainty ( Crosnier 1970 ), it is relevant to consider the material recorded in South Atlantic waters. Barnard (1950: 389) does indicate that P2 and P3 have spinulose meri. We have examined a few specimens from South Africa (ZRC 2012.0158 and MNHN-IU-2016-1336): they have spinulose meri and are most likely M. lanata . Dorippoides nudipes Manning & Holthuis, 1986 ( Manning & Holthuis 1986: 364 , fig. 1c; Holthuis & Manning 1990: 66, fig. 26) is another dorippid with unarmed P2 and P3 meri, which occurs in South Africa (see under Dorippoidinae n. subfam. ): it cannot be confused with a Medorippe species on account of its short G1 ending in horn-coloured projection with a twisted whip-like appendage, the carapace margin lacking epibranchial spine and the cardiac region without a V-shaped ridge ( Fig. 16C, D ). Similarly, Medorippe crosnieri cannot be confused with the West-African Phyllodorippe armata , known from the Spanish Sahara to Angola and never recorded further south, which also has unarmed P2 and P3 meri ( Fig. 29A, B ) and a slender, S-shaped G1 ending in a prominent subdistal lobe ( Fig. 30C, D ) ( Monod 1933b : fig. 3H; Manning & Holthuis 1981 : fig. 4k, l), contrasting with the typical short, stout G1 of Medorippe , abruptly turned outward and ending in a long, sharp setiferous apex ( Chen 1988 : fig. 2f; reproduced by Holthuis & Manning 1990: fig. 38g ). It proved difficult to conclude the value of the small differences reported by Chen (1988 : table 2) due to insufficient material in hand assignable to M. crosnieri . In addition, examination of M. lanata revealed a range of morphological variation (variations that also exist in the Mediterranean form), including spinulation of meri P2 and P3 that was more pronounced in smaller individuals and tended to decrease in larger adult males, particularly in the proximal half of the merus. Spinulation may also occasionally be missing completely, e.g. in a sample from the north of mouth of Congo River, MNHN-IU-2009-2001 (= MNHN-B13583), a large male 26.0 × 30.0 mm (with markedly asymmetrical chelae) that has abnormally smooth meri P2, P3; in contrast, another adult male of 28.2 mm cw (with less asymmetrical chelas) has spinulated meri, both specimens showing the typical G1 of Medorippe . For more details on the spinulation of P2, P3 meri of M. lanata , see Monod (1933b : fig.3C, D; 1956: 84, fig.103, as D. armata ); Manning & Holthuis (1981: 32 , figs 4a, c-f, table 1) and Holthuis & Manning (1990: 89, figs 36, 37). An ovigerous female collected in southern Madagascar - thus in proximity to Chen’s (1988) type series – by the ATIMO VATAE expedition in 2010 (MNHN-IU-2010-4307), which we had not previously examined, shows smooth meri P2, P3, which undermines Holthuis & Manning’s (1990) argument that the smoothness of P2 and P3 is a juvenile character, and leads us to believe that M. crosnieri is a valid species. Molecular analyses based on the COI carried on by J. C. Y. Lai were inconclusive, showing that the M. lanata from South Africa , with spinulated meri, was genetically almost identical to the M. crosnieri of the MAINBAZA Expedition in 2009, with smooth meri. The genetic evidence supported that M. lanata and M. crosnieri were closely related with little divergence (<3% divergence), insufficient to revalidate M. crosnieri , so the planned paper was not published, pending further research.