Genetic variation at 59 protein coding loci (16 polymorphic) and 25 microsatellite loci was analysed for 11 indigenous south‐east Asian goat populations, and the Australian feral population, to determine the magnitude of genetic differentiation and the genetic relationships among the populations. Significant deviations from Hardy–Weinberg equilibrium were detected in one or more populations for eight of the nine protein loci with codominant alleles, and for microsatellites for all except the two Sri Lankan populations and for all but four loci. For both marker types, average inbreeding coefficients (FIS) were exceptionally high. Heterogeneity of deviations from Hardy–Weinberg equilibrium for the microsatellites showed no differences for among loci within populations as compared with among populations within loci. For protein loci, however, the former was higher, indicating selection affecting allele frequencies at some loci. The variance among protein loci was significantly higher than among microsatellite loci, further indicating selection at some protein loci. There was significant differentiation among populations for both protein and microsatellite loci, most likely reflecting the geography of south‐east Asia, and the presumed spread of goats throughout the region. Phylogenies derived from pair‐wise genetic distance estimates show some similar clustering for the microsatellite and protein based trees, but bootstrap support was generally low for both. A phylogeny based on the combined set of 38 protein and microsatellite loci showed better consistency with geography and higher bootstrap values. The genetic distance phylogeny and the Weitzman diversity tree derived from microsatellite data showed some identical clusters, and both identified the Ujung Pandang and Australia populations as contributing most to overall genetic diversity.

Reaction norms across seven constant and one fluctuating temperature of development were measured for thorax length and several wing size traits for up to 10 isofemale lines of each of the cactophilic Drosophila species, D. aldrichi and D. buzzatii, originating from the same locality. Maximum thorax length was reached at different low to intermediate temperatures for the two species, whereas wing length was highest at the lowest temperature in both species. Various ratio parameters showed pronounced species differences. The reaction norm for the wing loading index (wing length/thorax length) decreased monotonically with temperature in both species, but was much steeper and spanned a wider range in D. aldrichi than in D. buzzatii, suggesting either that wing loading is not a good characterization of flight capacity or, more likely, that flight optimization does not occur in the same manner in both species. The vein ratio (distal length/proximal length of the third vein) increased with temperature in D. buzzatii but decreased in D. aldrichi. Wing development in the two species thus is very different, with the proximal part of the wing in D. buzzatii more closely allied to the thorax than to the distal part. Among line variation was significant for all traits in both species, and most pronounced for thorax length and the ratio parameters. Coefficients of variation were significantly different between the species for all traits, with those in D. aldrichi higher than in D. buzzatii. Genetic variance in plasticity was significant for all traits in D. buzzatii, but only for seven out of 12 in D. aldrichi. Additive genetic variances for all traits in both species were significantly larger than zero. Genetic correlations between thorax length and several wing length parameters, and between these and wing area, were positive and generally significant in both species. The genetic correlation between the distal and the proximal length of the third vein was not significantly different from zero in D. aldrichi, but negative and significant in D. buzzatii. Heritabilites varied significantly among temperatures for almost all traits in both species. Phenotypic variances were generally higher in D. aldrichi than in D. buzzatii, and commonly highest at the extreme temperatures in the former species. At the high temperature the genetic variances also were usually highest in D. aldrichi. The data clearly suggest that the process of thermal adaptation is species specific and caution against generalizations based on the study of single species.

In this study we describe an auditory processing front-end for missing data speech recognition, which is robust in the presence of reverberation. The model attempts to identify time-frequency regions that are not badly contaminated by reverberation and have strong speech energy. This is achieved by applying reverberation masking. Subsequently, reliable time-frequency regions are passed to a 'missing data' speech recogniser for classification. We demonstrate that the model improves recognition performance in three different virtual rooms where reverberation time T60 varies from 0.7 sec to 2.7 sec. We also discuss the advantages of our approach over RASTA and modulation filtered spectrograms.

No abstract