Seminário temp bov

16
Greenwood L. M. Cafe, D. L. Robinson, D. M. Ferguson, B. L. McIntyre, G. H. Geesink and P. L. efficiency, carcass and meat quality traits Cattle temperament: Persistence of assessments and associations with productivity, doi: 10.2527/jas.2010-3304 originally published online December 17, 2010 2011, 89:1452-1465. J ANIM SCI http://jas.fass.org/content/89/5/1452 the World Wide Web at: The online version of this article, along with updated information and services, is located on www.asas.org at UNESP on July 20, 2011 jas.fass.org Downloaded from

Transcript of Seminário temp bov

Page 1: Seminário temp bov

GreenwoodL. M. Cafe, D. L. Robinson, D. M. Ferguson, B. L. McIntyre, G. H. Geesink and P. L.

efficiency, carcass and meat quality traitsCattle temperament: Persistence of assessments and associations with productivity,

doi: 10.2527/jas.2010-3304 originally published online December 17, 20102011, 89:1452-1465.J ANIM SCI 

http://jas.fass.org/content/89/5/1452the World Wide Web at:

The online version of this article, along with updated information and services, is located on

www.asas.org

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 2: Seminário temp bov

Cattle temperament: Persistence of assessments and associations with productivity, efficiency, carcass and meat quality traits1

L. M. Cafe,*†2 D. L. Robinson,*† D. M. Ferguson,*‡ B. L. McIntyre,*§ G. H. Geesink,*# and P. L. Greenwood*†

*Australian Cooperative Research Centre for Beef Genetic Technologies, University of New England, Armidale, New South Wales 2351, Australia; †Industry & Investment New South Wales, Beef Industry Centre, Armidale,

New South Wales 2351, Australia; ‡CSIRO Livestock Industries, FD McMaster Laboratories, Armidale, New South Wales 2350, Australia; §Department of Agriculture and Food, Western Australia,

South Perth, Western Australia 6151, Australia; and #Department of Meat Science, University of New England, Armidale, New South Wales 2351, Australia

ABSTRACT: Relationships between temperament and a range of performance, carcass, and meat quality traits in young cattle were studied in 2 experiments conducted in New South Wales (NSW) and Western Australia (WA), Australia. In both experiments, growth rates of cattle were assessed during backgrounding on pasture and grain finishing in a feedlot. Carcass and objective meat quality characteristics were measured after slaughter. Feed intake and efficiency during grain finishing were also determined in NSW. Brahman (n = 82 steers and 82 heifers) and Angus (n = 25 steers and 24 heifers) cattle were used in the NSW experiment. In NSW, temperament was assessed by measuring flight speed [FS, m/s on exit from the chute (crush)] on 14 occasions, and by assessing agitation score during con-finement in the crush (CS; 1 = calm to 5 = highly agitated) on 17 occasions over the course of the ex-periment. Brahman (n = 173) and Angus (n = 20) steers were used in the WA experiment. In WA, tem-perament was assessed by measuring FS on 2 occasions during backgrounding and on 2 occasions during grain feeding. At both sites, a hormonal growth promotant (Revalor-H, Virbac, Milperra, New South Wales, Aus-tralia) was applied to one-half of the cattle at feedlot entry, and the Brahman cattle were polymorphic for 2

calpain-system markers for beef tenderness. Tempera-ment was not related (most P > 0.05) to tenderness gene marker status in Brahman cattle and was not (all P > 0.26) modified by the growth promotant treatment in either breed. The Brahman cattle had greater indi-vidual variation in, and greater correlations within and between, repeated assessments of FS and CS than did the Angus cattle. Correlations for repeated measures of FS were greater than for repeated assessments of CS, and the strength of correlations for both declined over time. Average FS or CS for each experiment and loca-tion (NSW or WA × backgrounding or finishing) were more highly correlated than individual measurements, indicating that the average values were a more reliable assessment of cattle temperament than any single mea-sure. In Brahman cattle, increased average FS and CS were associated with significant (P < 0.05) reductions in backgrounding and feedlot growth rates, feed intake and time spent eating, carcass weight, and objective measures of meat quality. In Angus cattle, the associa-tions between temperament and growth rates, feed in-take, and carcass traits were weaker than in Brahmans, although the strength of relationships with meat qual-ity were similar.

Key words: carcass, cattle, flight speed, meat quality, productivity, temperament

©2011 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2011. 89:1452–1465 doi:10.2527/jas.2010-3304

1 This work was possible because of the financial and in-kind support and efforts of many individuals from the Cooperative Re-search Centre for Beef Genetic Technologies (Armidale, New South Wales, Australia); Industry & Investment NSW (Armidale, New South Wales, Australia); Queensland Department of Employment, Economic Development and Innovation (City East, Queensland, Australia); CSIRO Livestock Industries (St. Lucia, Queensland, Australia); the University of New England (Armidale, New South Wales, Australia); Western Australia Department of Agriculture and Food (South Perth, Western Australia, Australia); Meat and Live-

stock Australia (North Sydney, New South Wales, Australia); South Australian Research and Development Institute (Urrbrae, South Australia, Australia); Victorian Department of Primary Industries (Melbourne, Victoria, Australia); the Australian Brahman Breeders’ Association (Rockhampton, Queensland, Australia); John Dee Ab-attoir (Warwick, Queensland, Australia); and Harvey Beef (Harvey, Western Australia, Australia).

2 Corresponding author: [email protected] July 8, 2010.Accepted December 8, 2010.

1452

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 3: Seminário temp bov

INTRODUCTION

Cattle vary in their behavioral response to stressful events, and this trait is defined as temperament. Ex-treme or reactive responses can be detrimental to cattle welfare and the safety of human handlers. Evidence is emerging that cattle with calmer temperaments have improved productivity; however, the effects of tempera-ment on economically important traits can be variable, and the biological basis for the effects is not well under-stood (Ferguson et al., 2006).

Several tests have been developed to measure tem-perament by using the escape and avoidance behaviors that cattle display when responding to stressful events, such as handling by humans (reviewed by Burrow, 1997). Two tests, which are simple and safe to mea-sure and are being used by the Australian beef cattle industry to select for calmer temperament, are flight speed (FS; Burrow et al., 1988) and crush score (CS; Grandin, 1993). It is likely that these tests measure dif-ferent combinations of aspects of cattle temperament, including general agitation and fear of humans, but this remains a topic of discussion (Burrow, 1997; Kilgour et al., 2006; Petherick et al., 2002, 2009a). Faster FS have been shown to lead to slower growth rates, particularly under more intensive management conditions (Burrow and Dillon, 1997; Petherick et al., 2009b); reduced feed conversion efficiency (Petherick et al., 2002); and re-duced yield of poorer quality meat (King et al., 2006). Similar effects have been shown with greater CS (Voi-sinet et al., 1997a,b).

The present study was conducted to investigate the persistency over time of cattle temperament, as assessed by FS and CS, and to assess relationships between tem-perament and a comprehensive range of performance traits in young Brahman and Angus cattle. The experi-mental design also allowed potential interactions be-tween temperament and tenderness gene marker status, sex, hormonal growth promotant (HGP) treatment, and cattle management and meat processing practices to be studied.

MATERIALS AND METHODS

Use of animals and the procedures performed in this study were approved by the Orange Agricultural Insti-tute Animal Ethics Committee of Industry & Invest-ment New South Wales, the Rockhampton Animal Ex-perimentation Ethics Committee of the Commonwealth Scientific and Industrial Research Organisation, and the Animal Ethics Committee of the Western Austra-lian Department of Agriculture and Food.

Animals and Experimental Designs

The present study was conducted as a part of 2 concurrent experiments designed to study the effects and mechanisms of action of tenderness gene markers, and their interaction with management and process-

ing practices. The Brahman cattle, treatments, data, and sample collection and their management through-out the experiments are described in detail by Cafe et al. (2010a,b). Results are also reported here for Angus cattle, which were treated identically to the Brahman cattle in both experiments, with the 2 breeds managed together in combined replicates throughout the experi-ments. The numbers of animals in the experiments are presented in Table 1.

Briefly, the experiments were conducted at Industry & Investment New South Wales, Agricultural Research and Advisory Station [Glen Innes, New South Wales (NSW); 29°44′ S, 151°42′ E, altitude 1,057 m, n = 213 cattle] and at the Western Australian Department of Agriculture and Food’s Vasse Research Station near Busselton, Western Australia (WA; 33°45′ S, 115°21′ E, altitude 25 m, n = 193). Brahman cattle were selected for experimental groups based on their genotype for the calpastatin (Barendse, 2002) and calpain 3 (Barendse et al., 2008) tenderness gene markers. At both sites, a small group of Angus cattle with only the favorable alleles for the calpastatin and calpain 3 gene markers were included as positive controls for biological studies on the calpain system. Equal numbers of heifers and steers were used in the NSW experiment, whereas only steers were used in the WA experiment. One-half the cattle in each experiment were treated with a HGP (Revalor-H, Virbac, Milperra, New South Wales, Aus-tralia) at feedlot entry.

The Brahman weaner cattle were sourced from re-search and commercial herds (n = 15 herds) in NSW, WA, and Queensland. Angus weaner cattle were sourced from research herds (n = 3) in NSW and WA. All cattle were weaned at approximately 6 to 8 mo of age, but be-cause of differing production systems in their regions of origin, the Angus cattle were approximately 2 mo older than the Brahman cattle at both sites. The cattle were grown (backgrounded) on pasture for approximately 6 mo, then grain finished in a feedlot for 80 d in WA and 117 d in NSW. In NSW, the cattle were trans-ported approximately 160 km to the Australian Coop-erative Research Centre for Beef Genetic Technologies “Tullimba” research feedlot near Kingstown (30°20′ S, 151°10′ E, altitude 560 m) for grain finishing. In WA, the cattle were transferred to the feedlot facility at the Vasse Research Station for grain finishing. Feed intake and feeding behavior were measured in the NSW feed-lot by using an automatic individual feeding system, as described by Bindon (2001), and feed efficiency traits were calculated as described in detail by Cafe et al. (2010a).

Cattle from each experiment were transported to their respective commercial abattoirs the day before slaughter, with no mixing of pens during transport or lairage. For each experiment, one-half of the replicates were slaughtered on each of 2 slaughter dates, with the remaining replicates slaughtered 2 d later. Slaughter was conducted through captive bolt stunning and ex-sanguination. Electrical stimulation of the carcasses

1453Cattle temperament and productivity

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 4: Seminário temp bov

was limited to that necessary for the hide removal pro-cess at both sites, plus immobilization during exsan-guination in WA. Standard AUS-MEAT carcasses were prepared (AUS-MEAT, 2007) and split into 2 sides, and the right sides were resuspended by the pelvis [ten-derstretch (TS) suspension method; Thompson, 2002]. Sides were graded according to Meat Standards Aus-tralia (2009) procedures, and at bone-out, the musculus longissimus lumborum (LLM) and musculus semiten-dinosus (STN) were taken from the Achilles-suspended (AT) sides, and the LLM were removed from the TS sides. These muscles were divided and aged at 1°C for either 1 or 7 d before freezing at −20°C. Sample prep-aration and measurement of texture (shear force and compression), cooking loss, CIELAB meat color, and intramuscular fat percentage (determined by near-in-frared spectrophotometry) were performed as described by Perry et al. (2001a).

Temperament Assessments

Temperament was assessed by FS (Burrow et al., 1988) in both NSW and WA and also by CS (Grandin, 1993) in NSW. The measurements were taken when the

cattle were being handled through the yards for other management or data collection purposes. Cattle were confined for a period of at least 5 s in a single-animal weighing crate before being released. Crush score was assessed visually during the period in the weighing crate, using a 5-point scale of agitation based on the behavioral scoring system described by Grandin (1993), which was applied to cattle restrained in a squeeze chute (crush) and head bail. Minor modifications were made so that it was more suitable for loosely restrained cattle. The scale used was as follows: 1= calm, stand-ing still, head mostly still, slow movements; 2 = slightly restless, looking around more quickly, moving feet; 3 = restless, moving backward and forward, shaking crate; 4 = nervous, continuous vigorous movement backward and forward, snorting; 5 = very nervous, continuous violent movement, attempting to jump out. All CS as-sessments throughout the experiment were made by the same person.

When the cattle were released from the weighing crate, flight time was measured over a distance of 1.7 m at both sites, and converted to FS (m/s) for analyses. Flight speeds of 1 to 1.5 m/s equated to cattle leav-ing the crush at a walk, FS of 2 to 2.5 m/s equated to

Table 1. Descriptive statistics for the major traits assessed in Brahman and Angus cattle in the New South Wales (NSW) and Western Australia (WA) experiments

Variable

NSW Brahman WA Brahman NSW Angus WA Angus

n Mean SD n Mean SD n Mean SD n Mean SD

Growth, kg                         Background start BW 164 218 36.0 173 208 59.2 49 295 28.3 20 293 9.4 Feedlot start BW 164 321 38.1 173 343 35.6 49 419 41.6 20 403 25.0 Feedlot end BW 164 435 55.8 173 449 51.1 49 578 59.1 20 520 28.5 Background ADG 164 0.72 0.119 173 0.64 0.169 49 0.71 0.134 20 0.52 0.097 Feedlot ADG 164 1.01 0.294 173 1.28 0.345 49 1.43 0.282 20 1.42 0.242Carcass                         Carcass wt, kg 164 244 32.3 173 242 25.7 49 321 36.6 20 270 15.1 Dentition1 164 0.05 0.310 173 0.72 1.032 49 0.61 0.931 20 1.90 0.447 LLM area,2 cm2 164 59.9 8.57 143 63.6 6.25 49 67.0 8.93 17 66.1 4.57 Rump fat, mm 164 12.0 2.61 173 8.0 2.56 49 18.3 5.44 20 8.9 2.67 Rib fat,2 mm 164 6.2 2.08 143 5.3 2.38 49 9.4 2.81 17 8.2 3.26 Marble score2 164 261 66.2 143 293 61.9 49 424 72.8 17 321 41.2 Meat color score2 164 2.8 1.07 143 2.7 1.05 49 3.2 0.74 17 2.8 0.83 Ultimate pH2 164 5.49 0.051 143 5.57 0.085 49 5.49 0.054 17 5.57 0.047Shear force,3 N                         AT 1-d aged LLM 161 78.2 18.53 140 52.2 11.46 49 61.7 11.78 16 44.1 7.44 AT 7-d aged LLM 161 68.1 17.63 133 49.5 10.25 46 51.7 10.79 15 44.4 7.80 TS 1-d aged LLM 164 47.2 5.61 141 51.6 11.81 49 37.0 4.43 17 41.2 6.39 TS 7-d aged LLM 163 45.6 5.56 128 46.0 9.63 49 36.5 4.01 14 40.4 5.29Feed intake and efficiency                         Feedlot DMI, kg of DM/d 160 8.0 1.36 — — — 49 11.0 1.34 — — — FCR, kg of DM/kg of BW gain 160 7.5 2.39 — — — 49 7.5 2.22 — — — NFI,4 kg of DM/d 160 −0.07 0.830 — — — 49 0.23 0.97 — — — Feedlot ADG, kg 160 1.13 0.314 — — — 49 1.53 0.322 — — — Feeding time, min/d 160 73.4 20.53 — — — 49 106.5 23.63 — — — Feeding sessions, n/d 160 11.7 6.14 — — — 49 8.8 3.90 — — —

1Dentition = number of erupted permanent incisors.2Meat Standards Australia (2009) grading data, where marble score is from 100 to 1,100 in increments of 10; meat color score is from 1 (lightest)

to 9 (darkest), and ultimate pH is the pH at grading.3AT = Achilles-suspended side; TS = tenderstretched side; LLM = musculus longissimus lumborum.4NFI = net feed intake.

Cafe et al.1454

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 5: Seminário temp bov

cattle leaving the crush at a trot, and FS of 3 to 3.5 m/s equated to cattle leaving the crush at a run. Dur-ing backgrounding in NSW, the yard design required the cattle to turn right at 90° into a side yard upon re-lease from the crate. In this case, the FS measurement began after the animals had made the turn and were traveling in a straight line. Portable yard panels were used in the side yard to narrow the exit sufficiently to keep the cattle moving in a direct route over the 1.7-m flight distance. At the feedlot in NSW and during both phases in WA, FS measurements were taken in a straight line directly ahead of the point of release from the weigh crate. In WA, the same set of yards was used to handle the cattle during the backgrounding and feedlot phases.

In NSW, FS was measured on 5 occasions during backgrounding and on 9 occasions during grain finish-ing (FS 1 to 14); CS was assessed on 6 occasions during backgrounding and on 11 occasions during feedlot fin-ishing (CS 1 to 17). The timing of the temperament as-sessments and the timing of the more invasive handling events are shown for the NSW experiment in Figure 1. Blood sampling from the tail vein was conducted in the race before weighing, and then the temperament measures were made as described above. Ultrasound scanning was conducted on 3 occasions with the cattle caught in the head bail; CS was assessed on each occa-sion during the final 30 s of the scanning process, which took approximately 2 min; and FS was measured after release from the crush on 1 of these occasions. Muscle biopsy was performed on the LLM, STN, and musculus semimembranosus under local anesthetic using a drill biopsy technique (Gardner et al., 2001) on 2 occasions with the cattle caught in the head bail. Temperament assessments were not made when biopsies were per-formed.

In WA, FS was measured on 2 occasions during back-grounding and on 2 occasions during grain finishing (FS 1 to 4). Flight speed 1 was measured 12 wk after the commencement of backgrounding, when the cattle were in the yard for weighing; FS 2 was measured after ultrasound scanning 10 wk later; FS 3 was measured at feedlot entry a further 8 wk later; and FS 4 was mea-sured after ultrasound scanning a further 10 wk later.

Statistical Analyses

The consistency of FS and CS in ranking animals throughout the experiment was assessed using Pearson correlations in Genstat (VSN International Ltd., Hemel Hempstead, UK). The consistency of FS was analyzed for both sites (NSW, 14 measures; WA, 4 measures), and the consistency of CS was analyzed using the 17 assessments made in NSW. The significance of the day of measurement effects for the repeated measures of FS at both sites and of CS in NSW were conducted in Genstat using linear mixed models and the REML

methodology (Robinson, 1987), with animal fitted as a random term.

The average temperaments during backgrounding and finishing (FS and CS for NSW, and FS alone for WA) were used in the analyses of temperament effects on other traits because of the differences in the way that FS and CS characterized the temperament of cat-tle; the changes in FS and CS between backgrounding and grain finishing; and the greater reliability of av-erages compared with individual assessments. The ef-fects of temperament on production, carcass traits, and meat quality traits were assessed using linear mixed models in Genstat. Separate analyses were carried out for each breed (Brahman and Angus) and experimental site (WA and NSW) combination because of the differ-ences in experimental designs and residual variances. To ensure all aspects of the experimental design were accounted for, the full models included the fixed effects of the tenderness marker genotypes, HGP treatment, and, for the NSW herd, sex. Random effects included in the models were property of origin, backgrounding rep-licate, feedlot replicate, slaughter group within slaugh-ter day, and the first-order interactions. The effect of temperament was fitted as a covariate in the full model for each site × breed combination, with the average temperament variables (FS or CS during background-ing or finishing) fitted as single covariates in separate analyses, and both the linear and quadratic fits were tested. Main effects and interactions were considered significant at P < 0.05 and were considered a tendency toward significance at P < 0.10.

RESULTS

The primary purpose of this paper is to report the assessments of temperament in Brahman and Angus cattle in NSW and WA and their relationships with productivity, carcass traits, and meat quality traits, for which descriptive statistics are presented in Table 1. Results for the effects of HGP, tenderness gene marker status, and sex on the measured traits in the Brahman cattle are presented by Cafe et al. (2010a,b).

Relationships Between Temperament, HGP Treatment, Tenderness Gene Marker Status, and Sex

No interactions were observed between HGP treat-ment and temperament assessments (all P > 0.26) in Brahman or Angus cattle at either site. No association was observed between CS and tenderness gene marker status (all P ≥ 0.12) in Brahman cattle in NSW, and no consistent association was observed between FS and tenderness gene marker status (Cafe et al., 2010a) in Brahman cattle at either site. Where there were indica-tions of sex differences in NSW, heifers always had nu-merically greater temperament scores than did steers,

Cattle temperament and productivity 1455

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 6: Seminário temp bov

but the differences were small and rarely significant. The effect of sex on FS was not significant for either breed (all P ≥ 0.09). Brahman heifers had greater CS than steers during backgrounding (2.15 vs. 1.98, SED = 0.079, P = 0.037) and in the feedlot (1.59 vs. 1.45, SED = 0.066, P = 0.045); no significant effect of sex on CS was observed in the Angus cattle (all P ≥ 0.18). Be-cause of the lack of interactions between temperament and these effects, further discussion on temperament is made without reference to them.

FS and CS over Time

FS in NSW. Means for all 14 FS measurements on NSW cattle are shown graphically in Figure 1a, with means and SD of a representative 8 (selected to provide an even spread over time) presented in Table 2. A sig-nificant effect of day of measurement on FS (P < 0.001) was observed in both breeds.

In the Brahmans, FS decreased during background-ing (FS 1 = 2.1 to FS 5 = 1.6 m/s, SED = 0.05 m/s, P

Figure 1. Mean (±SEM) a) flight speed (FS) and b) crush score (CS) for Angus (●, ○) and Brahman (■, □) cattle in the New South Wales experiment during backgrounding (solid symbols) and finishing in a feedlot (open symbols). Time of management, ultrasound scan (Scan), and tissue (Biopsy) and blood (Bleed) sampling events and are also shown.

1456 Cafe et al.

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 7: Seminário temp bov

< 0.001), with no pattern for the SD except that it was greater at FS1. In the feedlot (where FS was measured as the cattle exited the crush in a straight line, unlike backgrounding, where cattle had to turn right at 90° before measurement), the FS was slightly faster, but no consistent change over time was observed in means or SD. The first feedlot (FS 6) and FS 13 measure-ments (measured after the animals had been scanned and biopsied the previous week) were the greatest (P < 0.001). Angus cattle had slower FS than Brahmans. Flight speed in the Angus decreased during back-grounding (FS 1 = 1.3 to FS 5 = 1.1 m/s, SED = 0.07, P < 0.001), with no change in the SD over time. Flight speed of the Angus cattle was also faster in the feedlot, again with no pattern of change over time in means or SD. The fastest FS was FS 13 (P < 0.001), measured after the animals had been scanned and biopsied the previous week.

Correlations for 8 of 14 FS measurements (selected to provide an even spread over time) are presented in Table 2. The moderate to high correlations were all significant (all P < 0.001) for the Brahman cattle, and were greatest between measurements from the same lo-cation (i.e., backgrounding or finishing). For the Angus cattle, correlations were not as strong; 90% of the cor-relations of FS at the same location were significant (P < 0.001 to P < 0.21), but only approximately 30% at different locations were significant (P < 0.001 to P < 0.95). For both breeds, correlations decreased with increasing time between measures.

FS in WA. Flight speed was measured 4 times in the WA herd: 86, 155, 210, and 282 d after the begin-ning of backgrounding. Means and SD were 1.7 ± 0.45, 1.5 ± 0.52, 1.5 ± 0.52, and 1.5 ± 0.52 m/s for Brah-mans and were 1.7 ± 0.30, 1.4 ± 0.38, 1.5 ± 0.41, and 1.4 ± 0.45 m/s for Angus on these respective days. The effect of day of measurement was significant, with the second measurement being slowest in both Brahman (P

< 0.001) and Angus (P = 0.008) cattle. Little change was observed in SD over time in either breed. In Brah-man cattle, correlations ranged from 0.41 to 0.52 (all P < 0.001). In the Angus cattle, correlations ranged from −0.02 to 0.55 (P < 0.001 to P < 0.9), with the weakest being those involving the first measurement.

CS in NSW. Means for all CS assessments for the NSW cattle are shown graphically in Figure 1b, with means and SD for a representative 8 of 17 CS assess-ments (selected to provide an even spread over time) shown in Table 3.

A significant effect of day of assessment was ob-served. In the Brahman cattle, CS decreased during backgrounding (CS 1 = 2.5 to CS 6 = 1.7, SED = 0.07, P < 0.001), but the SD were variable. The second as-sessment (CS 2), when the animals were ultrasound scanned for the first time, was the greatest (P < 0.001). At the feedlot, CS were less in the second half of the feeding period, except for CS 15, which was assessed during ultrasound scanning (P < 0.001). In Angus cat-tle, CS decreased slightly during backgrounding (CS 1 = 1.6 to CS 5 = 1.4, SED = 0.10, P < 0.001), except for CS 2, taken during ultrasound scanning, which was the greatest. The SD was greater for CS 1 and CS 2. Crush scores were less at the feedlot, and again, they were slightly less in the second half of the feeding pe-riod. No pattern of change in SD in the feedlot was observed for either breed.

Correlations for 8 of the 17 CS assessments are pre-sented in Table 3. Overall, correlations for CS were less than for FS, and were less in the Angus cattle. In the Brahmans, correlations were generally greater between CS assessments at the same location, but most correla-tions between backgrounding and feedlot assessments of CS were also significant. Correlations were not as strong for the Angus cattle, with 30% of those for the same location being significant and 25% for different locations being significant.

Table 2. Mean (±SD) and timing of a spread of individual flight speed1 (FS, m/s) measurements taken during backgrounding or at the feedlot, and correlations between FS measurements for the Brahman (below diagonal, n = 164) and Angus (above diagonal, n = 49) cattle in the New South Wales experiment

Item FS 1 FS 2 FS 3 FS 4 FS 6 FS 9 FS 12 FS 14

Brahman 2.1 ± 0.99 2.0 ± 0.74 1.8 ± 0.75 1.5 ± 0.74 2.4 ± 0.78 2.1 ± 0.92 1.9 ± 0.92 2.1 ± 0.77Angus 1.3 ± 0.44 1.3 ± 0.53 1.5 ± 0.53 1.0 ± 0.42 2.0 ± 0.51 1.9 ± 0.62 2.0 ± 0.62 2.0 ± 0.49Day2 0 30 91 126 182 203 231 252FS 1   0.26† 0.43* 0.50* 0.33* 0.28† 0.26† 0.29†FS 2 0.65*   0.35* 0.54* 0.33* 0.15 −0.06 0.15FS 3 0.55* 0.67*   0.70* 0.19 0.27† 0.12 0.25FS 4 0.62* 0.66* 0.70*   0.40* 0.29† 0.14 0.25FS 6 0.48* 0.52* 0.54* 0.53*   0.51* 0.41* 0.30†FS 9 0.44* 0.45* 0.51* 0.52* 0.66*   0.50* 0.41*FS 12 0.37* 0.47* 0.54* 0.48* 0.61* 0.75*   0.49*FS 14 0.36* 0.45* 0.48* 0.50* 0.57* 0.63* 0.71*  

1FS 1 to 5 conducted during backgrounding, and FS 6 to 14 conducted at the feedlot. A subset of 8 FS were chosen to illustrate the range of correlations in the entire set of 14 measurements.

2Days from first FS measurement.†P < 0.10; *P < 0.05.

1457Cattle temperament and productivity

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 8: Seminário temp bov

Average FS and CS

Averages of FS and CS assessed during background-ing and at the feedlot in NSW, and average FS assessed during backgrounding and at the feedlot in WA are presented in Table 4. In line with the changes over time described above, average feedlot FS was faster than av-erage backgrounding FS in NSW (P < 0.001), whereas CS was less at the feedlot (P < 0.001). In WA Brah-mans, FS was slightly slower at the feedlot than during backgrounding (P = 0.024). In WA Angus cattle, no difference was observed in FS measured during back-grounding or at the feedlot (P = 0.35).

Correlations between backgrounding and feedlot aver-ages within assessment type (e.g., 0.69 for FS and 0.41 for CS in the NSW Brahmans; Table 5) were greater than for any individual pair of measurements from dif-ferent locations. This indicates that the averages gave a more accurate assessment of both FS and CS than did any single measure. Similarly, in WA the correlation between the average backgrounding and feedlot FS was 0.66 (P < 0.001) for Brahmans and 0.51 (P = 0.02) for Angus.

In the NSW Brahmans, correlations between average FS and CS, ranging from 0.41 to 0.49 (all P < 0.001), provide an indication that the 2 different temperament measurements ranked the cattle similarly. The relation-

ships were weaker for Angus cattle; only the correlation between average FS and CS during backgrounding was significant (r = 0.39, P = 0.006).

FS and Productivity Traits

The relationships between average FS and produc-tion, efficiency, carcass traits, and objective meat qual-ity traits are indicated by estimates of the linear covari-ates in Tables 6, 7, and 8; quadratic relationships are reported only when significant.

Growth, Intake, and Efficiency. Effects of av-erage backgrounding and average feedlot FS on produc-tion and efficiency traits are presented in Table 6. For NSW Brahmans, cattle with faster backgrounding FS had reduced BW at all times during the experiment (all P ≤ 0.046) and reduced ADG during backgrounding (P = 0.043) and finishing (P = 0.001). The quadratic relationship between background FS and background ADG was stronger (P = 0.009) than the linear relation-ship, with most of the decline in ADG occurring for background FS of >2.5 m/s. Increasing background FS was also related to reduced DMI (P = 0.012) and less time spent eating (P = 0.046). Increasing feedlot FS was related to reduced BW at the midpoint (P = 0.040) and at the end of the feedlot period (P = 0.030) and to decreased feedlot ADG (P = 0.007). Increasing feedlot

Table 3. Mean (±SD) and timing of a spread of individual crush score1 (CS, score 1 to 5) assessments taken dur-ing backgrounding or at the feedlot, and correlations between CS assessments for the Brahman (below diagonal, n = 164) and Angus (above diagonal, n = 49) cattle in the New South Wales experiment

Item CS 1 CS 2 CS 3 CS 4 CS 7 CS 10 CS 14 CS 17

Brahman 2.5 ± 0.93 2.8 ± 0.73 1.5 ± 0.63 2.0 ± 0.84 1.5 ± 0.59 1.6 ± 0.59 1.4 ± 0.57 1.4 ± 0.60Angus 1.6 ± 0.64 2.1 ± 0.69 1.3 ± 0.48 1.3 ± 0.48 1.2 ± 0.37 1.2 ± 0.43 1.1 ± 0.33 1.1 ± 0.35Day2 30 71 91 126 182 203 231 252CS 1   0.24† 0.31* 0.24† 0.27† −0.10 0.13 0.25†CS 2 0.20*   0.44* 0.25† 0.04 0.09 0.06 0.22CS 3 0.40* 0.19*   0.28† 0.37* 0.08 0.25† 0.32*CS 4 0.37* 0.21* 0.37*   0.03 0.08 0.25† 0.19CS 7 0.20* 0.08 0.26* 0.15   0.26† 0.17 0.29*CS 10 0.32* 0.23* 0.36* 0.27* 0.35*   0.37* 0.04CS 14 0.26* 0.18* 0.33* 0.22* 0.35* 0.47*   0.20CS 17 0.19* 0.21* 0.24* 0.32* 0.34* 0.48* 0.38*  

1CS 1 to 6 conducted during backgrounding, and CS 7 to 17 conducted at the feedlot. A subset of 8 CS were chosen to illustrate the range of correlations in the entire set of 17 measurements.

2Days from first flight speed measurement.†P < 0.10; *P < 0.05.

Table 4. Average flight speed (m/s) and crush score (1 to 5) during backgrounding and feedlot finishing in Brah-man and Angus cattle in the New South Wales (NSW) and Western Australia (WA) experiments

Item

Flight speed Crush score

NSW Brahman NSW Angus WA Brahman WA Angus NSW Brahman NSW Angus

Location               Background 1.84 1.24 1.61 1.49   2.12 1.56 Feedlot 2.09 1.97 1.54 1.42   1.58 1.21SED 0.042 0.063 0.028 0.080   0.034 0.04P-value <0.001 <0.001 0.024 0.35   <0.001 <0.001

Cafe et al.1458

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 9: Seminário temp bov

FS was also associated with reduced time spent eating (P = 0.040) and tended to reduce DMI (P = 0.07). In the WA Brahman cattle, increasing background FS was associated with reduced BW at the beginning and end of the feedlot period (both P = 0.008), reduced back-ground ADG (P = 0.025), and a tendency for reduced feedlot ADG (P = 0.07). In addition, increasing feedlot FS was associated with reduced BW at the beginning (P = 0.023) and end (P = 0.015) of the feedlot period and with a tendency toward reduced feedlot ADG (P = 0.07).

In the NSW Angus cattle, increasing background FS was related to lighter BW at the beginning of back-grounding (P = 0.045), and increasing feedlot FS was related to reduced time spent eating (P = 0.038) and a tendency toward a decreased feed conversion ratio (FCR; P = 0.07). In WA Angus cattle, no significant relationships were observed between background FS and any growth trait, but increased feedlot FS was as-sociated with reduced background ADG (P = 0.003), reduced BW at feedlot entry (P = 0.018), and a ten-

dency toward reduced BW at the end of the feedlot period (P = 0.06).

Carcass Characteristics. Effects of average backgrounding and feedlot FS on carcass traits are pre-sented in Table 7. In the NSW Brahman cattle, in-creasing background FS was associated with significant reductions in carcass weight (P = 0.001), rib fat (P = 0.016), and ultimate pH (P = 0.014), and an increase in the temperature at which the carcass reached pH 6 (P < 0.001). Carcass weight also tended to be reduced with increasing feedlot FS (P = 0.09). In the WA Brah-man cattle, increasing background FS was associated with reduced carcass weight (P = 0.013), reduced LLM area (P = 0.035), and darker meat color (P = 0.028). It also tended to influence carcass pH decline, with the carcasses tending to take longer to reach pH 6 (P = 0.07) and at reduced carcass temperatures (P = 0.09). Increased feedlot FS also tended to be associated with reduced carcass weight (P = 0.09). For the WA Brah-man cattle, the quadratic relationship between FS and carcass weight was slightly stronger than the linear re-

Table 5. Correlations between average flight speed (FS, m/s) and crush score (CS, score 1 to 5) determined during backgrounding and feedlot finishing in Brahman (be-low diagonal, n = 164) and Angus (above diagonal, n = 49) cattle in the New South Wales experiment

Item Background FS Feedlot FS Background CS Feedlot CS

Background FS   0.42* 0.39* 0.23Feedlot FS 0.69*   0.08 0.24Background CS 0.49* 0.41*   0.62*Feedlot CS 0.42* 0.41* 0.58*  

*P < 0.05.

Table 6. Significant effects of average flight speed (FS, m/s) determined during either backgrounding or feedlot finishing on live traits in Brahman and Angus cattle in the New South Wales (NSW) and Western Australia (WA) experiments

Breed Site Trait

Background FS Feedlot FS

Slope SE P-value Slope SE P-value

Brahman NSW (n = 164) Beginning background BW, kg −6.9 3.45 0.046          Beginning feedlot BW, kg −11.5 3.61 0.002            Mid feedlot BW, kg −19.1 4.48 <0.001   −9.3 4.50 0.040    End feedlot BW, kg −21.0 5.047 <0.001   −11.1 5.07 0.030    Background ADG, kg −0.02 0.011 0.043            Feedlot ADG, kg −0.08 0.025 0.001   −0.07 0.025 0.007    Feedlot DMI, kg of DM/d −0.37 0.146 0.012   −0.26 0.143 0.07    Feeding time, min/d −4.7 2.33 0.046   −4.68 2.259 0.040  WA (n = 173) Beginning feedlot BW, kg −14.3 5.26 0.008   −11.6 5.06 0.023  End feedlot BW, kg −20.9 7.78 0.008   −18.3 7.45 0.015    Background ADG, kg −0.05 0.020 0.025            Feedlot ADG, kg −0.10 0.056 0.07   −0.08 0.046 0.07Angus NSW (n = 49) Beginning background BW, kg −23.0 11.08 0.045          Feedlot FCR,1 kg of DM/kg of gain         −1.5 0.81 0.07    Feeding time, min/d         −17.6 8.19 0.038  WA (n = 20) Beginning feedlot BW, kg         −34.8 12.70 0.018  End feedlot BW, kg         −27.6 13.13 0.06    Background ADG, kg         −0.16 0.043 0.003

1FCR = feed conversion ratio.

Cattle temperament and productivity 1459

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 10: Seminário temp bov

lationship, with most of the decline in carcass weight occurring for cattle with a background or feedlot FS of >2 m/s (P = 0.012 and 0.06, respectively).

In the NSW Angus cattle, increasing background FS tended to reduce carcass weight (P = 0.07) and rump fat (P = 0.06) and significantly increased the time to reach pH 6 (P < 0.001), but no significant effects of feedlot FS on carcass traits were observed. In the WA Angus cattle, increasing background FS was associated with reduced marbling score (P = 0.020), and increas-ing feedlot FS was associated with reduced LLM area (P = 0.005).

Objective Meat Quality. Effects of average back-grounding and feedlot FS on objective meat quality traits are presented in Table 8. In the NSW Brahmans, increasing background FS was related to increased cook-ing loss in TS 1-d aged LLM (P = 0.036), and increas-ing feedlot FS tended to be related to increased shear force (SF) of AT 1-d aged LLM (P = 0.05). In the WA Brahmans, increasing background FS was related to in-creased SF of TS 7-d aged LLM (P = 0.044), increased compression in AT 1-d aged STN (P = 0.043), and in-creased cooking loss (P = 0.023) and darker meat color (P = 0.006) of AT 7-d aged LLM. Increasing feedlot

Table 8. Significant effects of average flight speed (FS, m/s) determined during either backgrounding or feedlot finishing on objective meat quality traits in Brahman and Angus cattle in the New South Wales (NSW) and West-ern Australia (WA) experiments

Breed Site Trait1

Background FS Feedlot FS

Slope SE P-value Slope SE P-value

Brahman NSW (n = 161) AT 1-d aged LLM SF, N         4.2 2.10 0.050  TS 1-d aged LLM cook, % 0.84 0.396 0.036          WA (n = 137) TS 7-d aged LLM SF, N 4.3 2.11 0.044   4.2 1.86 0.027  AT 1-d aged STN comp, N 1.4 0.70 0.043            AT 1-d aged LLM cook, %         0.65 0.292 0.028    AT 7-d aged LLM cook, % 0.65 0.284 0.023            TS 7-d aged LLM cook, %         0.70 0.286 0.016    AT 1-d aged LLM color L*         −0.84 0.461 0.07    AT 7-d aged LLM color L* −1.6 0.56 0.006   −1.1 0.53 0.034Angus NSW (n = 48) AT 1-d aged LLM SF, N 11.2 4.32 0.013          AT 7-d aged LLM SF, N 7.3 4.04 0.08          WA (n = 16) AT 7-d aged LLM SF, N 15.8 7.70 0.07          TS 1-d aged LLM SF, N 13.1 5.89 0.050   6.6 2.65 0.032    AT 7-d aged LLM comp, N 4.6 2.18 0.07            AT 1-d aged STN comp, N 7.1 3.06 0.045            TS 1-d aged LLM cook, %         2.4 0.92 0.027    AT 7-d aged LLM pH 0.10 0.042 0.049        

1AT = Achilles-suspended side; TS = tenderstretched side; LLM = musculus longissimus lumborum; STN = musculus semitendinosus; SF = shear force; comp = compression; cook = cooking loss; color L* = CIELAB color scale, where 0 = dark and 100 = light.

Table 7. Significant effects of average flight speed (FS, m/s) determined during either backgrounding or feedlot finishing on carcass traits in Brahman and Angus cattle in the New South Wales (NSW) and Western Australia (WA) experiments

Breed Site Trait

Background FS Feedlot FS

Slope SE P-value Slope SE P-value

Brahman NSW (n = 163) Carcass wt, kg −9.9 2.92 0.001   −5.0 2.93 0.09  Rib fat,1 mm −5.7 0.23 0.016            Ultimate pH1 −0.01 0.006 0.014            Temperature at pH 6, °C 0.81 0.068 <0.001          WA (n = 143) Carcass wt, kg −9.7 3.85 0.013   −6.4 3.73 0.09  LLM area,1 cm2 −2.67 1.20 0.035            Meat color score1 0.48 0.213 0.028            Temperature at pH 6, °C −1.1 0.64 0.09            Time to pH 6, h 0.22 0.120 0.07        Angus NSW (n = 49) Carcass wt, kg −27.0 14.47 0.07          Rump fat, mm −4.0 2.03 0.06            Time to pH 6, h 0.56 0.093 <0.001          WA (n = 17) LLM area,1 cm2         −5.9 1.61 0.005  Marble score1 −79.4 28.27 0.020        

1Meat Standards Australia (2009) grading data, where marble score is from 100 to 1,100 in increments of 10; meat color score is from 1 (lightest) to 9 (darkest), and ultimate pH is the pH at grading. LLM = musculus longissimus lumborum.

Cafe et al.1460

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 11: Seminário temp bov

FS was also related to increasing SF of TS 7-d aged LLM (P = 0.027) and to increased cooking loss in AT 1-d aged LLM (P = 0.028) and TS 7-d aged LLM (P = 0.016). Increasing feedlot FS was also related to darker meat color in AT 7-d aged LLM (P = 0.034) and tend-ed to be related to darker meat color in AT 1-d aged LLM (P = 0.07).

In the NSW Angus cattle, increasing background FS was associated with increased SF in AT 1-d aged LLM (P = 0.013) and a tendency toward increased SF in AT 7-d aged LLM (P = 0.08). No significant relationships were observed between feedlot FS and objective meat quality traits. In WA Angus cattle, increasing FS tend-ed to be associated with increased SF in AT 7-d aged LLM (P = 0.07), increased SF in TS 1-d aged LLM (P = 0.05), and increased compression in AT 7-d aged LLM (P = 0.07). It was also related to increased com-pression in AT 1-d aged STN (P = 0.045) and increased pH in the laboratory sample of the AT 7-d aged LLM (P = 0.049). Increasing feedlot FS was also associated with increased SF (P = 0.032) and cooking loss (P = 0.027) in TS 1-d aged LLM.

CS and Productivity Traits

Growth, Intake, and Efficiency. Effects of average backgrounding and feedlot CS on production and feed efficiency traits in the NSW experiment are presented in Table 9. In Brahman cattle, increasing background CS was associated with reduced mid feed-lot BW (P = 0.027), reduced background ADG (P = 0.016), and a tendency toward reduced BW (P = 0.09) at the end of the feedlot period. The relationship be-tween background CS and feedlot ADG and DMI was quadratic, with most of the decline in carcass ADG and intake occurring in cattle with a background CS >3 (P = 0.006 and 0.031, respectively). Increased feed-lot CS was related to reduced BW at the beginning of backgrounding (P = 0.034), mid feedlot (P < 0.001), and at the end of the feedlot period (P < 0.001), and to reduced feedlot ADG (P = 0.003) and DMI (P = 0.001).

In Angus cattle, increasing CS was related only to feeding behavior, with the number of daily feeding ses-

sions increasing with both increasing background (P = 0.016) and feedlot (P = 0.043) CS. Intake per ses-sion decreased (P = 0.020) with increasing background CS and tended to decrease (P = 0.05) with increasing feedlot CS.

Carcass Characteristics. Effects of average backgrounding and feedlot CS on carcass traits in the NSW experiment are presented in Table 10. In Brah-man cattle, increasing feedlot CS was associated with a reduction in carcass weight (P < 0.001), and increasing background and feedlot CS were associated with a re-duction in rib fat (P = 0.012 and 0.017). No significant effects of CS were observed on carcass traits in the Angus cattle.

Objective Meat Quality. Effects of average back-grounding and feedlot CS on objective meat quality traits in the NSW experiment are presented in Table 10. In Brahman cattle, as background CS increased, so did SF in TS 7-d aged LLM (P = 0.048) and com-pression in AT 1-d aged LLM (P = 0.019). As feedlot CS increased, SF in AT 1-d aged LLM increased (P = 0.024) and tendencies were observed for increased SF with increasing feedlot CS in AT and TS 7-d aged LLM (P = 0.08 and 0.09). Cooking loss in AT 1-d aged LLM also increased (P = 0.001) with increasing feedlot CS in the Brahman cattle.

In Angus cattle, greater background CS was associ-ated with increased compression in AT 7-d aged LLM (P = 0.04) and with a tendency toward increased SF and compression in AT 1-d aged LLM (P = 0.05 and P = 0.06). Increasing feedlot CS led to increased SF (P = 0.047) and compression (P = 0.045) in AT 1-d aged LLM and to a tendency toward increased SF in AT 7-d aged LLM (P = 0.09).

DISCUSSION

This study shows that the temperament of cattle, as assessed by FS and CS, was persistent over time, and that cattle with faster FS or greater CS (flightier tem-peraments) had inferior performance across a compre-hensive range of beef production traits. Flight speed (or its inverse, flight time) and CS are simple to conduct on farm, and their use is encouraged by various Aus-

Table 9. Significant effects of average crush score (CS, score 1 to 5) determined during either backgrounding or feedlot finishing on live traits in Brahman and Angus cattle in the New South Wales experiment

Breed Trait

Background CS Feedlot CS

Slope SE P-value Slope SE P-value

Brahman (n = 164) Beginning background BW, kg         −11.2 5.25 0.034Mid feedlot BW, kg −13.8 6.17 0.027   −30.5 6.80 <0.001

  End feedlot BW, kg −11.9 6.94 0.09   −30.0 7.83 <0.001  Background ADG, kg −0.04 0.015 0.016          Feedlot ADG, kg         −0.12 0.040 0.003  Feedlot DMI, kg of DM/d         −0.75 0.226 0.001Angus (n = 49) Feed intake per session, kg of DM −0.74 0.302 0.020   −0.98 0.484 0.05

Feeding sessions, No./d 3.7 1.46 0.016   4.9 2.32 0.043

Cattle temperament and productivity 1461

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 12: Seminário temp bov

tralian beef cattle breed societies [for example, Limou-sin (CS and pen score), Angus (CS and flight time), and Brahman (flight time); http://breedplan.une.edu.au] to allow selection for a calmer temperament or do-cility. Despite this, still relatively few published stud-ies have described relationships between temperament and commercially important traits, and the biological mechanisms that underpin these associations are not well understood (Ferguson et al., 2006).

Regular FS and CS measurements were taken throughout the NSW experiment to study the consis-tency of the measures over time with changes in loca-tion and during various husbandry and sample collec-tion procedures. Management was more intensive than for commercial herds because of the data and sample collection required for other aspects of the experiment, but all handling of the animals was conducted as calm-ly as possible.

Temperament over Time

The decreased average feedlot vs. backgrounding CS in NSW, and slower feedlot vs. backgrounding FS in WA indicate that the general response of the cattle to handling declined over the duration of the experiment. In contrast, FS in NSW was faster at the feedlot than during backgrounding, where FS was measured after the cattle had turned 90° after exiting the chute. The differences in the way in which backgrounding FS was measured resulted in slower speeds, but nonetheless provided a meaningful measure of FS.

Much of the decline in the mean and variation within FS was seen after the first 3 measurements, suggest-ing that the variation between animals stabilized after some initial familiarization with handling and the fa-cilities. The mean and variation for CS were also great-er during initial measurements, after which both were again very consistent for both breeds. Hence, the pat-tern of change in mean and variation for both FS and CS indicated that after a small number of consistent

handling events, the cattle showed calmer behavioral responses, presumably as they habituated to handling. This is consistent with results for repeated tempera-ment assessments reported by other authors (Burrow and Dillon, 1997; Curley et al., 2006; Kilgour et al., 2006; Petherick et al., 2009a). The day of measurement effect, significant for both FS and CS, can be attrib-uted, at least in part, to the fact that on some days, the data collection procedures involved closer and more prolonged handling.

Correlations between repeated measures for FS and CS in Brahman cattle were usually significant and were greater than for Angus cattle. The variation was also consistently greater for both FS and CS at each as-sessment in the Brahman than in the Angus cattle, indicating that the Brahman cattle had greater indi-vidual variation in temperament than the Angus cattle in the present study. This finding would account for the poorer correlations in Angus cattle among individ-ual measures of FS and CS, and between averages for FS and CS. For both breeds, the strength of correla-tions declined over time, indicating small, consistent changes over time. Because the behavioral response is a combination of genetic and environmental influences (Boissy et al., 2005), small changes over time would be expected. In this regard, it was also notable that the largest decline in the strength of correlations occurred with the change in location between backgrounding and feedlot finishing. It is also important to note that the use of average values for FS and CS resulted in greater correlations, indicating that average measures provided a more reliable assessment of cattle temperament than did any single measure, as suggested by Grandin (1993) and Burrow and Dillon (1997).

Relationships Between Temperament and Other Traits

Relationships between temperament and other pro-duction traits were assessed using average FS and CS

Table 10. Significant effects of average crush score (CS, score 1 to 5) during either backgrounding or feedlot finish-ing on carcass and meat quality traits in Brahman and Angus cattle in the New South Wales experiment

Breed Trait1

Background CS Feedlot CS

Slope SE P-value Slope SE P-value

Brahman (n = 161) Carcass wt, kg         −16.6 4.50 <0.001Rib fat, mm −0.77 0.302 0.012   −0.90 0.370 0.017

  AT 1-d aged LLM SF, N         7.6 3.34 0.024  AT 7-d aged LLM SF, N         5.5 3.10 0.08  TS 7-d aged LLM SF, N 1.5 0.74 0.048   1.6 0.97 0.09  AT 1-d aged LLM comp, N 0.8 0.35 0.019          AT 1-d aged LLM cook, %         0.98 0.293 0.001Angus (n = 48) AT 1-d aged LLM SF, N 9.4 4.64 0.05   16.0 7.78 0.047

AT 7-d aged LLM SF, N         12.1 6.98 0.09  AT 1-d aged LLM comp, N 1.4 0.72 0.06   2.4 1.16 0.045  AT 7-d aged LLM comp, N 1.1 0.53 0.04        

1AT = Achilles-suspended side; TS = tenderstretched side; LLM = musculus longissimus lumborum; SF = shear force; comp = compression, cook = cooking loss.

Cafe et al.1462

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 13: Seminário temp bov

during backgrounding or finishing. Where effects of temperament were significant, cattle with greater FS or CS grew more slowly, produced smaller carcasses with less fat cover, and had darker meat that was greater in SF and compression. It is important to note that all significant effects of a more reactive temperament (increasing FS or CS) on economically significant traits were detrimental.

Faster FS was associated with reduced BW and growth rates throughout the experiment in the Brah-man cattle in NSW and WA. Estimates of the reduc-tion in feedlot exit BW were 20.0 and 20.9 kg with each 1 m/s increase in background FS in the NSW and WA Brahman cattle, respectively. Increasing CS had similar effects in the NSW Brahmans, with a 1-score increase in background CS leading to an 11.9-kg decrease in feedlot exit BW. Carcass weights were reduced with increasing FS in the Brahmans, by 9.9 kg in NSW and 9.7 kg in WA for each 1 m/s increase in FS. The NSW Brahmans also had a 16.6-kg reduction in carcass weight per unit increase in feedlot CS. There was also some indication of reduced carcass fatness and LLM area with increas-ing FS, and of reduced carcass fatness with increasing CS, in the NSW Brahmans. The inclusion of carcass weight as a covariate in the statistical models showed that these differences in composition were mostly ex-plained by the differences in carcass weight. In the An-gus cattle, increasing FS also tended to reduce BW and growth rate, but the relationships were much weaker than in Brahmans. In line with the weak trends toward lighter BW with increasing FS in the Angus cattle, ten-dencies for reductions in BW-related carcass traits were observed. However, no significant relationships were ob-served between CS and BW-related traits in the Angus cattle in the NSW herd.

Previously reported results for relationships between cattle temperament and growth have been variable. Slower growth rates have been reported in cattle with faster FS, greater CS, or both in studies conducted un-der more intensive management systems (Burrow and Dillon, 1997; Voisinet et al., 1997b; Müller and von Keyserlingk, 2006; Behrends et al., 2009). Others have found little relationship between temperament and growth rates in herds in which the ranges in FS and CS were small and cattle were generally docile (Graham et al., 2001; Elzo et al., 2009). Our results are consistent with the above studies in that the Angus cattle were generally more docile than the Brahmans, and greater effects of temperament on growth were observed in the Brahmans.

It has previously been postulated that cattle with a more reactive temperament may grow more slowly because of the greater energy expenditure associated with, for example, more vigilant behavior, resulting in poorer FCR or net feed intake (NFI; Burrow and Dil-lon, 1997; Petherick et al., 2002). In the NSW Brah-mans, each meter per second increase in background FS was associated with a reduction in feed intake of 370 g of DM/d, and a reduction of 4.7 min/d in the amount

of time spent eating. A similar but slightly smaller ef-fect was obtained using feedlot FS as the measure of temperament. Increasing CS was also associated with a reduced feed intake (of 750 g of DM/d with each unit increase in feedlot CS) in the NSW Brahman cattle. These effects were not accounted for entirely by the BW differences at the beginning of the feedlot period, and they remained or tended to remain evident when feedlot entry BW was fitted as a covariate. However, there was no evidence of FS or CS being related to dif-ferences in FCR or NFI in the Brahman cattle. Togeth-er, this suggests that temperament plays a significant role in controlling feed intake and time spent eating, but that it has lesser effects on efficiency of utilization of feed; hence, poor temperament reduces DMI and ADG through behavioral rather than metabolic mecha-nisms. This conclusion is in agreement with recent work by Nkrumah et al. (2007) and Elzo et al. (2009), who found that young cattle of mixed breeds with faster feedlot FS had less feedlot DMI but showed no differ-ence in FCR or NFI.

In the NSW herd, temperament had less effect on feed intake in Angus than in Brahman cattle. However, in the Angus cattle, each meter per second increase in feedlot FS was associated with a 17.6 min/d reduction in feeding time and a tendency toward reduced FCR by 1.5 kg of DM/kg of BW gain. This reduction in FCR is the only result relating to poorer temperament within the present study that might be considered beneficial. Although caution is required, because of the P-value of 0.07, it is possible that among the Angus cattle, a slightly decreased intake with increasing FS allowed for more efficient digestion (Herd et al., 2004).

The determinants of eating quality of the meat are complex and multifactorial, and pre- and postmortem events can have major effects on beef tenderness (Mal-tin et al., 2003; Ferguson and Warner, 2008). Cattle temperament is assumed to be related to stress respon-siveness, and it is likely that the stress response to han-dling and transport is greater in temperamental cattle, resulting in depletion of muscle glycogen before slaugh-ter and hence reduced meat quality because of greater carcass pH and the associated darker color (Ferguson et al., 2006). In the present study, increasing FS or CS was related to darker LLM meat color and increased muscle pH, shear force, compression, and cooking loss, effects all considered detrimental to the eating quality of beef (Perry et al., 2001b).

The relationships between temperament and meat quality were strongest for the WA cattle. Differences between experimental sites could, at least in part, be due to processing differences resulting in differences in postmortem pH or temperature declines, as discussed by Cafe et al. (2010b). Greater effects of temperament on meat quality traits were evident in both breeds in WA, where the carcasses had much faster pH declines and slower cooling rates than in NSW. Relationships between temperament and shear force tended to be less significant with 7-d aging. This may indicate that

Cattle temperament and productivity 1463

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 14: Seminário temp bov

temperament was related to variation in tenderness be-cause of factors other than proteolysis. An unexpected result was that the Angus cattle, with smaller numbers of animals and less variation in temperament, showed stronger effects of temperament on SF than did the Brahman cattle in both herds.

Overall, most relationships with temperament were linear, indicating that selection of cattle with calmer temperaments, and not only culling of cattle with the most reactive temperaments, can improve productivity and the safety and welfare of cattle and can improve safety for human handlers. Other studies have shown both linear (Burrow and Dillon, 1997) and quadratic (Müller and von Keyserlingk, 2006) relationships be-tween temperament and productivity traits. However, differences in results between studies may simply de-pend on the extent to which extremes of temperament exist within the herd, with the presence of extremely reactive animals likely to result in nonlinear relation-ships.

Flight speed and CS were consistent measures of temperament during the experiment, even with varying management at the time of measurement. The earlier (backgrounding) assessments had stronger and more frequent relationships with productivity and carcass traits, but both the backgrounding and the finishing assessments showed significant relationships with meat quality. Other authors have reported weak to no re-lationships between temperament measurements taken later in life and growth and meat quality (Behrends et al., 2009). In the present study, the uniform measure-ment procedures and frequent handling of the cattle throughout the experiment for intensive data and sam-ple collection, and the young age of the cattle, may have allowed variation in temperament to continue to be detected at a later stage of production.

Conclusions

In this study, cattle with a calmer temperament, as measured by reduced FS and CS, had superior perfor-mance across a comprehensive range of beef produc-tion traits. Repeated assessments of temperament using FS and CS were correlated, with the strength of the correlations declining over time. Correlations between repeated measurements of FS were greater than be-tween repeated assessments of CS. A stronger relation-ship was observed between temperament and growth and carcass traits in Brahman than in Angus cattle, which, in our experiment, were more docile than the Brahmans. Similar strengths of relationships with meat quality were evident in both breeds. In general, cattle with poorer temperaments, as assessed by FS or CS, had consistently less feed intake and slower growth rates, which resulted in smaller carcasses with less fat cover and poorer objective meat quality characteristics. No relationship was observed between temperament and tenderness gene marker status, and temperament was not modified by HGP implants.

LITERATURE CITED

AUS-MEAT. 2007. AUS-MEAT National Accreditation Standards. 2007 Edition. AUS-MEAT Ltd., Brisbane, Queensland, Aus-tralia.

Barendse, W. J. 2002. DNA markers for meat tenderness. The Com-monwealth Scientific and Industrial Research Organization, The State of Queensland through its Department of Primary Industries, The University of New England, The State of New South Wales through its Department of Agriculture, and Meat and Livestock Australia Limited, assignees. Int. Patent No. W0 02/064820.

Barendse, W., B. E. Harrison, R. J. Bunch, and M. B. Thomas. 2008. Variation at the calpain 3 gene is associated with meat tenderness in zebu and composite breeds of cattle. BMC Gen-et. 9:41.

Behrends, S. M., R. K. Miller, F. M. Rouquette Jr., R. D. Randel, B. G. Warrington, T. D. A. Forbes, T. H. Welsh, H. Lippke, J. M. Behrends, G. E. Carstens, and J. W. Holloway. 2009. Re-lationship of temperament, growth, carcass characteristics and tenderness in beef steers. Meat Sci. 81:433–438.

Bindon, B. M. 2001. Genesis of the Cooperative Research Centre for the Cattle and Beef Industry: Integration of resources for beef quality research (1998–2000). Aust. J. Exp. Agric. 41:843–853.

Boissy, A., A. D. Fisher, J. Bouix, G. N. Hinch, and P. Le Neindre. 2005. Genetics of fear in ruminant livestock. Livest. Prod. Sci. 93:23–32.

Burrow, H. M. 1997. Measurements of temperament and their rela-tionships with performance traits of beef cattle. Anim. Breed. Abstr. 65:477–495.

Burrow, H. M., and R. D. Dillon. 1997. Relationships between tem-perament and growth in a feedlot and commercial carcass traits of Bos indicus crossbreds. Aust. J. Exp. Agric. 37:407–411.

Burrow, H. M., G. W. Seifert, and N. J. Corbet. 1988. A new tech-nique for measuring temperament in cattle. Proc. Aust. Soc. Anim. Prod. 17:154–157.

Cafe, L. M., B. L. McIntyre, D. L. Robinson, G. H. Geesink, W. Barendse, and P. L. Greenwood. 2010a. Production and pro-cessing studies on calpain-system gene markers for tenderness in Brahman cattle: 1. Growth, efficiency, temperament and car-cass characteristics. J. Anim. Sci. 88:3047–3058.

Cafe, L. M., B. M. McIntyre, D. L. Robinson, G. H. Geesink, W. Barendse, D. W. Pethick, J. M. Thompson, and P. L. Green-wood. 2010b. Production and processing studies on calpain-sys-tem gene markers for tenderness in Brahman cattle 2. Objective meat quality. J. Anim. Sci. 88:3059–3069.

Curley, K. O., Jr., J. C. Paschal, T. H. Welsh Jr., and R. D. Randel. 2006. Technical note: Exit velocity as a measure of cattle tem-perament is repeatable and associated with serum concentra-tion of cortisol in Brahman bulls. J. Anim. Sci. 84:3100–3103.

Elzo, M. A., D. G. Riley, G. R. Hansen, D. D. Johnson, R. O. Myer, S. W. Coleman, C. C. Chase, J. G. Wasdin, and J. D. Driver. 2009. Effect of breed composition on phenotypic residual feed intake and growth in Angus, Brahman, and Angus × Brahman crossbred cattle. J. Anim. Sci. 87:3877–3886.

Ferguson, D. M., D. Johnston, H. M. Burrow, and A. Reverter. 2006. Relationships between temperament, feedlot performance and beef quality. Australian Beef—The Leader! The impact of sci-ence on the beef industry. Cooperative Research Centre for Beef Genetic Technologies, Armidale, New South Wales, Australia.

Ferguson, D. M., and R. D. Warner. 2008. Have we underestimated the impact of pre-slaughter stress on meat quality in rumi-nants? Meat Sci. 80:12–19.

Gardner, G. E., B. L. McIntyre, G. D. Tudor, and D. W. Pethick. 2001. The impact of nutrition on bovine muscle glycogen me-tabolism following exercise. Aust. J. Agric. Res. 52:461–470.

Graham, J. F., A. J. Clark, K. Thomson, and G. Kearney. 2001. The relationship between temperament score and flight speed, and pre and post weaning growth of Angus, Hereford, Limousin and

Cafe et al.1464

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 15: Seminário temp bov

Simmental sired weaner cattle bred from Angus and Hereford dams. Proc. Assoc. Adv. Anim. Breed. Genet. 14:75–77.

Grandin, T. 1993. Behavioral agitation during handling of cattle is persistent over time. Appl. Anim. Behav. Sci. 36:1–9.

Herd, R. M., V. H. Oddy, and E. C. Richardson. 2004. Biological ba-sis for variation in residual feed intake in beef cattle. 1. Review of potential mechanisms. Aust. J. Exp. Agric. 44:423–430.

Kilgour, R. J., G. J. Melville, and P. L. Greenwood. 2006. Individual differences in the reaction of beef cattle to situations involving social isolation, close proximity of humans, restraint and nov-elty. Appl. Anim. Behav. Sci. 99:21–40.

King, D. A., C. E. S. Pfeiffer, R. D. Randel, T. H. Welsh Jr., R. A. Oliphint, B. E. Baird, K. O. Curley Jr., R. C. Vann, D. S. Hale, and J. W. Savell. 2006. Influence of animal temperament and stress responsiveness on the carcass quality and beef tenderness of feedlot cattle. Meat Sci. 74:546–556.

Maltin, C., D. Balcerzak, R. Tilley, and M. Delday. 2003. Determi-nants of meat quality. Tenderness. Proc. Nutr. Soc. 62:337–347.

Meat Standards Australia. 2009. MSA Standards Manual for Beef Grading. Meat Standards Australia, Fortitude Valley, Queen-sland, Australia.

Müller, R., and M. A. G. von Keyserlingk. 2006. Consistency of flight speed and its correlation to productivity and to personality in Bos taurus beef cattle. Appl. Anim. Behav. Sci. 99:193–204.

Nkrumah, J. D., D. H. Crews, J. A. Basarab, M. A. Price, E. K. Okine, Z. Wang, C. Li, and S. S. Moore. 2007. Genetic and phenotypic relationships of feeding behavior and temperament with performance, feed efficiency, ultrasound, and carcass merit of beef cattle. J. Anim. Sci. 85:2382–2390.

Perry, D., W. R. Shorthose, D. M. Ferguson, and J. M. Thompson. 2001a. Methods used in the CRC program for the determina-

tion of carcass yield and beef quality. Aust. J. Exp. Agric. 41:953–957.

Perry, D., J. M. Thompson, I. H. Hwang, A. Butchers, and A. F. Egan. 2001b. Relationship between objective measurements and taste panel assessment of beef quality. Aust. J. Exp. Agric. 41:981–989.

Petherick, J. C., V. J. Doogan, R. G. Holroyd, P. Olsson, and B. K. Venus. 2009a. Quality of handling and holding yard environ-ment, and beef cattle temperament: 1. Relationships with flight speed and fear of humans. Appl. Anim. Behav. Sci. 120:18–27.

Petherick, J. C., V. J. Doogan, B. K. Venus, R. G. Holroyd, and P. Olsson. 2009b. Quality of handling and holding yard environ-ment, and beef cattle temperament: 2. Consequences for stress and productivity. Appl. Anim. Behav. Sci. 120:28–38.

Petherick, J. C., R. G. Holroyd, V. J. Doogan, and B. K. Venus. 2002. Productivity, carcass and meat quality of lot-fed Bos in-dicus cross steers grouped according to temperament. Aust. J. Exp. Agric. 42:389–398.

Robinson, D. L. 1987. Estimation and use of variance components. Statistician 36:3–14.

Thompson, J. M. 2002. Managing meat tenderness. Meat Sci. 62:295–308.

Voisinet, B. D., T. Grandin, S. F. O’Connor, J. D. Tatum, and M. J. Deesing. 1997a. Bos indicus-cross feedlot cattle with excit-able temperaments have tougher meat and a higher incidence of borderline dark cutters. Meat Sci. 46:367–377.

Voisinet, B. D., T. Grandin, J. D. Tatum, S. F. O’Connor, and J. J. Struthers. 1997b. Feedlot cattle with calm temperaments have higher average daily gains than cattle with excitable tempera-ments. J. Anim. Sci. 75:892–896.

Cattle temperament and productivity 1465

at UNESP on July 20, 2011jas.fass.orgDownloaded from

Page 16: Seminário temp bov

Referenceshttp://jas.fass.org/content/89/5/1452#BIBLThis article cites 30 articles, 6 of which you can access for free at:

at UNESP on July 20, 2011jas.fass.orgDownloaded from