INSTRUMENTAL PREDICTION OF SOME CARCASS P ARAMETERS OF PIGS AT A PROGENY TESTING STATION

The objective of the paper was to evaluate opportunities for instrumen tal assessment of pigs at a progeny testing station while 216 pigs of five genotypes were used. These characteristics were determined by an instrument PIGLOG 105 a day before slaughter: backfat thickness betwee n 3rd and 4th lumbar vertebrae (x(1)), backfat thickness between 3rd a nd 4th last rib (x(2)) and m.l.d, thickness between 3rd and 4th last r ib (x(3)) at a distance of 7 cm laterally from the central dorsal line . Tab. I shows the figures of the investigated parameters of carcass v alue in pigs determined in vivo and/or post mortem. The percentage of valuable lean cuts (VLC) was 49.15 in the whole set, while the muscle percentage determined instrumentally in vivo made 50.98. Tab. II shows correlation coefficients calculated for the whole set, and/or for the Large White breed (LW). The correlations between the values of backfa t thickness determined in vivo and the percentage of VLCs were at the level r = -0.61 to -0.76, lower values of the relations were calculate d between m.l.d thickness on the one hand and VLC percentage and/or mu scle percentage in vivo on the other (r = 0.24 to 0.30, and/or 0.28 to 0.31). The correlation coefficient between the muscle percentage in v ivo and dissected VLC percentage made r = 0.81. It should be emphasize d that the analyzed set of pigs (216 animals) showed the above-mention ed value of correlation coefficient but the coefficient of variance up silon had the value 8.8%, which can be explained by low homogeneity of the investigated set, and/or by the fact that the shape of the origin al regression equation of PIGLOG 105 instrument did not describe accur ately enough the meatiness of the investigated genotypes. This regress ion equation usable for the investigated set of pigs was constructed b y mathematico-statistical methods: Y-1 = 55.0603 - (0.2426 x(1)) - (0. 3956 x(2)) + (0.1459 x(3)), where Y-1 is a predicted percentage of VLC and x(1), x(2) and x(3) are the values of backfat thickness and/or mu scle thickness determined by in vivo method. Tab. III shows the accura cy of an estimate in both equations as well as their usability for the particular cases. Applying the equation for Y-1 we constructed, an es timate of VLC percentage ranged within the values +/-2% in 67.1% of th e cases while it was in 45.8% of the cases when the original equation of PIGLOG 105 instrument was used. Considering the results obtained as well as the findings from our former studies that checked this instru mental method, it is possible to recommend a simultaneous use of PIGLO G 105 instrument and ALOKA sonograph with an optical control of record ed values on a display, which makes the estimate of measured backfat t hickness and/or m.l.d thickness approach the reality maximally. So it would be possible in future to use the PIGLOG 105 instrument at indivi dual selection during performance tests and/or fattening performance a cid carcass value tests at Pig Progeny Testing Stations while dissecti ons would be replaced by instrumental evaluation (a definite number of control dissections would be performed all the same). We believe on t he basis of this experiment that the in vivo muscle percentage out of the 100kg weight should amount to 52-53% in dam populations while it s hould be above 55-56% in sire lines.