If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Center for Animal Welfare, Department of Animal Science, University of California, Davis 95616Animal Behavior Graduate Group, University of California, Davis 95616
Many milk-fed dairy calves are not provided forage. In these settings, calves often perform abnormal repetitive behaviors (ARBs), including tongue rolling and nonnutritive oral manipulation (NNOM), which, based on their form, seem similar to movements used when processing feed. Feeding hay, typically presented as a short chop (≤5 cm) in a bucket, reduces ARBs. Our objective was to evaluate whether altering the presentation method of long hay (∼19 cm), by providing it in a bucket or a novel polyvinyl chloride (PVC) pipe feeder, could reduce ARBs. Holstein heifer calves were housed individually on sand and fed ad libitum starter grain and limited milk replacer (5.7–8.4 L/d step-up) via a bottle (Control, n = 9) or given access to mountaingrass hay in a bucket (Bucket, n = 9) or in a PVC pipe feeder (Pipe, n = 9). The 56 × 10.2 cm (length × diameter) PVC pipe feeder had 4 openings that were 6.35 cm wide, which required the calf to insert her tongue into the pipe and curl her tongue to extract hay. Treatments were applied from birth through 50 d of age, when step-down weaning began and TMR was provided to all calves. Calves were fully weaned at d 60. At wk 4 and 6, oral behaviors (eating, ruminating, drinking water, sucking milk, self-grooming, NNOM, tongue rolling, tongue flicking, and panting) were recorded by direct observation for 24 h using 1–0 sampling during 5-s intervals. Feeding long hay, regardless of presentation method, increased overall DMI, grain intake, and ADG compared with Control calves. Hay provision also increased rumination (25 vs. 15% of 24-h observations in Control) and eating time (5.5 vs. 2% in Control). Abnormal behaviors were seen in all calves. Hay provision reduced some of these, including NNOM (5 vs. 9% in Control). There was no difference in NNOM between calves fed hay in a pipe or bucket, even though Bucket calves consumed more hay (178 vs. 129 g/d in wk 6) and tended to spend more observations eating hay than Pipe calves (4.5 vs. 3%). Hay provision did not affect other behaviors: drinking water (0.5%), grooming (3%), or tongue flicking (3%). We also found evidence of other abnormal oral behaviors that have received less attention. Calves showed signs of polydipsia, and displayed excessive grooming, the latter indicated by overall duration, number of bouts per day, and duration of individual bouts (up to 25 min). Tongue rolling was expressed at low levels (up to 0.4% of intervals) but by 85% of calves. Feeding hay, both in a bucket and using novel methods, was not enough to counteract the welfare challenges associated with individual housing and limited ability to suck milk (<1% of time). Provision of long hay, regardless of presentation method, promotes rumination, improves performance (higher grain intake and ADG) and reduces at least some, but not all, of the considerable abnormal oral behaviors these calves performed.
). In many cases, it appears that these behaviors do not solely stem from hunger. Limited ability to perform the physical actions associated with food processing, including sucking, chewing, and ruminating, may also be a key component of these ARBs. Calves fed milk via esophageal feeders (
) increase oral manipulation of pen fittings following feeding compared with calves fed via teats, even when meal sizes and daily allowances are kept constant. Similarly, calves fed via slow-flow teats, which increase sucking time, show less nonnutritive oral manipulation (NNOM) than calves fed the same amount of milk at a faster flow rate (e.g.,
Replacing concentrates with a high-quality hay in the starter feed in dairy calves: I. Effects on nutrient intake, growth performance, and blood metabolic profile.
), and cattle seem to seek out opportunities for these behaviors. For example, when cows have their rumens artificially filled, they spend more time searching for feed and manipulating their environment than those who are able to ingest food themselves (
), which may suggest rumination is a highly motivated behavior. Calves that are able to consume hay, and engage in more chewing and ruminating than those reared without forage, in turn perform less NNOM (e.g.,
). Food processing behaviors may thus represent a behavioral need for cattle, that is, one that animals are highly motivated to perform regardless of functional consequences, which can have negative implications if prevented (e.g.,
), and provision of prechopped forage in a trough or bucket does not allow for this natural motion or effort. Indeed, prevention of these encircling motions during feeding has been hypothesized to be a factor influencing tongue rolling, a common abnormal behavior in cattle (e.g.,
). Tongue rolling is also seen in zoo-kept giraffes and okapi, who use their tongues when ripping browse in natural settings, but is not seen in species that bite grass, including goats and sheep (
). Although hay provision in a bucket can reduce tongue rolling and other ARBs in calves reared in farm settings, they are still performed by most animals (e.g.,
). It is possible that food acquisition may be part of a broader behavioral need to forage, along with food processing. Hay acquisition, and whether it encourages natural grazing motions, may thus be an important facet to further reduce ARB development and improve welfare that is not currently addressed by providing hay in a bucket.
When provided, forage is presented to young calves in different ways, but the influence these options have on ARB performance is not clear cut. It is common to study the effect of short chopped hay in a bucket (e.g., <5 cm;
Effects of feeding hay and calf starter as a mixture or as separate components to Holstein calves on intake, growth, and blood metabolite and hormone concentrations.
), likely because they require more manipulation. This should be associated with a decrease in ARBs, given previous findings, but this is not always the case. Both NNOM and tongue rolling are reduced when calves are fed coarse hay mixed with grain compared with ground (4 vs. 2 mm;
). Sometimes presentation of hay is combined with multiple other provisions that may reduce ARBs (e.g., dummy teats, socialization, grooming brushes, or scented hay;
), so it is not possible to isolate the effects of hay. Studies of feed presentation methods focus mostly on sorting behavior or physiological performance, and as such ARBs are not often scored (e.g.,
Effects of feeding hay and calf starter as a mixture or as separate components to Holstein calves on intake, growth, and blood metabolite and hormone concentrations.
). More work is needed to elucidate the link between forage presentation, natural food acquisition and processing motions, and ARB reduction.
This link between ARBs and methods of forage presentation is further complicated by how behavior is measured. Calves show circadian variation in behavior. Some behaviors peak around or after milk feeding (e.g., eating,
), which could skew interpretations about performance. This may be particularly important in light of recent evidence that, across 24-h periods, calves perform a wide array of possibly abnormal behaviors that are often not scored. In addition to possible sham rumination, calves show polydipsia (
). It is possible that providing more opportunities for natural forage acquisition and processing behaviors could have wide-ranging effects on all calf oral behaviors over 24 h.
We set out to design a novel hay enrichment device that encouraged natural forage acquisition movements and increased effort required to obtain hay before chewing it. Devices of this nature have been created to reduce oral ARBs, including tongue rolling, in giraffes (
) and were the basis for our design. We predicted that calves provided with hay would eat and chew more and perform fewer oral ARBs across 24 h relative to controls, and that presenting it in a pipe feeder would result in more marked changes in these behaviors. We also expected to see fewer possibly abnormal behaviors such as tongue flicks and long or repeated bouts of grooming in calves provided hay. Panting, another oral behavior performed by calves, typically reflective of heat stress (e.g.,
), was scored to add to a limited body of research on this behavior in calves. We did not expect that hay provision would reduce grain intake or ADG, given recent evidence that forage does not displace either one (e.g.,
This study was conducted from June to November 2019 at the University of California, Davis Dairy Facility. All procedures were approved by the University of California, Davis Institutional Animal Care and Use Committee (protocol #20466).
Animals and Housing
We enrolled all healthy female Holstein calves born between June 1 and September 21, 2019 (n = 27). Calves were housed individually in outdoor plastic hutches (2 × 1.5 m, length × depth) with an attached wire-fenced pen (2 × 1.5 × 0.9 m, length × depth × height). Hutches and pens were spaced ∼0.5 m apart, allowing calves to touch the muzzle of neighboring animals. The enclosures were bedded with sand approximately 12 to 17 cm deep that was spot-cleaned daily and topped-up as needed. The sand was covered with perforated rubber mats from d 0 to 5 ± 1 to limit unintentional inhalation of sand particulate (Supplemental Figure S1, https://doi.org/10.5281/zenodo.6618168;
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
Calves received colostrum twice a day for 5 d with each calf consuming 18.1 ± 1.1 L (mean ± SD) total. Colostrum meals were fed via a bottle and rubber teat (Connewango) or an esophageal tube feeder (2.1% of all colostrum feedings across calves). From 5 to 9 d of age, calves were fed 1.9 L of milk replacer (26% CP, 16% fat, 15% TS, mixed as indicated at a rate of 142 g/L of hot water; Calva Products Inc.) at each of 2 daily meals at approximately 0915 and 1615 h. From 10 to 23 d of age, calves received 2.4 L of milk replacer at each of the 2 meals, and from 24 to 49 d of age, 2.8 L of milk replacer per meal. All milk meals were fed via a bottle and rubber teat. Bottles were available inside the hutch 0.6 m above the ground. Weaning began at 50 d when the 0915 h meal was removed. Calves were fully weaned at 60 d when the 1615 h meal was removed. All colostrum and milk feedings were in accordance with farm protocol.
All calves were disbudded at 7 to 10 d (n = 14, mean birth weight = 38.7 kg) or 50 to 53 d (n = 13, mean birth weight = 38.6 kg) as part of a separate experiment balanced by feeding treatment and birth weight. Calves were given a cornual nerve block with 5 to 7 mL lidocaine hydrochloride on each side of the head and hair over the horn bud was trimmed before application of a disbudding iron until a copper ring formed. Oral meloxicam (0.5–1 mg/kg) was provided to all calves after disbudding. All calves were vaccinated with Bovishield 5 at 21 and 40 d, Spirovac at 30 d, and One Shot at 40 d (all Zoetis Inc.). Electrolyte treatments (1.9 L) were provided as per standard farm practice if loose feces were observed. Nine calves received electrolytes for 1 to 4 treatments total over the course of the experiment (2 Control, 4 Pipe, 3 Bucket); only 1 of these calves received electrolytes on consecutive days (2 feedings total). No observation days were affected by these.
Experimental Design
Calves were allocated to 1 of 3 treatments based on birth order, with each treatment represented in a given similarly aged cohort of 3 calves, using a random number generator. Adjustments were made to balance birth weight across treatments. Calves assigned to the Control treatment (n = 9, mean birth weight = 38.6 kg) received ad libitum water and grain (Table 1; Starter Calf Feed 901033, Associated Feed and Supply Co.; Supplemental Figures S1 and S2, https://doi.org/10.5281/zenodo.6618168;
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
) by bucket from birth. Calves assigned to the Bucket treatment (n = 9, mean birth weight = 38.6 kg) received ad libitum water, grain, and chopped (19 ± 4 cm) mountaingrass hay (mix of orchard, Dactylis glomerata, and fescue, Festuca arundinacea; Higby's Country Feed; Table 1; Table 2; Supplemental Figures S1 and S2) by bucket from birth. Calves in the Pipe treatment (n = 9, mean birth weight = 38.8 kg) received ad libitum water, grain, and the same chopped mountaingrass hay in a 56 × 10.2 cm (length × diameter) polyvinyl chloride (PVC) pipe feeder (Figure 1; Supplemental Figures S1–S3; Supplemental Videos S1–S4, https://doi.org/10.5281/zenodo.6618168;
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
). The pipe feeder had 4 equally spaced 6.35-cm (diameter) holes cut into it that were sanded down until smooth, and was fitted with 2 removable knockout end caps. Pipes were mounted 0.8 m from the ground using a 10.2-cm vent pipe hanger in all pens. All calves had 3 buckets available at all times: grain, water, hay (Bucket group), or empty bucket (Control and Pipe groups), left to right (Supplemental Figure S1), and a pipe feeder with hay (Pipe group) or kept empty (Control and Bucket groups). The assigned feeding method continued throughout the preweaning stage. On d 50, when weaning began, pipes were removed for all calves and ad libitum TMR (Table 1, Table 2; alfalfa, almond hulls, cottonseed, corn, barley, beet pulp) was provided for all calves via bucket. The TMR was placed in the bucket previously used for hay (Bucket group) or kept empty (Control and Pipe groups). All calves had access to grain and water in the same buckets as in the milk-fed period throughout weaning.
Table 1Chemical composition of solid feeds (mean ± SD, reported as % of DM)
Figure 1Calf feeding from a polyvinyl chloride (PVC) pipe feeder filled with chopped mountaingrass hay. The pipe was 56 × 10.2 cm (length × width) with 4 equally spaced 6.35-cm holes cut into it that were sanded down until smooth and was fitted with 2 removable knockout end caps. Pipes were mounted 0.8 m from the ground using a 10.2-cm vent pipe hanger in all pens.
Feed and water intake were recorded beginning on d 0 of life by providing preweighed fresh feed daily at 0800 h and subtracting the refusals on the following day at 0800 h from the initial provisions (GBK16a Bench Check Weighing Scale 8,000 g limit/0.1 g readability, Adam Equipment Inc.). Feed levels were checked at 1200, 1600, and 2000 h and fresh measured feed was added if necessary to maintain ad libitum levels. When possible, clean spilled hay was added back to the bucket at each feeding; spillage contaminated with feces or water was removed daily at 1600 h. Calves were fed to at least 115% of the previous day's intake throughout the experiment, with a minimum of 800 g of grain and 40 g of hay provided at each feeding. Water orts were recorded and fresh measured water provided at 0800, 1200, and 1600 h. If low, fresh water was added at 2000 h. If feces were present in any of the buckets, or intake did not represent a continuous 24-h period (e.g., for calves born after 0800 h), intake data for that calf for that day were discarded. Calves were weighed weekly to determine ADG during the experimental period (VS-660 Scale 300 kg limit/0.1 kg readability, A and A Scales LLC).
To calculate DM and nutritional protein of the feed, fresh feed was sampled once a week and sent to Cumberland Valley Analytical Services Inc. for analysis of DM (1358C; method 930.15;
; Leco FP-528 Nitrogen Analyzer, Leco]. The nutrient content of the solid feed is reported in Table 2. Samples of daily orts were taken per calf and combined weekly to determine DM. These combined samples were oven-dried at 100°C for 15 h to determine DM content. Weights were recorded with an AG104 101 g limit/0.1 mg readability scale (Mettler Toledo, LLC.). Particle sizes of mountaingrass hay and TMR were determined via a 3-sieve Penn State Particle Separator, used as directed by one trained researcher (40 total shakes, 5 in each direction;
Calf behavior was recorded using 1–0 sampling (present or not present) during 5-s intervals for a continuous 24 h beginning at 0800 h at wk 4 (31 ± 3 d) and 6 (45 ± 4 d) for each calf. This methodology was similar to our previous work (e.g., 1–0 sampling during 1-min intervals,
) but refined to 5-s intervals to improve estimates. Weeks 4 and 6 were chosen as calves are reported to consume nonnegligible amounts of hay during this period and proportion of time engaged in oral behavior stabilizes during this period (
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
) with continuous countdown timers (Pebble Technology Corp., Fitbit). Eating overall was the sum of eating hay and eating grain. NNOM: total was the sum of NNOM: pipe and NNOM: other. Calves were watched for a continuous 24 h using a total of 16 observers. An observer watched one calf at a time in 2-h shifts. Observers were not blind to treatment because feed availability could not be masked. Observers were first trained using 1–0 sampling during 5-s intervals using a total of 1.5 h of video from 9 unique calves not used in this experiment. All possible behaviors were exhibited in these videos. Observers were then trained in live sessions using 1–0 sampling during 5-s intervals for a minimum of 3 h across 3 separate days (maximum training time = 4.75 h across 5 sessions), and reliability values were taken from these. Tongue rolling and panting were rare in these live sessions and were assessed using a 30-question video test (15 yes, 15 no). All observers were trained to reliability ≥75% on tongue flicks (Cohen's kappa, irr package version 0.84.1,
For all relevant definitions, muzzle is defined as from the bottom of the eyes to the end of the mouth. See Supplemental Videos S1–S12 in https://doi.org/10.25338/B8V054, Downey et al., 2022b, and Supplemental Video S5 in https://doi.org/10.5281/zenodo.6618168, Downey and Tucker, 2022, for examples.
Behavior
Definition
Drinking water
Any part of the muzzle is in the water bucket for at least 1 s, or mouth or tongue visibly enters water.
Eating grain
Any part of the muzzle is in the feed bucket while grain is present for at least 1 s, or jaw movements while the head is above the bucket or within 1 head length of the feed bucket, or jaw movements while grain is visibly held within the mouth
Eating hay
Any part of the muzzle is in the feed bucket while hay is present for at least 1 s, or mouth is directed at one of the holes of the polyvinyl chloride (PVC) pipe of at least 1 s, or jaw movements while the head is within 1 head length of the feed bucket/PVC pipe, or jaw movements while hay is visibly held within the mouth
Grooming body
Touching hair with the tongue or mouth on calf's own body or a neighboring animal
Nonnutritive oral manipulation (NNOM) of the pipe was scored separately from all other NNOM because it was a novel addition to all pens.
Licking, chewing, or sucking directed toward the PVC pipe, but not at any of the 4 holes, unless the pipe is empty
NNOM: other
Licking, chewing, or sucking directed toward a nonnutritive item (includes bars, hutch, bedding, empty bucket; excludes PVC pipe): tongue or lips must be touching a nonnutritive item, or such item must be held inside the mouth
Panting
Breathing involves heavy movements of the abdomen and thorax; mouth may be open (space between the lips is visible)
Ruminating
Rhythmic circular jaw movements (at least 2) that happen anywhere in the pen except over the feed buckets/PVC pipe; if only neck is visible, can be identified by visual observation of a bolus moving down/up the neck
Sucking milk
Mouth around nipple of a bottle containing milk
Tongue flicks
Tongue extends out of the mouth without touching other objects or forming a full or partial circular motion, or extends up to the nose before retracting back into mouth and repeating at least once more within 1 s; can occur while eating and ruminating
Tongue rolling
Tongue is held in a full or partial circular position and/or moves in a full or partial circular motion; this can occur when the tongue is held within the border of the lips inside the mouth and/or extended outside the border of the lips. This cannot occur while any other behaviors are being performed (the tongue is not touching any feed/nonnutritive items), and does not need to repeat.
1 For all relevant definitions, muzzle is defined as from the bottom of the eyes to the end of the mouth. See Supplemental Videos S1–S12 in https://doi.org/10.25338/B8V054,
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
) with calf as the experimental unit. Model fit was checked for normality and homogeneity of variance using QQ plots and plots of residuals vs. fitted values (plot, boxplot, resid functions in base R, ggpubr package version 0.4.0,
) to obtain P-values. If significant (P < 0.05) treatment or interaction effects were detected, P-values for individual pairwise comparisons were obtained using estimated marginal mean contrasts (emmeans package v. 1.6.1,
). The interaction term is reported if P < 0.1, otherwise the individual significant fixed effects are reported.
Feed Intake
Some intake data were excluded: 9.4, 23.6, 4.9, and 7.7% of calf-day values for grain, hay, water, and TMR, respectively, due to contamination with feces or saliva, spillage by calves (e.g., entire bucket spilled or kicked over), or intake covering <24 h (e.g., d 0 of life for some calves). Most of the exclusions for grain (88% of the 9.4% excluded) and hay intake (76% of the 23.6% excluded) occurred within the first 2 wk when consumption was low (<0.09 kg/d) and saliva contamination had a large effect. Data were also excluded if intake could not be calculated on a DM basis, which was due to human error in drying individual ort samples.
Intake data were separated by age into preweaning (0–49 d) and weaning (50–59 d) periods. During the preweaning period, intake data were averaged by week. Data from d 49 were included in the wk 6 values. Grain and hay data were modeled with a linear mixed model using the REML method with an identity variance structure to account for heterogeneous variance across weeks (lme4 package version 1.1–26,
). Week was treated numerically as data were collected continuously over time. Week, treatment, and the interaction between week and treatment were fixed effects, whereas calf was a random effect.
During the weaning period, intake was assessed by day; day was treated as a numeral. Grain, TMR, and water intake were modeled with linear mixed models (lme4 package version 1.1–26,
). Week, treatment, and the interaction between week and treatment were fixed effects and calf was a random effect. ADG from wk 0 and 7 were not included in the model as not all calves these values, based on when they were first measured (wk 0) or weaned (wk 7), and early life stress and adjustment to the environment can lead to growth checks (e.g.,
). All calves thus had 5–6 ADG calculations, with the final preweaning measurement occurring on 45 ± 2 d.
Average daily gain during weaning was not assessed, as calves only had 1 or 2 measurements during this period, and measurements were not taken at a consistent point in the step-down process for all calves. Instead, final weight (measured on 57 ± 2 d) and overall ADG across the entire experimental period [(final weight − birth weight)/age in days at final weight] were assessed using a one-way ANOVA (aov function in base R). If a significant treatment difference was found, Tukey pairwise comparisons were calculated and adjusted P-values are reported.
Behavioral Observations
Proportion of time engaged in each behavior was analyzed using a generalized linear mixed model (glmmTMB package v 1.0.2.1.3,
) fit with a β distribution and logit link. Week was treated categorically as observations were conducted at 2 distinct time points. Week, treatment, and the interaction between week and treatment were fixed effects, and calf was a random effect. Tongue rolling and panting were rare and were not analyzed with a model. Behavioral data are reported as percentages hereafter in the text to facilitate readability; raw data (https://doi.org/10.25338/B8Z052,
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
Some data were analyzed for extreme expression. Outliers in water consumption could indicate polydipsia, whereas outliers in grooming bouts may provide insight into abnormal expression of a seemingly normal behavior. Grooming bouts were considered to be consecutive runs of 5-s intervals in which grooming occurred. For each calf in each week, we converted nongrooming intervals occurring singly or in consecutive runs of ≤3 (= 15 s) that were bordered by grooming to this behavior. Nongrooming gaps of ≥4 intervals (20 s) thus ended a consecutive run of grooming. This cutoff (20 s) was determined based on visual inspection of these data. Outliers in both water intake (preweaning and weaning, separately) and grooming were calculated using interquartile range criteria, such that values that fell more than 1.5× below or above the first and third quartiles, respectively (boxplot function in base R), were considered extreme.
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
During the preweaning period (d 0–49), there was a significant week by treatment interaction for grain (P < 0.001) and water intake (P = 0.004; Figure 2; Supplemental Table S1, https://doi.org/10.5281/zenodo.6618168;
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
). All calves consumed more grain and water over time, but calves fed hay (Bucket and Pipe groups) increased their consumption of these resources more than Control calves. Individual water intake varied from 0.1 to 11.05 L/d, with outliers considered to be those ≥7.35 L/d. Twelve calves (4 Control, 3 Bucket, 5 Pipe) had outlier values representing 2.5% of all calf-day measurements during the milk-fed period. For these 12 calves, these outliers represented 2 to 21% of all their values. There was also a significant week by treatment interaction for hay (P < 0.001), as calves in the Bucket group consumed more hay over time than calves in the Pipe group. Our intention was to feed hay ad libitum: 11% of calf-days for the Bucket treatment and 12% for Pipe had <10% hay refusals.
Figure 2Feed and water intake values throughout the treatment period of calves fed a diet of grain and milk replacer from birth (Control) or a control diet with additional mountaingrass hay from birth in a bucket (Bucket) or novel polyvinyl chloride (PVC) pipe feeder (Pipe). Total mixed ration was available for all calves at the start of step-down weaning (50–60 d), as indicated by the dashed line. Data are summarized by treatment (trt) and averaged by 7-d period (wk) across the preweaning period, and by day (d) during step-down weaning. Error bars represent SE. P-values <0.1 are reported for week (preweaning) or day (weaning), treatment, or their interactions.
During the weaning period, all calves consumed more grain (P = 0.018; Figure 2) and TMR (P < 0.001; Figure 2) as days passed, but there was no evidence that treatment affected intake (P ≥ 0.117). There was a day by treatment interaction for water consumption (P = 0.029; Figure 2), with Bucket calves consuming more water than Pipe and Control calves as weaning continued. Individual water intake during the weaning period varied from 1.00 to 17.15 L/d, with outliers considered to be those ≥16.05 L/d. Two calves (1 Bucket, 1 Pipe) had outlier values on 5 d (1.8% of all calf-day measurements) during the weaning period.
Average Daily Gain
Calves fed hay increased ADG at a faster rate than Control calves during the preweaning period (mean ± SE, Control: 0.50 ± 0.04 kg/d; Bucket: 0.61 ± 0.04 kg/d; Pipe: 0.60 ± 0.05 kg/d; week by treatment interaction P = 0.020; Supplemental Table S1), with all calves increasing ADG over time (P < 0.001). Across the entire experiment, Bucket calves had higher ADG than Control calves (Control: 0.54 ± 0.03 kg/d; Bucket: 0.68 ± 0.03 kg/d; P = 0.023) but not Pipe calves (0.63 ± 0.04 kg/d; P = 0.600). Pipe calves had a similar ADG to Control (P = 0.167). Similarly, Bucket calves tended to have a higher final BW than Control (Control: 69.1 ± 1.8 kg, Bucket: 77.9 ± 2.6 kg; P = 0.077) but not Pipe calves (75.0 ± 3.5 kg; P = 0.741), whereas there was no evidence that Control and Pipe calves differed (P = 0.290).
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
) and spent more observations ruminating compared with Control calves (P < 0.001). Rumination was consistent across weeks (P = 0.278). All calves spent similar percentages of observations eating grain across weeks (Ptrt = 0.122; Pwk = 0.223; Figure 3), though Bucket and Pipe calves spent significantly more observations eating overall (grain and hay) compared with Control (P < 0.001) due to the additional time spent consuming hay. There was no evidence that observations spent eating overall differed across weeks (P = 0.137).
Figure 3Mean percentage of observations engaged in eating grain, eating hay, eating overall, ruminating, drinking water, and sucking milk across 24 h of calves fed a control diet of grain and milk replacer from birth (Control) or a control diet with additional mountaingrass hay from birth in a bucket (Bucket) or novel polyvinyl chloride (PVC) pipe feeder (Pipe). Data were collected via 1–0 live sampling at 5-s intervals for a continuous 24 h in wk 4 and 6 and are summarized by treatment (trt) and week (wk). Boxplots represent the median (black line within box) and first and third quartiles (25 and 75% of data). Whiskers extend to the lowest and highest values that are not outliers (values that are 1.5× the interquartile range); outliers (o) and means (x) are also presented. Back-transformed model predicted estimates are represented by dashed lines (means) and shaded ribbons (SE). P-values <0.1 are reported for week, treatment, or their interactions.
Bucket calves spent more observations drinking water compared with Control calves (P = 0.031; Figure 3), but there was no evidence that either treatment differed from Pipe calves (P ≥ 0.317). Observations spent drinking water were not related to amount of water consumed (posthoc regression analysis, R2 = −0.01, Supplemental Figure S4, https://doi.org/10.5281/zenodo.6618168;
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
). All calves spent 0.7 ± 0.02% of observations (mean percentage of intervals ± SE) sucking milk in wk 4 (P ≥ 0.759; Figure 3) but by wk 6, Bucket calves tended to spend more observations sucking than Pipe calves (0.8 ± 0.03% vs. 0.7 ± 0.05%, respectively; P = 0.064) and significantly more than Control calves (0.6 ± 0.02%; P = 0.001).
Control calves consistently spent more observations manipulating the pipe feeder (NNOM: pipe) than Pipe calves (P = 0.007; Figure 4) and tended to spend more observations manipulating the pipe compared with Bucket calves (P = 0.078). There was no evidence that Pipe and Bucket calves differed in percentage of observations manipulating the pipe (P = 0.589), and all calves manipulated the pipe less across weeks (P = 0.008). Control calves spent more intervals performing NNOM: other and NNOM: total (pipe and other) than Bucket (P = 0.003, P = 0.002, respectively; Figure 4) and Pipe (P = 0.001, P < 0.001) calves. There was no evidence that Bucket and Pipe calves differed in NNOM: other (P = 0.934) or NNOM: total (P = 0.901). Pipe calves tended to spend fewer observations eating hay than Bucket calves (P = 0.068). All calves tended to spend more observations eating hay across weeks (P = 0.099). There was no evidence that week influenced observations spent performing NNOM: (P = 0.356), or total NNOM: total (P = 0.609).
Figure 4Mean percentage of time engaged in nonnutritive oral manipulation (NNOM): pipe, NNOM: other, NNOM: total, grooming, and tongue flicks across 24 h of calves fed a control diet of grain and milk replacer from birth (Control) or a control diet with additional mountaingrass hay from birth in a bucket (Bucket) or novel polyvinyl chloride (PVC) pipe feeder (Pipe). Data were collected via 1–0 live sampling at 5-s intervals for a continuous 24 h in wk 4 and 6 and are summarized by treatment (trt) and week (wk). Boxplots represent the median (black line within box) and first and third quartiles (25 and 75% of data). Whiskers extend to the lowest and highest values that are not outliers (values that are 1.5× the interquartile range); outliers (o) and means (x) are also presented. Back-transformed model predicted estimates are represented by dashed lines (means) and shaded ribbons (SE). P-values <0.1 are reported for week, treatment, or their interactions.
There was no evidence that treatment affected the percentage of observations performing tongue flicks (Ptrt = 0.184, Pwk = 0.530; Figure 4) and grooming (Ptrt = 0.735, Pwk = 0.815; Figure 4). Individuals varied in overall grooming performance (0.5–5.8% of 24-h observations). Individuals also varied in number of grooming bouts performed in a single day (31–274; average across both weeks = 150 bouts; Figure 5). Bouts ranged from 1 to 299 consecutive 5-s intervals (5 s to 24.9 min; Figure 5), with a median bout length of 2 intervals, or 10 s (mean = 4.3, or 22 s). Outliers were bouts of 12 to 299 consecutive intervals (1–24.9 min), which represented 7.3% of all grooming bouts performed across weeks. Individuals varied in how much they performed these outlier bouts (1.8–18.9% of total bouts in a 24-h day; Figure 5).
Figure 5Counts and durations of grooming bouts (individual dots) in wk 4 and 6 for 4 calves. Grooming bouts were evaluated over 24 h using 1–0 sampling at 5-s intervals; bouts were calculated as observations where grooming occurred consecutively, that is with 3 or fewer 5-s interval gaps of nonrooming behavior in between. Bouts with durations greater than or equal to 60 s were considered outliers, as they fell outside 1.5× the interquartile range below or above the first and third quartiles. The 4 calves chosen illustrate calves that reflect unique challenges with using repetition as a component to abnormal repetitive behaviors (ARBs). Calf 3030 has the longest outlier bout of all 27 calves sampled (1,494 s in wk 4). She spends a similar amount of time grooming overall as Calf 3038 (5.8% vs. 5.7%, respectively, in wk 6), but 3038 performs more grooming bouts, for shorter durations. Calf 3042 spends the least amount of time grooming (0.5%, wk 6), and has only 1 outlier grooming bout in each week. Calf 3043 spends a low percentage of time grooming overall (4.1% in wk 4, 3.8% in wk 6) but has the highest percentage of outlier bouts out of her total grooming bouts (18.9% in wk 4, 17.2% in wk 6).
Tongue rolling was seen in 85% of calves (8 Control, 7 Bucket, 8 Pipe) at least once during wk 4 and 6, but for low percentages of observations (Control: 0.02 ± 0.008%; Bucket: 0.05 ± 0.03%; Pipe: 0.006 ± 0.002%, Supplemental Figure S5, https://doi.org/10.5281/zenodo.6618168;
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
). Calves that tongue rolled performed this behavior for 0.006–0.4% of observations. Similar numbers of calves tongue rolled each week (wk 4: 18 calves; wk 6: 17 calves), but not all calves tongue rolled in each week. Tongue rolling occurred throughout the day (Supplemental Figure S6, https://doi.org/10.5281/zenodo.6618168;
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
). Panting was performed by 8 calves (2 Control, 4 Bucket, 2 Pipe) infrequently across both weeks (Control: 0.003 ± 0.003%; Bucket: 0.2 ± 0.1%; Pipe: 0.006 ± 0.005%). Panting was most common around 1400 h across weeks, as in
Calves fed hay consumed relatively high levels of it (Bucket: 0.086 kg/d; Pipe: 0.064 kg/d) and more grain (Bucket: 0.31 kg/d; Pipe: 0.3 kg/d) than Control calves (0.22 kg/d) throughout the milk-fed period. Providing forage alongside grain is often found to increase overall DMI (
). We also found that calves fed hay had higher ADG than Control during the milk-fed period (0.60–0.61 vs. 0.54 kg/d), similar to other findings (e.g.,
). This could be due, in part, to the low milk provision in our study (3.8–5.6 L/d, or 10% BW). Calves will drink 9 to 13 L milk/d when given the choice (
), and the low allowance in our experiment, likely left calves hungry, requiring increased solid feed intake in an attempt to compensate for low milk satiety. Indeed, the solid feed intake in this study is also higher than our own results using this same mountaingrass hay at a 2-cm chop length when milk was fed at 15% BW (grain: 0.18 kg/d; hay: 0.046 kg/d;
Hay intake increased over time, but Bucket calves consumed more as they aged than Pipe. Issues with maintaining ad libitum hay levels in the pipes were unlikely to play a role, as both Bucket and Pipe calves had a similar number of days (11–12% of calf-days) with orts <10%, suggesting neither treatment was more limited in total quantity of hay than the other. Instead, this could potentially reflect limitations in acquiring hay from the pipe. Hay rack slat sizes have been suggested to play a role in ease of obtaining forage and thus overall feeding time (
). Calves may have preferred feeding rates that they were unable to match given the hole size in the pipe feeder, leading to lower intake levels than when hay was available in a bucket.
Regardless of method of presentation, calves fed hay ruminated for 25% of the 24-h observations as early as 4 wk of age, compared with only 14% in Control calves. Rumination is reported to start in the first few days (e.g., <5 d;
Replacing concentrates with a high-quality hay in the starter feed of dairy calves: II. Effects on the development of chewing and gut fermentation, and selected systemic health variables.
). Some of this behavior may stem from feeding of a coarse, fibrous grain, instead of one that is pelleted or ground, and may prompt earlier rumination (e.g., 3.7 vs. 6 wk of age,
), though little is known about this. High levels may also indicate pseudorumination or sham chewing, which has been suggested by others to explain periods of jaw movement in ruminants when forage is withheld (
). Additional evidence for this may be found in instances where time spent eating is constant across treatments, but calves in intensive farm environments (e.g., individually housed) ruminate for longer than calves in pasture-based, dam-nursed settings (
) as a way of expressing a motivated behavior that is prevented by the current settings. Anecdotal evidence for pseudorumination may also be found in a description of early rumination-like behavior by
, which describes brief bouts of “grinding behavior” in some calves.
Hay reduced oral manipulation of nonfeed items (NNOM: total) regardless of presentation method compared with Control calves. A small percentage of this total NNOM was directed at the pipe (0.2–0.5% of observations), likely because it was a manipulable surface for all calves. Calves fed hay spent less time manipulating the pipe regardless of hay presentation method. This pattern was also true for other NNOM, which was more prevalent than pipe manipulation (5–9% of observations) and thus made up more of the total NNOM (5–10% of observations). Hay provision is well known to reduce NNOM (e.g.,
). Presenting solid feed in a method that promotes natural feeding behaviors has previously been found to further reduce ARBs in some cases (e.g., coarse vs. fine hay,
All calves demonstrated improved feeding and processing skills as they aged, as they consumed more solid feed, whereas observations spent eating overall and ruminating remained stable. This pattern was enhanced in calves fed hay. Faster processing time of solid feed has been documented as animals grow, due to increased experience and practice performing foraging motions (as reviewed in
). For example, veal calves fed limited quantities (250 g) of straw spend 13% of the time around milk meals chewing at 2 to 7 wk of age, but chew this same amount in <6% of time by wk 13 (
Replacing concentrates with a high-quality hay in the starter feed of dairy calves: II. Effects on the development of chewing and gut fermentation, and selected systemic health variables.
), which may suggest improved efficiency with time, or could be additional support for sham rumination in young animals. Bucket and Pipe calves consumed significantly more grain than Control, but all calves spent similar amounts of observations eating grain. This efficiency or feeding skill could stem from multiple facets of forage processing. Forage allows for more practice chewing and ruminating than grain. Processing forage can also stimulate mucosa development (e.g.,
), both of which lead to better rumen development. Together, these effects could help calves consume increasingly large quantities of complex feed.
In addition to changing presentation through provision in a bucket or pipe, we provided a relatively long chop of hay (∼19 cm) which may have affected behavior. Indeed, the magnitude of NNOM (all other) in our study was lower than in our previous work (
), but the percent reduction in NNOM in calves fed hay compared with Control was greater (40 vs. 18%). Most other studies provide short chopped forage (e.g., <5 cm,
Effects of feeding hay and calf starter as a mixture or as separate components to Holstein calves on intake, growth, and blood metabolite and hormone concentrations.
), as longer particles are correlated with more time chewing and ruminating than short ones, which may lead to greater ARB reductions. Using this long chop may have also been more effective at stimulating natural food acquisition behaviors than our pipe or bucket manipulation methods of presentation. We did not evaluate the specific movements used while acquiring hay during this experiment, and thus cannot determine if acquisition differed by treatments, as intended, or if long hay promotes functional tongue grasping motions itself.
Hay Presentation and Sensory-Specific Satiety
Presenting hay in a pipe feeder may have led to sensory-specific satiety. Pipe calves consumed less hay over time and tended to spend less time eating hay than Bucket calves. This may be because the extra effort and natural acquisition movements encouraged by feeding from the pipe led to satiety at a lower level of hay intake than feeding from the bucket, as may be expected if both food acquisition and processing are both part of a behavioral need to forage in cattle. Indeed, humans report altered levels of satiation and satisfaction when food is varied by texture, shape, or consumption method, even when nutrient levels are consistent (as reviewed in
). Animals similarly perform behaviors associated with satiety (e.g., voluntary cessation of feeding, postmeal rest) when fed in ways that encourage more species-typical feeding motions (e.g., fiber for pigs,
). Both Pipe and Bucket calves performed similar levels of NNOM: total, which was less than Control calves (5.2–5.5 vs. 9.7%). This meant hay provision conferred similar benefits to Pipe and Bucket calves, as they had equal reductions in NNOM compared with Control (∼4% fewer observations). However, given the differences in hay intake and eating time, Pipe calves were able to achieve this reduction with less effort (lower hay intake and less time eating) than Bucket calves. This could suggest that the extra engagement in feeding from the pipe feeder, and possible behavioral satiety due to more naturalistic tongue movements, may have reduced NNOM more efficiently.
also tested a PVC pipe feeder and found no overall ARB reductions. However, this design fed ground molasses through small holes and was intended to allow for oral manipulation. This was not comparable to our intended promotion of acquisition motions used in natural foraging.
Normal Behaviors Performed Abnormally?
Hay provision did not influence tongue rolling or tongue flicks. Tongue rolling was performed for short periods (0.006–0.02% of a 24-h day), but by most calves. This common prevalence but low performance of the behavior is similar to other findings (
), and thus it may still be developing during the milk-fed period and better captured by additional observation days. Indeed, veal calves are reported to tongue roll for 3 to 5% of observations when observed at 12 to 14 wk of age (
), suggesting tongue rolling may also relate to time kept in feed-restrictive settings, or may increase with age. It is possible that tongue rolling may develop from tongue flicks, which often share similar motions to tongue rolling, or that they may be different responses to a shared underlying problem. Calves spent 3% of observations performing tongue flicks. This is in contrast to the only other study known to measure tongue flicks, where calves performed this behavior for 13 to 18% of a 24-h day (
). There was no effect of hay on tongue flicks, despite our previous findings, which could stem from the low milk allowance provided in this study. Calves spent <1% of the day sucking milk, compared with 3% of time that they would naturally suckle the dam on pasture (as reviewed by
). This 67% reduction in time spent sucking may have been such a severe limitation for calves that it washed out any possible effects of hay on reducing tongue flicks, particularly because they occur for a small percentage of time. However, tongue flicks could also represent normal behavior, as animals may lick the nose to clear mucus, water, or feed from the nostrils (
). If tongue flicks fall on a spectrum of normal to abnormal behavior, they thus may not be consistently affected by hay provision or enrichment.
Percentage of time spent grooming also was not affected by treatment. Similar to tongue flicks, grooming is a normal behavior that may be performed at abnormal levels or out of context (e.g., displacement behavior) during periods of motivational conflict (e.g.,
). Extreme levels of self-grooming may thus indicate abnormality, reflecting redirected or displacement behavior, or a desire to self-soothe and reduce arousal during motivational conflict (e.g., when social grooming is restricted, as suggested in the review by
“Abnormal” grooming could be expressed in multiple dimensions, such as overall duration of grooming, location of grooming (e.g., repetitive focus on an easily reached body part), and number or duration of individual grooming bouts. Grooming in this study typically lasted for fewer than four 5-s intervals (∼20 s duration) suggesting “functional” grooming, as required to maintain cleanliness, could be reflected by bouts around or below four 5-s intervals. However, grooming duration is known to vary across individuals (e.g., 34–1,080 s, Horvath and Miller-Cushon, 2019), and we saw a similar pattern with grooming bouts (Figure 5). This led us to question what we consider to be “repetitive” in describing ARBs. Repetition within a single bout, leading to a long 24.9 min bout of grooming (Figure 5), may be considered a more concerning expression of behavior than short bouts, and indeed, ARBs are sometimes described by whether a motion in a bout repeats (e.g., “repetition of an identical pattern of movement … 2 or more times,”
). We did not apply formal bout criteria due to our 1–0 sampling methodology, but others using continuous durations have found that self-licking and scratching have a bout criterion of 50 s in milk-fed calves (
). This bout criterion was higher than our decision of using continuous grooming including any pauses of ≤3 intervals (15 s). Using a higher criterion would likely have increased all bout durations, including the extreme outliers, showing similar repetition. Repetition could also be described by a high number of bouts performed throughout a day, even if they are short (Figure 5). In contrast, some animals also performed low levels of grooming, and had very little repetition within bouts (Figure 5). Inactivity has been correlated with negative welfare states in other species (e.g., mink,
), but less is known about this than excessive repetition of behavior. Low grooming could thus reflect a concern related to inactivity, but more work is needed to determine if extreme low performance of behavior is a problem. These findings suggest that “normal” behaviors, such as grooming, may be expressed in a wide variety of potentially concerning and repetitive ways, and that analyzing overall group-level durations may miss important insights.
Defining abnormal behavior based on individual extremes is used in describing polydipsia, or excessive water drinking. Water consumption is an essential normal behavior but can reflect welfare concerns at high levels. For example, frustrated feeding opportunities have been suggested to lead to polydipsia in broiler breeders (
), and is thus higher in calves who are limit-fed milk compared with ad libitum provision. In addition, polydipsia has recently been identified in individual calves. We previously described 1 calf who consistently drank up to 20 L water/d (
), and in the present research found 13 calves that consumed outlier amounts of water (preweaning: ≥7.35 L/d, weaning: ≥16.05 L/d). The increased prevalence of polydipsic calves in the current study likely reflects feeding differences, as the 13 calves were in more restrictive environments than in our previous work (milk fed at 10% BW vs. 15%, <1% observations sucking vs. 3%, respectively). Polydipsia may serve as an additional indicator of compromised welfare in dairy calves kept in restrictive environments. However, time spent drinking water was not corelated to amount of water consumed (Supplemental Figure S4), suggesting polydipsia cannot be identified through behavioral observations alone.
CONCLUSIONS
Provision of long hay, regardless of presentation method, promoted rumination, improved performance, and reduced at least some, but not all, of the considerable abnormal oral behaviors these calves performed. Calves of all treatments performed a wide variety of abnormal behaviors, including possible pseudorumination, polydipsia, repetitive tongue movements and excessive grooming.
ACKNOWLEDGMENTS
We thank University of California Davis Dairy Facility (Davis, CA) manager Doug Gisi, assistant manager Maria Patino, and the dairy interns for animal care and support. We are grateful to those who assisted with data collection: Ana Calderon, Jesica Calderon, Ajanee Evans, McKenna Farnham, Rowan Farrell, Meliza Guox, Megan Harmon, Christopher Lingga, Alicia Marzolf, Isabelle McDonald-Gilmartin, Chelsea Morrow, Erin Nemivant, Theodore Oentoro, Gretchen Peckler, Alexis Roccia, Sabrina Sankus, Joshua Shaw, and Arden Uy, all affiliated with UC Davis at the time of the study, and Julie Gfeller, affiliated with AgroSup Dijon. Special thanks to Mark Rubio and José Villasenor (UC Davis Farm Crew) for helping build the pipe feeders. We also thank Margit Bak Jensen (Aarhus University, Aarhus, Denmark) for valuable feedback on earlier versions of this manuscript. This study was supported by USDA Multistate Research Project NC1029, and a Henry A. Jastro Research Scholarship Award to B.C.D. We gratefully acknowledge the infrastructure support of the Department of Animal Science, College of Agricultural and Environmental Sciences, and the UC Davis California Agricultural Experiment Station. The authors have not stated any conflicts of interest.
REFERENCES
Andrighetto I.
Gottardo F.
Andreoli D.
Cozzi G.
Effect of type of housing on veal calf growth performance, behaviour and meat quality.
Data from: Providing long hay in a novel pipe feeder or a bucket reduces some, but not all abnormal oral behaviors in milk-fed dairy calves. Dryad, Dataset.
Effects of feeding hay and calf starter as a mixture or as separate components to Holstein calves on intake, growth, and blood metabolite and hormone concentrations.
Replacing concentrates with a high-quality hay in the starter feed of dairy calves: II. Effects on the development of chewing and gut fermentation, and selected systemic health variables.
Replacing concentrates with a high-quality hay in the starter feed in dairy calves: I. Effects on nutrient intake, growth performance, and blood metabolic profile.
00012-7&id=gr1.jpg){kind=link}
00012-7&id=gr2.jpg){kind=link}
00012-7&id=gr3.jpg){kind=link}
00012-7&id=gr4.jpg){kind=link}
00012-7&id=gr5.jpg){kind=link}