The number of piglets in litter is an important trait to achieve economic success. Different breeds vary by litter size; breeders must carefully select breeds to realize heterosis. To select breeding gilts, it is important to consider all information a breeder has because the characteristic data about gilts are limited.
The aim of the research was to analyse different factors affecting the litter size of gilts.

2389 gilts were raised in 41 farms over Estonia at 1998 to 2003. The gilts average litter size at birth was 10.45 piglets and length of pregnancy 115.78 days. The gilts were inseminated at 180 to 290 days of age. Paterson (1989) recommended management strategy, by witch gilts are mated at about 200 days of age with a body weight of >100 kg.

Table 1. Characterization of the analyzed dataset (n = 2389)

Dataset was obtained from of Animal Recording Centre and included breed, sex, birth and testing date, weight, backfat thickness, area of loin eye and lean meat percentage, insemination and fertility data on gilts and their parents which was collected by PC program DB-Planer.
Gestation length was divided into three classes - 110…114, 115…117 and 118…122 days.
Meat traits were measured by ultrasonic equipment Piglog 105. Meat traits recorded were: backfat thickness at last (x1) and 11...12^th (x3) rib, 7 cm from midline (mm), and diameter of loin eye (x2), 7 cm from midline (mm). Lean meat percentage (y) was calculated using the formula (Piglog 105, 1991).
Calculating the effect of breed combination and gestation length on litter size of gilts, the following general linear model (GLM) was used (SAS, 1991):

Y= dependent variable;
μ = general mean;
T_i = breed combination (n=1...5);
M_j = insemination year (n=1...6);
K_k = insemination season (n=1...4);
S_e = gestation length classes (n=1...3);
e_ijkem = random residual effect

Calculating the effect of insemination traits and gilt breed on gestation length, the following GLM model was used:

Y_ijkem=μ+T_i+M_j+K_k+S_e+A_m+P_n+e_ijkem,

Y= dependent variable;
μ = general mean;
T_i = gilt breed (n=1...3);
M_j = insemination year (n=1...6);
K_k = insemination season (n=1...4);
S_e = insemination method (n=1...2);
e_ijkem = random residual effect

Calculating the effect of breed and technician on meat traits of live gilts, the following GLM model was used:

Y_ijklemn=μ+T_i+M_j+K_k+S_e+A_m+P_n+e_ijklemn,

Y= dependent variable;
μ = general mean;
T_i = gilt breed (n=1...3);
M_j = technician (n=1...7);
K_k = test year (n=1...6);
S_e = test season (n=1...4);
F_l = farm (1...41);
W_m = weight at test;
e_ijklemn = random residual effect

The results are given as least-square means (Parring et al., 1997). Level of significances expressed conventionally: *** - P<0.001, ** - P<0.01, * - P<0.05, # - P<0.1. a, b, c … – least square, within each effect with one letter in common do not differ significantly.

Usually farmers feed gilts intensively to prepare them for lactation period. According to the results, this could slightly decrease the litter size on the first parity as backfat thickness and number of live piglets in litter are negatively correlated (Table 2). Rozeboom’s (1996) results agreed with these results, where body composition at first mating did not affect litter size of primiparous sows. As expected, meat traits were significantly correlated. Gilts with thin backfat had somewhat larger loin eye. Cleveland (1988) concluded, that selection for lean growth should have little effect on litter size, but may have beneficial effect on carcass traits.

Table 2. Phenotypic correlations between meat and fertility traits

Due to heterosis effect, achieved through crossbreeding, significantly larger litters were found on EL and ELW gilts crossed with white boars (Tabe 3). Number of piglets in purebred litters of EL and ELW gilts was about same – 10.43 and 10.36 respectively. Significantly smaller litters were found in purebred P gilts.
Gestation length does not affect litter size significantly, although somewhat larger litters were found on gestation length 115...117 days.
Gestation length was highly influenced by insemination method, being longer in artificial insemination (Table 4). Significant breed effect was calculated on gestation length as well. Pietran gilts had shorter gestation period (115.03 days) than white breeds, whose gestation was shorter in Estonian Landrace breed (115.66 days). Gilts, inseminated in fall, had much longer gestation, than those inseminated in other seasons and shortest gestation was found in spring and summer.

Table 3. Effect of breed combination and gestation length on litter size of gilts

Table 4. Effect of insemination traits and gilt breed on gestation length

Gestation length increased from 114.82 days in 1998 to 116.03 days in 2003. One reason for this could be wider use of artificial insemination of pigs.
Pietran and Estonian Landrace gilts, whose lean meat percentage was over 6% had superior meat quality (Table 5).

Table 5. Breed effect on meat traits of gilts

Landrace pigs had thinner fat, but showed modest results in diameter of loin eye, compared with Pietrain gilts who were a little fatter, but with large loin eye. The worst results were shown by Estonian Large white breed. Superiority of Landrace breed was found also by Tummaruk et al. (2000).
Lean meat percentage, measured by different technicians, differed about 4% (Table 6).

Table 6. Technician effect on meat traits of gilts

Technician F measured the thinnest backfat and the smallest diameter of loin eye. Fatter were the gilts with technician C and loin eye was larger with technicians A and D.

Meat traits (backfat thickness and diameter of loin eye) do not affect litter size of gilts and they are not closely related to each other.
Large difference between white and colour breeds on litter size, whereas crossing white breeds give larger litters in the first parity.
Insemination method and gilt breed affect gestation length highly. Gestation length increases year by year.
Differences between colour meat type breeds and white breeds will decrease.

Cleveland, E.R., Johnson, R.K., Cunningham, P.J. 1988. Correlated response of carcass and reproductive traits to selection for rate of lean growth in swine. J. of Animal Sci. 66 (6):1371...1377.
Piglog 105. 1991. Piglog 105 User’s Guide. Soborg, Denmark: SFK - Technology, 14 pp
Rozebom, D.W., Pettigrew, J.E., Moser, R.L., Cornelius, S.G., El-Kandelgy, S.M. 1996. Influence of gilt age and body composition at irst breeding on sow reproductive performance and longevity. J. of Animal Sci. 74(1):138...150
Paterson, A.M. 1989. Age at mating and productivity of gilts. Manipulating pig production II. Proceedings of the Biennial Conference of the Australasian Pig Science Association, Albury, NSW on November 27-29, 1989. 310...314
SAS. 1991. SAS User’s Guide: Statistics. SAS Inst. Inc., GARY, NC. 305 pp.
Tummaruk, P., Lundeheim, N., Einarsson, S., Dalin, A.M. 2000. Factors influencing age at first mating in purebred Swedish Landrace and Swedish Yorkshire gilts. Animal Reproducton Science. 63 (3/4):241...253.