Changes in volatile compounds were monitored in whiting (Merlangius merlangus), cod (Gadus morhua) and mackerel (Scomber scombrus) and related to spoilage. Data are presented from headspace/mass spectrometric (HS/MS) analysis and solid-phase microextraction/gas chromatographic/mass spectrometric (SPME/GC/MS) analysis at two time points (day 0 and day 10) during storage at 4 • C. HS/MS revealed 24 ions that could be used as markers of spoilage. SPME/GC/MS identified 86 compounds, 20 of which could perhaps be used to characterize freshness: principally alcohols, ketones, aldehydes and C 2 -C 11 esters. Compounds common to the three species studied appear to be generated by microbial degradation.

Trimethylamine (TMA) and the total volatile basic nitrogen (TVBN) were determined in 169 samples of sea-fish (herring, cod, whiting, and mackerel) at all stages of decomposition. The comparison of these two parameters with the ratio P=TVBN/TMA (%) showed that P provides a useful index of freshness. It is relatively constant between species, its dispersion is less than that of TMA and it increases more rapidly than TVBN at the start of decomposition. Statistical analysis of the experimental results showed that there is a non-linear correlation between P and the decomposition index (i). The comparison of the line-equations defining log P as a function of i for different species of marine Teleostei led us to show that there is little or no intraspace variation of this correlation under determined conservation conditions. It is noted that for a given species, this correlation is temperature-dependant. Statistical exploitation of these results permit us to determine the maximal admissible values for P which can be used as a support for the development of new standards.

A simple, rapid and inexpensive method is proposed for determination of trimethylamine (TMA) in fish muscle. This procedure includes a deproteinization step with trichloroacetic acid (TCA) followed by blocking of primary and secondary amines using formaldehyde at alkaline pH and finally steam distillation of TMA. No statistically significant differences were found between this new optimized procedure and either the Conway microdiffusion method or the colorimetric method. Using the technique proposed here it is possible to assay both the total volatile basic nitrogen (TVBN) and the TMA in less than 30 min.

Freezing is an efficient way of storing fish. Objectively though, it is very hard to determine whether a fish has been previously frozen. Following an appraisal of various methods, we selected a physical determination (torrymeter), a physiological examination (eye lens) and three enzymatic assays (a-glucosidase, b-N-acetylglucosaminidase and b-hydroxyacyl-CoA-dehydrogenase) and applied them to three species: plaice (Pleuronectes platessa), whiting (Merlangus merlangus) and mackerel (Scomber scombrus). We also compared the results obtained following slow and rapid freezing and investigated how spoilage affects the torrymeter measurements and a-glucosidase assay values. For whole fish the physical method using the torrymeter is a reliable indicator. For fish fillets we recommend the enzymatic method using the a-glucosidase assay, which should be accompanied by measurement of the freshness to avoid confusing a frozen-thawed fish and a fish in an advanced stage of spoilage. The values noted for fresh and thawed whiting and plaice indicated cut-off values of 0.15 for whiting and 0.5 for plaice, above which it can be asserted that the sample had been frozen.

One hundred one strains of Listeria monocytogenes isolated from seafood and cheese industry samples and from patients with listeriosis were assessed using a microtiter plate method for adhesion to polystyrene and stainless steel surfaces. The adhesion rate for these strains ranged from 3.10 to 35.29% with an inoculum of 8 x 10(8) cells per well. A strong correlation was found between adhesion to polystyrene and stainless steel microtiter plates, indicating that the intrinsic ability of L. monocytogenes to adhere to inert surfaces is stronger than the influence of the surface's physicochemical properties. The clinical strains were less adherent to inert surfaces than were the industrial strains. By integrating other factors such as location of the industrial strains, contamination type of the clinical strains, serotype, and pulsotype into the analysis, some weak but significant differences were noted. For the industrial isolates, the number of cells attached to both surfaces differed significantly depending on whether they were isolated from food or food-processing environments in the seafood and cheese industry. For clinical isolates, sporadic strains exhibited greater adhesion to polystyrene than did epidemic strains. Strains belonging to the pulsed-field gel electrophoretype clusters A and M (lineages II and I, respectively) were less able to adhere to polystyrene and stainless steel than were strains in the more common clusters.