Study on Friesian Herds Raised in Egypt and Germany II . Genetic and phenotypic trends in estimated transmitting ability Summary

Original data consisted of 774 first lactation records of daughters of 124 sires of Friesian herds in Egypt and 9219 first lactation records of daughters of 679 sires of Friesian herds in Germany used to estimate genetic and phenotypic trends of initial milk yield in 70 days (IMY), 305-day milk yield (305-dMY), lactation period (LP) and calving interval (CI). Sire component of variance was consistently about 9.88%, 10.75%, 4.63% and 5.03% for IMY, 305-dMY, LP and CI, respectively of the total Variation in Friesian cows in Egypt and about 16.69%, 16.69%, 3.63% and 4.37%, for IMY, 305-dMY, LP and CI, respectively of the total Variation in Friesian cows in Germany. Annual genetic change was 29.98 kg, 112.99 kg, -0.92 d and 1.78 d for IMY, 305-dMY, LP and CI, respectively for Friesian cows in Egypt and was 70.41 kg, 200.38 kg, 0.12 d and -1.05 d for IMY, 305-dMY, LP and CI, respectively for Friesian cows in Germany. Annual phenotypic trend was 13.83 kg, 48.00 kg, -1.66 d and -1.82 d for IMY, 305-dMY, LP and CI, respectively for Friesian cows in Egypt and was 21.00 kg, 104.00 kg, 0.17 d and 0.78 d for IMY, 305-dMY, LP and CI, respectively for Friesian cows in Germany. The present results show that sires used in the later years were of superior genetic value than those used in the earlier years.


Introduction
Genetic and phenotypic changes in the Performance traits of dairy cattle are the ultimate indicator of progress in a herd.Such a change is measured as genetic and phenotypic trend.The measure of genetic and environmental changes from field data has problems since many factors are confound with both genetic and environmental changes (CANON and MUNOZ, 1991).Many studies showed more additive genetic control in the first lactation than later lactations.REGE (1991) reported that the accuracy of sire evaluation is increased by using more records than the first lactation.On the other side, WELLER et al., (1984) working on Holstein cows, reported that genetic trend for milk yield and fat corrected milk yield were decreased with parity.They concluded that it is useful to depend on the first lactation in estimating genetic trend.It, surely, depends on the expected length of the lifetime production of the cow.Sire evaluation on first lactation and initial milk yield (IMY) appears justified by the strong relationship between the yield of first lactation and the measures of lifetime production (HOQUE and HODGES, 1981;AGYEMANG et al., 1985 andBAFFOUR-AWUAH et al., 1996).In addition, the part lactation records could be very useful in early evaluation of sires, because of high genetic correlations between complete production records and initial milk yield (KHATTAB et al., 1993 andBAFFOUR-AWUAH et al., 1996).
Maximizing genetic gain is the objective of animal breeders.The most rapid genetic gain can be achieved by selecting and sampling young bulls, because younger animals are on average superior to the older animals.As genetic progress accelerates, the importance of young animals in breeding programmes increases.The objectives of this study were to: (1) estimate sire transmitting ability for initial milk yield (IMY), 305day milk yield (305-dMY), lactation period (LP) and calving interval (CI) for Friesian herds in Egypt and Germany, and (2) estimate genetic and phenotypic trends for these different traits.

Materials and Methods
The data used in this study were obtained from the milk records of Friesian cows in two locations.The first location was Dalla farm which is 130 kilometres from Cairo City, Egypt.The second location was Osnabrück in Germany, where data were collected from different farms.In Egypt data covered a period of 6 years, from 1987 to 1992.A total of 774 first lactation records of daughters of 124 sires were used.Each record included, cow, sire and dam identification, birth date, calving date, age at calving, parity, year, initial milk yield, 305-day milk yield, lactation period, days dry, calving interval, days open, and monthly milk yield (from first to tenth lactation month).The data was adjusted for year-season analysis.In Egypt, animals were kept loose under semi-open sheds all the year round.Cows were fed on alfaalfa and rice straw during the year with concentrate ration.Concentrates were offered twice daily before milking, about 4-5 kg, aecording to animals' body weight and production.Cows were machine-milked twice a day (at 7 a.m. and 4 p.m.).Cows producing more than 10 kg a day and those in the last two months of pregnancy were supplemented with extra concentrate rations.The nucleus of this herd was imported to Egypt from the USA as pregnant heifers in 1986.Also, imported semen every year from the USA was used in artificial Insemination (AI) at random.Heifers were served for the first time when they reached 24 months or 350 kg, therefore cows were usually served two months Postpartum.Pregnancy was detected by rectal palpation 60 days after the last service.In Germany, data was collected by VIT (Vereinigte Informationssysteme Tierhaltung, Verden) over a period of 15 years from 1979 to 1993.A total of 9219 first lactation records of daughters of 679 sires were used.Each record included the same division for Dalla farm data.Data originated from many farms in one sub-region in Osnabrück.It was the original data which was used for the estimation of the breeding values for the Friesian breed (Schwarzbunt).The animals were fed concentrates in addition to the grassland in spring, summer and autumn aecording to the Performance.In the winter they got concentrates and conserved feed.The data was adjusted for year-season analysis.Estimation of variance and covariance components depends mainly on Henderson's Method 3 (HARVEY, 1990).Accordingly, estimates of sire (CT 2 S ) and remainder (a 2 e ) components of variances and covariances were obtained.

Estimate of sire transmitting ability (ETA)
Sire transmitting ability (ETA) of different traits studied was estimated by using best linear unbiased prediction (BLUP) and was obtained from using the first lactation.In addition, 5 daughters per sire or more were examined.In matrix notation, the model can be written as: Y = Xf+Zs + Wb + e Where: Y was a vector of Observation for each trait, X was a known fixed design matrix, f was an unknown vector of fixed effect representing the mean and year of calving and month of calving, Z was a known design matrix, s was an unobservable vector of random sire effect, W was a vector of covariance variable (independent variable), b was a vector of regression of Y on W and e was unobservable random vector of error with mean zero and variance-covariance matrix I a 2 e.The mixed model equations (HENDERSON, 1973) are

W'Y
Where: K = (4 -h )/h , for each trait was added to the diagonal of sire effects in the matrix.The above analysis carried out to predict sire transmitting abilities (ETA) of sires was the sum of the sire Solution.The breeding value was twice the estimated transmitting ability.

Estimate of phenotypic and genetic trends
Annual phenotypic and genetic change for different traits studied was computed as the regression coefficients of trait values on the year of calving after adjusting the records for year and season.Trends in sires transmitting ability for different traits studied were estimated by regression ETAs for each year by year.For the p^ year, the average of sires transmitting abilities was ]>^,s, / n.p' where n ip is the number of daughters of sire / born in thep* year, and i, is the predicted transmitting ability of the z , th sire.

Random components
Estimates of variance components for sire (cr 2 s), cows within sires ((c 2 c:s) and residual effects (a 2 e) associated with IMY, 305-dMY, LP and CI for different parities are in (Table 1).For all these estimates, data were corrected for the systematic environment effects.Results obtained in this study show the possibility of the genetic improvement of milk traits through sire selection.ABUBAKAR et al., (1986); KHATTAB and SULTAN, (1990); TEMPELMAN and BURNSIDE (1990) and ABDEL GLIL et al., (1995) found that sires accounted for 7.39 to 10.21 % of the adjusted variance affecting milk yield.In addition, HAMED and SOLIMAN, (1994) with Fleckvieh cows found that sires accounted for 19.1 and 20.8% of the Variation for 100 dMY and 305 dMY, respectively.In contrast, BHAT (1980) andGARCHA andDEV (1985) working on Tharparkar and crossbred of Holstein, respectively reported no significant effect of sire on milk yield.The effect of cows is may be due to the permanent environmental effects which are transmitted from one record to another and cannot be corrected.ABUBAKAR et al., (1986) and ABDEL GLIL (1995) working on Friesian cows, found cows within sires was 17 and 11.2%, respectively for milk yield.KHATTAB and SULTAN (1990) working on Friesian cows in Egypt, found that there is no significant sire effect on the lactation period, but accounted for 2.01% variance on LP.The low genetic variance for the both lactation period and calving interval is also aecording to the limited lactation length of 305-days and due to the aim CI of 1 year.
Table 1 Estimates of sires' variance components (ah), cows within sires' components (a 2 c:s), residual variance components (a 2 e) and proportion of variance (V%) due to random effects for initial milk yield (IMY), 305-day milk yield (305-dMY), lactation period (LP) and calving interval (CI) of Friesian cows in Egypt and Germany (Schätzung der Varianzkomponente für Frühmilchleistung (IMY), 305-Tage-MiIchleistung (305 dMY), Laktationsdauer (LP) und Zwischenkalbezeit (CI) Schwarzbunter Rinder in Ägypten und Deutschland) Estimates of (a 2 e) is the residual variance component that contains the remainder of genetic and environmental variance which is the random effect, peculiar to the Observation of traits studied and ranged from 58.47 to 63.09% for Friesian cows in Egypt and from 25.56 to 58.84 for Friesian cows in Germany (Table 1).The present estimates are lower than those reported by STANTON et al., (1991)   Accordingly, it is safe to conclude that there is a high genetic potential for rapid genetic progress in traits studied through selection.
In Egyptian data, the present results indicate that the genetic differences between sires for the calving interval traits were small as compared with other traits studied.So there could possibly be an improvement in CI by better environmental factors and managerial system than selection, while in German data, the present results indicate that the genetic differences between sires for the lactation period trait were small compared to other traits studied.It is aecording to the limitation of LP to 305-days.Therefore an improvement of CI by better environmental factors and managerial system than selection could be possible.KAFIDI et al., (1992)   decrease the calving interval in the next generation and will tend to increase milk yield in the ränge from 189 to 529 kg.Table 4 presents German proofs for IMY, 305-dMY, LP and CI of thirty sires with the largest number of daughters in all data.Only 5 ofthose sires had a negative proof.The single most frequently used sire (247960) with a total of 406 daughters in the data had a positive (177.09kg, 743.19 kg, 0.46 d and 9.58 d) proof for IMY, 305-dMY, LP and CI.The BLUP estimates revealed that the percentage of sires with values 200 kg, 300 kg, and 400 kg more than the average of herd in 305-dMY were 22.41%, 14.65% and 11.21%, respectively and selecting those sires for breeding purposes may lead to rapid genetic improvement for milk yield in the next generation and will tend to increase milk yield in the ränge from 306 to 842 kg.KHATTAB and SULTAN (1991) in a study based on 1317 normal lactation records of Friesian cows, constructed four selection indices, indicated that the index which includes 305-dMY, LP and AFC was best (R = 0.69) and the expected genetic gain in 305-dMY increased by 163 kg/generation and LP increased by 25 d/generation.In addition, MANGUKAR et al., (1986) working on Holstein cattle, found that ranking of Holstein bulls on the basis of selection index incorporating lactation milk yield increased the precision of sire proof.

Genetic and phenotypic trends
Estimates of genetic and phenotypic trends for different traits studied for Friesian cows in Egypt and Germany are presented in Table 5 and 6.Annual genetic change for IMY, 305-dMY, LP and CI were 29.98 ± 11.21 kg, 112.99 ± 77.57kg, 129.08 ± 60.02 kg, -0.92 ± 1.75 d and 1.78 ± 0.64 d, respectively in Friesian cows in Egypt and were 70.41 ± 11.40 kg for IMY, 200.38 ± 69.16 kg for 305-dMY, 0.12 ± 0.16 d for LP and -1.05 ± 0.90 d for CI in Friesian cows in Germany, all of them were significant (PO.01 Table 5 and 6).The present results in Table 5 indicate that sires used in the following 2 years (1991 and 1992) were of superior genetic value than those used in the earlier years.It can be inferred that the sires used in later years (1991 and 1992) were proven to be superior sires.Also, Table 6 indicates that sires used in the later years were of superior genetic value than those used in the earlier years.Also, the present results show that sires used in the later years had, in most years, positive annual genetic gain in each of IMY, 305-dMY and LP, were shorter in calving interval, indicating, the use of those sires in purposes breeding will improve milk yield and lactation period and decrease calving interval, leading to a large calvers, big lactations and then, a large economic benefit from animals.NORMAN et al., (1991) working on Holstein, Jersey, Guernsey, Brown Swiss and Ayrshire cows, found that annual genetic trend in the breeding value for milk yield was 106, 79, 75, 85 and 44 kg for Holstein, Jersey, Guernsey, Brown Swiss and Ayrshire, respectively.POWELL (1990) indicated that genetic trend in breeding value for milk yield for Holstein by using the animal model evaluation was approximately 120 kg/year.Also, NIZAMANI and BERGER (1996) analysed 508828 American Jersey lactation records for 7942 sires from (1960 to 1990), and reported that the annual genetic response for milk yield was 232.00 kg/year.SHARABY and ELKIMARY (1982) on Friesian cows, estimated the genetic trend by using BLUP method and found that genetic change for LP was -0.30 d/year.On the other hand, negative genetic trends have been reported for milk yield by SHARABY and ELKIMARY (1982) on Friesian cattle in Egypt.Also, REGE (1991) working on Friesian cows in Kenya from 1978 to 1988, found that the genetic trends per year for 305-dMY and first lactation milk yield were negative and not significant, being -2.5 and -5.2 kg, respectively.In addition, POWELL et al., (1977) with Friesian cows, found strong negative genetic trends in two regions in the United States from 1961 to 1970 with average sire breeding values for milk (kg) dropping from 100 to -220 kg from 1961 to 1964 in the West region and from 25 to -150 kg during the same period of the time in the Midwest region.The annual phenotypic change for IMY, 305-dMY, LP and CI was 13.63 ± 1.50 kg, 48.00 ± 7.05 kg, 56.83 ± 5.64 kg, -1.66 ± 0.39 d and -1.82 ± 0.48 d, respectively for Friesian cows in Egypt (Table 5).These estimates were significantly varied (P<0.05 or P<0.01) except for LP and CI (Table 5).The observed negative phenotypic trends in the lactation period and calving interval are unexpected because the same data indicates positive phenotypic correlation between these two traits (LP and CI) and each of IMY and 305-dMY (Table 5).In Germany, The annual phenotypic change was 21.00 ± 0.51 kg for IMY, 104.00 ± 2.56 kg for 305-dMY, 0.17 ± 0.02 d for LP and 0.78 ± 0.14 d for CI.All those estimates were significant (PO.01 Table 6).The present results in (Table 5 and 6) indicate that the differences in Performance between years were mainly due to different nutritional, climatic and management practices prevalent over different times.
ABDEL GLIL (1985) working on Friesian cows in Egypt, found a positive annual phenotypic change for 305-dMY, being 33.29 ±13.50 kg.In addition, WELLER et al., (1984) working on Israeli Holstein Friesian cows, reported that the phenotypic trend for a 305-day milk yield was 173 kg/year.On the other hand, a negative annual phenotypic change for milk yield has been reported by CANON and MUNOZ (1991) working on Spanish Holstein in Kenya, being -78 ± 8 kg and REGE (1991), being -5.5 kg, with Friesian cows in Kenya.Data for Friesian cattle evaluated in this study suggest improving fodder supply in Egypt by 1) cultivation green fodder in the new land reclamation, 2) urea/ ammonia treated of poor quality roughages (agricultural by products) and 3) cultivation of more green forage (sorghum and darawa) in the summer season.The above measures together with selection programme could improve the producing ability of animals.It could also be concluded from the present study that producing ability in German Friesian cows was higher than Friesian cows in Egypt.Estimates of predicted transmitting ability (PTA) for IMY, 305-dMY and LP were larger for German Friesian cows than Friesian cows in Egypt, while PTA for CI in Egypt was longer than in Germany, annual genetic progress was higher in German Friesian cows as compared to Friesian cows in Egypt.A great improvement in LP and CI could be possible by improving feeding and management Systems in Egypt and Germany.
Number of Observation was 774 and 9219 for Friesian cows in Egypt and Germany, respectively.
** d.f of sires, cows:sires and residual components were 307,398 and 1694 respectively for Friesian cows in Egypt and were 658.8410 and 18667 respectively for Friesian cows in Germany.
ABUBAKAR et al., (1986) the residual components of variance for milk yield were79.88, 78.08 and  80.53% in USA, Colombia and Mexico, respectively.In addition, KHATTAB et al.,  (1993)obtained high residual components of variance for cumulative monthly milk yield and ranged from 73.0 to 89.90%.found that the decreased number of daughters per sire would reduce the accuracy of predicted sire breeding value.Also, SCHAEFFER et al., (1973)on Holstein sires concluded that the important culling decision usually occurred on young sires with fewer than 50 daughters.However,ABUBAKAR et al., (1986)in a study based on 15512 lactation records of daughters of 138 sires, concluded that 10 daughters per sire are considered a minimum number for the evaluation of sires using BLUP procedures in tropical areas.Then it is necessary to estimate the sire evaluation based on a large number of daughters per sire.Therefore, the sire transmitting ability of sires with at least 5 daughters of first lactation was examined.Estimates of sire transmitting abilities (ETAs) as deviation from the mean ranging from -22 to 36 kg for IMY, from -336 to 529 kg for 305-dMY, from -8 to 66 d for LP and from -0.69 to 1.56 d for CI, respectively in Friesian cows in Egypt, ranged from -201.39 to 179.71 kg for IMY, from -863.21 to 842.75 kg for 305-dMY, from -1.31 to 1.31 d for LP and from -8.28 to 12.73 d for CI in Friesian cows in Germany (Table2).The present results show large genetic differences between sires for different traits studied, which indicate the high potential for rapid genetic improvement in milk yield through sire selection.

Table 2 Range
CHAFIE (1981)working on Friesian x native cows in Egypt, found that the ränge of breeding values of 17 sires for 100-dMY, 305-dMY and LP were large and ranged from -56 to 86 kg for 100-dMY, from -802 to 344 kg for 305-dMY and from -37 to 47 d for LP.In addition, ABDEL GLIL (1991) analysed 1653 lactation records for daughters of 163 sires (each with 5 or more daughters), found that predicted sire value (BLUP) ranged from -466 to 68 kg for 305-dMY, and from -48.8 to 43.3 d for LP.However, BLUP estimates of Friesian sires as deviations from herd mean show that there was a difference of 58 kg, 865 kg, 933 kg, 74 d and 2.25 d in IMY, 305-dMY, LP and CI, respectively between the top and bottom sires in Friesian cows inEgypt and 381.1 kg, 1705.96kg,2.62 d and 21.01 d in IMY, 305-dMY, LP  and CI, respectively between the top and bottom sires in Friesian cows in Germany.