The increase in reports of novel diseases in a wide range of ecosystems, both terrestrial and marine, has been linked to many factors including exposure to novel pathogens and changes in the global climate. Prevalence of skin cancer in particular has been found to be increasing in humans, but has not been reported in wild fish before. Here we report extensive melanosis and melanoma (skin cancer) in wild populations of an iconic, commercially-important marine fish, the coral trout Plectropomus leopardus. The syndrome reported here has strong similarities to previous studies associated with UV induced melanomas in the well-established laboratory fish model Xiphophorus. Relatively high prevalence rates of this syndrome (15%) were recorded at two offshore sites in the Great Barrier Reef Marine Park (GBRMP). In the absence of microbial pathogens and given the strong similarities to the UV-induced melanomas, we conclude that the likely cause was environmental exposure to UV radiation. Further studies are needed to establish the large scale distribution of the syndrome and confirm that the lesions reported here are the same as the melanoma in Xiphophorus, by assessing mutation of the EGFR gene, Xmrk. Furthermore, research on the potential links of this syndrome to increases in UV radiation from stratospheric ozone depletion needs to be completed.

This study investigates potential causes of a novel blister-like syndrome in the plating coral Echinopora lamellosa. Visual inspections of this novel coral syndrome showed no obvious signs of macroparasites and the blisters themselves manifested as fluid-filled sacs on the surface of the coral, which rose from the coenosarc between the coral polyps. Histological analysis of the blisters showed that there was no associated necrosis with the epidermal or gastrodermal tissues. The only difference between blistered areas and apparently healthy tissues was the presence of proliferated growth (possible mucosal cell hyperplasia) directly at the blister interface (area between where the edge of the blister joined apparently healthy tissue). No bacterial aggregates were identified in any histological samples, nor any sign of tissue necrosis identified. We conclude, that the blister formations are not apparently caused by a specific microbial infection, but instead may be the result of irritation following growth anomalies of the epidermis. However, future work should be conducted to search for other potential casual agents, including viruses.

Published reports on the value of sodium valproate (Epilim) as an anticonvulsant drug have been encouraging.' Acquired platelet function defects ofdoubtful significance have been found and attributed to this drug,2 and there have been two reports of haemorrhages, prolonged bleeding times, and thrombocytopenia in patients on sodium valproate with other anticonvulsant drugs.3 4 The patient we report here developed thrombocytopenia when taking sodium valproate alone, and the investigations are recorded.Case report A 6-year-old boy weighing 20 kg had been having major and minor seizures for two years and had failed to respond to other anticonvulsant drugs. On treatment with sodium valproate 400 mg four times a day, all other drugs having been stopped, he was rendered fit-free for eight months. His mother then reported that he had developed hair loss and multiple spontaneous bruises. The DiscussionIn a further 32 of our patients on sodium valproate, including several who were on large doses and had high serum levels of the drug, and five who also complained ofhair loss, no other cases ofhaemorrhage or platelet counts below 100 x 109/l were detected.Apart from the sodium valproate with the very high serum level in our patient we found no other cause for the thrombocytopenia, which we therefore considered was probably due to the drug. The finding of increased numbers of megakaryocytes in the bone marrow, together with the rate ofplatelet recovery on discontinuing the drug, is suggestive of an immunological mechanism, though circulating platelet antibodies could not be detected in vitro. The negative results obtained during the thrombocytopenic phase may be explained by consumption of the antibody in vivo, though this mechanism cannot account for the negative findings in the recovery phase. A further possibility for the negative in-vitro results is that the circulating antibody is developed to a metabolic product of sodium valproate and not to the drug itself. It was considered unjustifiable to subject our patient to a further test dose of the drug, but it is of interest that in one of the previously reported cases of thrombocytopenia3 the platelet count recovered on stopping the sodium valproate but immediately fell on recommencing the drug.As a result of these findings we recommend that any excessive or unexplained bruising or bleeding in a patient on sodium valproate is an indication for immediate platelet assessment.We are grateful to the Regional Blood Transfusion Centre, Sheffield, for the platelet antibody studies. In April 1976 he developed an acute exacerbation of his bronchlitis accompanied by cor pulmonale. He failed to respond to treatment and died four days after admission to hospital. Necropsy examination was not performed. DiscussionIn view of the disappearance of this man's bovine cough at the same time as the improvement in his other Parkinsonian features during treatment with levodopa and benserazide, it is difficult to come to any other conclusion than that the bovine cough was the presenting featu...