The aim of the study was to propose and test formulae of feeding mashe
s for the fattening of turkeys, in which extracted soybean meal and ma
ize were fully or partially replaced by rapeseed, pea and wheat. The e
xperiment was carried out on 1 200 turkeys (both sexes) of two genotyp
es. They were housed in a windowless house on a deep litter with contr
olled environment according to the standard technological procedure of
the International Poultry Testing Station at Ustrasice and tested for
a period of 84 days. Within each genotype the turkeys were divided in
to three groups: each group was placed in one box with 200 birds in ea
ch (6 birds/m(2)). All turkeys were fed ad libitum with three diets (T
ab. I). The mixture KR IT was used as standard. Live weight at the age
of 84 days, mortality, feed consumption, carcass weight, abdominal fa
t and breast and leg muscles weight were monitored. Traits measured on
individuals were analyzed according to model (1), traits measured on
box basis were analyzed according to model (2), where y(ijkl) is the v
alue of the l-th individual of the k-th sex, receiving the j-th diet a
nd belonging to the i-th genotype, y(ij) is the value of the box where
birds from the i-th genotype are housed receiving the j-th diet, mu i
s the general mean, g(i) is the effect of the i-th genotype, v(j) is t
he effect of the j-th diet, p(k) is the effect of the k-th sex, (gv)(i
j), (gp)(ik), (vp)(ik), (gvp)(ijk) are the appropriate interaction eff
ects, e(ijkl) is the residual effect for models (1) and (2), resp. All
calculations were carried out using the procedure GLM of SAS (SAS Ins
titute Inc., 1988). From monitored characters only live body weight wa
s significantly influenced by the diet. Turkeys of experimental groups
(2 and 3) of both genotypes did not reach such high live body weight
(6.13 and 5.99 kg) as those of control group (6.32 kg), fed by mash KR
IT (1) with maize and soybean meal (Tab. II and III). From the econom
ic evaluation point of view (Tab. IV) the turkeys of genotype 2 of exp
erimental groups 2 and 3 are more effective than those of control grou
p 1.