Luciolinae

Martin, Gavin J., Stanger-Hall, Kathrin F., Branham, Marc A., Da Silveira, Luiz F. L., Lower, Sarah E., Hall, David W., Li, Xue-Yan, Lemmon, Alan R., Lemmon, Emily Moriarty & Bybee, Seth M., 2019, Higher-Level Phylogeny and Reclassification of Lampyridae (Coleoptera: Elateroidea), Insect Systematics and Diversity (AIFB) 3 (6), No. 11, pp. 1-15 : 5-7

Luciolinae

The Luciolinae are a subfamily of fireflies that have received much attention over the years (for a taxonomic treatment, see Ballantyne et al. 2015) and seem to share stable morphological characters such as a reduced number of abdominal ventrites in the males (see discussion of Lamprigera below).

Our analyses largely agree with recent morphological analyses in terms of the major groupings of the Luciolinae ( Ballantyne and Lambkin, 2013) . Thanks in large part to the effort of Ballantyne and Lambkin ( Ballantyne 1987 a, Ballantyne and Lambkin 2009, 2013), there also exists a discussion of the aedeagal sheath, which has provided evidence in support of the current taxonomic framework for these taxa. Many of the morphological characters used in these analyses ( Ballantyne and Lambkin 2009, 2013, Ballantyne et al. 2015, 2016) and the phylogenetic patterns they support, correlate well with the phylogenetic hypotheses generated from our molecular data. However, some minor differences between previous phylogenetic work and our overall topology do exist. For example, the Atyphella complex is not recovered as the the most highly derived lineage, as it is in Ballantyne and Lambkin (2013). Our molecular data set recovers this lineage as more basal compared with the other

Luciolinae . In addition, the lampyrine genus Lamprigera is recovered as a lineage within the lucioline clade.

Jeng et al. (2000) remarked that the systematic position of Lamprigera was uncertain and would hopefully be more accurately defined through future phylogenetic investigation. Previous analyses have recovered Lamprigera in various positions within the Lampyridae . In 2006, on the basis of the 16S mitochondrial marker, Li et al. performed a phylogenetic reconstruction of the Lampyridae . In this analysis, Stenocladius was recovered as the basal lineage, while Lamprigera was recovered in a surprising clade together with the amydetine genus Vesta sister to the remaining fireflies ( Li et al. 2006), albeit with low support. In his thesis, Jeng (2008) recovered Lamprigera sister to the Phausis + Lamprohiza using>400 morphological characters. Wang et al. (2017) recovered Lamprigera as a member of the Lampyrinae on the basis of 13 mitochondrial genes. It should be noted, however, that the Wang et al. (2017) analysis suffered from a limited taxon sampling (six lucioline taxa and two additional lampyrine taxa from a single genus). Martin et al. (2017) recovered Lamprigera as sister to the monophyletic Luciolinae with high support and placed it as Lampyridae incertae sedis.

Here we recover Lamprigera as a member of the Luciolinae for the first time with both strong support and congruence between all of our analyses. However, this placement is based on a single Lamprigera species and major morphological differences between Lamprigera and the luciolines, e.g., the number of abdominal ventrites, need to be addressed. A major , long-standing morphological synapomorphy for Luciolinae has been males with six abdominal ventrites, whereas all other Lampyridae have seven or eight abdominal ventrites. Males of Lamprigera exhibit the ‘typical’ abdominal morphology of Lampyridae with seven ventrites.

In certain elateroid lineages, ventrite number is known to vary greatly, even within recently derived tribes/subfamilies (Kundrata and Bocak 2019). In contrast, ventrite number has not been shown to vary in Luciolinae . To rigorously test the classification of Lamprigera relative to the Luciolinae , an expanded taxon sampling including deeper species coverage within the Lamrigera, combined with an in-depth morphological investigation, including the plasticity of ventrite number across these taxa will be needed. Until these analyses can be done, we elect to keep Lamprigera as Lampyridae incertae sedis ( Martin et al., 2017).

Pterotinae , Cyphonocerinae , and Psilocladinae ( McDermott,

1964) stat. nov.

Jeng et al. 1998, 2006, summarizing the work of McDermott (1966), Crowson (1972), and Nakane (1991), were the first to formally delineate the Psilocladinae (previously known as Cyphonocerinae ), by identifying the constituents of the group ( Cyphonocerus Kiesenwetter , Psilocladus , and Pollaclasis ) and laying out nine morphological features uniting the group. Jeng et al. (1998) also recognized Cyphonocerinae as a subjective synonym of Psilocladinae based on priority. In 2016, Silveira et al. treated Scissicauda as a member of the amydetine subtribe Psilocladina sensu McDermott 1964 , distinguishing Scissicauda from the other members of the group ( Ethra , Photoctus McDermott , Psilocladus , and Pollaclasis ). In 2017, another genus ( Araucariocladus Silveira & Mermudes ) was added to the Psilocladina sensu McDermott . As a consequence of recognizing Psilocladina sensu McDermott instead of Psilocladinae sensu Jeng/Nakane, Cyphonocerus was left as the sole member of the Cyphonocerinae .

Our phylogenetic analyses supports in part the classification presented by Crowson (1972) and Nakane (1991) and supports the classification of Psilocladus as a separate lineage from the Amydetinae . Without Cyphonocerus in our taxon sample, the placement of Psilocladus within the Cyphonocerinae cannot truly be tested; however, given that Psilocladus did not form a monophyletic lineage with Pollaclasis , we formally recognize the subfamily Psilocladinae , distinguished by the antennae with 11 articles, articles 2–11 with two weak, ciliate branches, with the only constituent genus being Psilocladus .

Following McDermott (1966) and Silveira et al. (2016), Pollaclasis is classified as a member of the Amydetinae . However, both our ML and coalescent analyses challenge this classification. The ML analysis places Pollaclasis as a sister taxon to the North American Pterotus , the sole member of the Pterotinae ( Fig. 1 ). In contrast, the coalescent analysis places Pollaclasis as sister to a Phausis + Lamprohiza clade (Supp Fig. 2 ). Hypothesized to be a close relative of Cyphonocerus , Pollaclasis has previously been classified in the Psilocladinae based on the morphology of antennae, mandibles, and abdominal segmentation (see Jeng et al. 1998). The present analyses do not support the monophyly of Pollaclasis and Psilocladus . On the basis of this evidence, as well as the absence of Cyphonocerus in our taxon sample, we transfer Pollaclasis to Lampyridae incertae sedis. Future efforts need to be made to ascertain whether Pollaclasis is indeed a member of the Cyphonocerinae , or as the ML analyses suggest, more closely related to Pterotus .

Figures

Fig. 1. ML reconstruction of full, untrimmed dataset aligned in MAFFT; log-likelihood = −26,321,164.240. The numbers at nodes indicate UFBoot support. Unlabeled nodes have UFBoot values of 100%,colors represent the newly proposed classification.Star at node depicts Lampyridae;A.-I.Dorsal habitus of genus representatives:A. Lamprigera yunnana; B.Luciola cyathigera; C. Stenocladius davidi; D. Amydetes fastigiata; E. Vesta chevrolati; F.Diaphanes sp.;G. Aspisoma gentile; H. Ellychnia aurora; I. Photinus laticollis;

Fig. 2. Lamprohizinae habitus images; A: Phausis reticulata dorsal; B: Phausis reticulata ventral; C: Lamprohiza delarouzei dorsal; D: Lamprohiza delarouzei ventral.