Macrobrachium suphanense, Saengphan & Panijpan & Senapin & Laosinchai & Ruenwongsa & Suksomnit & Phiwsaiya, 2018

Saengphan, Nukul, Panijpan, Bhinyo, Senapin, Saengchan, Laosinchai, Parames, Ruenwongsa, Pintip, Suksomnit, Auaree & Phiwsaiya, Kornsunee, 2018, Morphology and molecular phylogeny of Macrobrachium suphanense sp. nov. (Decapoda: Palaemonidae) from Thailand, Zootaxa 4482 (1), pp. 151-163 : 154-161

Macrobrachium suphanense , sp. nov.

Figs. (2–6)

Material examined. Holotype: male (cl 8.5 mm), Nikhom Krasiao subdistrict, Dan Chang district , Suphan Buri Province (14°49′50.3″ N, 99°34′52.5″ E), 28 February 2013 . Paratypes: 11 males (cl 2.5–7 mm), same location and date as holotype . Others: 15 males (cl 5–8 mm) and 8 females (cl 7–8.6 mm), Don Chedi subdistrict, Don Chedi district , Suphan Buri Province (14o37′37.7″ N, 100o0′43.2″ E), 21 March 2013 ; 15 males (cl 5–8 mm) and 8 females (cl 7–8.6 mm), Pho Muang Phan subdistrict, Samko district , Ang Thong Province (14o35′19.6″ N, 100o15′38.4″ E), 21 March 2013 ; 2 males (cl 9 and 11 mm), Chorakhe Sam Phan subdistrict, U Thong district , Suphan Buri Province (14o19′42.4″ N, 99o51′58.3″ E), 7 June 2016 ; 15 males (cl 5–8 mm) and 8 females (cl 7–8.6 mm), Hua Wiang subdistrict, Sena district , Ayutthaya Province (14o22′13.2″ N, 99o24′51.0″ E), 1 July 2012 ; 20 males (cl 3–6.5 mm) and 24 females (cl 4–8.2 mm), Srinagarindra dam, Si Sawat district , Kanchanaburi Province (14o35′50.1″ N, 99o6′33.0″ E), 13 December 2014.

Largest male: total length of 42 mm, carapace length 10 mm.

Ovigerous female. Largest: total length of 42 mm, carapace length 12.2 mm; smallest: total length of 27 mm, carapace 7.2 mm.

Description. Fully grown male (fig. 2 A): Rostrum: straight, elevating over eye sockets bending downward gradually to straight tip, extending beyond or as long as the antennular peduncle (but not reaching the end of scaphocerite), length about three quarters of carapace; upper part having 9–12 teeth, 2 teeth behind the eye socket (post orbital margin); lower part of rostrum having 3–4 teeth; distance between the first posterior tooth and the second tooth being longer than distances between the adjacent teeth in the middle (fig. 3).

Antennal spine: well developed with pointed end and situated below orbital angle (suborbital angle); hepatic spine of size similar to that of antennal spine and located slightly behind and lower than antennal spine; paired hepatic spines parallel to antennal spines situated between upper first and second teeth of rostrum on carapace.

Carapace: smooth with tiny protruding "spines" sparsely scattered but more densely populated at anterior and ventral region; thoracic sternite of fourth segment having conical median process; tiny spinules found on shell of abdominal pleurae; spinules sparsely scattered near lower edges of segments no. 3, 4, and 5; sternite no. 6 shell 1.8 times as long as that of sternite no. 5; upper telson (fig. 4) segment smooth and 1.44 times longer than segment no. 6 with 4 dorsal unpaired spines (paired ones in paratype) at the tail end; tail tapering toward the rear with a pointed end (median spine), each side flanked by 2 spines, outer lateral spine and the inner intermediate spine; several pairs of plumose setae found between the two flanking spines; lateral spines having similar size to dorsal spines; welldeveloped intermediate spines 3 times the length of median spine; abdominal sternites no. 1–3 having transverse ridges with median teeth on sternites no. 1 and 2; median tooth on sternite no. 2 larger than that on no. 1; abdominal sternite no. 3 having no median tooth; preanal region unarmed.

Eyes: well developed with cornea diameter shorter than stalk.

Basal segment of antennular peduncle: broad, stylocerite distinctly pointed, reaching to about one-third of basal segment; anterolateral tooth reaching to about middle of second segment; second segment as long as third segment.

Scaphocerite (fig. 2B): longer by 3.2 times than the width, outer lateral side straight.

Mandibular palp (fig. 2C): slender with three segments, incisor process robust.

Maxillular palp (fig. 2H): bilobed, upper lobe slender bearing 1–3 apical setae and 3–4 subapical ones at both lateral faces, lower lobe stout with distal blunt knob bearing a long sinuous apical seta.

Maxillar palp (fig. 2D): palp with numerous setae proximally, basal endite deeply bilobed, scaphognathite normal.

First maxilliped (fig. 2E): with setose palp, basal and coxal endites distinct, flagellum of exopod with numerous plumose setae distally, epipod deeply bilobed.

Second maxilliped (fig. 2F): with normal endopod, flagellum with numerous setae distally epipod simple, with well-developed podobranch.

Third maxilliped (fig. 2G): with robust endopod, exopod with numerous plumose setae distally, reaching to distal margin of ischiomerus.

First pereiopod: slender, reaching beyond scaphocerite by entire distal margin of carpus, equal in length, similar in form. Palm as long as finger, carpus 1.79 times as long as chela, merus shorter than carpus; both fingers ending in very small apical claw concealed by apical curved cutting edge. Male second pereiopod equal in length, similar in form.

Second pereiopod: reaching beyond scaphocerite by one-third of merus length, longer than total length; palm, carpus, menus, ischium covered with spinules with lower ones smaller than upper ones and bending towards legs; upper spinules obtuse. Movable finger curving toward the fixed finger tapering to a tip, 0.9 palm length, coated with long velvety hairs on proximal two-thirds (fig. 5), distal with some tiny or inconspicuous crescent-shaped tubercles present on inner side of the cutting edges; 2 conical strong teeth in proximal one-sixth of cutting edge of movable finger (fig. 6). Fixed finger ending in acute, downward curving apical claw, coated with velvety hairs on proximal two-thirds; proximal one-sixth of cutting edge armed with a strong conical tooth preceded by a low obtuse bifid triangular tooth. Both fingers curving inwards, tips crossing and not gaping when fingers closed; palm cylindrical, length 2.9 times of width; carpus distinctly shorter than chela (14:19), longer than palm.

Last three pereiopods: slender, similar in form; propodus, carpus, merus, covered with spinules and setae.

Third pereiopod: reaching beyond scaphocerite by one-fourth to two-fifths of propodus (part of propodus extending beyond the end of scaphocerite equal to one quarter to two-fifths of propodus length), propodus 3.33 times as long as dactylus with about 8 (paratype 12) movable spines on postero-lateral margin.

Fifth pereiopod: somewhat slenderer than the third, propodus ending beyond distal margin of scaphocerite (end of propodus and end of scaphocerite terminating in equidistance).

Endopod of male: first pleopod not reaching distal half of exopod, inner margin concave, outer margin slightly convex.

Appendix musculina of male: longer, stouter than appendix internal with numerous stiff setae.

Uropodial diaeresis (fig. 4): with inner movable spine longer than outer angle (in paratype as long as outer angle).

Ovigerous females: with eggs 1.0 × 1.3 mm to 1.1 × 1.5 mm in diameter (n=25).

Habitat. Inhabiting small stagnant water bodies, e.g. irrigation channels where the others not found.

Etymology. Location of first collections.

DNA sequence alignment and phylogenetic relationships. The COI sequence lengths were 668 base pairs (bp). The alignment of 18S sequences yielded a consensus whose length was 1453 bp while the majority of the 18S sequences were 1446 bp. Thus indels did not present much of a problem in the alignment. The combined dataset was 2121 bp. For Bayesian inference, GTR+G was selected as the nucleotide substitution model for all the partitions.

The phylogenetic trees reconstructed from the three datasets are shown in fig. 7. It can be seen that the partial COI and 18S rRNA genes were capable of grouping closely related species in this study together, as shown by high posterior probabilities supporting the clades toward the tips of the trees. For COI, the genetic differences within each clade, as shown by the length from the vertex of an isosceles triangle to its base (the height of the isosceles triangle) or the branch lengths from the node to the tips, tended to be small compared to the differences between clades, indicated by the branch lengths from the ancestral nodes to those taxa.

M. suphanense was most closely related to M. sintangense (de Man, 1898) and the two species formed a monophyletic clade with posterior probability of at least 0.98. This clade, in turn, form a monophyletic clade with M. nipponense (de Haan) with posterior probability of 1. The clade containing the three species will be called clade 1. M. niphanae Shokita & Takeda, 1989 and M. dienbienphuense Dang & Nguyen, 1972 also formed a monophyletic clade with posterior probability of at least 0.96. M. lanchesteri (de Man, 1911) and M. rosenbergii (de Man, 1879) , despite their differences in size, were closely related, with supporting posterior probability of 1. Unfortunately, COI and 18S genes yielded different relationship among clade 1 and the other two clades (figs. 7 A and 7B).

Remarks. Fully grown males of M. suphanense (wild and bred) had features similar to fully grown males (wild and bred) of M. sintangense , M. dolatum Cai et al., 2004 , M. hungi Xuan, 2012 and M. equidens ( Dana, 1852) , yet it was different from M. sintangense in that: 1) the form of the rostrum: narrower midway with straight end whereas that of M. sintangense broad midway with end curving upward; 2) the number of teeth on rostrum: 9– 10 (including teeth behind orbital margin 2) / 3 (ventral) versus 9–13 (including teeth behind orbital margin 2–4) / 3–5 (ventral) in M. sintangense (collected); 3) second pereiopods appeared less robust than those in M. sintangense ; 4) smaller body size: the largest male was 42 mm versus 71 mm in M. sintangense (collected) (fig. 8); 5) found mainly in small still water bodies vs running water for M. sintangense ; 6) fecundity of ovigerous female was lower: the number of eggs varied from 30–111 versus 159–272 in M. sintangense .

The new species differed from M. dolatum in that: 1) cutting edge of distal part of the fixed finger not razorlike versus M. dolatum with razor-like edge; 2) spine at the uropodal diaeresis equal to or longer than the outer angle versus shorter in M. dolatum .

The new species can be differentiated from M. hungi found in Cambodia by: 1) body size smaller: the largest male 42 mm versus 98.5 mm in M. hungi ; 2) rostrum shorter: not reaching scaphocerite versus reaching beyond the distal margin of scaphocerite in M. hungi ; 3) the number of teeth on rostrum fewer than in M. hungi : 9–10 / 3, versus 13–15 / 5–7 in M. hungi .

M. equidens preferring brackish water versus freshwater for M. suphanense .

In each population fully grown dominant males are few (from one to a few). In our hands, the wild caught fully grown males and the ones reared in the laboratory were virtually in the same size range regardless of whether only one-species population growing up by themselves or growing up with another species.

Molecular phylogeny. We have reconstructed phylogenetic trees (based on COI, 18S rRNA and combined sequence alignments) having M. suphanense in its relationship to extant, mostly local Macrobrachium species. It is clear that M. suphanense is a new species with the posterior probability supporting the clade in all the trees being 1 indicating clear genetic differences from its sister clade, M. sintangense . It is also clear that this new species is not a hybrid of extant species because its phylogenetic relationships to other closely related Macrobrachium species ( M. sintangense and M. nipponense ) from both the nuclear 18S and the mitochondrial COI genes agree (figs. 7 A and 7B). In fact, there are 2 specimens (M. sp. from Beungkan and Chainat) whose locations on the phylogenetic trees reconstructed from single genes differ, indicating that they are potential hybrids. For the specimen from Chainat, its COI gene indicates that it is an M. suphanense while its 18S gene falls within the clade M. niphanae . The other specimen’s COI gene forms a monophyletic clade with M. niphanae while its 18S gene is related to those of M. lanchesteri and M. rosenbergii .

As regards species identification by DNA sequences of other known species and their relative positions on the phylogenetic tree, our molecular phylogenetic results of those valid species based on the COI and combined datasets are consistent with those already published for M. sintangense , M. nipponense , M. niphanae , M. lanchesteri and M. rosenbergii ( Wowor et al., 2009) based on 18S+28S rRNAs (nuclear). Another noticeable feature common between our and Wowor et al.’s results is that the resolution of the phylogeny tends to be poorer toward the root of the Macrobrachium clade as shown by low posterior probabilities supporting them.

That the new species, M. suphanense , is not a hybrid of recent extant species is supported by over 10 generations of breeding resulting in only one type of prawns with identical and distinctive developmental and fully grown features (our observation). As indicated above, we are capable of detecting hybrids by DNA analysis, and are certain that M. suphanense is not a hybrid, but a pure species.