IMPORTANT LEGAL NOTICE - The information on this site is subject to a disclaimer and a copyright notice.
European Flag    Europa The European Commission Research Agriculture FAIR
index by sub-area Quality policy
index by country United Kingdom

Identifying and changing the qualities and composition of meat from different European sheep types which meets regional consumer expectations

Contract nr: FAIR-CT96-1768
Project nr: 1768
Project type: SC
Starting date: 01/01/1997
Duration: 36 months
Total cost: 1,831,000 EUR
EC Contribution: 1,647,000 EUR
Scientific Officer: Michel VAN DEN BOSSCHE
Research topic: Consumer's expectations
Acronym: OVAX

The carcass and meat quality of lamb types produced across the EU Member States are diverse. Types vary from the small, milk-fed lambs in the Mediterranean regions, which have pale, delicately flavoured meat, to the larger, fatter, more robustly flavoured type produced in more northerly areas. Although consumers of food products may generally exhibit preferences for varieties which are traditional and available locally, there are undoubted trends towards a more catholic attitude towards food among European consumers. The regional acceptability of different qualities of lamb meat has implications for its trade both within the EU and globally.

1. To characterise the diversity of sheepmeat from regional production systems across Europe, in terms of organoleptic properties and composition.
2. To identify the quality characteristics of sheepmeat preferred by trained sensory assessors and consumers in different European regions, and to define those regional needs.
3.To develop a rapid, objective, analytical method to classify sheep carcass meat according to its geographical origin, composition and sensory traits.
4. To establish the extent of change in sheepmeat quality attributes which can be manipulated through dietary means and different age or weight slaughter end-points.
5 To evaluate the implications for human health of producing a diverse range of sheepmeats, with particular reference to the fatty acid composition of lean tissue.

This project aimed to establish, on a scientific basis, the ratings of particular sensory qualities of different lamb meats awarded by assessors in different Member States tasting both home-produced and imported meat samples.
This was accomplished by each of the six Partners, extending from Iceland in the north to Greece in the South. Each procured nationally important, diverse, lamb types and exchanged samples for measurement and assessment by the other Partners. The organoleptic assessments were made both by a trained taste panel and a group of 36 families (consumers) in each country.
These assessments were underpinned by instrumental meat quality measurements, with particular emphasis on the fatty acid composition of the tissues and the characterisation of flavour and texture.
A rapid analytical method of 'fingerprinting' quality in the different lamb types was developed and applied.
Finally, the extent of change in meat quality which can be attained via dietary manipulation of lambs, or by changing slaughter age, was explored in order to produce meats of desired quality for specific markets.

Current situation/results:
One hundred and twenty lambs of each of two nationally important types were produced in each of the following countries in Phase 1 of the project: France, Greece, Iceland, Italy, Spain and the United Kingdom. They ranged from milk-fed lambs with 5kg carcass weight, through concentrate-fed lambs of intermediate weights, to grass-fed lambs with 30 kg carcass weight. This protocol was repeated in Phase 2 (not in France) but with lambs which differed from those in the first phase in one or more of the basic components of a `system` (integrated breed/feed/slaughter point). These lambs varied less in weight (11 -21kg carcass). Samples of meat from each lamb type were distributed to all partners.

Carcass dissection of a proportion of the lambs showed that leanness ranged from 55 to 63%, fat from 12 to 28%. The haem iron concentration and the proportion of insoluble collagen in the longissimus muscle were correlated with age of the lamb. However, there was no clear relation between collagen solubility and instrumental texture. There was good agreement between taste panels in different countries in their rating of several attributes on intensity scales, and combined panel means showed that tenderness, juiciness and several key flavours (but not odours) were significantly different between lamb types.

Multivariate analysis of these data showed that lambs reared principally on grass were grouped separately from concentrate-fed lambs, with milk-fed lambs forming a third grouping, even further away along an axis which mainly contrasted specific flavours. These differences between feeds were also evident in the fatty acid composition of the muscle, with grass-fed lambs having higher concentrations of linolenic and its longer chain n-3 derivatives than the concentrate-fed lambs which, in turn, had higher proportions of linoleic acid and its n-6 derivatives. The milk-fed lambs had high concentrations of n-6 and the longer chain n-3 fatty acids.
Analysis of variance of the consumer hedonic scores for texture, juiciness, flavour and overall liking showed there was significant variation attributable to country (of family assessors), lamb type and country x type interaction. However, much of the variation in this interaction was quantitative and analysis of the degree of agreement between rankings of lamb types within country showed that for texture and juiciness there was good agreement, less so for overall liking and poorest for flavour. Flavour preferences reflected, to some extent, accustomed lamb types. This was most marked in the first phase when consumers from more southerly situated countries, where milk-fed young lamb is traditionally consumed, rated highly this type or, in some cases, the concentrate-fed lambs, but rated poorly the grass-fed lambs. The consumers in the two more northerly countries gave low ratings to the milk-fed lambs although there was strong liking for some concentrate-fed and grass-fed lambs.

Novel methods using an adsorbent carrier for dissolved adipose tissue were devised to obtain successful ionic counts and reduce contamination of the instrumentation in the Curie point pyrolysis-mass spectrometry (Cp-PyMS). 120 samples were analysed from the Phase 1 lambs only. After processing the data, information for classifying the lambs according to type, geographic origin or feed consumed was extracted using discriminant factorial analysis. Lambs could not be classified by type or geographic origin but they could according to diet (forage, concentrate, milk) with 92.3% correctly classified using non-linear techniques.

Cp-PyMS was used for muscle samples to formulate predictions of three selected important quality attributes: lipid content, total collagen content and texture. The number of mass fragments had to be reduced to facilitate satisfactory predictions using neural networks, with further steps in the modelling involving subdivision of samples into learning, validation and test sets. Predictions were generally accurate with correlations between predicted and actual values of r= 0.78 (texture) to 0.89 (lipid). This demonstrates the potential of the technique to predict quality which may be commercially attractive if it could be broadened to include other compositional and quality aspects.

The volatile compounds extracted from subcutaneous fat samples of each lamb type in Phase 1 using a dynamic headspace method were separated by high resolution gas chromatography, then identified and quantified by mass spectrometry. The resulting profiles were related to the type of feed consumed by the lambs. Several chemical families were represented in the 114 compounds quantified. Principal component analysis showed that, with the exception of one sample (from a set of 30), `forage' and `concentrate' could be clearly separated. The `milk' samples were generally not separable from the `concentrate' samples and specific compounds were identified with the milk/concentrate and forage groups.
Cross-correlation between sensory, consumer, instrumental texture, composition and volatile data sets is being completed.


Alan Victor FISHER
University of Bristol
Senate House
UK-BS18 7DY Langford - Bristol
Tel: +44 1179 28 92 71
Fax: +44 1179 28 93 24

  • Philippe BERGE
    I.N.R.A. - Centre de Clermont-Theix
    F-63122 Saint-Genes-Champanelle
    Tel: +33 4 73 62 41 65
    Fax: +33 4 73 62 40 89

  • Dimitrios ZYGOYIANNIS
    Aristotelian University of Thessaloniki
    GR-540 06 Thessaloniki
    Tel: +30 31 99 99 54
    Fax: +30 31 99 98 92

    Agricultural Research Institute
    IS-112 Reykjavik
    Tel: +354 577 10 10
    Fax: +354 577 10 20

  • Carlos SANUDO
    Universidad de Zaragoza
    C/ Miguel Servet 177
    E-50013 Zaragoza
    Tel: +34 97 676 10 00
    Fax: +34 97 676 16 12

    Università degli studi di Udine
    Via San Mauro 2
    I-33010 Pagnacco (Udine)
    Tel: +39 043 265 01 10
    Fax: +39 043 266 06 14
  Search Top