Abstract. A heifer developed severe signs of acute gastrointestinal irritation 48 hr after ingesting fresh leaves of Colchicum autumnale growing on a damp meadow. Confirmation of the suspected toxicosis was obtained by detecting colchicine in serum and urine using liquid chromatography coupled with tandem mass spectrometry using atmospheric pressure chemical ionization. Although the serum colchicine concentration had declined to an apparently nontoxic level of 2.4 ng/ml, a more prominent concentration (640 ng/ml) indicative of colchicine poisoning was detected in the urine. This finding is consistent with the known toxicokinetic properties of colchicine, whereby a large volume of distribution results in low circulating blood concentrations and prolonged urinary excretion.

phase liquid chromatography (Hwang and others 2002), disclosed another 300 (36) mg/kg of HCN equivalents. Because the sample had been stored and shipped at room temperature in containers that were not tightly closed, it is likely that the original material contained even higher HCN levels and that much of this toxic gas had been lost from the specimen before analysis. The 2 kg or more of toxic feed ingested by each affected cow translates to at least 820 mg of HCN equivalents, approaching the reported lethal dose of 2 mg/kg bodyweight (Radostits and others 1999).The characteristic clinical signs, in conjunction with the consumption of appropriate plant material, are comparable to intoxication due to cyanogenic glycosides. As noted previously (Cran 1985, Roder 2001, Suter 2002, confirmatory HCN detection is a difficult and tedious procedure. Although toxic levels of HCN were detected in the present case, the volatility and rapid breakdown of HCN, combined with the time it normally takes for the samples to be received and analysed, may invariably lead to false-negative results. Therefore, in the search for an alternative approach, the authors took advantage of a more stable metabolite that can easily be monitored. The majority of HCN (80 per cent) is detoxified by the enzyme rhodanese in the liver, producing a thiocyanate metabolite that is excreted mainly through the kidneys with a half-life of up to three days in human beings (Gurnsey and others 1977, Erdman 2004). In the present study, the serum thiocyanate level was determined in blood samples obtained from three surviving cows 24 hours after the initial deaths. The procedure (Lundquist and others 1979) involved an ion exchange clean-up to remove interfering matrix components. After elution with sodium perchlorate, the thiocyanate concentration was determined by a quantitative colorimetric reaction in the presence of sodium hypochlorite and barbituric acid, yielding a derivative with peak absorption at a wavelength of 580 nm. The thiocyanate concentrations in the cows (145, 209 and 211 µmol/l) exceeded the reference range determined for human blood (15 to 70 µmol/l) and reached the same high concentrations as those found in severe cases of cyanide poisonings in human beings (Singh and others 1989). A healthy cow in the neighbourhood had a serum thiocyanate level of 52 µmol/l, suggesting a similar normal range in cattle to that in humans. This view was supported when, one month later, the thiocyanate serum concentration of the three cows was measured again and the values were 50 54, and 60 µmol/l. It was concluded that the conversion of HCN to thiocyanate provides a potentially useful diagnostic marker to confirm suspected exposures to cyanide or cyanogenic glycosides in

not available