A revision of the subgenus Stegana (s. s.) (Diptera, Drosophilidae) from mainland China Author Li, Tong Author Cao, Huazhi Author Gao, Jianjun Author Chen, Hongwei text Zoological Journal of the Linnean Society 2010 2010-04-30 158 4 726 739 http://dx.doi.org/10.1111/j.1096-3642.2009.00575.x journal article 10.1111/j.1096-3642.2009.00575.x 0024-4082 10114684 STEGANA ( STEGANA ) ROTUNDA CAO & CHEN SP. NOV. ( FIGS 36–40 ) Specimens examined: Holotype male, China : Guangxi , Guilin , Maoershan , 630 m , 21.ix.2006 , ex tree trunks, JJ Jiang ( SCAU , no. 120291). Etymology: From the Latin word rotundus , referring to the aedeagus being slightly rounded apically. Diagnosis: Katepisternum almost dark brown; paramere with 15–17 strong setae ( Figs 39, 40 ); surstylus without prensiseta ( Fig. 37 ). Description: Frons, face, and clypeus yellow. First flagellomere yellow on basal third, black on distal two-thirds. Gena brown. Mesonotum yellow, sublaterally with two brown stripes on each side. Scutellum brown. Legs yellowish white, fore leg brown on knee and tibia; mid and hind leg black on femora and basal three-quarters of tibiae. Fore femur with four setae on distal part of ventral surface. Abdomen with all tergites and sternites black. Male terminalia: hypandrium somewhat sclerotized anteromedially ( Fig. 38 ). Gonopods with one curved process posteromedially and two smaller triangular processes submedially ( Figs 39, 40 ). Aedeagus with several tentacle-like setae apically and spine-like processes ventrally, lacking pubescence ( Figs 39, 40 ). Measurements: BL = 2.46 mm in holotype ; ThL = 1.22 mm ; WL = 1.88 mm ; WW = 0.95 mm . Indices: arb = 7/5, avd = 0.73, adf = 1.67, flw = 2.22, FW/HW = 0.33, ch/o = 0.08, prorb = 1.00, rcorb = 0.80, vb = 0.29, dcl = 0.56, presctl = 0.70, sctl = 1.84, sterno = (damaged), orbito = 2.31, dcp = 0.24, sctlp = 1.50, C = 1.91, 4c = 1.25, 4v = 1.86, 5x = 1.18, ac = 9.26, M = 0.50, C3F = 0.63. Distribution: China ( Guangxi ). RESULTS OF MOLECULAR PHYLOGENETIC ANALYSIS Data analysis The alignment of the ND2 sequence was 1029 nucleotide sites in length, coding 342 amino acid residues, containing 183 parsimony-informative characters, with consecutive gaps at sites 267–269 in some sequences. The average base frequencies of the ND2 sequences (including outgroups) calculated in MEGA were 36.6, 45.0, 7.8, and 10.6% for A, T, G, and C, respectively, showing strong bias towards A and T. It was revealed by the Chi-square test that the base composition is not homogeneous among taxa ( P = 0.8208) at the third codon position, but homogeneous at the first and second codon positions ( P = 1.000 and 1.000, respectively), or even the average data ( P = 0.9988). The test of substitution saturation in DAMBE showed that the observed Iss index is 0.1707, significantly lower than the Iss.c