Search
Author
Title
Vol.
Issue
Year
1st Page

Journal of Animal Science - Animal Production

Life cycle efficiency of beef production: VIII. Relationship between residual feed intake of heifers and subsequent cow efficiency ratios12

 

This article in JAS

  1. Vol. 94 No. 11, p. 4860-4871
     
    Received: June 01, 2016
    Accepted: Aug 29, 2016
    Published: October 13, 2016


    3 Corresponding author(s): davis.28@osu.edu
 View
 Download
 Share

doi:10.2527/jas.2016-0690
  1. M. E. Davis 3*,
  2. P. A. Lancaster,
  3. J. J. Rutledge and
  4. L. V. Cundiff#
  1. * Department of Animal Sciences, The Ohio State University, Columbus 43210
     Darr School of Agriculture, Missouri State University, Springfield 65897
     Department of Animal Sciences, University of Wisconsin, Madison 53706
    # USDA, ARS, U.S. Meat Animal Research Center, Clay Center, NE 68933

Abstract

Data were collected from 1953 through 1980 from identical and fraternal twin beef and dairy females born in 1953, 1954, 1959, 1964, and 1969, and from crossbred females born as singles in 1974, and their progeny. Numbers of dams that weaned at least 1 calf and were included in the first analysis were 37, 45, and 56 in the 1964, 1969, and 1974 data sets, respectively. Respective numbers of dams that weaned 3 calves and were included in a second analysis were 6, 8, 8, 22, 33, and 33 in the 1953, 1954, 1959, 1964, 1969, and 1974 experiments. Individual feed consumption was measured at 28-d intervals from the time females were placed on the experiment until 3 calves were weaned or the dams had reached 5 yr of age. Residual feed intake (RFI) and residual gain (RG) of the heifers that subsequently became the dams in this study were determined based on ADG and DMI from 240 d of age to first calving. Various measures of cow efficiency were calculated on either a life cycle or actual lifetime basis using ratios of progeny and dam weight outputs to progeny and dam feed inputs. Residual feed intake was phenotypically independent of ADG and metabolic midweight (MMW), whereas the correlation between RFI and DMI was positive and highly significant (r = 0.67; P < 0.0001). Residual gain was highly correlated with ADG (r = 0.75; P < 0.0001) and had near 0 correlations with DMI and MMW. Correlations indicated that heifers that ate less and had smaller metabolic midweights from 240 d of age to first calving had superior efficiency ratios as cows. Residual feed intake was not significantly correlated with age at puberty, age at calving, or milk production. Results of this study do not indicate any serious antagonisms of postweaning heifer RFI with subsequent cow and progeny performance traits or with life cycle or actual lifetime cow efficiency. In addition, selection for increased RG would result in earlier ages at calving, but would also tend to result in taller and heavier cows.



INTRODUCTION

Approximately 60% to 70% of the energy for beef production is required by the cow herd (Shike et al., 2014). Of the energy needed for the cow herd, approximately 70% is used for maintenance (Ferrell and Jenkins, 1982). Thus, improvement in efficiency of energy utilization by the cow herd has the potential to greatly improve efficiency of beef production. A measure of feed efficiency often used in the beef industry is residual feed intake (RFI). Koch et al. (1963) first proposed the concept of RFI, which is defined as the difference between actual feed intake and predicted feed intake required for maintenance of live weight and measures of production such as observed rate of weight gain. A limited number of reports are available regarding the relationship between postweaning RFI in heifers and subsequent RFI as cows (e.g., Arthur et al., 1999; Archer et al., 2002; Basarab et al., 2007; Herd et al., 2011; Black et al., 2013). In addition, Bourg (2011) observed no relationship between RFI EPD and an energy efficiency index expressed as the ratio of cow ME to calf weaning weight. A 15% advantage in efficiency of low-RFI cows, expressed as the ratio of calf BW to cow feed intake, was observed using preliminary results of divergent selection for postweaning RFI in Australia (Herd et al., 2003). To our knowledge, no previous studies have examined the relationship between postweaning RFI of heifers and subsequent life cycle cow efficiency measured over several lactations. Therefore, the objective of this study was to determine the relationship of RFI estimated in heifers with their life cycle efficiency as cows, as well as the relationships of RFI with various traits of the cows and their progeny.


MATERIALS AND METHODS

Animal care and use committee approval was not required at the time the data were collected at the University of Wisconsin in the 1950s, 1960s, and 1970s.

Source of Data

Data were collected from 1953 through 1980 from identical and fraternal twin heifers born in 1953, 1954, 1959, 1964, and 1969, and from crossbred heifers born as singles in 1974, and their progeny (Christian et al., 1965; Kress et al., 1969, 1971a,b,c; Hohenboken et al., 1972, 1973; Towner, 1975; Baik, 1980). Numbers of dams that weaned at least 1 calf and were included in analysis I were 37, 45, and 56 in the 1964, 1969, and 1974 data sets, respectively. Respective numbers of dams that weaned 3 calves and were included in analysis II were 6, 8, 8, 22, 33, and 33 in the 1953, 1954, 1959, 1964, 1969, and 1974 experiments. The 1953, 1954, and 1959 twins were included only in analysis II, because in these data sets, feed consumption was available only for dams that completed 3 lactations. Breed composition and numbers of dams, along with single vs. twin status of the dams used in each of the 6 experiments, are shown in Table 1.


View Full Table | Close Full ViewTable 1.

Source of data for Analysis I and Analysis II

 
Experiment Breed composition N Singles/Twins Analysis I Analysis II
1953 Hereford 6 Twins No Yes
1954 Hereford 8 Twins No Yes
1959 Hereford 8 Twins No Yes
1964 Hereford 33 Twins Yes Yes
Hereford x Guernsey 1 Twins Yes Yes
Hereford x Shorthorn 1 Twins Yes Yes
Hereford x Holstein 1 Twins Yes Yes
Hereford x Brown Swiss 1 Twins Yes Yes
1969 Hereford 17 Twins Yes Yes
Hereford x Shorthorn 2 Twins Yes Yes
Hereford x Charolais 2 Twins Yes Yes
Holstein 24 Twins Yes Yes
1974 Hereford x Holstein 14 Singles Yes Yes
Angus x Holstein 14 Singles Yes Yes
Simmental x Holstein 15 Singles Yes Yes
Chianina x Holstein 13 Singles Yes Yes

Feeding and Management Systems

The heifers were purchased at 8 to 224 d of age and were placed on the experiments at 240 d of age for the 1953, 1954, and 1959 twins, 210 d of age for the 1964 and 1969 twins, and 168 d of age for the 1974 crossbreds. Females were randomly assigned to individual self-feeders where they were tied twice daily. Diets differed for the 1953, 1954, and 1959 birth year groups, but the same diet was fed to all females within a group (Christian et al., 1965). Heifers purchased in 1964, 1969, and 1974 were randomly assigned to either high- or low-energy diets. Females on the high-energy diet received a chopped mixed hay and concentrate diet, whereas those on the low-energy diet were fed chopped mixed hay. Individual feed offered was recorded daily and accumulated over 28-d periods from the time females were placed on the experiment until 3 calves were weaned or they reached 5 yr of age. Feed refusals were weighed at 28-d intervals. Feed consumption for each 28-d period was the difference between feed that was offered and the feed refusals. Procedures described by Kress et al. (1969) were used to predict the feed consumed by the dams from birth to 240 d of age.

Twin heifers were bred at the first observed estrus after 15 mo of age and at each succeeding estrus until conception occurred. Crossbred heifers purchased in 1974 were bred at first detected estrus (puberty) and at each subsequent estrus until pregnant. Following each calving, all dams were bred at first estrus and at each subsequent estrus until conception occurred.

The 1953 and 1954 dams were all mated to the same Hereford bull to produce all 3 calves. The 1959 twins were artificially inseminated using semen from 8 Hereford bulls chosen at random from several bull studs, whereas the 1964 twins were bred artificially to 1 of 4 Polled Hereford bulls. Semen from 4 Holstein bulls was used to artificially inseminate the 1969 Hereford twins. Holstein twins of the same year were mated artificially to 1 of 4 Hereford bulls. All matings in the 1974 experiment were monogamous; first calves were offspring of 1 of 56 Jersey bulls and second and third calves resulted from insemination with semen from 1 of 56 Charolais bulls.

Bull progeny were castrated shortly after birth. Progeny of the 1953, 1954, and 1959 twins were assigned randomly to individual self-feeders at 60 d of age, whereas progeny in the other experiments began receiving creep feed at 28 d of age. Individual feed offered to progeny was recorded daily and accumulated by 28-d periods from 60 to 240 d of age. Feed refusals were weighed at 28-d intervals. Feed consumption for each 28-d period was the difference between feed that was offered and the feed refusals. Dams and their progeny were weighed at 28-d intervals and wither heights were taken for dams and progeny every 56 d. Dams were also weighed and measured for wither height within 12 h after parturition. All weaning weights of progeny were adjusted to a 240-d basis using the equation: [(actual weaning weight – birth weight)/weaning age] x 240 + birth weight. Wither heights of progeny at 240 d of age and weights and wither heights of dams at 240 d of age and at 240 d after calving were obtained by linear interpolation for animals not measured at those times. All progeny weights and feed consumptions were also adjusted for sex using additive adjustment factors.

Length of lactation was 240 d, except for the 1974 crossbred cows whose progeny were weaned at 224 d of age. Milk production in each lactation was estimated by either machine or hand milking one half of the udder while the calf nursed the other half. Estimates of milk production and butterfat percentage (both estimates taken at least monthly) were used to compute milk production adjusted to a 4% butterfat equivalent as described by Morrison (1959). Milk production of the 1974 crossbred dams was extended from a 224-d to a 240-d lactation using ratio adjustment factors as described by Rutledge et al. (1972).

Estimation of Cow Efficiency

Cow efficiency ratios were calculated using 2 approaches: the first a life cycle approach (analysis I) and the second an actual lifetime approach for cows that weaned 3 calves (analysis II).

In analysis I, life cycle cow efficiency was expressed as the ratio of outputs to inputs, where outputs included weaning weights of all progeny and salvage value of the dams, and inputs were preweaning feed consumption of all progeny, as well as feed consumption of their dams:where PWi, DWi, PFi, and DFi are defined in Table 2 and ki, li, mi, and ni are weighting factors based on the age distribution of the cow herd and percentage calf crop. Derivation of the weighting factors is further described by Davis et al. (1983).


View Full Table | Close Full ViewTable 2.

Definitions of symbols and acronyms

 
Item Definition
PW1, PW2, PW3 Progeny weights. Sex-adjusted weaning weight (240-d weight) of the first, second, and third calf from each cow.
PF1, PF2, PF3 Progeny feed consumptions. Sex-adjusted feed consumption of the first, second, and third calf from 60 to 240 d of age.
DW1, DW2, DW3 Dam weights. Weight of the cow when her first, second, and third calf was weaned.
DF0 An estimate of the feed consumed by the dam from her birth to 240 d of age.
DF1, DF2, DF3 Dam feed consumptions. Feed consumed by the cow from 240 d of age to the weaning of the first calf, from the weaning of the first calf to the weaning of the second calf, and from the weaning of the second calf to the weaning of the third calf.
k1, k2, k3 Weighting factors to accumulate first, second, and third (subscripts 1, 2, and 3) progeny weights on a life cycle basis.
l1, l2, l3 Weighting factors to estimate average weight of the dam on a life cycle basis where subscripts 1, 2, and 3 denote first, second, and third parity, respectively.
m1, m2, m3 Weighting factors to accumulate first, second, and third (subscripts 1, 2, and 3) progeny feed consumptions on a life cycle basis.
n0, n1, n2, n3 Weighting factors to accumulate feed consumption of the dam on a life cycle basis where subscripts 0, 1, 2, and 3 denote periods from birth to 240 d and first, second, and third parity, respectively.
R1 Progeny and dam output divided by progeny and dam input computed on a life cycle basis (analysis I) as:
R2 Progeny output divided by progeny and dam input computed on a life cycle basis (analysis I) as:
R3 Progeny and dam output divided by progeny and dam input computed on an actual lifetime basis for cows weaning 3 calves (analysis II) as:
R4 Progeny output divided by progeny and dam input computed on an actual lifetime basis for cows weaning 3 calves (analysis II) as:

The weighted sum of the dam’s weights was multiplied by 0.5714 (i.e., 4/7), the approximate ratio of price per pound for cull cows to price per pound for feeder calves (Kress et al., 1969; USDA 1980; Feuz, 1995; Feuz and Burgener, 2005). Feed consumption was expressed in terms of Mcal of ME, whereas weights were expressed in kilograms. Thus, R1 estimates kilograms of weaning weight equivalent of beef produced/Mcal of ME consumed. Differences in estimates of R1 could be due to differences in fertility, weaning weights of progeny, salvage weights of dams, and feed consumption of dams and progeny.

A second measure of cow efficiency (R2 in Table 2) was calculated using the same approach as described above except that salvage value of dams was not taken into consideration. Therefore, R2 estimates kilograms of weaning weight of progeny produced/Mcal of ME consumed by the progeny and their dams.

In analysis II, actual lifetime cow efficiency was calculated for dams that weaned 3 calves by dividing the sum of the outputs by the sum of the feed inputs where all components were weighted equally (R3 in Table 2). The fourth measure of efficiency (R4 in Table 2) was the same as the third (R3) except that salvage weight of dams was not included in the numerator. Only dams that weaned 3 calves were included in the calculation of R3 and R4. Therefore, variation associated with rate of reproduction and calf survival was not included in R3 and R4. Variation in age at first conception and calving interval are the only components of reproduction remaining to influence R3 and R4.

Statistical Analysis

To account for differences among years, study diets, and breeds a year-diet-breed group variable was created based on year of birth and breed of heifers and dietary energy density of the diet fed to heifers from 240 d to first calving. Because some breed crosses had very small numbers, Hereford x Guernsey, Hereford x Holstein, and Hereford x Brown Swiss were grouped together in a Hereford x Dairy breed-type, and Hereford x Shorthorn and Hereford x Charolais were grouped together as a Hereford x Beef breed-type.

Dry matter intake was average daily feed intake of heifers from 240 d of age to first calving multiplied by DM percentage. Average daily gain was computed as BW at first calving minus BW at 240 d of age divided by number of days from 240 d of age to first calving. Midtest metabolic body weight (MMW) was computed as the average of BW at first calving and 240 d of age raised to the 0.75 power. Residual feed intake was computed as the residual from mixed model regression (PROC MIXED; SAS Inst. Inc., Cary, NC) of DMI on ADG and MMW having random intercept and slopes for the year-diet-breed group (Lancaster et al., 2009b). Residual gain (RG) was computed as the residual from mixed model regression of ADG on DMI and MMW having random intercept and slopes for the year-diet-breed group. The coefficients of determination for the RFI and RG regression models were 0.84 and 0.69, respectively, calculated by regressing an adjusted DMI trait (fixed effects plus residual) on ADG and MMW (Lancaster et al., 2009b).

All traits were adjusted to remove the random effect of year by diet group by breed using a mixed model (PROC MIXED). To accomplish this, dependent variables were analyzed using a 1-way random-effect treatment structure with year-diet-breed group as the random effect (Littell et al., 2006; Lancaster et al., 2009b). Phenotypic Pearson correlation coefficients (PROC CORR; SAS Inst. Inc.) were computed among the adjusted traits. Significance was set at P < 0.05 and tendencies if P > 0.05 and < 0.10.


RESULTS AND DISCUSSION

Numbers of observations, means, SD, and minimum and maximum values for each dependent variable are shown in Table 3. Means for RFI and RG equaled 0, as expected. Individual values for RFI and RG ranged from -1.48 to 3.21 and from -0.13 to 0.15 kg d-1, respectively.


View Full Table | Close Full ViewTable 3.

Means, SD, and minimum and maximum values for dependent variables1

 
Trait N Mean SD Minimum Maximum
PW1, kg 125 256 26 150 311
PW2, kg 152 285 28 163 341
PW3, kg 160 294 21 239 363
PF1, Mcal 125 1,239 243 691 1,847
PF2, Mcal 152 1,267 266 502 2,070
PF3, Mcal 160 1,316 229 620 1,939
DW1, kg 160 487 50 377 634
DW2, kg 160 534 53 413 702
DW3, kg 160 566 55 437 743
DF0, Mcal 160 1,626 194 838 2,171
DF1, Mcal 160 13,537 1,810 8,814 22,286
DF2, Mcal 160 9,361 1,437 5,905 17,353
DF3, Mcal 160 9,601 1,438 6,502 14,208
DMI, kg d-1 160 7.82 0.89 5.41 13.82
ADG, kg d-1 160 0.48 0.06 0.37 0.68
MMW, kg 160 75 5 58 94
RFI, kg d-1 160 0 0.54 -1.48 3.21
RG, kg d-1 160 0 0.04 -0.13 0.15
R1 160 0.0529 0.0037 0.0437 0.0664
R2 160 0.0244 0.0030 0.0161 0.0318
R3 110 0.0308 0.0019 0.0262 0.0363
R4 110 0.0223 0.0017 0.0185 0.0267
Feed consumption from 240 d to first calving, Mcal 160 8,078 1,495 4,090 15,503
Feed efficiency from 240 d to first calving, Mcal/kg 160 33.62 3.43 24.14 49.85
240-d weight, kg 160 196 22 113 255
Weight at first calving, kg 160 437 47 310 660
Weight at second calving, kg 160 494 51 359 640
Weight at third calving, kg 159 532 53 417 702
240-d height, cm 160 97 4 84 108
Height at first calving, cm 160 120 4 102 136
Height at second calving, cm 160 124 4 110 138
Height at third calving, cm 159 125 4 116 135
Height at first weaning, cm 160 122 4 109 137
Height at second weaning, cm 160 124 4 112 135
Height at third weaning, cm 160 125 4 110 137
240-d wt:ht ratio 160 2.00 0.18 1.28 2.36
Wt:ht at first calving 160 3.61 0.30 2.90 4.68
Wt:ht at second calving 160 3.98 0.33 3.09 5.04
Wt:ht at third calving 159 4.25 0.35 3.22 5.39
Wt:ht at first weaning 160 3.97 0.33 3.21 4.83
Wt:ht at second weaning 160 4.28 0.34 3.50 5.14
Wt:ht at third weaning 160 4.52 0.37 3.51 5.36
Change in wt:ht, lactation 1 160 0.36 0.22 -0.19 1.07
Change in wt:ht, lactation 2 160 0.30 0.21 -0.38 1.05
Change in wt:ht, lactation 3 159 0.25 0.25 -0.52 0.83
Weight gain, lactation 1, kg 160 50 27 -25 126
Weight gain, lactation 2, kg 160 39 26 -47 133
Weight gain, lactation 3, kg 159 31 30 -64 112
Age at puberty, d 138 406 53 208 592
Age at first calving, d 160 748 65 529 1,031
Age at second calving, d 160 1,123 88 895 1,421
Age at third calving, d 159 1,491 103 1,257 1,777
Milk production, lactation 1, Mcal ME 122 820 163 511 1,272
Milk production, lactation 2, Mcal ME 147 1,023 226 469 1,650
Milk production, lactation 3, Mcal ME 139 1,065 224 515 1,794
1See Table 2 for definitions of symbols and acronyms.

Correlations among RFI, RG, DMI, ADG, and metabolic weight are shown in Table 4. Residual feed intake was phenotypically independent of ADG and MMW, whereas the correlation between RFI and DMI was positive and highly significant (r = 0.67; P < 0.0001). These results were expected because the use of linear regression to compute expected DMI for estimation of RFI forces RFI to be phenotypically independent of the component traits. Basarab et al. (2007) reported a phenotypic correlation of 0.53 (P < 0.001) between RFI and DMI. Arthur and Herd (2012) examined genetic correlation estimates between RFI and feed intake in 11 different studies. With the exception of 1 negative estimate, values ranged from 0.41 to 0.81, indicating that cattle with low (desirable) RFI will produce offspring that consume less feed. Reduced DMI in more efficient RFI groups was also observed by Elzo et al. (2009), Lancaster et al. (2009a), Shaffer et al. (2011), and Basarab et al. (2011). RG was highly correlated with ADG (r = 0.75; P < 0.0001) and had near 0 correlations with DMI and MMW. A significant negative correlation was observed between RFI and RG (r = -0.47; P < 0.0001). Correlations among DMI, ADG, and MMW were large and positive (P < 0.0001). The large correlations of DMI with ADG and body weight are indicative of high-quality data (Basarab et al., 2011).


View Full Table | Close Full ViewTable 4.

Correlations1 among residual feed intake (RFI), residual gain (RG), dry matter intake (DMI), average daily gain (ADG), and metabolic weight (MMW)2

 
Variable RG DMI ADG MMW
RFI -0.47 0.67 0 -0.01
<0.0001 <0.0001 0.99 0.94
RG 0 0.75 -0.05
0.96 <0.0001 0.55
DMI 0.63 0.60
<0.0001 <0.0001
ADG 0.47
<0.0001
1Significance level for the test Prob > |r| under H0: ρ = 0 is presented below the correlation coefficient.
2See Table 2 for definitions of symbols and acronyms.

Correlations of RFI, RG, DMI, ADG, and MMW with cow efficiency ratios are presented in Table 5. Residual feed intake of heifers exhibited small negative (i.e., favorable) correlations (P > 0.06) with subsequent cow efficiency ratios. Correlations of RG of heifers with subsequent cow efficiency ratios were positive, but nonsignificant (P > 0.06). Basarab et al. (2007) reported a phenotypic correlation of 0.30 (P = 0.025) between progeny RFI and cow RFI measured in the same year, which indicates that postweaning RFI and cow RFI are different traits. Black et al. (2013) reported that the relationship between RFI measured in heifers and RFI measured in 3-yr-old lactating cows was not significant. On the other hand, Arthur et al. (1999) obtained a phenotypic correlation of 0.36 (P < 0.05) between postweaning NFI and NFI of 4- to 4.5-yr-old nonlactating cows, and Archer et al. (2002) reported a genetic correlation of 0.98 between postweaning RFI and RFI of the cow. Furthermore, heifers that were phenotypically superior for postweaning RFI on ad libitum feeding were also superior as lactating cows on medium-quality pasture and as dry cows on ad libitum feeding, but not on restricted feeding, in the study of Herd et al. (2011). Weight of calf weaned per cow exposed to the bull did not differ between high and low selection line Angus cows in Australia that were the result of 1 to 2.5 generations of selection (mean = 1.5 generations) for high vs. low RFI (Arthur et al., 2005). Dams that produced low-, medium-, and high-RFI progeny were also similar in production efficiency (kg of calf weaned per 100 kg of cow BW at weaning) and biological efficiency (ratio of calf weight at weaning to the sum of cow metabolic weight at weaning plus half of the calf’s metabolic weight at weaning) in experiments performed by Basarab et al. (2007). Medium- and high-RFI heifers had greater means for cow RFI than low-RFI heifers, indicating that heifers that consumed less feed than predicted during the postweaning period also ate less than predicted as 2-yr-old lactating cows in the study of Shike et al. (2014).


View Full Table | Close Full ViewTable 5.

Correlations1 of residual feed intake (RFI), residual gain (RG), dry matter intake (DMI), average daily gain (ADG), and metabolic weight (MMW) with cow efficiency ratios2

 
Variable RFI RG DMI ADG MMW
R1 -0.12 0.12 -0.09 0.03 -0.02
0.12 0.13 0.26 0.72 0.80
R2 -0.07 0.03 -0.23 -0.14 -0.27
0.37 0.72 <0.01 0.07 <0.001
R3 -0.18 0.18 -0.27 -0.05 -0.29
0.06 0.06 <0.01 0.61 <0.01
R4 -0.09 0.07 -0.29 -0.16 -0.40
0.36 0.49 <0.01 0.10 <0.0001
1Significance level for the test Prob > |r| under H0: ρ = 0 is presented below the correlation coefficient.
2See Table 2 for definitions of symbols and acronyms.

Bourg (2011) observed no relationship between postweaning RFI EPD and an energy efficiency index, which was expressed as the ratio of cow ME to calf weaning weight. A 15% advantage in efficiency of low-RFI cows, expressed as the ratio of calf BW to cow feed intake, was seen using preliminary results of divergent selection for postweaning RFI (Herd et al., 2003). These preliminary results indicate that a phenotypic association exists between postweaning RFI of heifers and later efficiency of the cow/calf unit on pasture.

Correlations of DMI and MMW with R2, R3, and R4 ranged from -0.23 to -0.40 and were significant, indicating that heifers that ate less and had smaller MMW from 240 d of age to first calving had superior cow efficiency ratios, although correlations of DMI and MMW with R1 were not significant (Table 5). A tendency (P < 0.10) existed for ADG to be negatively correlated with R2 and R4, the measures of cow efficiency that did not include salvage value of the cow. Black et al. (2013) also observed that ADG in heifers tended to be positively correlated with RFI in cows such that heifers that gained less weight had lower (more efficient) RFI values as cows. However, Arthur et al. (1999) reported a near-zero correlation of postweaning ADG with NFI of cows. Based on studies conducted in Australia, Herd and Arthur (2012) observed phenotypic and genetic correlations ranging from 0.28 to 0.70 and from 0.72 to 0.98, respectively, between ADG, metabolic weight, feed intake, and RFI measured during the postweaning period and the same traits measured at maturity. The authors concluded that selection for improved feed efficiency in young growing animals is also expected to result in improved efficiency of mature cows.

Correlations of RFI, RG, DMI, ADG, and MMW with component traits of cow efficiency ratios are presented in Table 6. Residual feed take was significantly correlated only with feed consumed by the cow from 240 d of age to weaning of the first calf (DF1; r = 0.31; P < 0.0001). A part-whole relationship existed between RFI and DF1 as RFI was determined in part by feed consumption from 240 d of age to first calving. Arthur et al. (1999) reported a phenotypic correlation of 0.30 (P < 0.05) between postweaning RFI and feed intake by the cow, and Archer et al. (2002) estimated a genetic correlation of 0.64 between these traits. Basarab et al. (2007) observed that cows that produced low-RFI progeny consumed less feed (P < 0.05) during their second trimester of pregnancy. Shike et al. (2014) reported that medium- and high-RFI heifers had greater DMI than low-RFI heifers in a follow-up RFI test as 2-yr-old lactating cows and as 2-yr-old nonlactating cows. In a study by Black et al. (2013), cows that were most efficient as heifers (low-RFI group) consumed approximately 11% less daily DM as lactating 3-yr-old cows than cows classified as medium- or high-RFI as heifers. Thus, evidence exists that heifers that consume less feed during the postweaning period will also consume less feed as cows most likely due to differences in BW.


View Full Table | Close Full ViewTable 6.

Correlations1 of residual feed intake (RFI), residual gain (RG), dry matter intake (DMI), average daily gain (ADG), and metabolic weight (MMW) with component traits of cow efficiency ratios2

 
Variable RFI RG DMI ADG MMW
DW1 0.04 0.07 0.56 0.51 0.80
0.62 0.36 <0.0001 <0.0001 <0.0001
DW2 0.00 0.12 0.51 0.51 0.76
0.97 0.13 <0.0001 <0.0001 <0.0001
DW3 -0.04 0.15 0.47 0.51 0.71
0.65 0.06 <0.0001 <0.0001 <0.0001
DF0 0.05 -0.24 0.38 0.13 0.73
0.52 <0.01 <0.0001 0.09 <0.0001
DF1 0.31 -0.21 0.65 0.33 0.73
<0.0001 <0.01 <0.0001 <0.0001 <0.0001
DF2 -0.01 0.10 0.19 0.22 0.27
0.86 0.22 0.02 <0.01 <0.001
DF3 -0.05 0.04 0.14 0.17 0.26
0.54 0.59 0.08 0.03 <0.01
PW1 0.13 -0.05 0.23 0.10 0.16
0.14 0.60 0.01 0.29 0.07
PW2 -0.03 0.05 0.18 0.18 0.26
0.68 0.56 0.02 0.03 <0.01
PW3 -0.09 0.09 0.03 0.13 0.12
0.25 0.26 0.68 0.11 0.12
PF1 0.17 0.01 0.18 0.10 0.02
0.06 0.95 0.04 0.26 0.82
PF2 0.02 0.06 0.06 0.08 -0.02
0.84 0.49 0.44 0.35 0.81
PF3 0.02 -0.10 -0.02 -0.07 -0.01
0.76 0.23 0.77 0.36 0.88
1Significance level for the test Prob > |r| under H0: ρ = 0 is presented below the correlation coefficient.
2See Table 2 for definitions of symbols and acronyms.

In our study, RFI also tended to be correlated with feed consumption of the first calf (PF1; r = 0.17; P = 0.06). Correlations of RFI with dam weights at weaning of their first, second, and third calves were nonsignificant. Genetic and phenotypic correlations between postweaning RFI and mature cow size were also low in the studies of Arthur et al. (1999) and Herd et al. (2003). Bourg (2011) reported no relationship between RFI EPD and cow weight at weaning in any year of a 4-yr study. Arthur et al. (2005) reported that cows in a low-RFI selection line were heavier than cows in a high-RFI selection line, but that the differences were not significant at any stage of the breeding cycle. Residual gain was positively correlated with DW3 (r = 0.15; P = 0.06) and was negatively correlated with DF0 (r = -0.24; P < 0.01) and DF1 (r = -0.21; P < 0.01). Dry matter intake, ADG, and MMW exhibited highly significant positive correlations with weights of the cows at weaning of their first, second, and third calves with correlations ranging from 0.47 to 0.80, were positively correlated with feed consumption of the cows during different phases of the life cycle with correlations ranging from 0.13 (P = 0.09) to 0.73 (P < 0.0001), and were positively, although not always significantly, correlated with weights of the progeny at first, second, and third weaning with correlations varying between 0.03 and 0.26. Correlations of DMI, ADG, and MMW with progeny feed consumption generally were not different from zero.

Correlations of RFI with various traits of the dam are shown in Table 7. Residual feed intake was correlated with feed consumption (r = 0.32; P < 0.001) and with feed efficiency (r = 0.73; P < 0.0001) from 240 d of age to first calving. These results are in agreement with those of Lancaster et al. (2009a), who reported correlations of 0.70 and 0.59 (P < 0.05) between RFI and DMI and between RFI and FCR, respectively, and with those of Herd and Arthur (2012), who estimated a genetic correlation of 0.66 between RFI and feed intake, as well as between RFI and FCR.


View Full Table | Close Full ViewTable 7.

Correlations1 of residual feed intake (RFI), residual gain (RG), dry matter intake (DMI), average daily gain (ADG), and metabolic weight (MMW) with traits of the dam2

 
Variable RFI RG DMI ADG MMW
Age at puberty -0.14 -0.04 -0.09 -0.05 0.08
0.11 0.62 0.30 0.58 0.37
Age at first calving -0.08 -0.33 0.03 -0.16 0.37
0.30 <0.0001 0.73 0.05 <0.0001
Age at second calving -0.11 -0.23 -0.04 -0.15 0.25
0.18 <0.01 0.63 0.05 <0.01
Age at third calving -0.01 -0.19 0.06 -0.08 0.20
0.92 0.02 0.42 0.33 0.01
Feed consumption, 240 d to first calving 0.32 -0.27 0.62 0.26 0.69
<0.001 <0.001 <0.0001 <0.01 <0.0001
Feed efficiency 240 d to first calving 0.73 -0.87 0.32 -0.47 0.11
<0.0001 <0.0001 <0.0001 <0.0001 0.15
240-d weight 0.05 -0.25 0.38 0.13 0.73
0.55 <0.01 <0.0001 0.10 <0.0001
240-d height -0.02 0.13 0.34 0.20 0.66
0.76 0.10 <0.0001 0.01 <0.0001
240-d wt:ht ratio 0.08 -0.29 0.34 0.07 0.56
0.33 <0.001 <0.0001 0.39 <0.0001
Wt at first calving -0.04 0.06 0.60 0.54 0.94
0.62 0.42 <0.0001 <0.0001 <0.0001
Wt at second calving 0.02 0.11 0.53 0.52 0.76
0.78 0.15 <0.0001 <0.0001 <0.0001
Wt at third calving -0.06 0.23 0.46 0.55 0.71
0.45 <0.01 <0.0001 <0.0001 <0.0001
Ht at first calving -0.09 0.10 0.41 0.43 0.69
0.27 0.20 <0.0001 <0.0001 <0.0001
Ht at second calving -0.08 0.14 0.39 0.44 0.62
0.32 0.07 <0.0001 <0.0001 <0.0001
Ht at third calving -0.09 0.20 0.36 0.44 0.57
0.27 0.01 <0.0001 <0.0001 <0.0001
Ht at first weaning -0.07 0.09 0.42 0.42 0.67
0.40 0.26 <0.0001 <0.0001 <0.0001
Ht at second weaning -0.08 0.18 0.40 0.47 0.62
0.30 0.02 <0.0001 <0.0001 <0.0001
Ht at third weaning -0.15 0.22 0.34 0.46 0.59
0.06 <0.01 <0.0001 <0.0001 <0.0001
Wt:ht at first calving -0.01 0.05 0.59 0.52 0.90
0.94 0.51 <0.0001 <0.0001 <0.0001
Wt:ht at second calving 0.07 0.09 0.50 0.46 0.68
0.39 0.27 <0.0001 <0.0001 <0.0001
Wt:ht at third calving -0.04 0.19 0.40 0.48 0.63
0.64 0.01 <0.0001 <0.0001 <0.0001
Wt:ht at first weaning 0.08 0.06 0.52 0.46 0.70
0.33 0.43 <0.0001 <0.0001 <0.0001
Wt:ht at second weaning 0.04 0.08 0.47 0.43 0.68
0.66 0.34 <0.0001 <0.0001 <0.0001
Wt:ht at third weaning 0.01 0.09 0.41 0.41 0.60
0.85 0.26 <0.0001 <0.0001 <0.0001
Change in wt:ht, lactation 1 0.12 0.03 -0.03 -0.02 -0.18
0.12 0.72 0.68 0.82 0.02
Change in wt:ht, lactation 2 -0.05 -0.02 -0.03 -0.03 0.03
0.54 0.81 0.72 0.72 0.71
Change in wt:ht, lactation 3 0.08 -0.11 0.01 -0.09 -0.04
0.32 0.18 0.93 0.26 0.64
Wt gain, lactation 1 0.14 0.03 0.01 0.01 -0.15
0.07 0.73 0.90 0.92 0.06
Wt gain, lactation 2 -0.05 0.02 0.00 0.01 0.05
0.54 0.84 0.96 0.87 0.50
Wt gain, lactation 3 0.05 -0.06 -0.01 -0.06 -0.05
0.57 0.45 0.95 0.42 0.57
Milk production, lactation 1 0.00 -0.10 0.17 0.05 0.24
0.97 0.30 0.07 0.60 0.01
Milk production, lactation 2 0.00 0.02 0.06 0.07 0.07
0.98 0.83 0.44 0.39 0.37
Milk production, lactation 3 -0.11 0.08 -0.06 0.04 0.01
0.18 0.37 0.48 0.67 0.88
1Significance level for the test Prob > |r| under H0: ρ = 0 is presented below the correlation coefficient.
2See Table 2 for definitions of symbols and acronyms.

Residual feed intake was not significantly correlated with age at puberty or age at calving. Shaffer et al. (2011) reported a small negative linear relationship between RFI and age at puberty (r = -0.16; P = 0.06). Arthur et al. (2005) observed a trend for low-RFI cows to calve an average of 5 d later than high-RFI cows. However, pregnancy, calving, and weaning rates were similar (P > 0.05) in the high and low selection lines. In a later study, involving cows from the same high- vs. low-RFI selection experiment in Australia, low-RFI line females calved significantly later in the calving season than high-RFI line females (d 35.7 ± 3.0 vs. d 27.6 ± 2.4), although pregnancy rates and calving rates once again did not differ between the lines (Donoghue et al., 2011). The results indicated that there was a delayed date of pregnancy during the first mating season, which led to a later calving date for the low-RFI line heifers. The later first calving date was maintained at subsequent calving. Basarab et al. (2007) also observed that low-RFI cows calved 5 to 6 d later, but pregnancy rate, calving rate, and weaning rate did not differ among cows producing low-, medium-, and high-RFI progeny. Shaffer et al. (2011) reported that heifers classified into a positive RFI group reached puberty earlier than those classified as having negative RFI. The authors also observed that each 1-unit increase in RFI was associated with a decrease of 7.54 d in age at puberty, but did not affect conception or pregnancy rates. Shaffer et al. (2011) suggested that low-RFI animals may metabolize reproductive hormones at a different rate, which could delay the ability of heifers to reach puberty. Shike et al. (2014) reported no differences in percentage of females kept as replacements, first AI conception rate, overall pregnancy rate, or age at calving among heifers classified into low-, medium-, or high-RFI groups. Black et al. (2013) reported no differences in days to calving among low, medium, or high cows as determined by their RFI values as heifers. Basarab et al. (2011) observed that prepubertal heifers consumed 4.7% less feed and had a 7.4% improvement in FCR compared with postpubertal heifers that had equal ADG, body size, and backfat thickness, and, therefore, suggested that calculating RFI for a mixture of pre- and postpubertal heifers will favor the later maturing heifers. Basarab et al. (2012) concluded that adjustment of feed intake for energy requirements associated with body size, growth, body composition, and sexual activity results in RFI being independent of pregnancy rate, calving rate, calf mortality, and weight of calf weaned per cow exposed to breeding.

In our study, RFI did not exhibit significant correlations with weight, height, or weight:height ratio at 240 d of age, at first, second, or third calving, or at weaning of first, second, or third calf, nor did it exhibit significant correlations with change in weight:height ratio or weight gain during first, second, or third lactation. Arthur et al. (2005) reported that high-RFI cows were generally fatter than low-RFI cows. In contrast to our results and to those of Arthur et al. (2005), Basarab et al. (2007) observed that cows that produced low-RFI progeny averaged 2 to 3 mm more backfat over the 12th and 13th ribs and lost less weight during early lactation than cows that produced high-RFI progeny. This effect was most pronounced in older cows, where low-RFI cows had much more backfat than high-RFI cows at weaning, precalving, and prebreeding stages of production. Positive RFI heifers had more backfat as estimated by ultrasound and tended to have greater LM area per 100 kg of BW than negative RFI heifers at the beginning of the trial conducted by Shaffer et al. (2011). No differences in final subcutaneous fat thickness or IMF were observed, whereas LMA per 100 kg of BW was significantly greater in positive than in negative RFI heifers.

Finally, RFI was not correlated with milk production during any of the 3 lactations in our study. Arthur et al. (1999) reported a near-zero correlation between postweaning NFI and milk yield of 4-yr-old cows. Bourg (2011) saw no relationship between RFI EPD and model-estimated peak milk yield in any year of a 4-yr study. Black et al. (2013) also observed that heifer RFI was not associated with energy-corrected milk production of cows. Shike et al. (2014) reported that heifer RFI classification did not affect cow body weight, hip height, backfat, or milk production at 60 d postpartum in lactating cows and did not affect cow body weight, hip height, body condition score, or backfat at 240 d postpartum in nonlactating cows. Likewise, Black et al. (2013) reported that performance of 3-yr-old lactating cows was similar among low, medium, and high groups as determined by heifer RFI with initial age, initial BW, final BW, mean body condition score, ADG, and energy-corrected milk production not differing among the groups. Divergent lines of cows selected to produce high- vs. low-RFI progeny also had similar milk yields and performance characteristics in the Australian study (Arthur et al., 2005).

Residual gain was negatively correlated with age at first, second, and third calving (P < 0.02). Residual gain was also negatively correlated with feed consumption (r = -0.27; P < 0.001) and with feed efficiency (r = -0.87; P < 0.0001) from 240 d of age to first calving, as well as with 240-d weight (r = -0.25; P < 0.01) and with weight:height ratio at 240 d of age (r = -0.29; P < 0.001). In addition, RG had significant positive correlations with weight and height at third calving, height at second and third weaning, and weight:height ratio at third calving. Correlations of RG with milk production did not differ from zero. Therefore, selection for increased RG would tend to result in earlier ages at calving, but would also result in taller and heavier cows by the time they gave birth to their third calves.

Dry matter intake of heifers was not correlated with age at puberty or age at calving. Average daily gain of heifers was negatively correlated with age at first and second calving (P = 0.05). However, correlations of MMW with age at calving were positive and highly significant, indicating that larger heifers were older at first, second, and third calving. Correlations of feed consumption from 240 d of age to first calving with DMI, ADG, and MMW were positive and highly significant, as would be expected. Correlations of feed efficiency from 240 d of age to first calving with DMI, ADG, and MMW were 0.32 (P < 0.0001), -0.47 (P < 0.0001), and 0.11 (P = 0.15), respectively. Correlations of 240-d weight and height with DMI and MMW were positive and highly significant, whereas correlations of these 2 traits with ADG tended to be positive. Weight:height ratio at 240 d of age had highly significant positive correlations with DMI and MMW but was not correlated with ADG.

Weight, height, and weight:height ratio at first, second, and third calving and weaning, and height at first, second, and third weaning all had strong positive correlations (P < 0.0001) with DMI, ADG, and MMW. Change in weight:height ratio and weight gain during the 3 lactations were not correlated with DMI, ADG, or MMW with the exception of the negative correlations with MMW in lactation 1. Milk production during lactation 1 was positively correlated with DMI (P = 0.07) and with MMW (P = 0.01), indicating that heifers that consumed more feed and had greater body weights from 240 d of age to first calving produced more milk during their first lactation. The remaining correlations of DMI, ADG, and MMW with milk production did not differ from zero. Black et al. (2013) reported that heifer DMI was positively correlated with d 70 energy-corrected milk production (P = 0.02) in 3-yr-old cows.

In summary, postweaning heifer RFI had small negative (i.e., favorable), but nonsignificant, correlations with cow efficiency ratios expressed on both a life cycle basis and on an actual lifetime basis. With the exception of a significant positive correlation with feed consumption from 240 d of age to weaning of the first calf, heifer RFI was not significantly correlated with subsequent dam weights at weaning, progeny weaning weights, dam feed consumption during various stages of the life cycle, or progeny feed consumption prior to weaning. In addition, significant relationships were not observed between heifer RFI and age at puberty or at calving, weight, height, or weight:height ratio at calving or at weaning, change in weight:height ratio or in weight gain during lactation, or in milk production during the first 3 lactations. Thus, results of this study do not indicate any serious antagonisms of postweaning heifer RFI with subsequent cow and progeny performance traits or with life cycle or actual lifetime cow efficiency. The findings also indicate that selection for increased RG would tend to result in earlier ages at calving, but would also result in taller and heavier cows.

 

References

Footnotes


Comments
Be the first to comment.



Please log in to post a comment.
*Society members, certified professionals, and authors are permitted to comment.