Phylogeny of Glaucosomatidae inferred from molecular evidence.

Most species of glaucosomatids (Teleostei: Glaucosomatidae) are endemic to Australia, except Glaucosoma buergeri that is widely distributed from Australia to Japan. This study elucidated phylogenetic relationships among glaucosomatids based on the morphological characters of the saccular-otolith sagitta, in addition to molecular evidence of mitochondrial 16S rDNA, cytochrome oxidase I (COI) and cytochrome b (cyt b) sequences, and nuclear rhodopsin sequences. The topologies of individuals' phylogenetic trees, based on 16S rDNA, COI and cyt b sequences, were statistically indistinguishable from one another, and were only slightly different from a tree based on rhodopsin sequences. These molecular tree topologies, however, differed from species relationships in morphology-based phylogenetic hypothesis proposed in previous studies. Specimens of G. buergeri from Australia and Taiwan showed differences in the sagitta and molecular differentiation at the four genes, suggesting a possible speciation event. Both molecular and morphological evidences indicate that Glaucosoma magnificum is the plesiomorphic sister species of other glaucosomatid species. Glaucosoma hebraicum is the sister species of a clade composed of G. buergeri and Glaucosoma scapulare. Molecular and morphological evidences also support the species status of G. hebraicum.

Spectral analysis of bubble sounds in decompressed guinea pigs.

Doppler ultrasonic monitoring was used to detect bubbles in sixteen guinea pigs subjected to a simulated air-dive profile of 9 ATA for 25 minutes. After completion of the decompression, eight subjects did not demonstrate any signs of decompression sickness (DCS), two developed paraplegia and six died. Under sedation, the Doppler ultrasonic bubble detector was placed precordially to record the bubble signals in both pre-dive and post-dive. The recorded signals were analyzed with a high resolution signal analyzer to compare changes between pre-dive and post-dive spectrograms. Bubble signals appeared in the frequency range between 0.64 +/- 0.02 KHz and 6.08 +/- 0.30 KHz. In terms of the net spectral level versus frequency areas (difference between pre-dive and post-dive spectrograms), the mean areas in the paraplegia group and the death group were significantly larger than those in the group without DCS. Therefore, we concluded that the spectral analysis may be an objective and quantitative adjunctive method to the interpretation of Doppler bubble signals.