Bioacoustics of Ochrilidia socotrae

Description . Lower margin of temporal foveolae is visible from above, foveolae are 4 times longer than high, pronotum keels more or less parallel in the prozona and just diverging in the metazona, pronotum much rugose with impressed points, interocular distance less than 1/2 width of fronto-vertical part of head in front of eyes in males, 1/2 to 3/ 5 in females ( Figs. 22-24 ); antennae not very long, reaching the mesonotum in males, the hind margin of the pronotum in females. Tegmina exceed apex of the abdomen, showing marginal area narrow and much close to submarginal ( Fig. 25 ). A small black spot is present on the upper side of fore and mid femora. Lower inner lobe of knee of hind femora shows a black spot, covering from 1/4 to 1/3 of the lobe in males, more extended on all the lobe in females. The inner side of male hind femora bears ca. 120-130 stridulatory pegs. Hind tibiae violet. Tegmina exceed the apex of hind femora ca. 2.9-3.2 mm in males and 4.1-4.5 mm in females. Male cerci are just longer than epiproctus, subgenital plate is a little pointed ( Fig. 27 ). Epiphallus (terminology according to Jago 1977 ) with erect ancorae, in dorsal view upper lophal lobe ( b ) large, median lobe ( c ) long and inclined inwards, and lower lobe ( d ) triangular and well separated from c ( Fig. 29 ). In lateral view lobe c is just visible between b (pointed) and d (very much protruded), about twice length of b ( Fig. 30 ). Aedeagal valves are directed dorsally and obliquely. Biometrics and comparison with O. geniculata . Socotra specimens resulted very small; I compared them with a long series coming from N Africa and Arabia and I found a very high reduction of the size. Tables 2 reports measurements of O. socotrae n. sp. compared with O. geniculata (specimens from Algeria , Tunisia , Libya and Arabia). My measurements of O. geniculata match those reported by Jago (1977) for gregarious populations. All pairs of data resulted statistically highly significant (test t of Student, P<0.0001). Additionally, I compared the ratio total length/length of tegmina ( O. socotrae : males 1.20 ± 0.07, females 1.21 ± 0.04; O. geniculata : males 1.11 ± 0.06, females 1.11 ± 0.05) and the ratio length of tegmina/length of hind femurs ( O. socotrae : males 1.65 ± 0.07, females 1.63 ± 0.04; O. geniculata : males 1.81 ± 0.10, females 1.76 ± 0.09). In both cases, I found statistically highly significant differences (test t of Student, P<0.0001 for all pairs, except length of tegmina/length of hind femurs for females, for which P=0.004).

Habits . Species of the genus Ochrilidia generally are associated with graminaceous plants; Ochrilidia socotrae n. sp. lives on Heleochloa dura linked to sand dunes of Ra’s Shoab, the East cape of Socotra , characterized also by the presence of Avicennia marina on the back of dunes. In April 2008 I found an abundant population, constituted only by adults; males stridulated and in a few cases I observed that they were moving closer to a female. When a male was much close to a female, it raised alternatively hind femurs, showing the black spot, which eventually may act as sexual recognition. Species and sex Total length Length of tegmina Length of pronotum Height of pronotum Length of hind femurs Height of hind femurs Ochrilidia socotrae males (n=11) 18.5 ± 0.64 15.5 ± 0.90 2.95 ± 0.14 2.14 ± 0.20 9.36 ± 0.39 1.92 ± 0.09 Ochrilidia geniculata males (n=25) 20.72 ± 1.21 18.64 ± 1.47 3.47 ± 0.26 2.55 ± 0.16 10.31 ± 0.61 2.27 ± 0.10 Ochrilidia socotrae females (n=5) 26.3 ± 0.77 21.8 ± 0.60 4.54 ± 0.15 3.52 ± 0.11 13.4 ± 0.40 2.80 ± 0.07 Ochrilidia geniculata females (n=25) 32.41 ± 1.73 29.23 ± 2.05 5.92 ± 0.38 4.52 ± 0.41 16.6 ± 1.18 3.45 ± 0.27 Acknowledgements I would like to thank very much Francesco M. Raimondo, manager of the Botanical Garden of Palermo and of the expedition to Socotra , Luigi Guiglia, Ahmed Saeid Suliman, director of the Conservation Unity of the Socotra Archipelago Conservation and Development Programme, who faciltated visit and collecting on the island. I am indebted to Attilio Carapezza, who collected together me insects at Socotra , and Pietro Lo Cascio and Flavia Griti who sent me some specimens collected in the Socotra archipelago between 20th and 27th February 2009 . I also thank very much Hendrik Devriese (IRSNB, Department Entomology, Bruxelles), who identified Paratettix subpustulatus and sent me interesting considerations. I also thank Maurizio Pavesi and Michele Zilioli for facilities and assistance during the study of specimens of La Greca’s collection preserved at the Museo Civico di Storia Naturale of Milan.

Taxonomic notes. Popov collected four Ochrilidia specimens at Shuab in 1953 (Fig. 51 ) and identified these as O. kraussi (Bolívar, 1913) (Popov in Uvarov and Popov (1957) ). However, he mentioned some differences with O. kraussi specimens from different African sites. These differences are mainly in the shape of the metazona of the pronotum: in the Socotran specimens, it is as long as wide, while in other kraussi specimens, it is often longer than wide. Specimens of O. kraussi from Somalia resemble the Socotra taxon (Popov in Uvarov and Popov (1957) ). In his revision of the genus Ochrilidia , Jago (1977) synonymised O. kraussi with O. geniculata , including Popov’s specimens of Socotra . Jago (1977) erroneously listed the four Socotran specimens under Ethiopia instead of Yemen . Massa (2009) , therefore, was misled when he interpreted Jago as not having studied the Socotran specimens. Massa (2009) described O. socotrae , based on specimens he collected at Shuab in 2008, the same site where Popov collected his specimens. He mentioned the following characteristics of O. socotrae that separate this taxon from O. geniculata from Africa and Arabia: overall smaller, shorter wings, a less marked black spot on the inner side of the hind knees, a less pointed subgenital plate, shorter cerci and less stridulatory pegs. According to Massa (2009) , the length and width ratio of the metazona does not differ between socotrae and geniculata , contrary to Popov’s statement about kraussi (in Uvarov and Popov (1957) ). After comparison, Popov’s specimens belong to O. socotrae , which is unsurprising since they come from the same site and habitat.

Bioacoustics. The calling song consists of 20–30 syllables repeated at 3–4 per second. Syllables last 90–110 ms and consist of a short, sharp tick-sound of about 25 ms followed by a weaker rustling sound (Fig. 53 ). The tick-sound is probably linked to the upward movement of both hind legs. During interaction with other males, a series of syllables may be shorter or may lead to the production of alternating syllables of the males involved. Additionally, some weaker shortly buzzing sounds can be heard, possibly linked to situations where a female is close to the male. The same syllabic structure can be found in the song of Ochrilidia sicula (Salfi, 1931) (Baudewijn Odé, XC 846260, accessible at https://www.xeno-canto.org/846260 ) and O. pruinosa Brunner von Wattenwyl, 1882 ( Willemse et al. 2018 ). However, in the latter species, the syllables are adjoined in a dense echeme.