Potential of in-plant intramuscular fat predictions to enable sheep breeders to incorporate consumer preferences in breeding programmes.

The inclusion of eating quality traits in sheep genetic improvement programmes is desirable. Intramuscular fat (IMF) plays a key role in ensuring consumer satisfaction when eating lamb, but genetic progress for IMF is constrained by a lack of routine data collection. This study investigated the potential for IMF predictor traits to substitute for measured IMF in genetic improvement programmes. Carcass and predicted IMF (near-infrared estimated IMF and marbling score) data were available on 10,113 New Zealand lambs, 1678 of which also had measured chemical IMF on a slice of M. longissimus lumborum on which the predictions of IMF had been made. Genetic antagonisms were observed between carcass lean traits and IMF. The genetic correlation between the predictors and measured IMF approached one, indicating that predictors of IMF can be used in genetic improvement programmes. Through using selection indexes, simultaneous increases in IMF and the existing terminal selection index are possible, provided all traits are measured. This study highlights the importance and potential of predicted IMF to achieve genetic improvement in traits of importance to consumers.

Evaluating the performance of a miniaturized NIR spectrophotometer for predicting intramuscular fat in lamb: A comparison with benchtop and hand-held Vis-NIR spectrophotometers.

This study compares a miniaturized spectrophotometer to benchtop and hand-held Vis-NIR instruments in the spectral range of 900-1700 nm for prediction of intramuscular fat (IMF) content of freeze-dried ground lamb meat; and their ability to differentiate fresh lamb meat based on animal age (4 vs 12 months). The performance of the miniaturized spectrophotometer was not affected by sample temperature equilibration time. Partial Least Square regression models for IMF showed Rcv2 = 0.86-0.89 and RMSECV = 0.36-0.40 values for all instruments. Day-to-day instrumental variation adversely affected performance of the miniaturized spectrophotometer (R2p = 0.27, RMSEP = 1.28). This negative effect was overcome by representing day-to-day variation in the model. The benchtop spectrophotometer and miniaturized spectrophotometer differentiated lamb meat by animal age. The miniaturized spectrophotometer has potential to be a fast, ultra-compact and cost-effective device for predicting IMF in freeze-dried ground lamb meat and for age classification of fresh lamb meat.

Application of Hyperspectral imaging to predict the pH, intramuscular fatty acid content and composition of lamb M. longissimus lumborum at 24h post mortem.

Cost-effective, rapid and objective measurement of lamb quality on a routine basis is an important step for lamb value chains wishing to manage lamb product quality. Hyperspectral imaging (HSI) technology has shown promise as a solution for objective non-invasive prediction of meat quality. The performance of HSI applied 24h post mortem to lamb M. longissimus lumborum (LL) within a processing plant environment was assessed over two sampling years to evaluate its suitability for an objective lamb meat quality assurance tool. Calibration and validation steps were undertaken to evaluate HSI prediction performance for predicting fatty acid content and composition (n=1020 lambs) and pH (n=2406 lambs). Practical considerations of reference meat quality data quality and validation strategies are discussed. HSI can be used to predict meat quality parameters of lamb LL with varying accuracy levels, but ongoing calibration and validation across seasons is required to improve robustness of HSI for objective non-invasive assessment of lamb meat quality.

Gas characteristics before and after biofiltration treating odorous emissions from animal rendering processes.

Studies of odour-control biofilters at two animal rendering plants were conducted. Biofilter influent and effluent gases were characterized using a gas chromatograph-mass spectrometer (GC-MS) and a GC fitted with an odour sniffing port. Overall odour-removal performance of the biofilters was evaluated using forced-choice dynamic-dilution olfactometers. GC-MS analysis revealed that the biofilter influent gases at each plant contained about 300 compounds. About 20% of these compounds were identified, and included hydrocarbons, ketones, heterocyclics, aldehydes, aromatics, carobxylic acids, esters, sulphur compounds, nitriles, an amine and an amide. About 20 identified compounds were present in both biofilter influent gases. The influent gas at Plant 1 contained more straight chain hydrocarbons, while the influent gas at Plant 2 contained more sulphur compounds and esters. Sensory evaluation of the gas samples using the GC-odour port technique indicated that about 45 compounds in the biofilter influent gases were odorous at Plant 1, and about 30 were odorous at Plant 2. A variety of odour characters was observed in the two biofilter influent gases. The biofilters reduced the concentration of most of the odorous compounds to levels that were not detectable by the GC-odour port analysis. The rendering process gases had odour concentrations of between 59,000 and 1,000,000 OU m(-1). The biofilters reduced the odour concentration by 82-99% and also reduced the "offensiveness" of the odour. Uneven gas distribution affected biofilter odour-removal performance.

A survey of the natural occurrence of Fusarium mycotoxins in cereals grown in New Zealand in 1986-1989.

Fusarium mycotoxins, principally trichothecenes, occurred commonly in grain from crops in the North Island of New Zealand, but were much less common and also at the much lower levels in grains from South Island regions. The principal contaminants were trichothecenes of the nivalenol (NIV) and deoxynivalenol (DON) types. Trichothecenes derived from scirpentriol and T-2 tetraol were not common. Moniliformin occurred very rarely, while zearalenone contamination was not uncommon, but the levels were generally low. Maize kernels were commonly contaminated by moderate levels of both NIV- and DON-type trichothecenes, with levels up to 3.6 mg/kg and 11.95 mg/kg respectively recorded. The occurrence of NIV-type trichothecenes as a general contaminant in the range of 0.3-0.8 mg/kg, and frequently as the main contaminant, is unusual.

Carcinogenicity of bracken fern (Pteridium esculentum) in New Zealand.

Bracken fern (Pteridium esculentum) was harvested from two sites LB and TB one of which (TB) was on a central North Island New Zealand farm where bovine enzootic haematuria (BEH) was known to occur. The fern was dried, ground and incorporated (25% w/w) into a pelleted diet and fed to female rats for a total of 162 days. Fifteen weeks later when the rats were autopsied it was found that numerous tumours, mainly of the ileum and urinary bladder were present in the animals fed the bracken fern from the TB site. Neoplasms were found in 85% of rats from the TB group compared with 11% in the LB group while only a single tumour (a haemangioma of the uterus) was observed in the controls. In all, there were neoplasms in 42, 5 and 1 organ/s from the TB, LB and control rat groups respectively (p<0.001). Analysis of the fern and pellets for ptaquiloside, the carcinogen in bracken fern, showed much higher levels present in the material from the TB site. There was 26 and 2270 microg ptaquiloside/g of dried fern and, for pellets from the same fern, 6.5 and 355 microg ptaquiloside/g of pellets, for one collection from the LB and TB sources respectively.