Analysis of milking characteristics in New Zealand dairy cows.

The objective of this study was to describe the variation in milking characteristics, and factors associated with these traits, in grazing dairy cows milked without premilking stimulation. Milk yield, duration, and average and maximum milk flow rate data were collected from 38 farms in New Zealand at 2 time points (spring and autumn) during the 2010 to 2011 season. Subsequently, a second data set, allowing the generation of daily milk flow profiles, was collected from 2 farms in the 2011 to 2012 season. Corresponding animal data, such as breed, date of birth, and ancestry information, were extracted from the New Zealand Dairy Industry Good Animal Database (New Zealand Animal Evaluation Ltd., Hamilton, New Zealand). Residual milking duration (deviation from the regression line of milk yield on milking duration) was calculated, allowing the identification of fast-milking cows independent of milk yield. Variance components for the milking characteristics traits were estimated using an animal linear mixed model. The average milk yield was 10 kg/milking and the average milking duration was 360 s. The average milk flow rate was 1.8 kg/min and maximum milk flow 3.3 kg/min, with 44% of milk flow curves being classified as bimodal. Primiparous animals exhibited different milk flow profiles, with a lower maximum flow, than multiparous animals, possibly due to differences in cisternal capacity. Residual milking duration was shortest (-10s) in mid-lactation (121-180 d) and was 13s longer for Jersey compared with Friesian cows; however, it was 19s shorter when adjusted for energy content. Residual milking duration had a negligible genetic correlation (-0.07) with milk yield, indicating that selection for cows with shorter residual milking duration should have a negligible effect on milk yield. A heritability of 0.27 indicated that residual milking duration could be valuable as part of a breeding program. Knowledge of the distribution of milking durations for a given milk yield allows farmers to choose appropriate cluster-on time when using a maximum milking time strategy to improve milking efficiency.

Using activity-based monitoring systems to detect dairy cows in oestrus: a field evaluation.

AIM: To assess the use and performance of activity-based oestrus detection systems (ODS) on two commercial dairy farms using a gold standard based on profiles of concentrations of progesterone in milk, artificial insemination (AI) records and pregnancy diagnosis results. METHODS: Two activity-based ODS were evaluated in mature cows on two large pasture-grazed dairy farms (>500 cows) over the first 3 weeks of AI. Farm 1 (n=286 cows) used a leg-mounted device and cows were drafted automatically based on activity alerts. Decisions regarding AI were then made based on tail-paint and cow history for these cows. Farm 2 (n=345 cows) used a collar-mounted device and activity alerts were used in conjunction with other information, before the farmer manually selected cows for AI. The gold standard to define the timing of oestrus was based on profiles of concentrations of progesterone in milk measured twice-weekly, used in conjunction with AI records and pregnancy diagnosis results. Sensitivity and positive predictive value (PPV) were calculated for the activity-based ODS data only, and then for AI decisions, against the gold standard. RESULTS: Farm 1 had 195 confirmed oestrus events and 209 activity alerts were generated. The sensitivity of the activity-based ODS was 89.2% with a PPV of 83.3%. Using tail-paint and cow history to confirm activity-based alerts 175 cows were inseminated, resulting in a sensitivity of 89.2% and an improved PPV of 99.4%. Farm 2 had 343 confirmed oestrus events, and 726 alerts were generated by the activity-based ODS, giving a sensitivity of 69.7% with a PPV of 32.9%. A total of 386 cows had AI records, giving a sensitivity of 81.3% and PPV of 72.3%. CONCLUSIONS: The two activity-based ODS were used differently on-farm; one automatically selecting cows and the other supporting the manual selection of cows in oestrus. Only one achieved a performance level suggested to be acceptable as a stand-alone ODS. Use of additional tools, such as observation of tail paint to confirm activity-based oestrus alerts before AI, substantially improved the PPV. CLINICAL RELEVANCE: A well performing activity-based ODS can be a valuable tool in identifying cows in oestrus prior to visual confirmation of oestrus status. However the performance of these ODS technologies varies considerably.

Applying additive logistic regression to data derived from sensors monitoring behavioral and physiological characteristics of dairy cows to detect lameness.

The hypothesis was that sensors currently available on farm that monitor behavioral and physiological characteristics have potential for the detection of lameness in dairy cows. This was tested by applying additive logistic regression to variables derived from sensor data. Data were collected between November 2010 and June 2012 on 5 commercial pasture-based dairy farms. Sensor data from weigh scales (liveweight), pedometers (activity), and milk meters (milking order, unadjusted and adjusted milk yield in the first 2 min of milking, total milk yield, and milking duration) were collected at every milking from 4,904 cows. Lameness events were recorded by farmers who were trained in detecting lameness before the study commenced. A total of 318 lameness events affecting 292 cows were available for statistical analyses. For each lameness event, the lame cow's sensor data for a time period of 14 d before observation date were randomly matched by farm and date to 10 healthy cows (i.e., cows that were not lame and had no other health event recorded for the matched time period). Sensor data relating to the 14-d time periods were used for developing univariable (using one source of sensor data) and multivariable (using multiple sources of sensor data) models. Model development involved the use of additive logistic regression by applying the LogitBoost algorithm with a regression tree as base learner. The model's output was a probability estimate for lameness, given the sensor data collected during the 14-d time period. Models were validated using leave-one-farm-out cross-validation and, as a result of this validation, each cow in the data set (318 lame and 3,180 nonlame cows) received a probability estimate for lameness. Based on the area under the curve (AUC), results indicated that univariable models had low predictive potential, with the highest AUC values found for liveweight (AUC=0.66), activity (AUC=0.60), and milking order (AUC=0.65). Combining these 3 sensors improved AUC to 0.74. Detection performance of this combined model varied between farms but it consistently and significantly outperformed univariable models across farms at a fixed specificity of 80%. Still, detection performance was not high enough to be implemented in practice on large, pasture-based dairy farms. Future research may improve performance by developing variables based on sensor data of liveweight, activity, and milking order, but that better describe changes in sensor data patterns when cows go lame.

Milking efficiency for grazing dairy cows can be improved by increasing automatic cluster remover thresholds without applying premilking stimulation.

It was hypothesized that streamlined premilking stimulation routines are effective at reducing cow cluster-on time but are not required to maintain milk yield or quality when increasing the automatic cluster remover (ACR) threshold above 0.4 kg/min. This was tested by examining the effect of 3 premilking treatments and 4 ACR thresholds over an 11-wk period with 96 mixed-age New Zealand Friesian-Jersey cross cows during peak lactation. Three premilking treatments were chosen: attach cluster immediately (control), attach cluster immediately and apply 30s of mechanical stimulation (Stim), and remove 2 squirts of milk from each quarter and attach cluster (Strip). Four ACR milk flow rate thresholds were imposed: 0.2 kg/min (ACR2), 0.4 kg/min (ACR4), 0.6 kg/min (ACR6), and 0.8 kg/min (ACR8). Measurements included individual cow milk yield, cluster-on time, average milk flow rate, maximum milk flow rate, time to average milk flow rate, time from maximum milk flow rate to end of milking, and the milk flow rate and cumulative yield at predetermined intervals during each milking session. Milk composition and somatic cell count (SCC) were determined on composite milk samples, collected weekly. Postmilking strip yield was measured at the end of each treatment period. Cows receiving the Strip treatment had a 3 to 4% shorter cluster-on time than did cows on the control treatment, but cows receiving Stim were not different from the control cows. Milk yield, SCC, and postmilking strip yield were not different between the 3 premilking treatments. Cluster-on time of the ACR8 cows was 18 to 26% less than that of the ACR2 cows, but SCC and milk production variables did not differ between the 4 end-of-milking treatments, despite higher strip yields as the ACR threshold increased. Increasing the ACR threshold is an effective strategy to improve milking efficiency (cows milked per operator per hour) in situations where the work routine times of dairy operators can be accelerated. To achieve the greatest milking efficiency, clusters should be attached immediately without premilking manual or mechanical stimulation.

Relationships between residual feed intake, average daily gain, and feeding behavior in growing dairy heifers.

Residual feed intake (RFI) is a measure of an individual's efficiency in utilizing feed for maintenance and production during growth or lactation, and is defined as the difference between the actual and predicted feed intake of that individual. The objective of this study was to relate RFI to feeding behavior and to identify behavioral differences between animals with divergent RFI. The intakes and body weight (BW) of 1,049 growing dairy heifers (aged 5-9 mo; 195 ± 25.8 kg of BW) in 5 cohorts were measured for 42 to 49 d to ascertain individual RFI. Animals were housed in an outdoor feeding facility comprising 28 pens, each with 8 animals and 1 feeder per pen, and were fed a dried, cubed alfalfa diet. This forage diet was chosen because most dairy cows in New Zealand are grazed on ryegrass-dominant pastures, without grain or concentrates. An electronic feed monitoring system measured the intake and feeding behavior of individuals. Feeding behavior was summarized as daily intake, daily feeding duration, meal frequency, feeding rate, meal size, meal duration, and temporal feeding patterns. The RFI was moderately to strongly correlated with intake in all cohorts (r=0.54-0.74), indicating that efficient animals ate less than inefficient animals, but relationships with feeding behavior traits (meal frequency, feeding duration, and feeding rate) were weak (r=0.14-0.26), indicating that feeding behavior cannot reliably predict RFI in growing dairy heifers. Comparison of the extremes of RFI (10% most and 10% least efficient) demonstrated similar BW and average daily gain for both groups, but efficient animals ate less; had fewer, longer meals; shorter daily feeding duration; and ate more slowly than the least-efficient animals. These groups also differed in their feeding patterns over 24h, with the most efficient animals eating less and having fewer meals during daylight (0600 to 2100 h), especially during the afternoon (1200 to 1800 h), but ate for a longer time during the night (0000-0600 h) than the least-efficient animals. In summary, correlations between RFI and feeding behavior were weak. Small differences in feeding behavior were observed between the most- and least-efficient animals but adverse behavioral effects associated with such selection in growing dairy heifers are unlikely.

Short-term application of prestimulation and increased automatic cluster remover threshold affect milking characteristics of grazing dairy cows in late lactation.

It was hypothesized that reducing cow cluster-on time by increasing automatic cluster remover (ACR) thresholds above 0.4 kg/min would require premilking stimulation of the mammary gland to maintain milk yield. This was tested by examining the interaction between 4 ACR thresholds and 3 premilking treatments over an 8-wk period with 96 mixed-age Friesian-Jersey cross cows being milked twice per day in late lactation (average production: 13.9 kg/d). The 3 premilking treatments were attach cluster immediately (control), attach cluster 60s after entering the dairy (delay), or remove 2 squirts of foremilk from each quarter and attach cluster 60s after entering the dairy (Prep). Four ACR thresholds were chosen, where the cluster was removed after the milk flow rate was less than 0.2 (ACR2), 0.4 (ACR4), 0.6 (ACR6), and 0.8 kg/min (ACR8). Measurements included individual cow milk yield, cluster-on time, average milk flow rate, maximum milk flow rate, time from cluster attachment to average milk flow rate, milk yield in the first 2 min, time from maximum milk flow rate to end of milking, and the milk flow rate at predetermined intervals during each milking session. Composite milk samples were collected weekly at a.m. and p.m. milkings to determine composition and somatic cell count (SCC). On 3 occasions during the experiment, postmilking strip yield was measured. No interactions were detected between premilking treatment and ACR threshold in any of the measured variables. Cows receiving the Prep treatment had a 5 to 9% shorter cluster-on time than the control treatment. Milk yield, SCC, postmilking strip yield, and maximum flow rate were not different between the 3 premilking treatments. Cluster-on time of the ACR8 cows was 21 to 29% less than ACR2, but SCC and milk production variables were not different between the 4 end-of-milking treatments despite higher strip yields as ACR threshold increased. Increasing ACR threshold offers the potential to reduce the duration of milking without detriment to overall productivity. The results of the premilking treatments indicate that to achieve the most efficient routine, the operator should attach clusters as close as possible to the first bail in rotary dairies to increase bail utilization in pasture-based systems. If cluster attachment can be sped up and ACR threshold lifted, significant potential exists to decrease herd milking duration and improve labor productivity.

Associations between herd size, rate of expansion and production, breeding policy and reproduction in spring-calving dairy herds.

Dairy herd size is expected to increase in many European countries, given the recent policy changes within the European Union. Managing more cows may have implications for herd performance in the post-quota era. The objective of this study was to characterise spring-calving herds according to size and rate of expansion, and to determine trends in breeding policy, reproduction and production performance, which will inform industry of the likely implications of herd expansion. Performance data from milk recording herds comprising 775,795 lactations from 2,555 herds for the years 2004 to 2008 inclusive were available from the Irish Cattle Breeding Federation. Herds were classified into Small (average of 37 cows), Medium (average of 54 cows) and Large (average of 87 cows) and separately into herds that were not expanding (Nil expansion), herds expanding on average by three cows per year (Slow expansion) and herds expanding on average by eight cows per year (Rapid expansion). There was no association between rate of expansion and 305-day fat and protein yield. However, 305-day milk yield decreased and milk protein and fat percentage increased with increasing rate of expansion. There were no associations between herd size and milk production except for protein and fat percentage, which increased with increasing herd size. Average parity number of the cows decreased as rate of expansion increased and tended to decrease as herd size increased. In rapidly expanding herds, cow numbers were increased by purchasing more cattle. The proportion of dairy sires relative to beef sires used in the breeding programme of expanding herds increased and there was more dairy crossbreeding, albeit at a low rate. Similarly, large herds were using more dairy sires and fewer beef sires. Expanding herds and large herds had superior reproductive performance relative to non-expanding and small herds. Animals in expanding herds calved for the first time at a younger age, had a shorter calving interval and were submitted for breeding by artificial insemination at a higher rate. The results give confidence to dairy producers likely to undergo significant expansion post-quota such that, despite managing more cows, production and reproductive performance need not decline. The management skills required to achieve these performance levels need investigation.

Effect of setting a maximum milking time, from peak lactation, on production, milking time and udder health.

AIM: To examine the effect of setting a maximum milking time, from peak lactation until drying-off, on production, duration of milking, and udder health of dairy cows. METHODS: Forty cows were assigned in twin-pairs to be either milked until cups were removed at a milk flow-rate threshold of 0.35 kg/minute (Control), or until cups were removed at a milk flow-rate threshold of 0.35 kg/minute, or maximum time, whichever came first (MaxT). The maximum time was set by determining the milking time of the 70th percentile cow when ranked from fastest to slowest, irrespective of yield. The milking routine was typical of that practised on dairy farms in New Zealand, and involved no pre-milking preparation. The study began at peak lactation (68 (SD 7) days in milk; DIM) and continued for 26 weeks. Duration of milking and milk yield were measured for each milking. Composition of milk was determined from weekly herd tests, and milk quality from fortnightly somatic cell counts (SCC). Completeness of milking and teat condition were assessed during the study. The bacterial status of quarter milk samples was determined at the beginning and end of the study, and all treated cases of clinical mastitis recorded. ANOVA was used to examine the effect of treatment group on variables of interest. RESULTS: Total milk, fat and protein yields during the study period did not differ between treatments. On average, 30.3% of the morning and 27.6% of the afternoon milkings of MaxT cows reached the maximum time at which cups were removed, and were therefore shortened. While the average milking time of the slowest-milking cow was longer for the Control compared with MaxT group in Weeks 1-18, the average milking time did not differ between treatments. There was no difference in overall SCC, and the incidence of clinical mastitis, or the percentage of infected quarters at drying-off, was similar for the MaxT and Control cows. CONCLUSION: The results show that setting a maximum milking time can reduce the milking time of slower-milking cows in a herd without compromising overall herd production and udder health. CLINICAL RELEVANCE: Although the numbers of cows in the study were small there was no evidence of a major increase in SCC, or subclinical or clinical mastitis when a maximum milking time was set for slower-milking cows.

Effect of automatic cluster remover settings on production, udder health, and milking duration.

This study evaluated the effect of 4 criteria for determining the end-point of milking on milk yield, milk composition, completeness of milking-out, teat skin condition, somatic cell count (SCC), and the incidence of clinical mastitis (CM) in pasture-based dairy cows milked over 35 wk. The objective was to reduce milking duration without affecting milk production, SCC, or CM. Milking end-point treatments were as follows: cluster removed at a milk flow rate of 0.2 kg/min (ACR200); cluster removed at a milk flow rate of 0.4 kg/min (ACR400); cluster removed at a milk flow rate of 0.2 kg/min or at a maximum cluster attachment time from d 5 of lactation (MaxTEarly); and cluster removed at a milk flow rate of 0.2 kg/min until an average of 63+/-21 d in milk, then cluster removed at a milk flow rate of 0.2 kg/min or a maximum cluster attachment time (MaxTPeak). Maximum cluster attachment times were set at 7.5 min and 5.4 min for morning and afternoon milkings, respectively. Cows (approximately 94/treatment) were assigned to treatment at calving and milked twice daily at intervals of 9 and 15 h. Milking duration was shorter for ACR400, MaxTEarly, and MaxTPeak compared with ACR200. During wk 1 to 15, milk, protein, and lactose yields were less for MaxTEarly than for ACR400 and MaxTPeak, but not different from ACR200. During wk 16 to 35 and over the entire experiment, total milk, fat, protein, and lactose yields did not differ among treatments. Teat condition did not differ among the 4 treatments. Postmilking strip yield in wk 12 was greatest for MaxTEarly and least for ACR200; at wk 27, however, treatment had no effect on the completeness of milking-out. No differences were observed in either teat condition or the proportion of cows with at least 1 case of CM during the 35 wk. Somatic cell count was low across all treatments, but highest for ACR400. Increasing the automatic cluster remover threshold setting from 0.2 to 0.4 kg/min decreased milking duration without affecting milk production, CM, or teat condition. Combining a cluster removal milk flow threshold setting with a maximum cluster attachment time, when applied from either early lactation or from peak lactation, reduced milking duration without affecting milk production, CM, or SCC. Both strategies have potential to improve milking efficiency in dairy herds in which premilking preparation is minimal.

Capturing urine while maintaining pasture intake, milk production, and animal welfare of dairy cows in early and late lactation.

Capturing urine and spreading it evenly across a paddock reduces the risk of nitrogen loss to the environment. This study investigated the effect of 16h/d removal from pasture on the capture of urination events, milk production, pasture intake, and animal welfare from cows grazing fresh pasture in early and late lactation. Forty-eight Holstein-Friesian cows in early [470+/-47kg of body weight (BW); 35+/-9 days in milk] and late (498+/-43kg of BW; 225+/-23 days in milk) lactation were allocated to 3 treatment groups. Cows had access to pasture for either 4h after each milking (2 x 4), for 8h between morning and afternoon milkings (1 x 8), or for 24h, excluding milking times (control). When not grazing, the 2 x 4 and 1 x 8 groups were confined to a plastic-lined loafing area with a woodchip surface. In early lactation, the proportion of urinations on pasture and laneways was reduced from 89% (control) to 51% (1 x 8) and 54% (2 x 4) of total urinations. The 1 x 8 cows ate less pasture [10.9kg of dry matter (DM)/cow per day] than the control (13.6kg of DM/cow per day) and 2 x 4 (13.0kg of DM/cow per day) cows, which did not differ from each other. The 1 x 8 and 2 x 4 cows produced less milk (21 and 22kg of milk/cow per day, respectively) compared with control cows (24kg of milk/cow per day). There were no differences in BW or body condition score (BCS) change across treatment groups, with all groups gaining BW and BCS during the experimental period. In late lactation, there was no difference in pasture intake (mean=8.8kg of DM/cow per day), milk production (mean=10kg of milk/cow per day), and BW or BCS change (mean=3.7kg and -0.2U/cow per week, respectively) between treatment groups. As in early lactation, urinations on pasture and laneways were reduced from 85% (control) to 56% (1 x 8) and 50% (2 x 4) of total urinations. These findings highlight an opportunity to maintain performance and welfare of grazing cows in early and late lactation while capturing additional urine. This can subsequently be spread evenly across pasture to minimize nitrogen loss to the environment.

Restricting time at pasture: effects on dairy cow herbage intake, foraging behavior, hunger-related hormones, and metabolite concentration during the first grazing session.

This study investigated the effect of restricting grazing time on circulating concentrations of ghrelin, nonesterified fatty acids (NEFA), and glucose before, and foraging behavior of dairy cows during, the first grazing session of the day (GS, 0800-1200 h). Forty-eight Holstein-Friesian cows (470 +/- 47 kg of BW; 35 +/- 9 d in milk) were strip-grazed on a perennial ryegrass pasture for either 4 h after each milking (2 x 4), 8 h between milkings (1 x 8), or the 24-h period excluding milking times (CTL). Cows were bled before the GS; plasma was analyzed for ghrelin and serum for glucose and NEFA. Herbage mass was measured pregrazing (0730 h), during and at the end of the GS (1200 h), and postgrazing (24 h after the first measurement). Herbage mass data were fitted to a model to estimate herbage disappearance rates. Herbage intake and bite mass were calculated using herbage mass disappearance and behavioral measurements. Bite rate, eating, searching, ruminating, and idling time were determined during the GS for each cow. No difference in glucose concentration was found between treatments. Concentrations of NEFA and ghrelin were the greatest for cows in the 1 x 8 treatment. Daily herbage intake did not differ between treatments; however, during the GS 1 x 8 had a greater herbage intake than 2 x 4 and CTL. Bite mass differed between treatments and throughout the GS. Bite mass was smallest for CTL during the first 60 min and greatest during the last 90 min, when cows in the 2 x 4 treatment had the smallest bite mass. Cows in 1 x 8 spent the longest time eating and the least time searching and ruminating. Eating time was greatest for 1 x 8 during the first 60 and last 90 min of the GS. Searching time only differed in the second 60 min, when it was the lowest for 1 x 8. Cows from all treatments did not ruminate during the first 120 min. Cows in CTL had the greatest rumination time during the last 90 min. The model fitted to represent dynamics of herbage mass disappearance presented differences in the fractional herbage disappearance rate. There was an interaction between treatment and time in herbage depletion rate. The results of this study present a fuller picture of foraging dynamics during the first 4 h of grazing and its potential relationship with physiological markers of hunger as affected by grazing management.

The effect of prepubertal immunization against gonadotropin-releasing hormone on the development of sexual and social behavior of bulls.

To determine the effect of prepubertal immunization against GnRH on the development of sexual and social behavior of Friesian bulls, 90 calves were randomly assigned to five treatments: 1) I2, immunized against GnRH at 2 and boosted at 2.5, 4, and 7.5 mo of age, n = 2 x 10; 2) I4, immunized against GnRH at 4 and boosted at 4.5 and 7.5 mo of age, n = 2 x 10; 3) I7.5, immunized against GnRH at 7.5 and boosted at 8 mo of age n = 2 x 10; 4) S, steers castrated at 2 mo of age, n = 10; and 5) B, intact bulls, n = 2 x 10. Blood samples were collected initially every 2, then every 3 wk. Plasma was analyzed for anti-GnRH titers and plasma testosterone concentration. Sexual and agonistic behavior, male-male mounting, and damage to paddocks was assessed throughout the experiment. All immunized calves developed antibodies against GnRH (32.3 +/- 2.0% bound at a 1:10 plasma:PBS-BSA dilution, 14 d after first boost). Plasma testosterone concentrations were < 1 ng/mL for all immunized animals until 11 mo of age, when they increased to levels found in intact bulls at 14 mo of age. At slaughter, testes and seminal vesicle weights were 38.3 and 31.6% lighter, respectively, for all immunized treatments compared to B. There were no significant differences between I2, I4, and I7.5 in any of the sexual or agonistic behavior tests. Bulls scored higher than steers in all sexual behavior tests. Immunized bulls scored lower than bulls in sexual behavior tests from 10 to 17 mo of age. The proportion of immunized animals that serviced an estrous cow was lower than the proportion of intact bulls at 10, 12.5, 14, and 17 mo of age. Immunized animals scored lower than bulls in bull challenge tests at 8.5, 11.5, 13, 14.5, and 17 mo of age. Paddock damage by animals on the three immunization treatments was lower than that by bulls from 7 to 14.5 mo of age, as were leg were scores (an indicator of male-male mounting behavior) from 9 to 14 mo of age. There was no difference in sexual behavior between immunized bulls (I2, I4, and I7.5) and bulls while held in lairage pens for 16 h before slaughter, but all treatment groups scored higher than steers. There was a similar trend for agonistic behavior, although I4 bulls were no different from steers. Prepubertal immunization against GnRH at 2, 4, and 7.5 mo of age impaired testes function and affected the development of social and sexual behavior of young bulls.

Risk factors associated with bruising in red deer at a commercial slaughter plant.

Bruising of slaughter animals has both economic and welfare implications. In order to identify potential risk factors we surveyed bruising of red deer carcasses from a single Deer Slaughter Plant (DSP) over a three year period. Downgrading due to bruising did not vary between the three years surveyed [1991, n = 21 454, 6.5% downgraded (dg); 1992, n = 22 289, 7.9% dg; 1993, n = 22 262, 6.1% dg]. From an intensive examination of one year's kill (1991) we related the incidence of bruising to various preslaughter transport, seasonal and animal variables. Bruising varied significantly with time of year, hot carcass weight, distance transported, carrier company and carcass fatness. Bruising was related to duration of lairage for some periods of the year which varied with sex. The majority of bruising was on one or other hindquarter implying a consistent cause of the damage. Carcass quality and deer welfare can be best protected by careful driving, by only transporting animals in good condition and avoiding transporting males during the rut.