Resolving a century-old case of generic mistaken identity: polyphyly of Chitoniscus sensu lato resolved with the description of the endemic New Caledonia Trolicaphyllium gen. nov. (Phasmatodea, Phylliidae) Author Cumming, Royce T. https://orcid.org/0000-0001-7930-1292 Montreal Insectarium, 4101 rue Sherbrooke est, Montre ́ al, Que ́ bec, H 1 X 2 B 2, Canada & Richard Gilder Graduate School, American Museum of Natural History, New York, NY 10024, USA & Biology, Graduate Center, City University of New York, NY, USA roycecumming@gmail.com Author Tirant, Ste ́ phane Le Montreal Insectarium, 4101 rue Sherbrooke est, Montre ́ al, Que ́ bec, H 1 X 2 B 2, Canada Author Bu ̈ scher, Thies H. Department of Functional Morphology and Biomechanics, Zoological Institute, Kiel University, Am Botanischen Garten 9, 24118, Kiel, Germany text ZooKeys 2021 2021-08-05 1055 1 41 http://dx.doi.org/10.3897/zookeys.1055.66796 journal article http://dx.doi.org/10.3897/zookeys.1055.66796 1313-2970-1055-1 DB8C0779E75744EDB9B2397EA2EF3BAE 69CACA7352355D15AE1D785BE3CA28C5 Trolicaphyllium sarrameaense ( Groesser , 2008) comb. nov. Figures 1 , 3 , 4A , 4C , 5A, C , 6A, C , 7A , 8A , 9A , 10A , 11A , 12B , 13A, B , 15A-C , 16A-C , 17A-D , 18A , 19 , 20 , 29 Material examined. ( 8 ♀♀ , 9 ♂♂ , 3 eggs ): Holotype and paratypes examined: 1 ♀ , 1 ♂ , 3 eggs : " Chitoniscus , sarrameaensis, Neu Kaledonien, Sarramea, Sep. 2006 , det. Groesser" (SDEI: HT ♀ , DEI Hemimetabola #100215; PT ♂ , DEI Hemimetabola #100214; PT eggs, DEI Hemimetabola #100216); (SDEI; Figs 7A , 8A , 13A, B , 29 ). See Suppl. material 1 for additional specimens reviewed, their collection data, and depositories. Remarks. As it was only described in 2008, this was the most recently described species from New Caledonia with type material originally collected by Sigetake Suzuki in 2004 from Sarramea ( Groesser 2008a ). Within the original description this species was not explicitly compared with the sympatric and morphologically very similar Trolicaphyllium erosus comb. nov. but was instead only differentiated from Chitoniscus lobipes Redtenbacher, 1906, where most features given for differentiation were simply the features we discuss above as significant for differentiating the two genera. Other lobed specimens have been recovered from throughout New Caledonia, but unfortunately most have been nymphs (such as several from within the QM collection) and therefore they could not be confidently identified as Trolicaphyllium sarrameaense comb. nov. or as Trolicaphyllium erosus comb. nov. nymphs. Unfortunately, in Groesser (2008b) the key to species of Chitoniscus sensu lato tried to use the female tegmina radial and media venation pattern to differentiate species, but mixed up the species. Within the key Trolicaphyllium erosus comb. nov. and Trolicaphyllium brachysoma comb. nov. (from New Caledonia) instead key out as the Fijian population and Chitoniscus lobiventris (Blanchard, 1853) and Chitoniscus lobipes Redtenbacher, 1906 (from Fiji) key out as the New Caledonian population. We have reviewed the type specimen photos available on the Phasmid Species Files ( Brock et al. 2021 ; http://phasmida.speciesfile.org) as well as numerous museum specimens, and always the female tegmina venation allowed accurate distinction of these two genera. Even if you look past this inaccuracy within the key, unfortunately no additional features can be gleaned from the further couplets to allow differentiation of Trolicaphyllium sarrameaense comb. nov. from Trolicaphyllium erosus comb. nov. (as the further couplets discuss abdominal shape, which in these two species is identical/variable). At this moment in time, we still lack significant details about the population of Trolicaphyllium gen. nov. on Grande Terre as material is limited and molecular data has yet to be compared across a wide sampling on the island. With phylliid abdominal shapes sometimes rather variable within a single species, this makes us wonder if Trolicaphyllium sarrameaense comb. nov. is in fact a valid species, or simply a synonym of Trolicaphyllium erosus comb. nov. which was described more than 100 years previously from the same island. Our examination of all type specimens which could be traced has not yet revealed additional features for morphological differentiation besides the overall size of these two species. Hopefully, future molecular analyses from across New Caledonia will reveal if there are several species present on Grande Terre or if it is simply a single species which can range in size from smaller (ca. 40 mm; Trolicaphyllium erosus comb. nov.) to larger (ca. 60 mm; Trolicaphyllium sarrameaense comb. nov.). It is due to this lack of sound molecularly based evidence and the propensity for phylliids to be morphologically variable that we refrain from synonymizing Trolicaphyllium sarrameaense comb. nov. with Trolicaphyllium erosus comb. nov. as we feel a significant decision such as this should be based upon a solid foundation. If future molecular analyses reveal that there is only a single morphologically variable species of Trolicaphyllium gen. nov. on Grande Terre based upon a sampling throughout the island, then we feel a synonymization will be necessary, but not until that time. The etymology given in the original description is that this name is a toponym, named after the type locality, Sarramea, New Caledonia ( Groesser 2008a ). The original combination was with the masculine genus ( Chitoniscus ) and therefore in order for the species name to agree in gender with our newly erected genus, the spelling of " sarrameaensis " is changed to the neuter gender " Trolicaphyllium sarrameaense ". Differentiation. Females can be differentiated from Trolicaphyllium brachysoma comb. nov. based upon abdominal shape, as Trolicaphyllium brachysoma comb. nov. are considered to have a spade-shaped abdomen, with smooth margins, versus Trolicaphyllium sarrameaense comb. nov. which has a broad abdominal shape with parallel sides, ending in lobed segments VII and VIII. From Trolicaphyllium erosus comb. nov. the only feature we have been able to identify as useful is the overall length, with Trolicaphyllium erosus comb. nov. ca. 40 mm long versus Trolicaphyllium sarrameaense comb. nov. ca. 60 mm long. Unfortunately, males of Trolicaphyllium brachysoma comb. nov. and Trolicaphyllium erosus comb. nov. have never been confidently confirmed through breeding or molecular comparison. Based upon the confidently confirmed male/female Trolicaphyllium sarrameaense comb. nov. however, we expect that the male morphology should mirror the female morphology. Most likely the male Trolicaphyllium brachysoma comb. nov. will lack prominent abdominal lobes and the male Trolicaphyllium erosus comb. nov. will have distinct lobes to match with their female counterparts. Based upon the female Trolicaphyllium erosus comb. nov. smaller size, we expect that the male must also be rather small, which could likely be used as a feature for differentiation. Distribution. To date we have only confirmed adult specimens and observations which are the correct morphology and size of Trolicaphyllium sarrameaense comb. nov. from central Grande Terre (Fig. 21 ). We have however seen nymph specimens which had characteristically lobed abdomens which may represent this species from other locations on Grande Terre, so we expect this species may be widespread throughout the island. Within the MZPW collection there is a pair of Trolicaphyllium sarrameaense comb. nov. specimens with the data of simply "Lifu" , which if true could lend credibility to the hypothesis that perhaps these species are all variable in their abdominal shape (if there is only one species present on Lifou island), but as these are antique and give no other data, we do not take these as highly credible, and therefore exclude this record from further discussion and they are not included within the distribution map (Fig. 21 ). Or it is possible Lifou island holds several morphologically different species.