Maternal behavior of dairy cows and suckling behavior of dairy calves in different cow-calf-contact conditions

Cow-calf contact (CCC) has been suggested as an alternative to the common practice of early separation of dairy cow and calf. Increasing CCC poses challenges by reducing the machine milk yield and increasing separation stress. Part-time CCC may prove more feasible than full-time contact for farmers but may affect the maternal behavior of the cows. Gradually reducing the daily contact could gradually wean the calf off milk, preparing it for separation. This study aims to investigate 1) how cows’ maternal behaviors toward their own calf are affected by part-time contact, and 2) whether calves’ suckling behavior is reduced by reducing daily cow contact. Fifty-six dairy cows and their calves were assigned to either full-time (23 h/d) or part-time (10 h/d) cow-calf contact for 7 weeks (main rearing period). This was followed by 2 weeks of either reduced contact (50% in wk 8, then 25% of original contact time in wk 9) or unchanged contact. First, the maternal behavior of full and part-time cows was observed for 24 h in the third week of the main rearing period. Part-time cows spent less time nursing and grooming their own calf than full-time cows did. However, the amount of nursing in the inverse parallel position and the probability of a cow nursing a calf other than her own did not differ between the 2 treatments. Second, calf suckling behavior was observed for 24 h in both weeks of the reduced contact period. Averaged across both observations, total suckling time (i.e., time spent suckling any cow) was lower in part-time calves that experienced reduced contact compared with part-time calves with unchanged contact. In wk 9, time spent suckling only the dam was lower for full-time calves with reduced contact compared with part-time calves with unchanged contact. In conclusion, although we found a lower nursing and grooming time in part-time cows compared with full-time, we found no difference between treatments with regards to nursing in the inverse parallel position and nursing a calf other than the cow’s own. The effect of gradually reducing cow-calf contact on calves’ suckling time requires more research


INTRODUCTION
In semi-natural conditions, the dam (i.e., the mother cow) is the calf's main social contact during the first weeks of life.Initially, the dam initiates nursing bouts (Lidfors et al., 1994; nursing defined as a cow allowing a calf to suckle her udder, Sirovnik et al., 2020).As the calf grows older, dam and calf increase the time spent away from each other, and nursing bouts become fewer but longer (Vitale et al., 1986;Kerr and Wood-Gush, 1987;Kour et al., 2021a, but see Lidfors and Jensen, 1988).Additionally, calves increase their solid feed intake, possibly facilitated through learning from their mother or other older animals (Whalin et al., 2021).During the day, calves typically stay together in crèches, while the cows graze nearby (Vitale et al., 1986) , and cow and calf could therefore be naturally adapted to temporary separations from each other.
The calf is considered weaned when it no longer depends on its dam's milk, and when its suckling attempts are rejected by the dam (reviewed by Weary et al., 2008; suckling defined as the calf obtaining milk from a cow's udder, Sirovnik et al., 2020).The natural weaning age is unclear but estimated to be between 6 and 11 mo (Flower and Weary, 2001;Johnsen et al., 2015).However, in conventional dairy production systems, cow and calf are separated shortly after calving to ease the management of both (Flower and Weary, 2003).Furthermore, calves are weaned off milk at between 7 to 12 weeks of age (Vasseur et al., 2010;Klein-Jöbstl et al., 2015).Similar to natural conditions, artificially reared calves can be encouraged to eat solid

Maternal behavior of dairy cows and suckling behavior of dairy calves in different cow-calf-contact conditions
feed by gradually reducing milk availability (Khan et al., 2007a;b;Jasper et al., 2008).
An alternative to conventional dairy calf management is cow-calf contact systems.Cow-calf contact refers to systems where the calf is reared for weeks to months after calving by either its dam or an unrelated cow (Sirovnik et al., 2020).The amount of contact between cow and calf may vary (Barth, 2020;Bertelsen and Jensen, 2023a).For some farmers, part-time contact (Sirovnik et al., 2020) may be more feasible than full-time contact, due to for instance increased machine milk yield (Bertelsen and Vaarst, 2023).Furthermore, part-time contact may also reflect the temporary separations that older calves could experience in natural conditions (Vitale et al., 1986).Part-time contact could better prepare cows and calves for weaning and improve handleability of calves (Johnsen et al., 2016).On the other hand, the repeated separations could be perceived as stressful by cow and calves and thus raise the question of whether part-time contact is a viable solution regarding the animals' welfare.Little is known of dairy cows' behaviors and possible benefits in cow-calf contact systems, as most research has been on calves (reviewed in Beaver et al., 2019 andMeagher et al., 2019).Most studies on dairy cow maternal behavior focus on the first few days after calving (Edwards and Broom, 1982;Illmann and Špinka, 1993, but see Jensen, 2011; Johnsen et al., 2021) revealing a research gap on this specific topic.
Recent studies (Bertelsen and Jensen, 2023b;Neave et al., 2023a) did not find part-time contact to reduce calves' response to weaning and separation from their dam compared with full-time contact.Because the reaction of both cow and calf to separation after prolonged contact is a welfare concern (Johnsen et al., 2015), research on how to best mitigate the stress of separation is called for.Some studies have found that the calf's reaction to separation may be reduced by de-coupling the loss of milk and the loss of the dam.For instance, stress reactions are reduced by fence-line separation, which allows cow and calf some contact but prevents nursing (beef cattle: Price et al., 2003;dairy cattle: Johnsen et al., 2015;Bertelsen and Jensen, 2023b).However, such methods entail an abrupt loss of milk for the calf, and therefore methods to gradually reduce milk intake of calves in cow-calf systems are needed but not yet fully developed (Sirovnik et al., 2020, but see Sørby et al., 2024a;b).
In this paper, we aim to investigate first, how maternal behavior in dairy cattle is affected by part-time contact early in the calf's life, and second, whether gradually reducing the daily cow-calf contact time can reduce calves' suckling behavior.
Previous studies have found that cows most often nurse their own calf in inverse parallel position (IPP, Figure 1), while nursing of unrelated calves most often occur in other positions (Waltl et al., 1995;Fröberg and Lidfors, 2009).IPP enables the cow to sniff or lick the calf's ano-genital area, allowing for the cow to identify whether the calf is her own or an unfamiliar calf (Waltl et al., 1995).Furthermore, nursing in IPP has been used as an indication of cows adopting, rather than just tolerating, foster calves (Le Neindre, 1982).Since a part-time contact schedule implies less time together and repeated separations, we hypothesized that parttime cows would spend less time nursing their own calf, reflecting a reduction in maternal behavior.Additionally, we expected part-time cows to spend less time grooming their calf, to be less "selective" when nursing [i.e., allowing calves other than their own (alien calves) to suckle], and to nurse their calf less in the inverse parallel position, compared with full-time cows.
In an attempt to reflect a "natural" weaning situation, we gradually reduced the time cow and calf spent together before the 2 were completely separated.We hypothesized that the reduction in cow-contact, and thus reduced time available for suckling, would reduce suckling time, both in total (i.e., time spent suckling any cow) and time spent suckling only the dam.In addition, and similar to the effect of part-time contact, the calves may try to compensate for reduced time available for suckling by suckling alien cows (i.e., cows other than the dam, Víchová and Bartoš, 2005).We therefore additionally expected that the reduction in cow-contact would increase probability of calves suckling alien cows, thus reducing their selectivity.

MATERIALS AND METHODS
The study was conducted at the Danish Cattle Research Centre, Aarhus University (Tjele, Denmark), from September 2021 to August 2022.All animal procedures were approved by the Danish Animal Experiments Inspectorate in accordance with the Danish Ministry of Environment and Food Act No. 474 (May 15, 2014).

Experimental design
Fifty-six Danish Holstein-Friesian dairy cows and their calves were enrolled in 7 blocks of 8 cow-calf pairs each.Cows calved in an individual maternity pen, where cow and calf remained for the first approximately 48 h postpartum (range 42-66 h).Pairs were only enrolled in the experiment if both cow and calf were healthy, the calving was unassisted and of a single calf, and if the calf was observed to suckle the cow without assistance within 48 h of birth.Following the first 48 h, 2 pairs at a time were assigned to one of 2 cow-calf contact treatments: part-time contact treatment (10 h calf contact per day; cow and calf were together during the day between morning and afternoon milking but separated during the night) or full-time contact treatment (23 h calf contact per day; cow and calf together during day and night and only separated when the cow was milked in the milking parlor).The full-time contact treatment functioned as a control for part-time contact in the current study.Cows were allocated treatments based on their calving date, and order of allocation to treatment was balanced between blocks.Sample sizes were based on cow and calf availability and were supported by post hoc power calculations to detect significant differences in nursing and suckling behavior (5% significance level, power level > 90%).
The distribution of calf sex and cow parity was not fully counter-balanced across cow-calf contact treatments, but the final distributions were similar between the treatments (17 heifer-and 11 bull calves on fulltime contact; 14 heifer-and 14 bull calves on part-time contact; and 11 primiparous and 17 multiparous cows on full-time contact; 12 primiparous and 16 multiparous cows on part-time contact).A third treatment, where dam and calf were separated 48 h after birth, was also enrolled in the study, but not included in the present study and is therefore not described any further.
Each block consisted of a 2-week enrolment period, followed by an experimental period of 10 weeks.At wk 8 of the experimental period, cow-calf pairs were randomly assigned to one of 2 additional treatments with either gradually reduced contact (RC) or unchanged contact (UC).Four different contact treatment groups were thus created for the final 2 weeks of the experiment.The treatment groups with unchanged contact functioned as controls for the treatments with reduced contact.Cows and calves were completely separated by the tenth experimental week; this part of the experiment is beyond the scope of the current study.

Cow-calf contact treatments during the main rearing period
During the first 7 weeks of the experimental period, the animals were housed in groups of 4 cow-calf pairs in pens deep-bedded with straw (see Figure 2 for measurements).Each pen was fitted with 2 rotating brushes ('Dan-Swing', Dan Egtved, Egtved, Denmark), 2 selffilling water bowls (type 130P, Dan Egtved, Egtved, Denmark), and 2 purpose-made metal feed troughs.The feed troughs provided ad libitum access to TMR (approximately 50:50 concentrate to roughage ratio) for both cows and calves.Additionally, calves had access to 2 calf creeps, one big and one small.Both creeps provided ad libitum access to concentrate from a bowl and hay from a rack, with the larger creep also containing a self-filling water bowl.Solid feeds were replenished twice daily.Calves received no supplementary milk; all milk was obtained by suckling any cow present in the home pen.Straw bedding was added once daily, and the bedding was completely changed out every 4 weeks.Twice daily, at approximately 0500 and 1530, cows were away from the pen for about 30 min to be milked in a milking parlor.
Between afternoon and morning milking, part-time cows were housed in a free-stall pen (6.5 × 18.64 m) in a separate barn, while their calves remained in the deep-bedded pens.This prevented any visual, auditory, and olfactory contact between the part-time cows and their calves during this separation period.The free-stall pen contained 14 lying stalls, which were fitted with mattresses and topped daily with sawdust.Cows received the same TMR ration delivered at a feed bunk.Water was provided in 2 self-filling troughs.

Cow-calf contact treatments during the reduced contact period
Two days before the reduced contact period was initiated at wk 8, the deep-bedded pens were split into 2 pens of equal size (Figure 2).The new pens each pro- Left: the calf is suckling, while its body is completely parallel to the cow's.Right: The calf is suckling while the angle formed between the midline of the cow and calf's bodies is 45 degrees.In both ends of the range, the cow is able to reach the hind quarters of the calf and sniff it, and she is thus able to recognize whether it is her own or another calf.All nursing taking place outside of this range was categorized as not in IPP.Nursing position was not registered when cows nursed alien calves.
vided access to one rotating brush, one self-filling water bowl, and one feed trough.One new pen contained the small calf creep and the other the larger creep.Randomly, the 4 cow-calf pairs in each original pen were divided into 2 groups, with 2 pairs housed in each of the new pens.The pairs housed in the new pens with small creeps experienced no change in cow-calf contact schedule (unchanged contact, UC), while the pairs housed in the new pens with large creeps experienced a gradual reduction in cow-calf contact time (reduced contact, RC).Reduced contact was achieved by confining the calves inside the creep.Cow and calf thus still had visual, auditory, and limited physical contact, but suckling was prevented while calves were confined in the calf creep.During the first week of reduced contact (wk 8), the calves were confined for approximately 50% of their original cow-contact period (corresponding to 9.5 h contact for full-time RC pairs and 4.5 h contact for part-time RC pairs).Confinement was increased to approximately 75% of the original cow-calf-contact in the second week of reduced contact (wk 9, corresponding to 4.5 h contact for full-time RC pairs and 2 h contact for part-time RC pairs, see Figure 3).

Behavioral observations
Behavioral observations were conducted at the end of wk 3 of the experimental period (the main rearing period), and 48 h after each stepwise reduction in contact in wk 8 and 9 (the reduced contact period).These observation days were chosen, as cows concurrently took part in behavioral tests for other studies; however, on these days, cows and calves would be undisturbed.One cow-calf pair on the part-time treatment in block 4 was excluded before any of the observations began, as the calf had to be euthanized due to disease.
Duration of behaviors were recorded from videos obtained by surveillance cameras (DS-2CD2143G2-I 4MP, Hikvision, Hangzhou, China) placed above the pens.The behaviors were recorded continuously using focal animal sampling (Bateson and Martin, 2021) of either each cow (the main rearing period) or calf (the reduced contact period).The calf was chosen as the focal animal during the reduced contact period to get a more representative measure of whether milk intake (measured as suckling time) would be reduced with reduced cow-calf contact.All behaviors were mutually exclusive.Animals could be identified by symbols painted on their backs (using Inebrya© oxidizing emulsion cream, San Martino di Lupari, Italy).All recordings were made in the program BORIS (Friard and Gamba, 2016).Due to the experimental set-up, it was not possible to blind the observers to the treatments; however, not all observers  were aware of the hypotheses tested.The inter-observer reliability for each set of observations was assessed by calculating Cohen's Kappa (Bateson and Martin, 2021) in BORIS with a 5 s interval.
The main rearing period: maternal behavior.Cows from block 1 to 6 (total 47 cows; 24 full-time and 23 part-time) were observed for 24 h on the Sunday of wk 3 when calves were on average (mean ± SD) 28.4 ± 6.6 d old.The behavioral recordings were undertaken by 2 observers (A and B), after achieving inter-observer reliability during block 1 (κ = 0.872; near-perfect agreement, Ranganathan et al., 2017).Observer A undertook observations on block 1 to 3, while observer B undertook observations on block 4 to 6.
For full-time cows, behavioral recordings started at 00:00:00 and ended at 23:59:59.For part-time cows, behavioral recordings started the moment the cow entered the home pen (both front legs inside pen) after morning milking (approximately 05:30:00) and ended the moment the cow left for afternoon milking (both front legs outside of pen; approximately 15:30:00).
Behaviors recorded included the cow nursing her own calf exclusively (i.e., only her own calf was suckling) while the calf was in inverse parallel position (IPP); cow nursing her own calf exclusively while the calf was in any other position than IPP; cow nursing an alien calf (i.e., a calf other than the cow's own); cow simultaneously nursing multiple calves including her own calf; cow simultaneously nursing multiple calves not including her own calf; and cow sniffing or licking her own calf (see Table 1 for ethogram).When cows were nursing calves other than their own, the nursing position was not registered.
The reduced contact period: suckling behavior of calves.Calves from block 2 to 7 (total 47 calves; 12 full-time RC, 12 full-time UC, 13 part-time RC, and 10 part-time UC) were observed during 24 h beginning 48 h after each reduction in cow-calf contact was initiated.Calves were on average (mean ± SD) 57.0 ± 5.9 d of age on the first observation day.Observations were not performed for block 1 as reduction in contact occurred earlier than in the other blocks.Instead, block 7 was observed to ensure similar sample size as the observations in the main rearing period.Calves in the full-time treatment of block 6 (both UC and RC) were observed a day earlier than planned, because one cow jumped the fence on the scheduled observation day.The behavioral recordings were conducted by 3 observers (C, D, and E) after achieving inter-observer reliability during block 2 (κ = 0.95 between C and D; 0.98 between C and E).Observer C undertook observations on block 2 to 5, while observers D and E observed block 6 and 7.
Calves with UC were observed either from 00:00:00 to 23:59:59 (full-time contact treatment) or from the moment their dam returned to the home pen after morning milking (both front legs inside home pen, approximately 05:30:00) until the moment the dam left for afternoon milking (both front legs outside of home pen, approximately 15:30:00, part-time contact treatment).Calves with RC were observed from when they were released from the calf creep at 11:00:00 and until they were re-confined in the calf creep in accordance Cow's own calf has its head underneath the cow's belly near the udder.The calf's body is positioned at an angle from inverse parallel to 45 degrees, measured as the angle between the axes drawn from the midline of the cow's and calf's bodies.The cow is standing still, but may move her leg(s), or take 1 to 2 steps in any direction.No other calf is suckling.Exclusively nursing own calf, not inverse parallel position Cow's own calf has its head underneath the cow's belly near the udder.The calf's body is positioned at an angle of more than 46 degrees, measured as the angle between the axes drawn from the cow's and calf's bodies.The cow is standing still, but may move her leg(s), or take 1 to 2 steps in any direction.No other calf is suckling.

Exclusively nursing one alien calf
A calf other than the cow's own calf has its head underneath the cow's belly near the udder.The cow is standing still, but may move her leg(s), or take 1 to 2 steps in any direction.No other calf is suckling.Simultaneously nursing multiple calves, own calf not included Two to three calves other than the cow's own have their heads underneath the cow's belly near the udder area.The cow is standing still, but may move her leg(s), or take 1 to 2 steps in any direction.Simultaneously nursing multiple calves, own calf included Cow's own calf and one to three other calves have their heads underneath the cow's belly near the udder area.The cow is standing still, but may move her leg (s), or take 1 to 2 steps in any direction Grooming own calf Cow's head (muzzle, or tongue) is touching any part of own calf's head or body, while sniffing or licking.Touching is defined as no visible gap between the cow's head and the calf's body or head.Reduced contact, focus on calf Suckling dam The calf has its head underneath the dam's belly near the udder area.This may be preceded by the calf moving toward the udder with a stretched neck and ears back.The dam is standing still, but may move her leg(s), or take 1 to 2 steps in any direction.

Suckling alien cow
The calf has its head underneath an alien cow's belly near the udder.This may be preceded by the calf moving toward the udder with a stretched neck and ears back.The alien cow is standing still, but may move her leg(s), or take 1 to 2 steps in any direction.
with their contact-treatment schedule (Figure 3).The behaviors recorded included calf suckling dam (i.e., the calf's mother); and calf suckling an alien cow (i.e., a cow other than the calf's dam; Table 1).

Statistical analyses
All statistical analyses were carried out in R (R Core Team, 2022).Continuous variables were assessed for normal distribution using the Shapiro-Wilk test and by visual inspection of the distribution of the residuals.The homogeneity of variance was assessed by visual inspection.Continuous variables were all analyzed using general linear mixed models, while binary variables were analyzed using negative binomial mixed models.All models were fitted using the package glmmTMB (Brooks et al., 2017) and were evaluated based on a type II sum of squares ANOVA using the function Anova (Fox and Weisberg, 2019).Estimates and t-ratios were obtained using the package emmeans (Lenth, 2022).In both sets of observations, the statistical unit was the individual animal; for the observations in the main rearing period this was the cow, while it was the calf for the observations during the reduced contact period.
Main rearing period.For each cow (n = 47 cows and observations), the total duration of exclusively nursing own calf was calculated as the sum of the duration of exclusively nursing own calf in IPP and the duration of exclusively nursing own calf in any other position than IPP.Additionally, the percentage of the total duration of exclusively nursing own calf taking place in IPP was calculated.Only 17 of 47 cows nursed an alien calf, and nursing an alien calf was therefore transformed into a binary variable.Similarly, only 15 cows nursed their own and one or more alien calves simultaneously; this variable was also transformed into a binary variable.Only one cow simultaneously nursed multiple calves that were not her own, therefore this behavior was not analyzed.The percentage of nursing in IPP was arcsine square root transformed to achieve an approximate normal distribution.The linear mixed or mixed negative binomial regression models used to analyze the behaviors (duration of exclusively nursing own calf, the percentage of nursing own calf taking place in IPP, the probability of exclusively nursing an alien calf, and the probability of simultaneously nursing multiple calves including the cow's own) included the fixed effects of cow-calf contact treatment (full-vs.part-time), cow parity (primiparous vs. multiparous), calf sex (bull vs. heifer), the mean-centered calf age at observation (Becker et al., 1988), the interaction between treatment and parity, as well as the random effects of block (1 to 6) and pen group (1 to 12).The interaction between treatment and parity was not significant for any of the variables and was removed from the final models.Additionally, a model containing the treatment groups of the reduced contact period was run to control for any baseline differences in these groups before the reduced contact treatments were applied.
To create an overview of when nursing takes place during the day for full-and part-time cows, respectively, a descriptive graph was created.In R, the average cumulative nursing of own calf exclusively for each hour of the day was calculated, as were the standard deviations.
Reduced contact period.The recordings of 2 calves (part-time, RC, block 7) were excluded from the analyses.The dam of one of the calves developed mastitis, which likely affected the suckling behavior of both calves in the pen.The resulting sample size was n = 45 calves (12 full-time RC, 12 full-time UC, 11 part-time RC, and 10 part-time UC), with 90 observations.For each calf, the total duration of suckling per observation day was calculated as the sum of the durations of suckling dam and suckling alien cows.Only 14 calves suckled an alien cow on both observation days, and 9 calves did not suckle an alien cow on any of the 2 d resulting in a zero-inflation of 43.3%.This variable was therefore transformed into a binary variable.Total suckling duration and time spent suckling only the dam were analyzed using linear mixed models, while the probability of suckling an alien cow was analyzed using a mixed negative binomial regression model.The linear mixed or mixed negative binomial regression models used to analyze the behaviors (total suckling duration, time spent suckling only the dam, and the probability of suckling an alien cow) included the fixed effects of the combined cow-calf contact treatment (full-time RC vs. full-time UC vs. part-time RC vs. part-time UC), dam parity (primiparous vs. multiparous), calf sex (bull vs. heifer), observation week (wk 8 vs. wk 9), and the mean-centered calf age at the first observation.The interaction between treatment and week, as well as the random effects of block (2 to 7), pen group (11 groups), and calf were also included.

The main rearing period: maternal behavior
Estimates and test statistics are summarized in Table 2. Our main outcome of observations in the main rearing period was time spent exclusively nursing own calf.Full-time cows spent more time exclusively nursing their own calf than part-time cows.Additionally, primiparous cows spent more time exclusively nursing their own calf than multiparous cows.The cumulative duration of exclusively nursing own calf over the observed 24 h is shown in Figure 4 for full-and parttime cows.Descriptively, this figure shows a rebound in nursing for part-time cows upon reunion with their calf, resulting in a cumulative nursing duration until the afternoon milking similar to that of the full-time cows.No differences between the future reduced contact treatment groups could be found for nursing time in the main rearing period.
Our secondary outcomes were the percentage of nursing own calf taking place in IPP and the probability of nursing an alien calf.Contact treatment had no effect on the percentage of nursing own calf taking place in IPP.However, there was an effect of calf sex, with mothers of bull calves spending a higher proportion of nursing own calf in IPP compared with mothers of heifer calves (Table 2).None of the included explanatory variables affected the probability of a cow nursing an alien calf, but part-time cows tended to be more likely to simultaneously nurse their own calf and one or more alien calves (Table 2).Full-time cows spent more time grooming their calf than part-time cows did; additionally, mothers of bull calves tended to perform more maternal grooming than mothers of heifer calves (Table 2).Calf age had no effect on any of the observed behaviors.

Reduced contact period: suckling behavior of calves
The main outcome of the observations taking place in the reduced contact period was time spent suckling either any cow or only the dam.The combined contact treatments applied in the reduced contact period affected total suckling duration (i.e., the time spent suckling any cow; Χ 2 = 8.7, df = 3, P = 0.034).Averaged across both observation days, part-time calves with unchanged contact spent more time suckling any cow than part-time calves with reduced contact (Figure 5).For time spent suckling only the dam, there was an interaction between the combined contact treatment and observation week (Χ 2 = 7.9, df = 3, P = 0.047).Part-time calves with unchanged contact spent more time suckling in wk 9 compared with part-time calves with reduced contact in wk 8.In wk 9, part-time calves with unchanged contact spent more time suckling compared with full-time calves with reduced contact (Figure 6).Additionally, time spent suckling the dam was negatively related to calf age (−3.9 min/d of calf age, Χ 2 = 5.7, df = 1, P = 0.017).
Dam parity tended to affect calf selectivity, which was a secondary outcome of these observations.Calves of multiparous cows tended to be more likely to suckle an alien cow (probability and CI: 0.65 [0.50-0.78] vs. 0.46 [0.30-0.63]for multi-and primiparous cows, respectively).

DISCUSSION
In accordance with the hypothesis, part-time cows spent less time exclusively nursing their own calves compared with full-time cows in the main rearing period.Additionally, they also spent less time on maternal grooming, supporting that part-time cows perform less maternal behavior than full-time cows.However, we did not find the expected effect of contact treatment on the amount of nursing occurring in inverse parallel position (IPP), or on cows' selectivity, i.e., the probability of a cow nursing alien calves.In the second set of observations, we did not consistently see the hypothesized effect of reduced contact on calf suckling time.Furthermore, we did not find the expected effect of contact amount in the reduced contact period on the probability of a calf suckling an alien cow.

Maternal behavior in the main rearing period: Effect of contact treatment
In our study, cow-calf pairs on the part-time treatment should have had enough time together for nursing and grooming during the daytime hours to compensate for the time spent apart at night (Table 2).We did observe a rebound (Nicol, 1987) in nursing for part-time cows during the daytime.However, for the remainder of the hours between morning and afternoon milking, cows on full-time and part-time treatments showed a similar increase in cumulative nursing duration.Thus, part-time cows did not compensate for the time spent apart in term of nursing behavior.Diurnal variation in nursing behavior has previously been reported.Nursing rarely takes place during the night; however, peaks in nursing activity have been observed around sunrise and sunset (Reinhardt and Reinhardt, 1981;Lewandrowski and Hurnik, 1983;Johnsen et al., 2021).The reunion of part-time cows and calves following morning milking coincides with the morning-period of high nursing activity observed in previous studies.This could explain the steep rebound effect we observed.Nevertheless, part-time pairs are separated in the afternoon before the expected nursing peak in the early evening.Perhaps the motivation to nurse is lower for cows before this expected peak, explaining why we did not observe cows compensate for the approaching separation period.The time spent exclusively nursing own calf during 24 h in the current study is lower than what was reported in a study with a similar set-up (37.0 and 25.7 min for full-and part-time, respectively; Bertelsen and Jensen, 2023a).On the other hand, the amount of grooming in the current study is somewhat more than previously observed (16.9 and 8.5 min for full-and part-time, respectively; Bertelsen and Jensen, 2023a).The lack of compensation by part-time pairs could be caused by a faster rate of milk ingestion during suckling (Von Keyserlingk et al., 2006).
We found a tendency for part-time cows to be more likely to simultaneously nurse multiple calves including their own.A previous study using a similar experimental set-up found that calves on a part-time contact schedule were more likely to suckle alien cows immediately after the animals were reunited.The calves often suckled the first cows that entered the home pen irrespective of whether it was their dam or not (Bertelsen and Jensen, 2023a).These authors suggested that calves were very hungry after more than 12 h without access to milk.Calves experiencing a nutritional deficiency are more likely to suckle multiple cows (Víchová and Bartoš, 2005).Additionally, cows have been reported to be less likely to reject alien calf if they are already nursing their own calf (Fröberg and Lidfors, 2009), which the first cows entering the home pen after the separation period likely would be.There may also be an effect of housing and especially aspects of the lying surface.Dairy cows show preference for both straw-bedded lying areas and larger lying areas than free stalls (Shewbridge Carter et al., 2021).Part-time cows therefore likely preferred to lie in the straw-bedded home pens compared with the free stalls in the separation pen.Some part-time cows may have compensated for their potentially lower lying time in the separation pen by lying down shortly after returning to the home pen.This would affect the calves' access to the cows' udders, possibly causing them to suckle other cows.However, we did not find an effect of contact treatment on the probability of a cow nursing alien calves (i.e., being less selective), which could indicate that the calves were still able to drink milk to satiation, even when they spent less time doing so.Furthermore, the repeated relocation experienced by the part-time cows may have been stressful, which may have affected their maternal behaviors; however, no studies to our knowledge have examined this.
In our study, we did not find an effect of part-time contact on the amount of nursing taking place in IPP or on the probability of a cow nursing an alien calf, i.e., being less selective.These 2 behaviors could be seen as proxies of maternal bond strength (Le Neindre, 1982, 1989).Therefore, we found no indication of a difference in maternal bond strength between part-time and full-time cows, despite part-time cows having less time with their calves and spending less time on maternal behaviors.That part-time cows appeared to have formed a similar strong maternal bond can be inter- , time spent suckling any cow) for 47 Danish Holstein-Friesian calves with different amounts of cow-contact.Suckling behavior was observed twice over 24 h following each reduction in contact during the reduced contact period.Calves were on average 57.0 ± 5.9 d of age on the first observation day.Estimates are averaged across both observation days.RC and UC refer to reduced contact and unchanged contact.Part-time calves with unchanged contact spent more time suckling than part-time calves with reduced contact did (t-ratio = −2.9,P = 0.027).Estimates with different superscripts are significantly different from each other (P < 0.05).Figure 6.Interaction between contact amount and observation week on time spent suckling the dam.Observations were made on 47 Danish Holstein-Friesian calves when they were on average 57.0 (wk 8) and 64.0 ± 5.9 d old (wk 9).RC and UC refer to reduced contact and unchanged contact.Part-time calves with UC spent more time suckling their dam in wk 9 than part-time calves with RC did in wk 8 (t-ratio = −3.3,P = 0.029) and full-time calves with RC in wk 9 (t-ratio = −3.8,P = 0.007).Estimates with different superscripts are significantly different from each other (P < 0.05) preted as positive.However, experiencing the repeated daily separations on the part-time treatment may lead to stress and frustration, precisely because cow and calf have formed a bond to each other (Roadknight et al., 2022).In our set-up, the daily separation duration was approximately 14 h, which is within the range of peak post-separation reaction of cows following their first separation from their calf (9 to18 h; Weary and Chua, 2000).Although previous studies have found that the proximity between cow and calf varies during the day (Kour et al., 2021b), the intervals between periods of close proximity do not reach 14 h (max 100 min for beef cattle 1 to 107 d postpartum; Hirata et al., 2003).After approximately 4 weeks of repeated separations, the cows may have become accustomed to the daily separations, but the reduced contact time may influence the cows' affective states negatively.Neave et al. (2023b) did indeed find that part-time cows had a more negative bias in a judgement bias test compared with full-time cows.Unfortunately, we did not observe the cows' behavior during the nightly separation period, which could otherwise have given an indication of the possible stress caused by the separation.A previous study found signs of cows reacting more to the daily separations in the beginning of the experimental period, possibly indicating that cows learn to expect being reunited with their calf with time (Bertelsen and Jensen, 2023a).The relationship between maternal bond strength and the experience of repeated separations requires more research before part-time cow-calf contact can be recommended to farmers.Furthermore, validation of nursing in IPP and cow-calf selectivity as indicators of bond strength could prove valuable in future studies.
Part-time calf-contact reduced the time cows spend on nursing and grooming their calves but did not affect indicators of maternal bond strength, namely nursing in IPP and the probability of nursing other calves.

Maternal behavior in the main rearing period: Effect of parity and calf sex
In addition to the effects of our contact treatment, we found that dam parity affected the time spent exclusively nursing own calf.Primiparous cows spent more time nursing their calves than multiparous cows.This may be explained by primiparous cows having lower milk flow and smaller cisternal capacity (Edwards et al., 2014;Vrhel et al., 2021), requiring that the calves suckle for longer to ingest similar milk amounts.Additionally, we saw that mothers of bull calves spent a significantly larger proportion of nursing in IPP and tended to groom their calves more than mothers of heifer calves.This indicates a stronger maternal bond between bull calf and dam than between heifer calf and dam.There could be an evolutionary explanation for cows bonding more strongly with their male offspring.The Trivers-Willard model suggests that a male offspring in good condition should have a higher reproductive potential than a female offspring in good condition, therefore mothers should invest more in their male offspring (Trivers and Willard, 1973).Female calves of beef and zebu breeds are more likely to suckle alien cows and are weaned earlier than bull calves, similarly indicating a weaker bond between the female calf and her mother (Reinhardt and Reinhardt, 1981;Víchová and Bartoš, 2005).The motivational background of cows forming a stronger bond to male than female calves deserves future research.

Suckling behavior in the reduced contact period: Effect of contact
Averaged across the 2 observation days, total suckling time (i.e., the sum of time spent suckling any cow) was reduced for part-time calves with reduced contact compared with part-time calves with unchanged contact.However, no difference was found between fulltime calves with either reduced or unchanged contact.This indicates that reduced contact only affected total suckling time for calves with part-time contact.When considering the time spent suckling the dam, the results are more complex.Within the wk 8 observations, no treatment group differences could be detected.In wk 9, part-time calves with unchanged contact suckled for longer than did full-time calves with reduced contact.In other words, calves with 10 h of cow contact suckled more than calves with 4.5 h of cow contact.This is in accordance with our expectations, as the possibility of suckling is reduced.However, we did not find a difference between full-time calves with unchanged contact and part-time calves with reduced contact, i.e., between calves with 23 h and 2 h cow contact, respectively.Comparing across wk 8 and 9, part-time calves with reduced contact spent less time suckling in wk 8 compared with part-time calves with unchanged contact in wk 9. Nevertheless, this difference was not seen in wk 9.These results indicate that full-time calves with reduced contact were not able to compensate for the time spent apart from their dam, whereas part-time calves were better able to compensate.This could be because the part-time calves had experienced some restriction throughout the main rearing period, while the full-time calves needed to suddenly adjust their time schedule in the reduced contact period.
To our knowledge, the current study is among the first to investigate the effect of gradually reducing cow-calf contact on time spent suckling (Sirovnik et al., 2020;Jensen et al.: Maternal and suckling behaviors of dairy cow and calf Sørby et al., 2024a;b).A previous study with a similar main rearing period investigated a stepwise weaning strategy (Bertelsen and Jensen, 2023b), first weaning calves off of milk before cow contact was removed.The authors found that part-time calves had higher average daily gains during weaning and spent more time eating solid food in the main rearing period (Bertelsen and Jensen, 2023b;a).Part-time calves thus appeared to be better prepared for separation from their mothers.This could also explain the reduction in total suckling time observed for part-time calves with reduced contact in the current study; their solid feed intake may have been higher already in the main rearing period, making it easier for them to adapt to less suckling opportunity than full-time calves.
In the current study, calves were still able to suckle cows and thus still had access to milk even with reduced contact.The purpose of the gradual reduction in contact time was to both reduce milk intake and reduce dam contact to prepare the pair for complete separation.However, suckling time does not necessarily indicate milk intake (Cameron, 1998), and calves are able to increase their rate of milk ingestion in response to increased competition (Jensen, 2004;Jensen and Budde, 2006).Furthermore, calves have been reported to consume up to 10 L of milk even with just 2 30 min periods of suckling opportunity (Fröberg et al., 2008).Therefore, it is difficult to determine whether calves with reduced suckling times also consumed less milk.Another option could have been to measure the calves' solid feed intake, as reduction in milk intake stimulates an increase of solid feed intake (Khan et al., 2007b;Jensen et al., 2020).With no direct measurements of milk intake or observations on solid feed intake, it is difficult to determine whether calves with reduced cow contact were successfully prepared to be completely weaned off milk and separated from their mothers.Future studies are encouraged to investigate this further by including measurements of intake.Additionally, the gradual contact reduction could be prolonged or have more steps of reduced contact added (Sørby et al., 2024b); for instance, keeping cow and calf completely fence-line separated for an amount of time (i.e., weaned off milk), before they are fully separated (Bertelsen and Jensen, 2023b).Such initiatives may reduce separation stress and would thus be of interest to farmers practicing cowcalf contact.
Reducing contact time reduced total suckling time for part-time calves and affected time spent suckling only the dam for both part-and full-time calves, though this effect depended on observation week.However, whether the reduction in suckling time also reduced milk intake, and thus better prepared calves for complete separation from their dams, remains unclear.

Suckling behavior in the reduced contact period: Effect of calf age and parity
Natural weaning is estimated to take place much later than in our experiment (6 to 11 mo after calving; Weary and Chua, 2000;Johnsen et al., 2015), and the reduction in suckling time is not expected to decrease much from approximately 3 to 6 mo of age (Waltl et al., 1995).That we did see some effect of reduced contact on time spent suckling could then be a sign of the calves relying less on milk for their energy intake, as their time together with the cows should allow them to suckle as much as calves with unchanged contact.Additionally, we found a negative relationship between calf age and time spent suckling the dam, indicating an age effect even earlier than previously reported (Paranhos da Costa et al., 2006;Kour et al., 2021a, but see Vitale et al., 1986).
We only detected a tendency for dam parity to affect the probability of calves suckling alien cows.Calves of multiparous cows tended to be less selective of their choice of cow to suckle than calves of primiparous cows, possibly indicating a weaker maternal bond between older cows and their calves.However, we found no effect of parity on selectivity between calf and cow in the main rearing period, cautioning further interpretation of this tendency.

Limitations and future directions
Overall, the focus of this paper has been on maternal behavior of cows and suckling behavior of calves under different cow-calf contact schedules.The reaction to the actual separation has not been within the scope of this paper but is investigated in another study (Neave et al., 2023a).Further research on other aspects of cow-calf behavior, such as proximity to each other and reaction to daily separations, could further enlighten the effects of part-time contact and gradual weaning.Additionally, including measures or observations of solid feed intake could give indications of the calves' level of nutritional independence.Including more observations days, especially during the main rearing period, could provide more details on the development of the maternal bond.A more detailed mapping of the timing of nursing and other maternal behaviors during the day could reveal possible circadian rhythms.This could be used when designing part-time contact schedules to best accommodate cow and calf's needs, both nutritionally and behaviorally.Because potential effects of housing and diurnal variation in maternal behaviors are confounded within the contact treatments, we cannot rule out that the observed effects of part-time contact may be influenced by other factors than reduced cow-calf contact.Finally, as natural weaning is expected to take place much later than our experiment allowed for (Flower and Weary, 2001;Johnsen et al., 2015), extending the main rearing period could reveal when dairy cows and calves themselves reduce nursing and start spending more time apart.

CONCLUSIONS
This study revealed that cows with part-time calf contact spent less time nursing and grooming their calf than cows with full-time calf contact.Nevertheless, we found no indication that full-time cows form stronger bonds to their calves than part-time cows do, as measured by percentage of nursing taking in inverse parallel position and cows' selectivity.However, more research is required on this aspect.When contact time was gradually reduced, part-time calves reduced their time spent suckling both dam and alien cow(s); however, full-time calves appear less able to compensate for reduced contact when considering time spent suckling only the dam.As calves still spent considerable time suckling regardless of contact time, it is unclear whether this reduced contact is enough to prepare calves for complete weaning and separation, and more research on gradual weaning is thus needed.

Figure 1 .
Figure 1.Nursing in inverse parallel position (IPP).Illustrations left and right indicate the range within which nursing was recorded as happening in IPP.Left: the calf is suckling, while its body is completely parallel to the cow's.Right: The calf is suckling while the angle formed between the midline of the cow and calf's bodies is 45 degrees.In both ends of the range, the cow is able to reach the hind quarters of the calf and sniff it, and she is thus able to recognize whether it is her own or another calf.All nursing taking place outside of this range was categorized as not in IPP.Nursing position was not registered when cows nursed alien calves.
Figure 2. Schematic presentation of one of the home pens.Indicated are position of feed troughs (F), water bowls (W), brushes (B), concentrate bowls (C), hay rack (H), calf creep (S), and temporary fence (TF).Following the installation of the temporary fence, pairs on the unchanged contact treatment are kept on the side with the small creep (UC), while pairs on the reduced contact treatment are kept on the side with the big creep (RC).

Figure 3 .
Figure 3. Hours of cow-calf contact per day for each of the 4 contact/weaning treatment combinations across the experimental period.F refers to full-time, P to part-time, and RC and UC refer to reduced contact and unchanged contact.In wk 1 to 7, 24 full-time cow-calf pairs and 23 part-time pairs were included in the observations.In wk 8 and 9, 12 full-time RC cow-calf pairs, 12 full-time UC pairs, 11 part-time RC pairs, and 10 part-time UC pairs were included in the observations.
Jensen et al.: Maternal and suckling behaviors of dairy cow and calf Jensen et al.: Maternal and suckling behaviors of dairy cow and calf Figure 4. Descriptive results of the cumulative duration of exclusively nursing own calf for 47 Danish Holstein-Friesian cows over a 24 h observation period.Observations took place during the main rearing period, on average 28.4 ± 6.6 d postpartum.Blue lines indicate milking time.Part-time cows could not nurse until they were reunited with their calves after morning milking and could not nurse after they were separated following afternoon milking.Shortly after the reunion with their calves, part-time cows reached a similar level of cumulative nursing to full-time cows, and the cumulative nursing of the 2 treatments increased over the day.Part-time cows showed a greater individual variation than full-time cows.

Figure 5 .
Figure5.Total time spent suckling (i.e., time spent suckling any cow) for 47 Danish Holstein-Friesian calves with different amounts of cow-contact.Suckling behavior was observed twice over 24 h following each reduction in contact during the reduced contact period.Calves were on average 57.0 ± 5.9 d of age on the first observation day.Estimates are averaged across both observation days.RC and UC refer to reduced contact and unchanged contact.Part-time calves with unchanged contact spent more time suckling than part-time calves with reduced contact did (t-ratio = −2.9,P = 0.027).Estimates with different superscripts are significantly different from each other (P < 0.05).
Jensen et al.: Maternal and suckling behaviors of dairy cow and calf

Table 1 .
Jensen et al.: Maternal and suckling behaviors of dairy cow and calf Ethograms for both observations in the main rearing period and the reduced contact period Main rearing, focus on cow Exclusively nursing own calf, inverse parallel position

Table 2 .
Least squares means with 95% confidence limits of maternal behaviors of cows, recorded during the main rearing period.47 Danish Holstein-Friesian cows were housed with either full-time (23 h/d, 24 cows) or part-time (10 h/d, 23 cows) calf contact.Cows were observed for 24 h 28.4 ± 6.6 d after calving Behavior