M.A. Desnoyers
K.L. Butler
J.L. Barnett
Victoria Dept. Primary Industries,
Werribee Vic 3030
Australia
The interpretation of hen behaviour in the pre-laying period for hen welfare is contentious. This preliminary experiment of 18 pairs of hens aged 45-48 weeks examined the effects of cages with and without a nest box (NB) on behaviour in the "searching" and "sitting" phases of nesting using hens with or without previous experience of a NB.
Half of the 18 hens with access to a NB chose to lay in the NB (nest layers). For the other nine hens with a NB and the 18 hens without a NB, oviposition occurred on the wire cage floor (floor layers).
Nest layers were less active and performed more sitting behaviour than floor layers in the hour before laying. A behaviour termed "following behaviour" was recorded for over half of the floor layers but none of the nest layers. The interpretation of pre-laying behaviour for hen welfare requires further investigation.
Introduction
From studies of the pre-laying activities of domestic hens, it is generally accepted there are two phases of behaviour involved in oviposition (Sherwin and Nicol, 1992). Beginning about one hour prior to oviposition, the activity level of hens increases in a phase of behaviour termed "searching" in which hens appear motivated to seek a nest site. In this phase hens increase locomotion and perform behaviours such as inspection of potential nests. Once hens have selected the preferred nest site the "sitting" phase commences, which includes the adoption of a sitting posture interspersed with nest-building activities such as scratching the floor/litter, rotating the body on the nest and collecting litter if available.
Activities performed in the searching phase are goal-directed or appetitive behaviours, occurring when hens are motivated to find a suitable nest for oviposition (the consummatory behaviour). Thus, Appleby and McRae (1986) and Duncan and Kite (1989) showed that hens were motivated to lay their egg in a nest box, and if a nest box was not available hens performed more nest-searching behaviour (Cooper and Appleby, 1995; Freire et al., 1996). While an increased occurrence of appetitive behaviour may indicate a stronger motivation to achieve the consummatory phase, it does not necessarily indicate that increased pre-laying locomotion reflects increased frustration and thus a potential welfare problem.
For example, using an aversive task approach, Freire et al. (1997) suggested that hens were only weakly motivated to reach the nest site during the searching phase, although the motivation to gain access to a nest site increased near the start of the sitting phase. Further, Cooper and Appleby (1997) compared hens that were consistent and inconsistent in their use of nest boxes. While no apparent difference between the two classes of hens was detected in hens' motivation to use a nest box, the inconsistent hens were less responsive to the cues provided by nest boxes than consistent hens. Based on evidence of motivation of hens to lay in a nest box, increased time spent in pre-laying behaviours in the absence of a nest and increased vocalisations indicative of frustration when access to a nest is blocked, it has been concluded that there is convincing evidence of the importance of a suitable nest site (see Keeling, 2004). However, it is also known that even in the presence of a nest not all birds lay in the nest with reports of 0-80% of eggs being laid on the floor (Sherwin and Nicol, 1993) and the question remains of what do hens perceive as a suitable nest site?
There is reasonable knowledge of the environmental cues that influence pre-laying behaviour by hens, including the presence of a nest box, litter compared to wire flooring, level of light, genotype and social position (Appleby et al., 1993; Freire et al., 1996). However, the importance of pre-laying searching behaviour and sitting posture to the welfare of laying hens is less well understood. The objectives of this preliminary experiment in hens that were experienced with laying with or without a NB were to measure activity levels of hens around oviposition, the performance of key behaviours in the "searching" and "sitting" phases of nesting and to examine the use of a NB.
Methods
Commercial Hyline Brown hens aged 45-48 weeks were used in this preliminary experiment, in a controlled climate shed. The hens were part of a larger experiment investigating the effects of different cage "furniture" on hen welfare (i.e. nest box, dust bath and perch; Victorsson Trivselburen 8-bird Furnished Cages, Sweden, measuring 1.2 m wide, 0.5 m deep and 0.45 m high at the rear of the cage). The NB measured 0.24 m wide, 0.5 m deep and 0.27 m high at the front of the cage and had a solid ceiling, rear and sides, apart from an entrance opening in one sidewall. A blue vinyl flap covered the front of the NB while the NB floor was overlain with "astro turf" (0.37 m x 0.22 m x 15 mm thick).
For the present experiment two Victorsson Trivselburen Furnished Cages positioned back-to-back and with all furniture removed were used as the experimental cages. The 36 focal hens (18 pairs) selected for the experiment were from six specific "home" cages (8 hens per cage) that were without dust baths or perches. Three of these "home" cages however contained a nest box (NB treatment) and three were without a NB (No NB treatment). There were nine replicates in time and the NB was randomly allocated to one of the 2 experimental cages for each replicate. At about 1600 h on day 1 of each replicate, a pair of hens was selected on an ad hoc basis from a home cage containing a NB and transferred to the experimental cage in which an identical NB had been fitted. Similarly, a pair of hens from a home cage without any furnishings was moved into the other experimental cage.
Video recording commenced from the time the hens entered the cages using black and white video cameras with in-built infra-red (IR) lights and time-lapse video recorders. The cameras were positioned to provide views of the two cages from overhead, in front and inside the cages, as well as inside the NB. The two focal hens per cage were marked on the back with a carbon-based dye for individual identification under IR light. The Observer behaviour-recording program supplemented with the Support Package for Video Analysis (version 4.0 for Windows; Noldus Information Technology, 1997; Wageningen, The Netherlands) was used to continuously record the activity and location of each focal hen.
To allow habituation to the cages there was a minimum of 24 h from placement in the cages to the commencement of data collection from 4 h of video records, commencing 2 h pre-laying on the day that both hens in the pair laid an egg on the same day. The mean (± std dev) time to the first egg laid on the observation day was 47.0 ± 10.7 h (min. 38.4 h, max. 67.5 h). Most hens laid one egg during this "habituation" period, although only the times and locations of oviposition were recorded from the video record. As the experimental cages were of the same design as the home cages, and the pairs of hens originated from the same home cage, it was assumed that this would be a sufficient period of habituation.
A comprehensive list of behaviours was developed to enable the measurement of hen activities, while hen movement was tracked by dividing the video image into nine similar-sized areas excluding the NB or the equivalent area in the No NB cage and the frequency of hens occupying the different locations in the cage. The 4 h observation period was divided into eight, 30-min periods for analysis. Statistical analysis was conducted using a restricted maximum likelihood analysis using the GenStat Committee (2000) statistical package on individual hen data with random effects for home cage, replicate and pair of birds and fixed effects of laying status of the bird and its cage-mate and presence of a NB. The random effect of pair of birds accounted for any correlation between the two birds in each pair.
Results
Of the 18 focal hens in the NB treatment, nine laid in the NB and nine laid outside the NB on the wire floor. These hens were thus termed "nest layers" and "floor layers", respectively. Based on observation of these hens on consecutive days, only one hen was observed to change her location of laying between days, i.e. laid in the NB and on wire floor on different days. Every combination of "pairing" was observed, with both hens laying in the NB or on the cage floor, one of each laying in the NB or on the cage floor and overlap or separation of both oviposition and nesting behaviour within a pair of hens (ie both hens laid eggs within the same 4-h period or with a clear separation of time periods).
Hen activity level around oviposition was assessed by the time hens spent walking and the frequency of entering areas 1-9 of the cages. Darkness strongly reduced hen activity. As the 2-h pre-laying period included darkness for six hens, the time in the dark was used as a co-variate in the analysis. These six hens alternated between standing and sitting in the dark and none were observed to locomote. After adjusting for darkness, there were no differences in the activity measures in the period 120-60 min pre-laying. However, in the 1 h before laying, nest layers were less active than floor layers; nest layers performed less (P<0.01) walking behaviour (excluding following behaviour) and entered areas 1-9 less frequently (P<0.01) than floor layers. No other treatment effects were observed. A behaviour termed "following behaviour" was observed. Following behaviour, defined as the focal hen walking or running but following the other hen as she moved about the cage, was only observed in the floor layers (16 of the 27 hens) and occurred during ~5% of the total time in the period 60 min pre-laying to 30 min post-laying and the peak frequency of occurrence was in the 30 min prior to oviposition.
Discussion
This small experiment that involved observations on egg-laying behaviour in 18 pairs of hens has provided more questions than answers. An interesting observation was "following behaviour" which only occurred in hens that laid eggs on the wire floor, irrespective of the presence of a nest box. This behaviour occurred during pre- and post-laying periods and involved the hen (follower) appearing to attempt to remain close to the other hen (followed), including when the followed hen was locomoting. When the followed hen was stationary and standing, the follower would often sit next to her and the follower would put her head under the body of the followed hen. A similar behaviour was described by Kite et al. (1980) in which hens appeared to follow and attempt to crawl underneath pen-mates. The reasons for this behaviour are unknown as are the reasons why it occurred in 59% of floor layers and occurred both pre- and post-oviposition. One explanation is that when a nest box was present which was utilised for egg laying, this environment provided appropriate cues for nest site selection. Another explanation is the follower hen may have derived cues for nest site selection from the followed hen.
This experiment has shown, as reported in other studies (Cooper and Appleby, 1997), that the use of the NB for egg laying is highly variable between birds. In this experiment where experienced (with a NB) hens were housed in pairs in a cage with a NB, 50% of hens laid in the nest box and 50% laid on the wire cage floor outside the NB. This is similar to the daily egg production records for the three NB treatment home cages, recorded over 10 months, in which 55.8% of eggs were laid in the NB. These data raise the question of hens' preference for egg laying location. While the literature suggests hens are motivated to seek a "preferred" location for egg laying, the data from this experiment could be interpreted to suggest that either the nest box or the wire cage floor may be preferred locations. Alternatively, by one hen making a choice, this may or may not force the other hen into a less-preferred location. This experiment only used pairs of hens and the possible combinations for preference presumably become more complex in commercial settings of group sizes of 5 to 20 hens, particularly when the number of nest sites (boxes) is limited.
Clearly further research is required to answer the following types of questions: What is the biological significance of following behaviour, including any relationship with nest site selection? Is consistency of nest site selection associated with a preference for that site, or are some birds forced to choose a less-preferred site? Are these behaviours the same with larger group sizes? Are there any implications for welfare?
References
Appleby, M.C. and McRae, H.E. (1986). Applied Animal Behaviour, 15:169-176.
Appleby, M.C., Smith, S.F. and Hughes, B.O. (1993). British Poultry Science, 34:835-847.
Cooper, J.J. and Appleby, M.C. (1995). Applied Animal Behaviour, 42:283-295.
Cooper, J.J. and Appleby, M.C. (1997). Animal Behaviour, 54:1245-1253.
Duncan, I.J.H and Kite, V.G. (1989). Animal Behaviour, 37:115-231.
Freire, R., Appleby, M.C. and Hughes, B.O. (1996). Applied Animal Behaviour, 48:37-46.
Freire, R., Appleby, M.C. and Hughes, B.O. (1997). Animal Behaviour, 54:313-319.
Keeling, L.J. (2004). Welfare of the Laying Hen, CABI, Wallingford (Ed. G.C. Perry) pp. 203-213.
Kite, V.G., Cumming, R.B. and Wodzicka-Tomaszewska, M. (1980). Reviews in Rural Science IV, Uni. of New England, Armidale (Eds. Wodzicka-Tomaszewska, M., Edey, T.N. and Lynch, J.J.) pp. 93-96.
Sherwin, C.M. and Nicol, C.J. (1992). Applied Animal Behaviour, 35:41-54.
Sherwin, C.M. and Nicol, C.J. (1993). Applied Animal Behaviour, 36:211-222.
From Proceedings of the "17th Australian Poultry Science Symposium", New South Wales, Australia.
