MILK YIELD AND ITS REPEATABILITY IN JAPANESE BLACK COWS

Daily inilk yield estimates were obtained on 74 lactations of 35 Japanese Black cows on weeks 1 to 9, 11, 13, 17, 21 and 26 postpartum using the weigh-suckle-weigh method. The data obtained were analyzed by least-squares procedures according to the models including the following effects; dam, lactation number, season of calving, week of experiment and partial regressions on the lactation number. The overall mean and standard error for daily milk yield was 4.55 士 0.04 kg.The milk yield declined essentially linearly throughout the experimental period. The main effects of lactation number, season of calving and week of experiment were highly significant. Differences in daily milk yield among cows were significant, and the repeatability was estimated as 0.60 ± 0.06. The overall means(kg) and repeat ­ abilities of cumulative milk yield for 1,4, 9, 13, i7, 21 and 26 weeks postpartum were 41,0.08; 166, 0.40; 361,0.61; 503, 0.66; 632, 0.67; 749, 0.65 and 884, 0.58, respectively. The means of calf weight at birth and 26 weeks weight for different lactations ranged from 25.5 to 33.6 kg and 145.6 to 185.4 kg, respectively. Calf daily gains between experiment weeks were low in earlier stages of lactation.


Introduction
It is well established that the milk yield of beef cow has a large effect on calf growth (Somerville et al., 1983; Clutter and Nielsen, 1987).McMorris et al. (1986) reported in their economic study that increased milk yield had a positive linear effect on the gross margin of the integrated production system.Recently in Japan, the importance of milk yield has been recognized in Japanese Black cows.However, there is a lack of detailed know ledge on milking ability of this breed.
The objectives of this study were to describe the lactation curve of Japanese Black cows and to estimate repeatability of milk yield.Additionally, the relationship between milk yield and calf growth was determined.

Animal and management
Thirty-five Japanese Black cows derived from a Address reprint requests to K.

Milk production and live weight
Daily milk yield estimates were obtained during weeks 1 to 9, 11, 13, 17, 21 and 26 postpartum for first to third lactations, and in weeks 1 to 9, 13 and 26 for subsequent lactations using the weigh-suckle-weigh method.In order to start all cows under equal conditions, cows and calves were separated at 900 h (13.00h for 1-2 week post partum) on the first day, and allowed to suckle at 16.00 h.The following two consecutive days, calves were weighed to the nearest 100 g and allowed to suckle at 9.00 h and at 16.00 h.The four calf weight increments were summed and divided by two for the estimate of daily milk yield.
Cow and calf were weighed within 24 h post partum and at the beginning of each milk record ing week.Calf weight was calculated as the mean of weights before and after suckling.

Statistical methods
In order to calculate cumulative milk yield for each cow, an equation developed by Wood (1969) was fitted to the estimates of daily milk yield.
The data obtained were analyzed by least squares procedures (Harvey, 1977).The daily milk yield was analyzed according to the follow The standard error of repeatability was estimat ed according to Swiger et al. (1964).

Cow live weight
The average live weights and standard devia tions of cows during lactation were 367 ± 34, 402 ± 44, 433 ± 44 and 458 ± 50kg for first, second, third and fourth to ninth lactations, respectively.For third lactations only, live weight decreased slowly throughout the lactation from 456±49kg at calving to 419±44kg at 26 weeks; for other lactations, weight generally was maintained.

Daily milk yield
The least-squares analysis of variance (Model I), and least-squares means and standard errors for daily milk yield are presented in table 1 and 2, respectively.The overall mean and standard error for daily milk yield was 4.55 ± 0.04kg.The main effects of lactation number, season of calv ing and week of experiment were highly signi ficant.Heifers (first lactation) and cows in eighth and ninth lactations gave significantly less milk than cows in second to seventh lactations, and spring-calving cows produced more milk than cows calving in other seasons.
Daily milk yield changes for different lacta tions are presented in figure 1.The milk yield of cows in these five lactation groups (first, second, third to fourth, fifth to seventh and eighth to ninth) declined essentially linearly, and the first and eighth to ninth lactations produced consistently less milk throughout the experiment.
Differences in daily milk yield among cows were significant, and the repeatability was esti mated as 0.60 士 0.06 from the Model II analysis  of variance (table 3).Linear, quadratic and cubic partial regression coefficients were 1.8632, -0.3549 and 0.0175, respectively.These values were different from zero (P < 0.001).
From Model III analysis of variance, overall means and standard errors of cumulative milk yield from parturition to 1, 4, 9, 13, 17, 21 and 26 weeks were 41 ± 2, 166 士 7, 361 ± 15, 503 ± 21, 632 ± 27, 749 ± 32 and 884 ± 40 kg, respec tively.Differences among cows were highly signi ficant except first week.Repeatabilities for cumu lative milk yield for each week are presented in table 3.As these estimates increased from 0.08 in 1 week to a maximum estimate 0.67 in 17 weeks, those of standard errors decreased from 0.17 to 0.09.Afterwards repeatability estimates decreased slightly.Quadratic partial regression coefficients in 4 to 26 weeks and cubic ones in 9 to 26 weeks were significant.

Calf growth
Least-squares means of calf birth weight for different lactation groups were 25.5 ± 1.3, 31.2 ± 1.0, 32.7 ± 0.9, 33.6 ± 1.2 and 33.6 ± 1.9 kg,  Calf daily gains adjusted to a male calf basis between experiment weeks are shown in figure 2. At the start of the experiment, calf daily gains were higher than next several weeks except the eighth to ninth lactation group.Minimum daily gains were recorded from 1 to 6 weeks postpar tum, and thereafter daily gains increased slowly.After 7 weeks, calf daily gains for first lactation were consistently lower than those for the other lactations.cant positive relationships were found between daily milk yield of each week and calf daily gain from the previous experiment week.The regres sion and correlation coefficients reached maxi mum in 3 to 4 weeks, and then decreased.In 13 to 17 and 17 to 21 weeks the correlation values tended to be negative.

Relationship between milk yield and calf growth
The correlation coefficients between cumula tive milk yield and calf daily gain from birth to each experiment week, ranged from 0.43 to 0.90, and were higher than those between daily milk yield and calf daily gain between experiment weeks.The maximum correlation coefficient was recorded between 8 weeks cumulative milk yield and calf daily gain from birth to 8 weeks.
Figure 3 shows the least-squares means for daily milk yield and calf daily gain for different lacta tions.

Discussion
In Japanese Black cows, the shape of the lac tation curve was essentially a straight line.This linear decrease in daily milk yield without a peak lactation was similar to that in studies in cross bred beef cows (Gaskins and Anderson, 1980) and in Hereford cows (Holloway et al., 1985).Time of peak lactation occurred several weeks postpartum for beef breeds in the reports of Russel et al. (1979), Somerville et al. (1983) and Jenkins and Ferrel (1984).In a previous study, Shimada et al. (1987) reported time of peak lactation as approximately 5 weeks in second lactations for Japanese Polled cows developed from a cross between Japanese Black cows and Angus bulls in the earlier part of the 20th cen tury; this breed also had a greater daily milk yield after the time of peak yield than Japanese Black cows.These results suggested that different breeds have characteristic lactation patterns.
The levels of daily milk yield recorded in this study were lower than those in Hereford x British Friesian cows (Somerville et al., 1983), Simmental x Angus and Jergey x Angus cows (Gaskins and Anderson, 1980), but were similar to levels in Hereford and Angus cows (Holloway et al., 1985).
Daily milk peak yields for first, second, third, fourth and fifth to ninth lactations ranged from 2.1 to 6.6, 4.5 to 8.0, 4.5 to 8.6, 4.3 to 8.2 and This difference between cows within lactations was one of the reasons for high repeatability estimate of 0.60 (table 3).The repeatability of daily milk yield estimated in this study was similar to the estimates of 0.55 to 0.79 reported by Williams et al, (1979), The lower calf birth weight and milk yield of autumn-calving cows were probably related to lower cow weights compared to those calving in the other seasons.This in turn appeared to be related to insufficient nutrition before calving because of lower yield of pasture due to low rain fall and high temperature in summer.Russel et al, (1979) suggested that severe undernourishment in late pregnancy reduced calf birth weight and subsequent production of the cows.
The decrease in calf daily gain between earlier experiment weeks suggested an insufficient level of nutrition because of the linear decrease in daily milk yield.Although the pattern of daily milk yield change for first lactation was similar to that for eighth to ninth lactation, the pattern of calf daily gain was different for two groups.
Calves for eighth to ninth lactations grew more rapidly in the earlier stages of lactation than those for first lactation.The main factor affecting this increase in the earlier stage could be the differ ence of approximately 8 kg in calf birth weight.
The maximum correlation coefficient between daily milk yield and calf daily gain recorded in 3 to 4 weeks was 0.73; correlations decreased as lactation progressed.While calf feed intake was not tested in the present study, the negative cor relation coefficients suggested that calves from lower milk yield cows consumed more hay and creep ration during the time of peak daily gain at 17 to 21 weeks.Gutter and Nielsen (1987) showed that the conelation of milk intake with subsequent postweaning gain tended to be nega tive.The correlation coefficients between cumu lative milk yield and calf daily gain from birth were higher than those between daily milk yield and calf daily gain between experiment weeks.The latter seemed to be influenced primarily by scours, general health of calf and short term weather.
It was concluded that the milk yield in effect of the Zth lactation fixed effect of the /th season of calving fixed effect of the fcth week of the ex periment the residual error term In order to estimate the variance component of dam, the following two mixed models were adoptedthe /th dam = fixed effect of the/th season of calving = fixed effect of the fcth week of experi ment ijkl -b3 = linear, quadratic and cubic partial regression coefficients on the lactation number (L) e uki = the residual error term and Yz/ = M + bi L1 + b2 L2 + b3 L3 + e .. (Ill) where 丫毎=cumulative milk yield The other elements were as defined for Model JI.Repeatability estimate was obtained as follows: r =曲/(免+後) where r = repeatability estimates of daily or cumu lative milk yield 说 = variance component of dam 徒 =the residual error term
Figure 3. Least-squares means for daily milk yield (A) and calf daily gain (B) for different lactations in Japanese Black cows and calves.
Shimada, Chugoku National Agricultural Experiment Station, Kawaicho, Oda-shi, Shimane-ken 694, Japan bPresent address: National Institute of Animal Indus try, Tsukuba Norindanchi, 305, Japan Received February 23, 1988 Accepted April 20, 1988 the experimental herd maintained at Chugoku National Agricultural Experiment Station (lati tude 35°10'N, longitude 132°30'E), were involved in the study; data were obtained from 74 lacta tions.Of these, 11 were first lactation, and tiie others were second to ninth.The experiment was carried out between 1982 and 1986.Cows were allowed to graze from April to November until approximately two weeks before calving.After parturition, cows and calves were maintained in pens.Cows were fed 20 kg of silage, 2 kg of hay and 2 kg of concentrate per day.Calves were fed hay ad libitum and a pelleted creep ration of not more than 2 kg per day.

TABLE 2 .
LEAST-SQUARES MEANS AND STAND ARD ERRORS FOR DAILY MILK YIELD (KG) a,b,c,d,e,f^eansdifferent superscripts within main effects differ significantly (P <0.05).

TABLE 4 .
LINEAR RELATIONSHIPS BETWEEN MILK YIELD (Y, KG) AND CALF DAILY GAIN (X, KG)