Tritonus, Mulsant, 1844

Archangelsky, Miguel, Martínez Román, Nicolás Rafael & Fikáček, Martin, 2021, Larval chaetotaxy and morphology are highly homoplastic yet phylogenetically informative in Hydrobiusini water scavenger beetles (Coleoptera: Hydrophilidae), Zoological Journal of the Linnean Society 192 (2), pp. 416-416 : 416-

POSITION OF TRITONUS

An unexpected result of the analyses based on all larval characters and non-chaetotaxic larval characters ( Fig. 99 ) is the position of Tritonus as a sister clade to the Hydrobiusini (excluding Hybogralius ). Four larval characters support this clade: (1) character 16(1), inner margin of epistomal lobes with cuticular spines (also present in other dissimilar genera, e.g. Phaenonotum Sharp, 1882 , Chaetarthria , Tropisternus Solier, 1834 , convergence); (2) character 35(0), femur as large as trochanter (also shared with unrelated Phaenonotum , Chaetarthria and Tormus , convergence); (3) character 49(0), ratio distance between both FR8/distance between FR8–FR9 less than 0.6 (also shared with larvae of the tribe Hydrophilini and Tormus , a probable convergence); and (4) character 99(1), a short antennal sensorium (SE1), not shared with any other genera included in the analysis and can be considered a synapomorphy for this clade; however, Tropisternus and Sternolophus Solier, 1834 (Hydrophilini) have an extremely short sensorium. The topology-constrained analysis revealed Tritonus as sister to the Hydrobiusini only in the analysis based on non-chaetotaxic characters; in all other cases it was revealed as sister to the Hydrophilini . This position is supported by three larval characters: (1) 57(1), sensillum PA11 developed as long trichoid seta (also present in Amphiops , Derallus and Ametor ); (2) 65(1), pore AN2 closer to distal margin of the antennomere than to AN1 (weak character with many reversals across the tree); and (3) 96(0) third antennomere long (ratio AN2/AN3 <1.5) (only present in the Hydrophilini, Tritonus and Hydramara in the taxa included in the analysis). All these results are in contrast to those analyses based on molecular data ( Toussaint et al., 2016; Toussaint & Short, 2018) in which Tritonus is revealed as closely related to Tormus and Paracymus Thomson, 1867 . Larvae of the latter two genera differ from those of Tritonus in many aspects [discussed by Fikáček et al. (2017)] and from the morphological point of view, their close relationship seems unlikely. Our analyses indicate that the same may be the case for adult morphology. We hence consider the sister position of Tritonus and Hydrophilini as an alternative hypothesis worthy of further testing.

Figures

Figures 99–106. Results of the phylogenetic analyses. 99–105, topologies obtained using different datasets: 99, larval morphology only (majority rule consensus); 100: larval chaetotaxy only (implied weighted tree, k = 20). 101, all larval characters (implied weigthed tree, k = 20). 102, larval and adult characters (most parsimonous tree). 103, reference topology based on DNA data, adopted from Toussaint & Short (2018). 104, part of the tree based on all larval characters (same as on Fig. 101) with characters mapped. 105, Hydrophilinae molecular reference topology with updated position of Tritonus with larval characters mapped. 106, alternative position of Tritonus in constrained topology search using different datasets.