Relationship Between Production and Days Open at Different Levels of Herd Production

GENETICS AND BREEDING Relationship Between Production and Days Open at Different Levels of Herd Production C. F. MART11 and D. A. FUNK University of Wisconsin Madison 53706 ABSTRACT

cows and between winter and summer calvings. Within herd, days open were longer for cows with higher production. Regressions of days open on production records that were not adjusted for days open were significantly larger for the herds at lower production. (Key words: days open, heritability, parity, season)

Five production variables and days open were analyzed using 611,680 records from 348,243 cows in 5694 herds enrolled in the Wisconsin DHI program. Production variables included 305-d milk production and several production measures adjusted for combinations of mature equivalent, fat and protein content, and effects of days open. Herds were divided into four groups by herd production. Heritability estimates for production variables ranged from .27 for mature equivalent milk that was corrected for fat and protein content and adjusted for days open to .34 for mature equivalent milk and for mature equivalent milk that was adjusted for days open. Adjustment of production records for days open had little impact on heritability estimates of production traits. Heritability for days open was approximately .05. Heritability estimates were larger for all measures of production and for days open for the herds with higher mean production. The effects of parity and season were important for both adjusted and unadjusted measures of production. After production variables were adjusted for mature equivalent factors, large differences remained between REML estimates of fixed effects of parity and season. For days open, estimates were larger for later parity cows and were highest for cows calving during spring. For measures of production, estimates were largest between first and second parity

Received September 13, 1993. Accepted January 25, 1994. 'Present address: American DeFomt. WI. 1994 J Dairy Sci 77:1682-1690

Breeders

Abbreviation key: DO = days open, ME = mature equivalent, MEC = MEM corrected for fat and protein content, MEC-DO = MEC adjusted for DO, ME-DO= MEM adjusted for DO, MEM = ME MP, MP = 305-d milk production, PG = herd production group. INTRODUCTION

Service,

Dairy cattle profitability is influenced by herd production and reproduction. Mean milk production per cow in Wisconsin doubled from 1950 to 1990 (31). Genetic and phenotypic antagonism of high milk production and measures of reproduction have been reported (2, 8, 13, 14, 23, 28, 29). Reproductive failure accounts for approximately 12% of disposals of US Holsteins (4). Shanks and Freeman (24) estimated that reproductive disorders accounted for 21% of the direct health cost of herds for US dairy producers. To improve the reproductive performance of their herds, dairy producers must understand the complex interactions of milk production, reproduction, nutrition, genetics, and management. Most estimates of heritability for days open (DO) or for calving interval are <5% (2, 10, 11, 12, 17, 20, 22, 25, 28). leading some to question whether selection for reproductive performance is worthwhile. However, Hermas et al. (12) and Philipsson (19) state that substantial additive genetic variance exists and that profitable breeding programs should include selection for reproductive traits. Others (9, 13,

1682

RELATIONSHIP BETWEEN PRODUCTION AND DAYS OPEN

28) recommend improvements in management, such as detection of estrus, insemination technique, herd health programs, and nutrition, to improve reproductive performance at higher production. Reproductive efficiency may differ for cows of different production levels. Faust et al. (8) found that reproductive performance declined as production increased, especially for cows at highest production. Foster et al. (9) analyzed within-herd average production and reproduction over time and found that reproductive parameters changed little when either the genetic milk production or the environmental milk production of the herd increased. Short et al. (25) found more unfavorable relationships between milk production and reproduction for cows in herds that had larger variation of milk production than for cows in herds with smaller variation of milk production. Laben et al. (13) analyzed reproduction both across herds and within herds and found that average reproductive performance of herds improved as herd production increased, although the relationship between production and reproduction became increasingly more antagonistic as cow production increased within herd. Schaeffer and Henderson (22) reported that age at calving had no effect on DO. Stevenson et al. (27) reported that older cows had more days to first breeding and more DO; however, the average production per cow was approximately 1400 kg more than the average production per cow in the earlier study of Schaeffer and Henderson (22). Reproductive performance may be affected by season. Schaeffer and Henderson (22) found that DO were longer for cows freshening during summer than for cows freshening during winter or spring. Faust et al. (7) reported that genetic antagonism between reproduction traits and fat-corrected milk was greater for cows calving from April to July than for cows calving during cooler seasons for herds in North Carolina. Faust et al. (8) observed that COWS calving during summer had reduced reproductive performance compared with that of cows calving during spring, winter, and fall. The purpose of this study was to estimate variance components for both adjusted and unadjusted production records for different levels of herd production and to examine the relationship between days open and production

1683

within herd production levels. As mean production per cow continues to get larger over time, the impact of various adjustments on production records needs to be examined because of the potential impact on genetic evaluations. MATERIALS AND METHODS Data

Production and reproduction data were obtained from Wisconsin DHI Cooperative for all cows on DHI production testing in Wisconsin with records completed from January 1987 to December 1991. Both official and ownersampler records were used. All records, regardless of lactation number, were included, which may have introduced selection bias because first records were not required. Inclusion of all records greatly increased the number of records available for later lactations. Lactation records c305 d of lactation were projected (6). Lactations were projected to 305 d so that the mature equivalent (ME)factors (18) that are used by USDA for genetic evaluations could be applied to all records. The ME factors are used by USDA to standardize actual or projected 305-d production (MP) for age at calving, region, and month of calving to arrive at ME MP W M ) . Adjustment factors for DO developed by Sadek and Freeman (21) from projected and actual 305-d records were used to account for the effects of pregnancy on current MEM (ME-DO). The MEM records were standardized for total energy output with adjustment factors reported by Van Arendonk et al. (28) for fat and protein content (MEC). Finally, MEM records were adjusted for both DO and €at and protein content (MEC-DO). Gestation length was assumed to be constant (282 d) for all cows. Breeding dates reported to DHI are often incomplete, so DO was estimated by subtracting 282 d from the calving interval between successive calving dates. Because the DO calculation required a subsequent calving date, lactation records without a subsequent calving date were removed. Exclusion of records without a subsequent calving date may introduce selection bias, but was necessary to verify DO. Lactations initiated by parity 26 and records from cows other than Holstein were excluded. Journal of Dairy Science Vol. 77, No. 6 . 1994

1684

MARTI AND FUNK

Further edits deleted records initiated by abortions and records from embryo transfer donor cows. Records with <120 d of lactation were deleted so that miscoded records initiated by abortions were not included. If a cow had a portion of her lactation in more than one herd because she had changed herds, both partial records for that lactation were removed. Records with missing information for protein production were deleted. The remaining 878,267 records from 12,574 herds were used to calculate herd production groups (PG)for MEC. Herds were equally divided into four groups based on mean of herd production: PG 1 (<7818 kg), PG 2 (7818 to 8409 kg), PG 3 (8410 to 8955 kg), and PG 4 (A955 kg). Four 3-mo seasons were defined: 1) winter (December through February), 2) spring (March through May), 3) summer (June through August), and 4) fall (September through November). Data used to calculate heritability estimates required further edits. Herds with <25 cows among all years were eliminated. Sires were required to have a minimum of 10 daughters in 22 herds. These final edits resulted in 611,680 records from 348,243 cows in 5694 herds. A total of 1422 sires remained in the final data. Models

Analysis of variance was performed using REML procedures developed by Meyer (16). Five production measures (MP, MEM, MEC, ME-DO, and MEC-Do) and DO were analyzed using Model [l] (shown in Table 3), which included fixed effects of herd-years (absorbed), parities (1 to 5), seasons (spring, summer, fall, and winter), and random effects of sires and cows nested within sires. Herd-yearseason effects were not considered in order to evaluate effects that were due to seasons, especially for DO. Additionally, herd-year-season subclasses could be quite small or empty for many of the smaller Wisconsin herds. A numerator relationship matrix added 113 ancestral sires to the original 1422 sires. These 113 sires had no daughters with records in the data and were assumed to be unrelated. Cows were also assumed to be unrelated. A univariate iterative algorithm, described by Smith and Graser (26),was used to estimate variances for the four PG and the entire data. Journal of Dairy Science Vol. 77, No. 6, 1994

TABLE 1. Number of records, cows, and herds within each herd production group (pG)' and for entire data (total). PG ~~

Records

cows

Herds

99.9 15 136,291 166,547 208,927 611,680

55,738 71,762 94,987 119,756 348,243

1142 1366 1490 1696 5694

~

PGl PG2 FG3 €434

Total

1PG 1, e7818 kg; PG 2, 7818 to 8409 kg; PG 3, 8410 to 8955 kg; and FG 4, >8955 kg.

Heritability was estimated as four times the sire variance divided by the phenotypic variance. Repeatability was estimated as the sum of sire and cow variance divided by the phenotypic variance. The dependent variable of DO was analyzed with Model [2] (shown in Table 5), which included all independent variables from Model [l]plus milk production as a covariant. Three separate analyses were made with covariables of either MEM, MEC, or MEC-DO. Pairwise comparison of linear estimates was used to determine whether regressions of DO on MEM, MEC, or MEC-DO differed among PG. RESULTS AND DISCUSSION

The numbers of herds, cows, and records that remained after final edits are listed in Table 1. The mean number of cows per herd was larger for higher PG. The final edit, which required herds to have a minimum of 25 cows and required sires to have a minimum of 10 daughters in 22 herds removed more herds from the lower PG than from the higher PG. The lowest PG, however, still contained nearly 100,OOO records from cows in 21100 herds. Means and standard deviations for DO and the five measures of production are in Table 2. Mean for DO was highest for PG 1. Mean for DO was slightly less for PG 2 and PG 3 than for PG 4, but differences were very small. Summary statistics (15) for all Holstein herds on DHI test in Wisconsin show that herd average DO decline from 148 d for herds with mean production of 5910 kg per cow to 125 d for herds with mean production of 10,455 kg per cow, but then increase to 129 d for herds with mean production >10,455 kg per cow.

1685

RELATIONSHIP BETWEEN PRODUCTION AND DAYS OPEN

TABLE 2. Means and standard deviations for production measures and days open for the entire data (total) and for herd production groups (pG).* Trait2

-XTotal

DO MP MEM MEC ME-DO MEC-DO

115.3 7973 9029 8637 9020 8625

PGl

SD 59.6 1571 1526 1388 1500 1363

Tz 117.6 6994 7681 7343 7669 7328

SD 64.5 1312 1240 1122 1217 1100

-XP G 2 114.6 7707 8555 8169 8549 8160

PG3

SD 59.7 1339 1220 1080 1191 1052

-X 114.6 8154 9098 8700 9092 8690

SD 59.1 1406 1276 1121 1244 1090

-XP G 4

SD

115.1 8870 9926 9511 9915 9469

57.5 1542 1411 1251 1376 1217

'PG 1. 4'818 kg; PG 2, 7818 to 8409 kg; PG 3, 8410 to 8955 kg; and PG 4, 28955 kg. 2D0 = Days open, MP = 305d milk production, MEM = mature equivalent MP, MEC = MEM conected for fat and protein content, ME-DO = MEM adjusted for DO, and MEC-DO = MEC adjusted for DO.

Laben et al. (13) reported that herd mean for DO consistently was smaller for herds with higher mean production, but the highest herd mean in their study was only 8165 kg per cow. Herd mean for DO appears to get smaller for herds with higher production up to a point, and then herd mean for DO appears to stay about the same or to get slightly larger for herds at highest production. Standard deviations for DO became slightly smaller for herds with higher production, suggesting that herd management for DO may be more consistent for higher producing herds. For all measures of production, PG 4 had the largest standard deviations, in agreement with results from previous studies (3, 5).

the entire data, estimates within PG may still be biased slightly because of unequal distribution of daughters of sires among PG. Furthermore, Mantysaari and Van Vleck (14) suggested that single-trait estimates of heritability for DO may be biased downward because of selection for milk. Production. For all production measures, estimates of phenotypic variance, heritability, and repeatability were larger for herds with higher mean production (Table 3). Trends from low to high PG were similar for all measures of production, so estimates are only listed for MP in Table 3. Heritability estimates for PG 4 were consistently similar to heritability estimates for the entire data. For three of the five production measures, the percentage of phenoVariance Components typic variance because of sire effect was larger DO. Estimates of phenotypic variance were for the entire data than for PG 4. This differsmaller for herds with higher production, and ence may indicate that heritability estimates phenotypic variance estimates for the entire for each PG are underestimated, perhaps bedata were intermediate to estimates for PG 2 cause daughter of sires were not distributed and 3 (Table 3). However, heritability esti- equally among PG. Similar trends for estimates were larger for herds with higher mates of phenotypic variance, sire variance, production and ranged from .037 for PG 1 to and heritability among production levels have .047 for PG 4. Heritability estimates are simi- been reported by Boldman and Freeman (3) lar to previous results (10, 11, 17, 23). Repeat- and De Veer and Van Vleck (3,but neither ability estimates for DO were larger for herds study examined variance components for all with higher production. Repeatability estimates levels of herd production combined. Stratifying were similar in value to those reported by data by herd production, as in the present Hayes et al. (11). Results suggest that DO are study, may reduce estimates of heritability less variable, but more heritable and repeata- within production levels. Repeatability estimates were larger for ble, for the higher production herds, perhaps because of better, more consistent management herds at higher production. For each of the five practices among all cows. Although heritability production measures, repeatability was higher estimates for DO were larger for PG 4 than for for PG 4 than for the entire data. The percentJournal of Dairy Science Vol. 77, No. 6, 1994

1686

MARTI AND FUNK

age of phenotypic variance from cow effects was always higher for PG 4 than for the entire data which, when combined with sue variance, resulted in repeatability estimates for the entire data that were intermediate to estimates for high and low PG. Estimates were close to the .55 used by USDA for animal model evaluation in the US (30). Heritability estimates for the entire data ranged from .27 for MEC-DO to .34 for MEM and ME-DO. Estimates for MEM are slightly higher than those from other studies (1, 2, 10, 20), although higher estimates are reported in more recent studies. Possible explanations for the trend toward higher estimates of heritability over time include better statistical models to estimate heritability, less residual variance because of more consistent management, and effects of heterogeneous variance. The largest estimates of heritability were obtained from data that were preadjusted for effects of age and season. When MEM was

adjusted for DO, estimates of sire and phenotypic variance were smaller than when unadjusted records were used, but estimates of heritability changed little. Seykora and McDaniel (23) reported that estimates of heritability for three measures of yield were smaller by either .01 or .02 when records were adjusted for effects of DO. Adjustment of records for DO appears to have little impact on heritability estimates for production. Phenotypic variance and heritability were smaller when production was corrected for fat and protein content (MEC and MEC-DO). Others (2, 10,23) have reported smaller phenotypic variance and heritability estimates when records are expressed on an energy-corrected basis. In contrast, Van Arendonk et al. (28) found that estimates of heritability were similar for actual milk and milk corrected for fat and protein content; however, the DutchFriesian population may differ from North American Holsteins.

TABLE 3. Estimates of phenotypic variance, heritability, and repeatability for production traits and days open (Do)' for entire data (total) and for herd production groups (PG)? Trait3 ~

~~~~~~~

~

~

(a2, DO PG 1 ffi2 PG 3 PG 4 Total

3752 3261 3223 3045 3257

.037 ,042 .045 .047 .045

.005 ,005 ,005 ,005 ,004

.I35 .138 .I38 .153 .143

1.01 1,783 1,105,120 1,198,649 1,346,096 1,214,821

.21 .25 .27 .30 .30

.03 .03 .03 .03 .02

S I .54 .55 .57 .56

1,476,827

.34

.03

.58

l,I31,320

.28

.02

.53

1,393,665

.34

.03

.56

1,060,260

.27

.02

.51

(k92, MP PGl PG2 PG3

ffi4 Total MEM Total MEC Total ME-DO Total MEC-DO Total

*Model [l]: Trait = p + herd-year + season + parity + sire + cow(sire) + error, where p = overall mean. ZPG 1, <7818 kg; PG 2, 7818 to 8409 kg; PG 3, 8410 to 8955 kg; and PG 4, >8955 kg. 3MP = 305-d Milk production, MEM = mature equivalent MP, MEC = MEM corrected for fat and protein content, ME-DO = MEM adjusted for DO, and =-DO = MEC adjusted for DO. Journal of Dairy Science Vol. 77, No. 6, 1994

1687

WATIONSHIP BETWEEN PRODUCTION AND DAYS OPEN Parity and Season Effects

were expected to be large for MP, because MP records are not adjusted for age effects. Mature The effect of parity was significant (P< .01) equivalent factors adjust records for age at for all five production measures and DO. The calving, and, although parity solutions were REML estimates for fixed effects of parity are shown as a deviation from first parity in Table smaller for the four ME measures of produc4. For DO, estimates for parity effects were tion, parity effects were still significant. larger in later parities, which is similar to the Trends from low to high PG were similar trend reported by Stevenson et al. (27). Possi- among all four ME measures of production, so ble explanations for more DO for older cows estimates are only listed for MEM in Table 4. are a higher incidence of reproductive diseases, Differences were less among parities 2, 3, and more stress associated with higher production, 4 than between parity 1 and any of the later or management decisions to delay intentionally parities. The REML estimates for fixed effects of the breeding of the highest producing cows in season are shown as a deviation from winter the herd. Fifth parity cows produced 1945 kg more season in Table 4. Estimates for season effects than first parity cows (Table 4). Parity effects show that cows calving during spring (most

TABLE 4. Solutions for effects' of parity and season for entire data (tdal) and for herd production groups (pG).2 Parity4 Trait3

2

3

season5

4

5

Spring

Summer

Fall

\-,

DO PGl PG2 PG3 PG4

Total

-1.49 .70 1.27 2.61 1.13

1.a7

3.74 4.19 6.16 4.26

3.54 6.33 7.28 10.22

7.45

6.79 8.61 9.27 11.71 9.57

4.44 6.34 6.21 6.56 5.93

-4.16 -2.01 -1.81 -2.59 -2.66

-3.74 -3.84 -3.08 -3.17 -3.53

MP PGl PG 2 PG 3 PG 4 Total MEM PGl PG2 PG3 PG4 Total MEC Total ME-Do Total

968 1092 1187 1327 1 I83

1433 1591 1719 1895 1712

1579 1767 1915 2115 1907

1585 1802 1945 2166 1945

-230 -227 -217 -227 -226

4 2 -455 -468 -48 1 -467

-19 -27 -41 -43 -36

357 398 449 522 456

385 372 399

356 338 343 385 385

-5

420

367 335 355 381 383

33

91 I25 138 179 140

53 56 37 46 45

446

401

370

380

47

204

51

409

343

288

277

6

152

60

402

330

283

283

22

214

65

440

21 43 51

MEC-DO Total

lModel [l]: Trait = p + herd-year + season + parity + sire + cow(sire) + error, where p = overall mean. IPG 1, <7818 kg; PG 2, 7818 to 8409 kg; PG 3, 8410 to 8955 kg; and PG 4, >8955 kg. 3D0 = Days open, MP = 305d milk production, MEM = mature equivalent MP, MEC = MEM corrected for fat and protein content, ME-DO = MEM adjusted for Do, and MEC-DO = MEC adjusted for Do. dSolutions for parity effects an shown as a deviation from parity 1. 5Solutions for season effects are shown as a deviation from winter season. Journal of Dairy Science Vol. 77, No. 6. 1994

1688

MARTI AND FUNK

likely bred during summer) had most DO followed by cows calving during winter. Cows calving in summer and fall had fewest DO. These seasonal effects for DO are fairly consistent among all PG. Faust et al. (8) found that days to first breeding and conception rate from first breeding were poorer for cows calving from April to July. Seasonal differences in the present study may be because of heat stress during summer but may also reflect the shortage of labor and time that many Wisconsin dairy producers typically experience during spring and summer. Estimates for MP indicate that cows calving during winter produced the most milk, followed by production of cows calving in fall, spring, and then summer. Although season solutions were smaller for all four ME measures of production than for MP, the effect of season was still significant (P < .01). Because trends from low to high PG were similar among all four ME measures of production, only estimates for MEM are listed in Table 4. The greatest discrepancy occurred for winter and summer calvings, and the discrepancy was larger for herds with higher mean production. Perhaps dairy producers in Wisconsin have modified management to cope better with the heat of summer and cold of winter, so that ME factors based on old production records may be outdated. The ME factors used for Holstein cows in Wisconsin did not completely remove effects of season for these data, particularly for the highest producing herds.

TABLE 5.Linear regression coefficient of days open (DO) on production1.2 for entire data (total) and for herd production groups (PG)? Production Covariant4

Linear

SE

1.2788 1.193b 1.13s 1.052d 1.135

.018 ,014 ,012 ,010

1.41Sa 1.3758 1.326b 1.25lC 1.303

.02 1 .016 .013 ,011 .007

,2778 .329a .298a .29S8 .294

,022 ,017 ,014 ,012 ,008

MEM PGl PG2 PG3 PG4 Total

MEC PGl PG2 PG3 PG4

Total MEC-DO PGl PG2 PG3 PG4 Total

,006

kb.c.dRegressions coefficients among PG within production covariate with different superscripts differ (P e .OS).

'Model [2]: DO = p + herd year + season + parity + sire + cow(sire) + production covariate + error, where = overall mean. zProduction covariant expressed in 100-kg units. 3PG 1, 4 8 1 8 kg; PG 2, 7818 to 8409 kg; PG 3, 8410 to 8955 kg; and PG 4, >8955 kg. 4MEM = Mature equivalent 305-d milk production, MEC = MEM corrected for fat and protein content, and MEC-DO = MEC adjusted for DO.

were smallest for PG 4 and largest for PG 1 (P < .OS). For all herds, DO were longer for cows with higher production, but to a lesser degree Three covariants (MEM, MEC, and MEC- for the higher producing herds. Perhaps herd DO) were evaluated independently to inves- management for both production and reproductigate the relationship between DO and tion is of higher quality and is more consistent production within each PG. Regression coeffi- among cows in the higher producing herds. cients were different from zero (P c .01)for all Within the eight classes of herd production of three milk covariants. Within a herd, the Laben et al. (13), trends were for largest regreshighest producing cows had the most DO, sion coefficients (more antagonism) for herds which supports the antagonistic relationship with highest production, but the trend was between milk production and reproduction ematic, and the regression of DO on reported by others (2,8, 10, 12, 14, 23, 28,29). 180-d milk production was only significant for For each 100-kg increase in MEM or MEC, the highest class of herd production. Faust et DO increased approximately 1.1 to 1.3 d p a - al. (8) found that days to first breeding were ble 5). Pairwise comparisons of regression longer, conception rate from first breeding was coefficients from different PG revealed that the smaller, and number of breedings were larger effects on DO of 100 kg more MEM and MEC for cows with higher production. However, the Production and Reproduction Relationship

Journal of Dairy Science Vol. 77. No. 6. 1994

RELATIONSHIP BETWEEN PRODUCTION AND DAYS OPEN

study by Faust et al. (8) consisted of only six experimental herds, and the study by Laben et al. (13) included only 201 California herds divided into eight categories of herd production, but the present study examined 5694 Wisconsin herds, with over 1100 herds in each category of herd production. Antagonism between DO and measures of production was considerably less when Model [2] included the production covariant for MEC-DO. Days open increased only .29 d per 100-kg increase in MEC-DO; however, the relationship between DO and MEC-DO was expected to be small. Regression estimates were similar among PG and are in agreement with those reported by Van Raden et al. (29). Small regression coefficients for the covariant for MEC-DO indicate that adjustment factors for DO account for a major portion of the effect of DO on production. CONCLUSIONS

Heritability estimates for milk production were largest for MEM and ME-DO and when estimated from data from highest producing herds. Heritability estimates for DO were 4% and were slightly larger for herds with higher production. Low estimates of heritability for DO suggest that the trait is largely environmental. Adjustment of milk records for DO had almost no impact on heritability estimates for production traits. Adjustment of records for fat and protein content reduced heritability estimates considerably. Effects of parity and season were significant for all production measures and for DO. The ME factors for production traits were unable to remove effects of parity and season completely. Differences between REML estimates for first and second parities and winter and summer calvings were large for both adjusted and unadjusted traits. Older cows had more DO, and cows that calved during spring and, therefore, were most likely bred during summer, had the most DO. Regression coefficients of DO on milk production indicated that cows had 1.1 to 1.3 more DO for each 100 kg more milk. Regression coefficients from all PG indicate an antagonistic relationship between DO and production. On a within-herd basis, DO were longer for cows with higher production. The

1689

antagonism was significantly more severe for lower producing herds than for higher producing herds. Perhaps management for both production and reproduction is of higher quality and more consistently applied to the entire herd for higher producing herds. When production records were adjusted for DO, regression coefficients of DO on production were small (.3 d for each 100 kg more milk adjusted for DO) and were similar in different PG. Results from this study indicate that adjustment of production for DO removed much of the variation in production associated with DO. REFERENCES

1 Batra. T. R., A. J. Lee, and A. J. McAUister. 1986. Relationships of reproduction traits, body weight and milk yield in dairy cattle. Can.J. Anim. Sci. 6653. 2 Berger, P. J., R. D. Shanks,A. E. Freeman, and R. C. Laben. 1981. Genetic aspects of milk yield and reproductive performance. J. Dairy Sci. 64:114. 3Boldman, K. G., and A. E. Freeman. 1990. Adjustment for heterogeneity of variances by herd production level in dairy cow and sire evaluation. J. Dairy Sci. 73503. 4Dentine. M. R., B. T. McDaniel, and H. D. Norman. 1987. Comparison of culling rates, reasons for disposal, and yields of registered and grade Holstein cattle. J. Diury Sci. 70:2616. 5 De Veer, J. C., and L. D. Van Vleck. 1987. Genetic parameters for first lactation milk yields at three levels of herd production. J. Dairy Sci. 70:1434. 6Eastwocd, B. R. 1967. Factors for extending incomplete lactations to 305 days. Iowa State Univ. Coop. Ext. DyS-718. Iowa State Univ., Ames. 7Faust, M. A., B. T. McDaniel, and 0. W. Robison. 1989. Genetics of reproduction in primiparous Holsteins. J. Dairy Sci. 72:194. BFaust, M.A., B. T. McDaniel, 0. W. Robison, m d J. W. Britt. 1988. Environmental and yield effec:s on reproduction in primiparous Holsteins. J. Dairy Sci. 71:3092. 9Foster, W. W., M. L. McGilliard, and R. E. James. 1988. Associations of herd average genetic and environmental milk yield with dairy herd improvement variables. J. Dairy Sci. 71:3415. loHansen, L. B., A. E.F ~ w n a nand , P. J. Berger. 1983. Yield and fertility relationships in dairy cnttle. J. Dairy Sci. 66:293. 11 Hayes, J. F., R. I. Cue, and H. G. Monardes. 1992. Estimates of repeatability of reproductive measures in Canadian Holsteins. J. Dairy Sci. 791701. 12Hermas. S. A., C. W. Young, and J. W.Rust. 1987. Genetic relationships and additive genetic variation of productive and reproductive traits in Guernsey dairy cattle. 1. Dairy Sci. 701252. 13Laben, R. L., R. D. Shanks, P. J. Berger, and A. E. Freeman. 1982. Factors affecting milk yield and reproductive performance. J. Dairy Sci. 65:1003. 14Mantysaari. E., and L. D. Van V k k . 1989. !&tima-

Journal of Dairy Science Vol. 77, No. 6, 1994

1690

MART1 AND FUNK

tion of genetic parameters for production and reproduction in Finish Ayrshire cattle. J. Dauy Sci. 72:2375. 15 McCullough, D. 1993. Wisconsin DHI Holstein Herd Summary Statistics: March 1993. Dairy Profit Report. Vol. 5, No. 1. Dauy Sci. Dep.. Univ. Wisconsin, Madison. 16Meyer. K. 1986. Restricted maximum lielihood for data with a hierarchical genetic shucture. Proc. 3rd World Congr. Genet. Appl. Livest. Prod., Lincoln, NE XIIk454. 17 Moore, R. K., B. W. Kennedy, L. R. Schaeffer, and J. E. Moxley. 1990. Relationships between reproduction traits, age and body weight at calving, and days dry in first lactation Ayrshires and Holsteins. J. Dairy Sci. 73:835. 18Norman. H. D., P. D. Miller, B. T. McDaniel, F. N. Dickinson, and C. R. Henderson. 1974. USDA-DHIA factors for standardizing 305-day lactation records for age and month of calving. ARS, Northeastern Region 40. USDA, Beltsville, MD. 19 Philipsson, J. 1981. Genetic aspects of female fertility in dairy cattle. Livest. Prod. Sci. 8:307. 20Raheja K. L.. E. B. Burnside, and L. R. Schaeffer. 1989. Relationships between fertility and production in Holstein dairy cattle in different parity lactations. J. Dairy Sci. 72:2670. 21 Sadek, M. H., and A. E. Freeman. 1992. Adjustment factors for previous and present days open considering all lactations. J. Dairy Sci. 75:279. 22 Schaeffer, L. R., and C. R. Henderson. 1972. Effects of days dry and days open on Holstein milk production. J. Dairy Sci. 55:107.

Journal of Dairy Science Vol. 77, No. 6, 1994

23 Seykora, A. J., and B. T. McDaniel. 1983. Heritabilities and correlations of lactation yields and fertility for Holsteins. J. Dairy Sci. 661486. XShanks, R. D., and A. E. Freeman. 1979. Postpartum distribution of health costs and health disorders in Holstein cows. J. Dairy Sci. 62(Suppl. 1):87.(Abstr.) 25 Short., T. H., R. W. Blake, R. L. Quaas, and L. D. Van Vleck. 1990. Heterogeneous within-herd variance. 2. Genetic relationships between milk yield and calving interval in grade Holstein cows. J. Dairy Sci. 73:3321. 26Smith, S. P., and H. U. Graser. 1986. Estimating variance components in a class of mixed models by restricted maximum likelihood. J. Dairy Sci. 69:1156. 27 Stevenson, J. S.. M. K. Schmidt, and E. P. Call. 1983. Factors affecting reproductive performance of dairy cows first inseminated after five weeks postpaaum. J. Dairy Sci. 66:1148. 28 Van Arendonk, J.A.M., R. Hovenier, and W. DeBoer. 1989. Phenotypic and genetic association between fertility and pduction in dairy cows. Livest. Prod. Sci. 21:l. 29Van Raden, P. M., A. E. Freeman, and P. J. Berger. 1987. Relationships among production and reproduction in nulliparous and first lactation Holstein cattle. J. Dairy Sci. 7qSuppl. 1):231.(Abstr.) 3OWiggans, G. R., I. Misztal, and L. D. Van Vleck. 1988. Implementation of an animal model for genetic evaluation of drury cattle in the United States. J. Dairy Sci. 71(Suppl. 2):54. 31 Wisconsin Agriculture Statistics Service. 1991. Wisconsin 1991 Dairy Facts. Wisconsin Dep. Agric., Trade, and Consumer Protection, Madison.