96 Original Paper Czech J. Anim. Sci., 50, 2005 (3): 96–103 Mainly for historical reasons, genetic evaluation for pigs is carried out separately for production and reproduction traits in most pig breeding as- sociations. As a rule, the animal model for produc- tion traits was developed first and reproduction traits were added later to genetic evaluation. Computer resources were limited in the starting phase of animal model based genetic evaluation, which was probably another reason for separate analyses. Recently, the joint evaluation of production and reproduction traits has become possible and should be used as the method of choice being a logical ex- tension. Joint genetic evaluation of production and reproduction traits has been implemented in the Czech Republic and in Slovakia since 1999 (Wolf et al ., 1999; Peškovičová et al ., 2002). Currently, in the Czech Republic routine testing has shifted completely from station to large-scale field testing with only a small number of animals being tested on station for monitoring the development of carcass and meat quality traits. Therefore, a revision of the models used and a new estimation of genetic parameters became necessary. MATERIAL AND METHODS The analyses were based on performance test data for the breeds Czech Landrace and Czech Large White. The time period considered was 1995 to 2004, both for production and reproduction traits. The analysed traits were as follows: LM – Lean mean percentage at the end of the per- formance test, estimated from ultrasonic meas- urements unadjusted for live weight Genetic parameters for a joint genetic evaluation of production and reproduction traits in pigs J. WOLF 1 , E. ŽÁKOVÁ 1 , E. GROENEVELD 2 1 Research Institute of Animal Production, Prague-Uhříněves, Czech Republic 2 Institute for Animal Breeding, Mariensee, Federal Agricultural Research Centre (FAL), Neustadt, Germany ABSTRACT: The covariance structure was estimated by REML for a joint genetic evaluation of production and reproduction traits for Czech Landrace (58 258 records) and Czech Large White (167 161 records) pigs using four- trait animal models. The following traits were analysed: lean mean percentage at the end of the performance test in the field, estimated from ultrasonic measurements unadjusted for live weight (LM), average daily gain in field test (in g/day) calculated as weight at the end of the test divided by age at the end of the test (ADG), number of piglets born alive in parity 1 (NBA1) and number of piglets born alive in parity 2 and subsequent parities (NBA2+). The heritabilities were in the range from 0.30 to 0.37 for LM, from 0.13 to 0.18 for ADG, from 0.09 to 0.13 for NBA1 and from 0.10 to 0.14 for NBA2+, depending on the breed and on the model (herd-year-season random or fixed). Genetic correlations between production and reproduction traits were estimated to be non-zero. Correla- tions between traits caused by the herd-year-season effect were mostly positive. As a general conclusion, the joint genetic evaluation of production and reproduction traits is recommended. The herd-year-season effect should be preferably considered as random. Keywords: pig; production traits; reproduction traits; multi-trait animal model; genetic parameters Supported by the National Grant Agency of the Ministry of Agriculture of the Czech Republic (Project QC1235 NAZV) and by Project No. 2/04 of German-Czech Cooperation in the field of agricultural research.