Scholarly article on topic 'The effect of nursing on the cow–calf bond'

The effect of nursing on the cow–calf bond Academic research paper on "Veterinary science"

CC BY-NC-ND
0
0
Share paper
Academic journal
Applied Animal Behaviour Science
OECD Field of science
Keywords
{"Maternal bond" / "Dam rearing" / "Animal welfare" / Dairy / Growth}

Abstract of research paper on Veterinary science, author of scientific article — Julie Føske Johnsen, Anne Marie de Passille, Cecilie Marie Mejdell, Knut Egil Bøe, Ann Margaret Grøndahl, et al.

Abstract Dairy calves are often separated from the cow soon after birth and prevented from nursing, but little is known about the effects of nursing on the development of the cow–calf bond. This study evaluated the effect of nursing on affiliative behaviours between the dam and her calf including allogrooming, proximity and latency to reunite after a period of separation. Holstein cow–calf pairs were randomly allocated to three treatments differing only in nutritional dependency on the dam: milk feeder (n =10 pairs), combined (n =10 pairs) or nursing (n =10 pairs). Milk feeder calves could feed ad libitum from an automated milk-feeder, combined calves could suckle from their dams at night and could feed ad libitum from an automated milk feeder and nursing calves could suckle from their dams at night. Cows and calves were kept together during the night (between 20:00h and 08:00h) and were housed adjacent to each other during the day (between 08:00h and 20:00h). Direct live observations were performed 2h following the opening of the gate that allowed calves to mix with cows at night. All pairs spent more time (% of observations) allogrooming each other (i.e. own cow/calf) than they did grooming other cows and calves within the same group (10±0.8% vs. 0.4±0.7%, t 29 =168.8, P <0.001). The time cow–calf pairs spent allogrooming did not vary with treatment; 10.0±0.8%, F 2, 27 =0.4, P =0.696). Similarly, time spent in close proximity without nursing did not differ among treatments; 31±2.6%, F 2, 27 =0.6, P =0.543). The percentage of occasions a pair did not reunite (i.e. not observed within 1m of each other within 3min) was 23% for the milk feeder pairs, 38% for the combined pairs and 32% for the nursing pairs (χ 2 =3.9, P =0.415). Nevertheless suckling from another cow than own dam was observed at least once by 19 of the 20 calves that were allowed to suckle. Latency to reunite (among pairs that did so within 180s) was highest for the combined pairs and tended to be lower for milk feeder and nursing calves; 52.5±16.0, 23.3±8.3 and 12.9±5.8; F 2, 24 =3.1, P =0.062). These results indicate that dam and calf form a bond independent of nursing.

Academic research paper on topic "The effect of nursing on the cow–calf bond"

Contents lists available at ScienceDirect

Applied Animal Behaviour Science

ELSEVIER journal homepagewww.elsevier.com/locate/applanim

The effect of nursing on the cow-calf bond

Julie F0ske Johnsen2*, Anne Marie de Passilleb, Cecilie Marie Mejdell Knut Egil B0ec, Ann Margaret Grendahl3, Annabelle Beaverb1, Jeffrey Rushenb, Daniel M. Wearyb

a Norwegian Veterinary Institute, Department of Health Surveillance, P.O. Box 750, 0106 Oslo, Norway b University of British Columbia, Faculty of Land and Food Systems, 2357 Main Mall, Vancouver, BC V6T1Z4, Canada c University of Life Sciences, Institute of Animal and Aqua cultural Sciences, Box 5003, 1432 Äs, Norway

CrossMark

ARTICLE INFO

ABSTRACT

Article history:

Accepted 7 December 2014

Available online 24 December 2014

Keywords: Maternal bond Dam rearing Animal welfare Dairy Growth

Dairy calves are often separated from the cow soon after birth and prevented from nursing, but little is known about the effects of nursing on the development of the cow-calf bond. This study evaluated the effect of nursing on affiliative behaviours between the dam and her calf including allogrooming, proximity and latency to reunite after a period of separation. Holstein cow-calf pairs were randomly allocated to three treatments differing only in nutritional dependency on the dam: milk feeder (n = 10 pairs), combined (n = 10 pairs) or nursing (n = 10 pairs). Milk feeder calves could feed ad libitum from an automated milk-feeder, combined calves could suckle from their dams at night and could feed ad libitum from an automated milk feeder and nursing calves could suckle from their dams at night. Cows and calves were kept together during the night (between 20:00 h and 08:00 h) and were housed adjacent to each other during the day (between 08:00h and 20:00h). Direct live observations were performed 2 h following the opening of the gate that allowed calves to mix with cows at night. All pairs spent more time (% of observations) allogrooming each other (i.e. own cow/calf) than they did grooming other cows and calves within the same group (10± 0.8% vs. 0.4±0.7%, t29 = 168.8, P<0.001). The time cow-calf pairs spent allogrooming did not vary with treatment; 10.0 ± 0.8%, F2,27 = 0.4, P=0.696). Similarly, time spent in close proximity without nursing did not differ among treatments; 31 ± 2.6%, F2,27 =0.6, P =0.543). The percentage of occasions a pair did not reunite (i.e. not observed within 1 m of each other within 3 min) was 23% for the milk feeder pairs, 38% for the combined pairs and 32% for the nursing pairs (x2 =3.9, P=0.415). Nevertheless suckling from another cow than own dam was observed at least once by 19 of the 20 calves that were allowed to suckle. Latency to reunite (among pairs that did so within 180 s) was highest for the combined pairs and tended to be lower for milk feeder and nursing calves; 52.5 ±16.0, 23.3 ± 8.3 and 12.9 ± 5.8; F2,24 = 3.1, P=0.062). These results indicate that dam and calf form a bond independent of nursing.

© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

* Corresponding author. Tel.: +47 23 21 64 47/95 49 61 60; fax: +47 23 21 63 01. E-mail addresses: Julie.johnsen@vetinst.no (J.F. Johnsen), passille@mail.ubc.ca (A.M. de Passille), Cecilie.mejdell@vetinst.no (C.M. Mejdell), Knut.boe@nmbu.no (K.E. Boe), Ann-Margaret.grondahl@vetinst.no (A.M. Grandahl), ab2368@cornell.edu (A. Beaver), rushenj@mail.ubc.ca (J. Rushen), danweary@mail.ubc.ca (D.M. Weary). 1 Present address: Cornell University, College of Agriculture and Life Sciences, 149 Morrison Hall, Ithaca, NY 14853, USA.

http://dx.doi.org/10.1016/j.applanim.2014.12.003

0168-1591/© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/3.0/).

1. Introduction

Under natural conditions, the survival of the new-born calf depends on the establishment of a strong and lasting social bond with the dam (Enriquez et al., 2011), but on dairy farms calves are often separated from the cow within hours of birth. Early separation is often perceived to be unnatural and problematic for the welfare of the cow and calf (Ventura et al., 2013), and research has shown that nursing can provide health and welfare benefits (Krohn, 2001; Flower and Weary, 2003). For example, calves reared with the dam stand earlier after birth (Lidfors, 1996), gain more weight than calves reared artificially with restricted milk allowances (Flower and Weary, 2001) and show reduced signs of distress during an isolation test (Duve et al., 2012). Long-term positive effects of staying longer with the cow can be seen when heifers are introduced into the dairy herd (Wagner et al., 2012). Some farms have adopted alternative management systems that allow for some contact between the calf and cow. For example, in organic dairy production in Sweden, Denmark and Norway, it is mandatory to let the calf nurse at minimum 1-3 d after birth (Vidensenteret for okologisk landbrug, 2012; Debio, 2005; KRAV, 2012).

The social bond between mother and young is described as a preferential mutual, affectionate, emotional attachment that is relatively long lasting and survives temporary separations (Newberry and Swanson, 2001). The criteria for attachment are preference for one individual over another, seeking and maintenance of proximity, as well as showing a differential response to short-or long-term separation from and reunion with that specific individual (Gubernick, 1981). The bond is characterized by affiliative behaviour such as allogrooming, provision of nourishment, warmth and protection, resting in contact, synchronized activity, and maintained proximity. Bonded individuals exhibit reinstatement behaviour when motivated to reunite after a period of separation, and greeting behaviour upon reunion (Newberry and Swanson, 2008). Dams and their calves often synchronize their activities more than unrelated animals in the same group (Veissier et al., 1990). The mother-young bond is typically stronger in beef than in dairy breeds (Le Neindre, 1989), and varies among dairy and dual-purpose breeds (Sandem et al., 2002).

Harlow (1959) found that contact between mother and young, with or without feeding, elicited maternal attachment in rhesus monkeys. In cattle, the relative importance of milk provision versus that of caregiving in the formation of a bond may be reflected in the behavioural response to weaning and separation (Weary et al., 2008). Weaning comprises two elements: the loss of milk provided and loss of care from the dam. The reaction to separation may be caused in part by hunger (Flower and Weary, 2003), as calves newly separated from their dam call less frequently when fed more milk (Thomas et al., 2001). Similarly, by first reducing the calves' dependency on milk by preventing access to the udder the distress response at separation may be reduced (Haley et al., 2005). These results suggest that the nutritional component of the bond is important.

Krohn et al. (1999) tested the effect of maternal presence per se and found that calves housed with the dam

but prevented from suckling had a higher weight gain than calves given the same amount of milk but isolated from the dam. Thus the presence of the dam may have a positive effect on weight gain independent of milk transfer.

To the best of our knowledge, no research has attempted to evaluate the effect of nursing on the cow-calf bond. The aim of the current study was to test this effect. We hypothesized that nursing is essential for bond formation and predicted that if a cow and her calf is allowed to nurse they would reunite faster after a period of separation and show more affiliative behaviour than would a calf allowed contact with the mother but prevented from suckling.

2. Materials and methods

The trial took place at the University of British Columbia's Dairy Education and Research Centre, Agassiz, Canada, from June to October 2012. The university's Animal Care Committee approved the study.

2.1. Animals and experimental design

Holstein cow-calf pairs were assigned to treatment at calving in randomized blocks of three. The three treatment groups were kept as a dynamic group (Fig. 1) as described below. The three treatments differed only in nutritional dependency on the dam: milk feeder (n = 11 pairs), combined (n = 11 pairs) and nursing (n = 11 pairs). Calves were allowed access to the cow pen at night (between 20:00 h and 08:00 h) and cows and calves were kept in adjacent pens during the day (between 08:00 h and 20:00 h) with visual and auditory contact. Milk feeder calves could feed ad libitum from an automated milk-feeder throughout the day and night. Combined calves could suckle from their dam at night and feed ad libitum from an automated milk feeder throughout the day and night. Nursing calves could suckle from their dam at night only; the automated feeder was programmed to provide no milk access to these calves. We predicted that daily milk intakes of nursing calves would be comparable to that of the other treatments based on earlier research showing high milk intakes of calves suckling the dam twice daily for 2 h (de Passillé et al., 2008).

Each treatment used one heifer and 10 bull calves. Throughout the trial, cows were milked twice daily (at approximately 08:00 h and 18:00 h) in a double 12 stall parallel milking parlour. Three first parity cows were included in the milk feeder treatment and four in each of the combined and nursing treatments. The median value of parity for all three treatments was two. Because cow-calf pairs were enrolled over a 2-month period, and cows stayed in the pen for a little more than 6 weeks, a maximum of 24 cows were housed in the group pen at any one time. Both the cow pen and calf creep only housed animals included in the experiment.

2.2. Calving management

Calving took place in a 4 m x 4 m individual indoor maternity pen bedded with sand and covered with straw into which cows were moved within 24 h of calving. Each cow-calf pair spent 27-88 h (mean ± SD of 52 ±12.7 h) in

Fig. 1. The three treatments were kept as one dynamic group in a group pen (9.5 m x 36.9 m). The group pen was in a naturally ventilated wooden frame barn (width=38 m, length = 156 m) with curtained sidewalls, and had 36 sand-bedded free stalls. Individuals from all treatments were present at all times. Cows and calves were kept together at night (between 20:00 hand 08:00 h)and housed adjacent to each other during the day (between 08:00 hand 20:00 h) with visual and auditory contact. During the day, calves were housed in a calf creep (10.0 m x 3.0 m) located adjacent to the cow-pen. Cow-calf separation occurred when calves were 6 weeks of age when the calves were moved to the separation pen (7.5 m x 3.0 m). MF = Milk feeder, CF=concentrate feeder, W=water.

the maternity pen, with no difference between the treatments (F2,32 = 0.0, P =0.99). Established nursing (i.e. calf suckles without assistance) for nursing and combined was a prerequisite for moving a cow-calf pair to the group pen. All calves were bottle-fed 4L of colostrum within 6 h after calving. Thereafter, milk feeder calves were hand fed whole pasteurized milk ad libitum four times daily from nipple bottles during the period that cow-calf pairs spent in the maternity pen. Within the first 12 h of life, the calves were weighed and fitted with an ear-transponder. Milk feeder cows were fitted with an udder net (model Nr. 87355301, DeLaval, Tumba, Sweden) directly after calving; this covered the teats and prevented the calf from suckling. Calves were assisted to stand (milk feeder treatment) or stand and suckle (nursing and combined treatments) if the calves were not seen to do so on their own within 6 h of birth. Three cow-calf pairs, one from each treatment group were excluded from the analysis due to severe mastitis and indigestion during the first week post-partum.

water, and concentrate (barley based, 21.6% crude protein; Unifeed Calf Tex®, Chilliwack, BC, Canada) as well as the cows' Total Mixed Ration (TMR; 22% corn silage, 19% grass silage, 10% alfalfa hay, and 49% concentrated mix). Daily concentrate and milk intakes from the feeders were measured throughout the study. After evening milking, cows were given 2 h to feed and rest before the gate was opened and calves were allowed to enter the cow pen. At gate opening the experimenter whistled to signal that the gate was open. During the 3 d following first entry of a new cow-calf pair to the experiment, the experimenter gently led the calf through the gate at opening time and helped it find the dam. Milk feeder calves had access to their dam but were prevented from suckling by the udder net. Combined and nursing calves could suckle their own dam after entering the cow pen. All calves were able to suckle other cows.

Cows had ad libitum access to water, TMR (same as above) and orchard grass hay (same as above) 24 h/day. At night calves were also able to access the cows' feed.

2.3. Feeding

Milk feeder calves and combined calves were allowed access to 12 L of pasteurized whole milk provided by a CF1000CS-Combi automatic feeder (De Laval, Tumba, Sweden); this feeder also provided free-access to concentrate for all treatment groups. Milk feeder and combined calves were trained by the caretakers to drink from the feeder during the first two days after they were moved to the group pen, and all calves were trained to use the concentrate feeder. Once moved to the group pen, all calves had ad libitum access to orchard grass hay (90.5% DM),

2.4. Health and performance

Calves were weighed at birth and twice weekly using a portable walk-on scale (Smart 1, Western Scale Inc., Port Coquitlam, Canada). Weights were taken three times during each measurement time and the median weight was recorded. Calf health checks were performed twice a week by a veterinarian. The health check included general state of the hair coat and signs of dehydration, provoked coughing and lung auscultation and presence of ocular and nasal discharges as well as rectal temperature. Additionally, faecal consistency was monitored daily and recorded as diarrheic

Table 1

Behaviours recorded for both cow and calf once they were reunited after daytime separation in the partial suckling system.

Behaviour Definition Modifier

Lying Lying down Distance between dam and calf: close; <1 m, far; >1 m or calf creep

Standing or Standing on all 4 feet or

moving moving one or more extremities either in a forward or reverse motion

Allogrooming Initiating or receiving Initiator/receiver of the

licking (tongue touches behaviour: own

cow/calf)/sniffing (nose calf/dam, other

<5 cm from cow/calf air dam/calf)

is pulled through nose)

or rubbing (nose or

other body part

touches any other body

Nursing Cow is standing in nursing position, or calf is standing in nursing position

Other Drinking water, defecating, urinating, eating concentrate (calf) licking itself or being social

Ruminating Moving jaw in a rhythmic manner, and/or regurgitating swallowed material

Eating Taking feed (TMR or Type of food and

hay) into mouth or location in the pen

positioned with head

<5 cm from TMR/hay in

feed alley or calf creep

or normal based on visual appearance (Svensson et al., 2006). Some calves had symptoms of diarrhoea and coughing, but this was rarely observed after 3 week of age. Cow health was recorded as part of the standard procedures at the farm during weekly veterinary health checks. Common cow production diseases were recorded; health status of the test cows was comparable to the rest of the herd. Cow milk weights were collected automatically twice daily.

2.5. Behaviour observation

Each cow-calf pair was observed on nine different days, spread over the period when calves were between 3 and 6 week old. Direct live, focal animal observations were carried out using the Observer software (version 11; Noldus Information Technology, Wageningen, The Netherlands) which was loaded onto laptops. By using the start/stop option in Observer, durations (s) for each behaviour was calculated by the software. Before the study began the seven observers were trained to record the behaviours listed in Table 1 using representative video recordings and live observations. Blinding the observers to treatment was not possible.

Behaviours were recorded during the first 2h after cow-calf pairs were allowed to reunite (hereafter called the

gate opening). Animals were identified by large letters dyed into the hair coat. One observer followed three cow-calf pairs that were comparable in terms of calf age, one from each treatment. During each of the nine observation days, each animal within this triplet was observed six times for 3 min each on a rotatingbasis (6 rotations = 18 min/animal). Thus total observation time per day was 36 min for each cow-calf pair, and a total of 324 min over the nine days of observation. Allogrooming and nursing were recorded for a given animal regardless of it being the initiator or receiver. In addition, we recorded which animal was the reciprocal initiator/receiver of the behaviour.

Animals were habituated to the presence of the observers at the time the gate was opened by their presence (recording behaviour for the sake of a different experiment) 2 h prior to the start of the observations. The sequence, in which the animals were observed, was always cow, cow, cow, calf, calf, calf. The order with which the pairs were observed was randomized at each observation day. In addition, any case of a milk feeder calf suckling another dam was noted whenever it was seen.

2.6. Statistical analyses

The experimental unit was the cow-calf pair. Residuals were visually assessed for normality and Levene's test was used to assess homogeneity of variances. Some variables (allogrooming other dam/calf and other) were square root transformed to normalize variance. One-way ANOVA (SPSS v. 21, IBM) was used to test treatment differences in these behaviours and in average daily weight gain and milk production. Significant treatment differences were investigated using Tukey's post hoc comparisons. Milk intake from the calf feeder was not normally distributed. Differences in intake between the two treatments allowed to access the feeder (i.e. the milk feeder and combined treatments) were analyzed using a Mann-Whitney U test.

For all behaviours except latency to reunite, the duration of dam and calf behaviour was summed for each observation day in order to obtain one result for each cow-calf pair. Thereafter, the behaviours were averaged over the nine observation days. During observation a cow or calf was recorded with one of two mutually exclusive position behaviours; standing or moving and lying. In addition, the activity of the animals was recorded; e.g. rumination or allogrooming independent on whether the animal was lying or standing/moving. Results are presented as percentage of total observed time. Since the position behaviours and activity behaviours were not mutually exclusive, the percentage may add up to >100%. Although milk feeder calves were prevented from suckling their own dam, they were (according to the definition in the ethogram) recorded as suckling when they stood in suckling position. To evaluate if allogrooming occurred preferentially within the cow and calf pair, the total time spent grooming the subject's own cow/calf was subtracted from the time spent grooming unrelated cows/calves; this difference was compared with null expectation of 0 (no preference) using a one sample t-test. The same procedure was repeated specifically for the milk feeder only pairs and for combined and nursing

pairs with respect to nursing own cow/calf vs. nursing other cow/calf.

Latency to reunite was calculated as the time from gate opening until cow and calf were in close proximity (<1 m apart). Cow-calf pairs not reunited by >180 s were defined as non-reunited. Associations between treatment and reunion/non reunion were explored with a Chi-square test. Latency to reunite was averaged for each cow-calf pair and square root transformed to normalize variance. Levene's test was used to assess homogeneity of variances and treatment differences were investigated with one-way ANOVA. One cow-calf pair was removed from this analysis because of a missing value. Additionally, two (combined) pairs were not observed to reunite at any of the observational sessions, so final analysis of mean latency to reunite included data from 27 cow-calf pairs.

Descriptive results are presented as means ±SD; comparisons are presented as means ±SEM. Effects were considered as significant when P< 0.05 and as a tendency when P <0.10.

3. Results

3.1. Cow and calf performance

Calf weight gain during the experiment (calf age 0-6 weeks) averaged (±SD) 0.9 ±0.22 kg/day and did not vary among treatments (F2 29 = 0.6, P =0.524). Calf daily milk intake from the milk feeder averaged 8.2 L/day for calves that only had access to the feeder (Fig. 2a), and the minimum average daily intake for calves in this treatment was 5.2 L/day. In contrast, 6 of 10 calves in the combined treatment (that were also able to suckle from the dam) often consumed less than 1.5 L/day from the milk feeder, and intake for the four remaining calves averaged just 2.3 L/day (U =7188, P<0.001, r =0.8). As shown in Fig. 2b, the daily milk intake of the combined calves also varied from day to day.

As expected, cows that were able to nurse their calves provided less milk in the milking parlour (F2 29 =8.2, P = 0.002): yield for the milk feeder cows (i.e. those unable to nurse their calves) averaged 40.5 ± 3.63 L/day, versus just 25.3 ± 2.70 L/day and 26.3 ± 2.41 L/day for cows in the combined and nursing treatments.

3.2. Affiliative behaviours

Cow-calf pairs in all three treatments were observed allogrooming own cow/calf (i.e. licking, sniffing and rubbing own calf, own dam or mutual). Time engaged in this behaviour averaged approximately 10% and did not differ among treatments (Table 2). All pairs spent more time allogrooming each other (i.e. own cow/calf) than they did grooming other cows and calves within the same group (10 ± 0.8% vs. 0.4 ± 0.7%, t29 = 168.8, P < 0.001). Specifically, the milk feeder pairs also spent more time allogroom-ing each other than unrelated individuals (11 ±2.0% vs. 0.5 ±0.4%, t9 = 5.2, P<0.001).

During our observations cows and calves spent approximately 30% of the available time in close proximity without suckling. Calves in the nursing treatment tended to spend

Fig. 2. Daily milk intakes (L) during the nursing period (0-6 weeks) presented as median with 25th and 75th percentiles. Values>3.0, and >1.5 interquartile ranges from the nearest edge of the box are indicated by "x" and " respectively. Whiskers show minimum and maximum values. The letters indicate the two treatments: (a) milk feeder (n = 10) and (b) combined (n =10). Calves in both treatments had access to 12 L/day of fresh, warm, pasteurized whole cow's milk from a milk feeder positioned in the calf creep. Combined calves could also suckle their dams at night (between 20:00 h and 08:00 h).Calves in the third treatment (nursing) were not allowed access to the milk feeder and are not shown.

more time suckling from their own dam as did the combined calves (Table 2, P =0.062). Milk feeder calves were not able to suckle, but were occasionally observed standing in a suckling position.

Of the cow-calf pairs able to nurse (combined and nursing), 19 of 20 pairs (95%) were observed on at least one occasion to nurse other dam/calf. In contrast, this behaviour was only observed in 2 of 10 milk feeder pairs. However, there was no effect of treatment, and combined and nursing pairs spent much more time nursing own cow/calf than nursing other cow/calf (10 ± 0.8 vs. 2 ± 0.4, t19 = 10.6, P <0.001).

3.3. Latency to reunite

Five of the milk feeder pairs, three combined pairs and two nursing pairs were observed to reunite (i.e. cow and calf observed in close proximity; <1 m apart) within 180 s during all observation sessions. Two combined pairs were

Table 2

Affiliative behaviours recorded during the first 2 h afterthe cow-calf pairs were reunited in a partial suckling system: five response measures are shown: (mean ±SEM) % (of total time available) the pair spent within 1 m without nursing, % the pair was allogrooming, % the pair was nursing or suckling, % time nursing or suckling another calf or cow, % time allogrooming with another calf or cow. Responses are shown separately for the three treatments: milk feeder (n = 10), combined (n = 10), and nursing (n =10).

Treatment

Milk feeder

Combined

Nursing

P-value

Behaviour within a cow-calf pair

Time spent close without nursing, % 33.3 ± 6.14

Allogrooming OWN dam/calf, % 10.8 ± 1.96

Nursing OWN dam/calf, % 0.3 ± 0.22

Behaviour directed towards alien cow/calf

Nursing OTHER dam/calf, % 0.5 ± 0.47

Allogrooming OTHER dam/calf, % 0.5 ± 0.12

26.8 ± 2.65 9.1 ± 1.03 8.9 ± 1.05

2.0 ± 0.56 0.3 ± 0.08

32.6 ± 4.14 10.2 ± 0.99 11.8 ± 1.02

2.0 ± 0.67 0.4 ± 0.17

0.6 0.4 48.4

2.1 1.0

0.543 0.696 0.000

0.141 0.381

never observed to reunite within 180 s. The number of occasions the cow-calf pairs did not reunite is shown in Fig. 3a. Accordingly, the cow-calf pairs did not reunite in 23%, 38% and 32% of the occasions for milk feeder, combined and nursing pairs respectively (x2 =3.9, P = 0.415).

Latency to reunite (among pairs that did so within 180 s) was highest for the combined pairs and tended to be lower for milk feeder and nursing calves; Fig. 3b; F2 24 =3.1, P= 0.062).

3.4. Other behaviours

There were no significant differences in general activity between the three treatments. For example cow-calf pairs spent on average 36 ±1.7% of their time lying (F2,27 = 1.3, P =0.302); 11 ±1.0% eating (F2,27 = 1.5, P =0.237), and 19 ±1.1% ruminating (F2,27 = 1.1, P =0.353).

4. Discussion

We hypothesized that nursing was essential for the formation of the cow-calf bond, and predicted that in comparison to cow-calf pairs allowed to nurse, the milk feeder pairs would spend less time close to one another, less time allogrooming and have a longer latency to reunite after a period of separation. Our results provide little support for these predictions. Time in proximity, allogrooming, reuniting after a period of separation and preference for the related cow/calf over an unrelated pen mates all indicate that a bond formed (Gubernick, 1981; Newberry and Swanson, 2008), even in the absence of nursing.

This study is the first to disentangle the effects of affil-iative behaviours and nursing with respect to the maternal bond in dairy cattle. All cows accepted the calf, and the cow and calf usually reunited rapidly when allowed to do so. We found no differences among treatments in the amount of time dam and calf spent close to one another. Young dairy calves have earlier been shown to choose the dam as the closest neighbour; this affiliative behaviour is thought to indicate attachment (Froberg and Lidfors, 2009), perhaps especially in the absence of milk transfer. Affiliative behaviour is characterized by maintaining proximity, providing food, protection or allogrooming between specific individuals (Boissy et al., 2007). Maintaining close proximity might provide opportunities for social transmission of information from a mother to her young such as

Milk feeder Combined Nursing Treatment

1501? и 'с

Ü 100-

<1> ra _J

o-1-1-1-1-

Milk feeder Combined Nursing Treatment

Fig. 3. The figure shows the number of instances cow-calf pairs within each treatment reunited within 180 sand the latency to reunite. The three treatments differed in the calves' nutritional dependency on the dam: Milk feeder calves could feed ad libitum from an automated milk-feeder, combined calves could suckle from their dams at night and could feed ad libitum from an automated milk feeder and nursing calves could suckle from their dams at night. (a) Observational sessions (milk feeder; n = 28, combined; n = 29and nursing; n = 31)inwhichacow-calfpairdidordid not reunite (i.e. approach within 1 m and 180 s after daytime separation). (b) Mean latency (±SEM)(s) to reunite for cow-calf pairs (milk feeder (n =10), combined (n = 8) and, nursing (n = 9) that did reunite within 180s after a gate was opened after separation during the day.

information about food sources and predators (Thorhallsdottir et al., 1990).

Allogrooming was recorded for all pairs. Allogrooming was performed more within family pairs than unrelated cows and calves indicating that the bond is specific to the dam and her calf. Cow-calf pairs prevented from nursing also showed preference for one another as an allogrooming partner. Nursing and licking often occur together; one study reported that 56% of licking bouts were associated with nursing (Keeling, 2001). Licking seems to be important in the relationship between a cow and her calf (Vitale et al., 1986), and is considered essential in establishing the maternal filial bond (von Keyserlingk and Weary, 2007). Allogrooming is also thought to play a key role in reinforcing social bonds and is associated with positive emotions (Boissy et al., 2007).

An existing bond between the cow and calf did not eliminate nursing by other calves; indeed, the majority of combined and nursing cows were seen to nurse other calves. Indeed, 95% of pairs were recorded suckling/nursing a cow/calf other than own at least once. This percentage is somewhat higher than that reported by Froberg and Lidfors (2009) and by Spinka (1992). This behaviour was infrequent for milk feeder pairs, perhaps because these calves had less experience with suckling behaviour.

The high percentage of combined and nursing calves observed to at least once suckle a cow other than their own dam may been the result of these calves being highly motived to access milk when first reunited. Indeed, these calves seemed to approach and suckle the first available cow, before moving on to find their own dam. However, the pairs allowed to nurse spent more time nursing own cow/calf than nursing a dam/calf other than own. Other work has found that young calves commonly attempt to suckle from other cows, but strongly prefer to suckle from their own mother (Spinka and Illmann, 1992).

Cows and calves normally reunited soon after the gate opening, indicating both recognition and motivation to reunite. We had predicted that the milk feeder cow-calf pairs would reunite later after a period of separation. However, milk feeder pairs did on most occasions reunite, and many pairs reunited at all occasions. In each of the treatments there were some cow-calf pairs that did not reunite within the first 3 min after the gate was opened at night. Combined pairs tended to reunite later than nursing pairs. That most combined and nursing calves at some occasions suckled a cow other than own dam can explain why the pairs reunited later as discussed above. Interestingly, combined calves tended to spend less time suckling from their own dam (compared to nursing calves) perhaps explained by milk intake from the milk feeder. However, spending less time suckling from own dam and an increased latency to reunite may both demonstrate a high flexibility of the nutritional component of the bond. Two combined pairs were never recorded to reunite within 180 s, but still spent substantial amounts of time allogrooming and being close without nursing.

Calf weight gains in the current study were comparable to those reported in previous work on the same farm for calves fed large quantities of milk (Sweeney et al., 2010). A study on suckling calves with free access to the dams has

shown even higher gains; in excess of 1 kg/day (Grondahl et al., 2007). Nursing calves had access to milk only at night and the cows were milked twice daily, perhaps limiting milk availability. Milk transfer from cows to calves was not measured, but evidence from the weight gains suggests that milk intakes of nursing calves were similar to that of the other treatments. Treatment differences in daily milk yield in the parlour likely reflect the high milk intake by calves allowed to suckle. Even in a restricted suckling systems allowing calves to suckle twice daily for 2 h, calves consumed 12.5 kg by 9 week of age (de Passille et al., 2008); this result suggests that calves in a partial suckling system drink as much milk as those fed ad libitum from a teat (Appleby et al., 2001).

The combined pairs were allowed access to milk from both the cow and the milk feeder, but 6 of 10 calves drank very little (<1.5 L/day) from the automatic milk feeder. The reason for the daily variation in milk intakes of these calves is unknown. However, similar findings have been reported for calves fed ad libitum from an artificial teat (Appleby et al., 2001), and may reflect difficulties in regulating feed intake over longer periods. Integrating a milk feeder into a partial suckling system could meet individual preferences both in milk intake and patterning across the day. Compared calves in the milk feeder treatment, combined calves drank less milk from the feeder; the weight gains of these calves indicate that they made up for the shortfall in intake from the milk feeder by suckling the dam. As combined calves were trained to use the milk feeder, this result suggests that some calves prefer nursing to drinking from an automatic feeder, and that no calves have the opposite preference.

5. Conclusion

These results indicate that bonding occurs between calf and cow even in the absence of nursing. Preference for own cow/calf as an allogrooming partner, time spent allogrooming, proximity between cow and calf and reunion after a period of separation, all indicate that a maternal bond developed and persisted regardless of how much calves were nursed. These results can inform the development of housing and management systems that allow continued cow-calf contact even in the absence of nursing.

Acknowledgements

We thank the project funders: Foundation for Research Levy on Agricultural Products (FFL) and the Agricultural Agreement Research Fund (JA) (Norwegian Research Council Project Number 190424), and the NSERC Discovery Grants program.

We thank the staff and students at University of British Columbia's Dairy Education and Research Centre and the University's Animal Welfare Program. We are especially grateful to Louise Buxant, Zoe Cocker, andJohan Rojham-mar for their work during the trial and to Gosia Zdanovich, John Luu, Rebecca Wright, Joao Henrique Cardoso Costa and Dhavan Vora for technical assistance.

References

Appleby, M.C., Weary, D.M., Chua, B., 2001. Performance and feeding behaviour of calves on ad libitum milk from artificial teats. Appl. Anim. Behav.Sci. 74,191-201.

Boissy, A., Manteuffel, G., Jensen, M.B., Moe, R.O., Spruijt, B., Keeling, L.J., Winckler, C., Forkman, B., Dimitrov, I., Langbein, J., Bakken, M., Veissier, I., Aubert, A., 2007. Assessment of positive emotions in animals to improve their welfare. Phys. Behav. 92,375-397.

de Passille, A.M., Marnet, P.G., Lapierre, H., Rushen, J., 2008. Effects of twice-daily nursing on milk ejection and milk yield during nursing and milking in dairy cows. J. Dairy Sci. 91,1416-1422.

de Passille, A.M., Marnet, P.G., Lapierre, H., Rushen, J., 2008. Effects of twice-daily nursing on milk ejection and milk yield during nursing and milking in dairy cows. J. Dairy Sci. 91,1416-1422.

Debio, N., 2005. 2005-10-04: Forskrift om okologisk produksjon og merking av okologiske landbruksprodukter og nsringsmidler. www.lovdata.no

Duve, L.R., Weary, D.M., Halekoh, U., Jensen, M.B., 2012. The effects of social contact and milk allowance on responses to handling, play, and social behavior in young dairy calves. J. Dairy Sci. 95, 6571-6581.

Enriquez, D., Hotzel, M.J., Ungerfeld, R., 2011. Minimising the stress of weaning of beef calves: a review. Acta Vet. Scand., 53.

Flower, F.C., Weary, D.M., 2001. Effects of early separation on the dairy cow and calf: 2. Separation at 1 day and 2 weeks after birth. Appl. Anim. Behav. Sci. 70, 275-284.

Flower, F.C., Weary, D.M., 2003. The effects ofearly separation on the dairy cow and calf. Anim. Welf. 12,339-348.

Froberg, S., Lidfors, L., 2009. Behaviour of dairy calves suckling the dam in a barn with automatic milking or being fed milk substitute from an automatic feeder in a group pen. Appl. Anim. Behav. Sci. 117, 150-158.

Grondahl, A.M., Skancke, E.M., Mejdell, C.M., Jansen, J.H., 2007. Growth rate, health and welfare in a dairy herd with natural suckling until 6-8 weeks of age: a case report. Acta Vet. Scand., 49.

Gubernick, D.J., 1981. Parental care in mammals. In: Gubernick, D.J., Klopfer, P.H. (Eds.), Parent and Infant Attachment in Mammals. Plenum Press, New York, pp. 246-305.

Haley, D.B., Bailey, D.W., Stookey, J.M., 2005. The effects of weaning beef calves in two stages on their behavior and growth rate. J. Anim. Sci. 83, 2205-2214.

Harlow, H., 1959. Love in infant monkeys. Sci. Am. 200,68-74.

Keeling, L.J., Gonyou, H.W. (Eds.), 2001. Social Behaviour in Farm Animals. CABI Publishing, Wallingford.

KRAV, S., 2012. In: KRAV, S. (Ed.), Regler för KRAV-certifierad produktion utgäva 2012., www.krav.se

Krohn, C.C., 2001. Effects of different suckling systems on milk production, udder health, reproduction, calfgrowth and some behavioural aspects in high producing dairy cows - a review. Appl. Anim. Behav. Sci. 72, 271-280.

Krohn, C.C., Foldager,J., Mogensen, L., 1999. Long-term effect of colostrum feeding methods on behaviour in female dairy calves. Acta Agric. Scand. A: Anim. Sci. 49, 57-64.

Le Neindre, P., 1989. Influence of cattle rearing conditions and breed on social relationships of mother and young. Appl. Anim. Behav. Sci. 23, 117-127.

Lidfors, L.M., 1996. Behavioural effects of separating the dairy calf immediately or 4 days post-partum. Appl. Anim. Behav. Sci. 49, 269-283.

Newberry, R.C., Swanson, J.C., 2001. Breaking social bonds. In: Keeling, L.J., Gonyou, H.W. (Eds.), Social Behaviour in Farm Animals. CABI Publishing, Wallingford, UK, pp. 307-331.

Newberry, R.C., Swanson, J.C., 2008. Implications of breaking mother-young social bonds. Appl. Anim. Behav. Sci 110,3-23.

Sandem, A.I., Braastad, B.O., Boe, KE., 2002. Eye white may indicate emotional state on a frustration-contentedness axis in dairy cows. Appl. Anim. Behav. Sci. 79,1-10.

Spinka, M., 1992. Intersucking in dairy heifers during the 1st 2 years of life. Behav. Process. 28,41-50.

Spinka, M., Illmann, G., 1992. Suckling behavior ofyoung dairy calves with their own and alien mothers. Appl. Anim. Behav. Sci. 33,165-173.

Svensson, C., Hultgren,J., Oltenacu, P.A., 2006. Morbidity in3-7-month-old dairy calves in south-western Sweden, and risk factors for diarrhoea and respiratory disease. Prev. Vet. Med. 74,162-179.

Sweeney, B.C., Rushen, J., Weary, D.M., de Passille, A.M., 2010. Duration of weaning, starter intake, and weight gain of dairy calves fed large amounts of milk. J. Dairy Sci. 93,148-152.

Thomas, T.J., Weary, D.M., Appleby, M.C., 2001. Newborn and 5-week-old calves vocalize in response to milk deprivation. Appl. Anim. Behav. Sci. 74,165-173.

Thorhallsdottir, A.G., Provenza, F.D., Balph, D.F., 1990. The role of the mother in the intake of harmful foods by lambs. Appl. Anim. Behav. Sci. 25, 35-44.

Veissier, I., Lamy, D., Leneindre, P., 1990. Social behaviour in domestic beef-cattle when yearling calves are left with the cows for the next calving. Appl. Anim. Behav. Sci. 27,193-200.

Ventura, B.A., von Keyserlingk, M.A.G., Schuppli, C.A., Weary, D.M., 2013. Views on contentious practices in dairy farming: the case of early cow-calf separation. J. Dairy Sci. 96, 6105-6116.

Vidensenteret for okologisk landbrug, Oversikt over regler for okologisk landbruk, Danmark, 2012,www.okolandmand.nu

Vitale, A.F., Tenucci, M., Papini, M., Lovari, S., 1986. Social behaviour of the calves of semi-wild Maremma cattle, Bos primegenius taurus. Appl. Anim. Behav. Sci. 16, 217-231.

von Keyserlingk, M.A.G., Weary, D.M., 2007. Maternal behavior in cattle. Horm. Behav. 52,106-113.

Wagner, K., Barth, K., Palme, R., Futschik, A., Waiblinger, S., 2012. Integration into the dairy cow herd: long-term effects of mother contact during the first twelve weeks of life. Appl. Anim. Behav. Sci. 141, 117-129.

Weary, D.M., Jasper, J., Hotzel, M.J., 2008. Understanding weaning distress. Appl. Anim. Behav. Sci. 110, 24-41.