Construction of the molecular dataset: regional and continental sampling

Sample sizes were restricted because several Indian Ocean Orthoptera species are threatened [23,24], and in view of conservation priorities, research permits restrict sampling for island endemics with small populations. Further, many species occur in treetops and are highly time-consuming to collect. Depending on numbers of individuals available, between two and four individuals from each natural song group in the western Indian Ocean were sampled from across that group’s range (be it restricted to one island, or found across many), and including at least one individual per island.

Ornebius has a wide distribution across the tropics, with most described species occurring in Australasian and Oriental regions, and a minority of occurrences in the Americas and islands of the Pacific and Indian Ocean [25]. A single species is described from Africa (Ornebius acutus from Ghana), but probably belongs to a distinct genus (S. Hugel, unpublished data). We therefore obtained Ornebius samples from Australia, French Polynesia, Borneo and Central America as potential source areas for the colonisation of the western Indian Ocean, as well as outgroup samples from two other Mogoplistinae genera (Ectatoderus and Cycloptiloides) and one other Grylloidea (Creolandreva).

Each individual was vouchered and stored in Ethanol. DNA was extracted from finely sliced leg tissue using a DNeasy Blood and Tissue Kit (Qiagen Inc., Valencia, CA, USA), and following the manufacturer’s protocol. For all samples (i.e. 2–4 individuals per song group in the Indian Ocean, and 1 individual per continental species), a total of 1594 bp of sequence data were obtained from the mitochondrial COI gene, and ribosomal RNA genes 12S and 16S. Primers used for amplification and sequencing of these three genes were LCO1490 and HCO2198 [26], SR-J-14233 and SR-N-14588 [27], and 16Sa and 16Sb [28], respectively. The polymerase chain reaction (PCR) program began with denaturation for 5 min at 94°C, followed by 35 cycles of 1 min at 94°C, 1 min at 49°C, 1 min at 72°C, and a final extension step at 72°C for 5 min. To guard against the possibility of introgression obscuring true lineage history and improve the resolution of phylogenetic relationships, the mitochondrial data were supplemented with nuclear EF1α and H3 sequences from a reduced sample set (1–3 individuals from each terminal monophyletic group in the mitochondrial dataset), providing a five-gene dataset totalling 2352 bp. Primers used for amplification and sequencing of EF1α and H3 were M51F and M53R [29], and 5’-ATGGCTCGTACCAAGCAGACVGC-3’ and 5’-ATATCCTTRGGCATRATRGTGAC-3’ [30], respectively. The PCR program began with denaturation for 4 min at 94°C, followed by 40 cycles of 30 sec at 94°C, 40 sec at 54°C, and 40 sec at 72°C, and a final extension step at 72°C for 7 min. Amplified products were purified and sequenced by Macrogen (Seoul, Korea) using the standard-seq single service. Sequences have been deposited in Genbank (Accession numbers {"type":"entrez-nucleotide-range","attrs":{"text":"KC494774-KC494868","start_term":"KC494774","end_term":"KC494868","start_term_id":"469475666","end_term_id":"469475789"}}KC494774-KC494868, {"type":"entrez-nucleotide-range","attrs":{"text":"KU597648-KU597710","start_term":"KU597648","end_term":"KU597710","start_term_id":"1000478036","end_term_id":"1000478122"}}KU597648-KU597710).