Coeloperix sinica, Pan & Jiang & Fan & Al-Farraj & Gao, 2017

COMMENTS ON D. COMPRESSA View in CoL

In the Chinese population, two morphotypes (with or without a spine) of D. compressa were observed, but they had no differences with respect to other live morphology and infraciliature. The same phenomena were also observed in D. brasiliensis and D. crassipes by Gong et al. (2007). Therefore, they are likely conspecific.

Dysteria compressa ( Fig. 9A) was originally reported by Gourret & Roeser (1886) under the name ‘ Aegyria compressa ’, but subsequently Kahl (1931) transferred it to the genus Dysteria . After that, no studies on its morphology have been reported. Although few valuable morphological characters have been presented in previous studies, we can identify our isolate as D. compressa by its shape and the dorsal spine.

Hitherto, there are six Dysteria species ( Fig. 9B–E, G–K) possessing spines, namely, D. spinifera Dragesco, 1966 , D. aculeate Claparède & Lachmann, 1859 , D. spinigera Claparède & Lachmann, 1859 , D. marioni ( Gourret and Roeser, 1886) Kahl, 1931 , D. crassipes

CV, contractile vacuoles; Fvk, frontal ventral kineties; IRK, inner-most right kinety; –, data not available.

Claparède & Lachmann, 1859, and D. brasiliensis Faria et al., 1922 . However, they can be separated from D. compressa by cell size, body shape, the number of right kineties, the number of spines, and the number of contractile vacuoles ( Table 6).

Concerning five right kineties, four more species ( Fig. 9F, L–R; Table 6), in which the spine is absent, should be compared with D. compressa , namely, D. monostyla (Ehrenberg, 1838) Kahl, 1931 , D. antarctica Gong et al., 2002 , D. calkinsi (Calkins, 1902) Kahl, 1931 , and D. nabia Park and Min, 2014 . In addition to the characters listed in Table 6, D. monosyla and D. antarctica can be distinguished from D. compressa by the pattern of the posterior ends of their innermost right kineties (terminating caudally in the same level vs. shortened and terminating at the level of the podite); D. calkinsi can be identified by the presence of two longitudinal grooves (vs. absence; Song & Wilbert, 2002); D. nabia can be characterized by the position of the left frontal kineties (on the right of the circumoral kineties vs. on the left; Park & Min, 2014).

PHYLOGENETIC ANALYSES

The overall topology of our phylogenetic tree is congruent with previous research ( Gong et al., 2008; Gao et al., 2012; Pan et al., 2013a; Qu et al., 2015b, c; Chen et al., 2016): (1) Cyrtophoria is monophyletic with full supports; (2) Chlamydodontida is paraphyletic, within which the families, that is Lynchellidae , Chlamydodontidae , and Chilodonellidae , are monophyletic; (3) the systematic position of D. compressa , O. alpestris biciliata , and C. sinica sp. nov. is stable.

The present work indicates that Chlamydonella has a closer relationship with Chlamydonellopsis than Atopochilodon , although both Atopochilodon and Chlamydonella have Y-shaped perioral kineties ( Qu et al., 2015b). This topology is in agreement with the pattern of ventral kineties (continuous in Chlamydonella and Chlamydonellopsis , but discontinuous in Atopochilodon ; Gong & Song, 2006). It also indicates that the Y-shaped perioral kineties are likely a plesiomorphy in the Atopochilodon + Chlamydonella + Chlamydonellopsis clade.

In the family Chilodonellidae , the phylogenetic tree revealed a closer relationship between Odontochlamys and Chilodonella than with the other two genera, Trithigmostoma and Pseudochilodonopsis . This is consistent with their morphological data, that is Odontochlamys and Chilodonella have discontinuous somatic kineties (vs. continuous in Trithigmostoma ) and an unsegmented preoral kinety (vs. segmented in Pseudochilodonopsis ) ( Foissner et al., 1991). However, there is also a morphological difference between Odontochlamys and Chilodonella , that is the position of the terminal fragment (apical position in Odontochlamys vs. subapical position in Chilodonella ) ( Blatterer & Foissner, 1992). In addition, the SSU rRNA gene sequence of Odontochlamys differs from that of Chilodonella by 128–134 bp. These results support that Odontochlamys is a valid genus different from Chilodonella .

It is quite surprising that Aporthotrochilia pulex was nested in Coeloperix and differs from C. sleighi in only 5 bp. We checked the ABI files of the sequence but they showed clear and well-defined peaks, indicating that this sequence was more likely from a wrong cell than from a mixed PCR product. Owing to the small size, A. pulex is difficult to pick as a live cell for DNA extraction under a stereo microscope. We suspect the ‘sequence of A. pulex ’ might be from another species on two grounds: (1) A. pulex and C. sleighi differ from each other in many morphological characters such as infraciliature and presence/absence of a podite ( Pan et al., 2012); (2) these two species cluster with C. sinica .