ENVIRONMENT, WELL-BEING, AND BEHAVIOR Dustbathing in food particles does not remove feather lipids B. Scholz,1 J. B. Kjaer, S. Petow, and L. Schrader Institute of Animal Welfare and Animal Husbandry, Friedrich-Loeffler-Institut, Doernbergstrasse 25/27, 29223 Celle, Germany allowed pair-wise access to a dustbath for 2.5 h and 3 hens were sampled without access to a dustbathing tray (control). After 10 wk of free access to the dustbathing trays, all hens were sampled again. In trial 2, an additional third sampling was made after dustbaths had been closed again for 6 wk. Here, 6 hens per compartment were sampled immediately before and after a dustbath. Dustbathing in F resulted in higher FLC compared with L and control (P < 0.001), whereas no significant difference was found between L and control (P = 0.103). When open access to litter was provided, hens had higher FLC in F compared with L (P < 0.001). The FLC immediately after dustbathing in F was higher compared with the level before dustbathing (P < 0.001), whereas it was lower after dustbathing in L (P = 0.006). These results show that F are not suitable litter material for laying hens because they lead to lipid accumulation on the plumage.
Key words: dustbathing, food particle, laying hen, feather lipid concentration, welfare 2014 Poultry Science 93:1877–1882 http://dx.doi.org/10.3382/ps.2013-03231
INTRODUCTION The provision of dustbathing material has been recognized as important in laying hen husbandry systems. Since 2012, dustbathing substrate has become compulsory in all types of housing systems for layers within the European Union to allow hens the ability to “satisfy their ethological needs” (Article 2, § 2b, European Union, 1999). The purpose of dustbathing behavior has been examined in a variety of studies. Apart from possibly serving to remove skin parasites (Borchelt et al., 1973), dustbathing in laying hens was found to regulate the amount of feather lipids (Van Liere et al., 1990; Van Liere, 1992a,b), and importantly, to maintain plumage condition (Van Liere and Bokma, 1987; Olsson and Keeling, 2005). It is well known that the uropygial gland is responsible for the production of ©2014 Poultry Science Association Inc. Received April 9, 2013. Accepted April 14, 2014. 1 Corresponding author: [email protected]
preen oil, which is distributed onto the plumage during preening. Additionally, sebaceous secretions from the skin most likely add to lipid contents of the feathers (Sandilands et al., 2004a). Plumage lipid concentration was found to be strongly influenced by the presence or absence of litter material with hens housed on wire having higher feather lipid concentration (FLC) compared with hens kept on litter (Sandilands et al., 2004b). In a study by Van Liere and Bokma (1987), hens prevented from dustbathing showed an increase in lipids on their back feathers, whereas the original level of lipids was restored after they were given access to river sand as litter material. Particularly the removal of stale lipids was found to play a central role in dustbathing duration. Dustbath duration increased after stale lipids had been artificially distributed on breast feathers. This increase was due to higher numbers of side-lying and side-rubbing bouts, which are functionally crucial to remove lipids (Van Liere et al., 1991). To serve the function of lipid removal adequately, the type of dustbathing litter is important. Van Liere
1877
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
ABSTRACT Within the European Union, dustbathing material in cage-housing systems for laying hens became compulsory in 2012. In practice, most producers use food particles as litter substrate. The feed is dropped in small amounts on scratching mats by an automatic transporting system. However, because dustbathing behavior is meant to remove stale lipids from hens’ plumage, food particles may not be a suitable substrate due to their fat content. This study analyzes feather lipid concentration (FLC) of laying hens with access to food particles (F) or lignocellulose (L) as litter substrates. In each of 2 identical trials, 84 laying hens of 2 genotypes (Lohmann Selected Leghorn, Lohmann Brown) were kept in 12 compartments (7 hens each). Compartments were equipped with a grid floor and additionally contained a closed dustbathing tray holding F or L. Feather samples (150 feathers) were taken 2 times throughout the experiment. At 23 wk of age, 4 hens per compartment were sampled after they were
1878
Scholz et al.
MATERIALS AND METHODS Experimental Design In each of 2 experimental trials, 84 laying hens (42 Lohmann Selected Leghorn and 42 Lohmann Brown) were used. Hens were reared in an aviary system (trial 1, access to litter material (chopped straw) from wk 6 to 10; no access to any other litter outside of these times) or cages (trial 2, without access to litter substrate). Hens were transferred to the experimental stable at the age of 19 wk and individually marked with wing badges. In the experimental stable, hens were kept in 12 compartments, which were arranged in a single row.
Within each compartment, hens were kept in groups of 7, randomly selected from the same genotype (6 compartments per genotype). Genotypes were arranged in a random order. Compartments (floor space: 2.0 × 3.0 m, height: 2.0 m) provided 0.86 m2 space per hen. They contained a plastic grid floor, a circular feed trough, a water dispenser with drinking nipples, a perch (100 cm length), and a nest box without litter filling. Hens were offered food (trial 1: 11.0 MJ of ME, 3.40% crude lipid, 16.2% CP, 3.89% Ca, 0.66% P; trial 2: 11.0 MJ of ME, 3.40% crude lipid, 17.0% CP, 4.35% Ca, 0.50% P) and water ad libitum. From wk 22 (trial 1) and wk 24 (trial 2), the light period was 14 h and was set from 0330 to 1730 h. Each compartment contained a litter tray (750 × 800 mm), which was either supplied with food particles (F) or lignocellulose (L; SoftCell, Agromed Austria GmbH, Kremsmünster, Austria). Substrates were randomly assigned to compartments balanced for genotype, and 3 compartments of each genotype received each substrate. The F were identical to the food provided at the feed troughs. An analysis of litter crude lipid content was conducted before the experiment using fresh substrate. A crude lipid content of 3.40% was measured for F and 0.50% for L. Litter substrates were provided at approximately 15 mm height and fully covered the floor of each litter tray. In both trials, the proportions of litter particle sizes of approximately 110 to 120 g of litter material each were examined with help of a particle sieve (according to DIN 4188) at 3 times throughout each trial and data on particle sizes for each substrate were averaged across all 3 samples of the same trial. The first sample per tray was taken immediately before substrates were distributed on the trays (12 samples). Throughout the experiment, a second sample of each tray was taken on the first day after hens were given free access to the litter trays (12 samples) and a third sample immediately before the second feather sampling period (12 samples, 36 samples per trial in total). The density of substrates was estimated by weighing 1 L of fresh particles per substrate (trial 1: L: 516 g/L, F: 695 g/L; trial 2: L: 534 g/L, F: 734 g/L). Throughout the experiment, the filling level of each tray was controlled each time before hens were placed into the particular trays or on a daily basis when free access to substrate was provided. Substrate was added when necessary and soiled litter substrate was removed.
Feather Sampling and Feather Lipid Analysis Before feather sampling was conducted, hens were given a habituation period of approximately 4 wk in which they could adapt to the experimental compartments. Throughout this period, litter trays were closed with a grating, but litter could be seen from the experimental compartments.
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
(1992a) found an effect of litter type on FLC. In his study, hens were offered wood-shavings, sand, or peat as litter substrate and hens on wood-shavings showed higher lipid concentrations compared with those on sand or peat. Wood shavings may not have had a fine enough grain to infiltrate layers’ plumage and absorb lipids. In practice, however, the type of dustbathing litter in cage-systems is often determined by the technical properties of the system. Due to frequent coupling of food chain and litter supply pipe, hens in enriched cages are predominantly offered food as litter substrate, which is automatically supplied in small amounts on scratching mats up to a few times per day. Hens will preferentially dustbathe on substrates other than food particles, even if the alternatives are of a coarser grain (Scholz et al., 2010). It was assumed that the fat content of food may possibly inhibit lipid removal from hens’ plumage and may therefore lead to an avoidance of food as a litter source for dustbathing. When food particles of different experimentally varied lipid contents were provided as litter material, hens most frequently chose the feed with the lowest lipid content for dustbathing behavior (Scholz et al., 2011). Although food particles are already a widely used litter material in cage systems and have gained even more importance from 2012 due to the legal requirements mentioned above, the adequacy of food particles as dustbathing material related to their effect on FLC in laying hens has not been investigated. The aim of our study was to measure, for the first time, the effect of food particles on FLC in laying hens either having access to food particles or lignocellulose as dustbathing material. Lignocellulose is a woodderived material of pelleted lignin, hemicellulose, and cellulose, which is naturally almost fat-free in its composition and is highly preferred for dustbathing behavior compared with wood shavings and food (Scholz et al., 2010). Furthermore, it is used as litter for turkeys (Berk, 2010). We hypothesized that hens had increased FLC when food particles were provided as litter substrate and decreased FLC when they were given access to lignocellulose compared with nondustbathing control hens.
FEATHER LIPID CONCENTRATION IN LAYING HENS
thermore, hens with intact feather cover were excluded from feather lipid analyses. A total of 68 hens [n = 32 (F), n = 36 (L)] were sampled. At each sampling, 150 feathers per hen were taken from 3 defined locations. Fifty feathers were taken from the breast region and 50 feathers were taken from each of both sides lateral to the body. Only apparently clean feathers with a maximum length of 7 cm were carefully cut off at their base and filled in plastic bags. Feather samples were dried in an oven for 15 h at 40°C. Feather lipids were then extracted by the Soxhlet extraction method. After extracting lipids with hexane for at least 6 h (69°C), the solvent was removed with a rotary evaporator. The lipid residue was dried for 6 h (40°C) and put in an exsiccator for a minimum time of 3 h. Lipid residues were weighed and FLC were given as a percentage of dried feather weight.
Video Analysis Each compartment was equipped with a video camera (S/W-CCD-Mini-Camera C3172, ELV Elektronik AG, Leer, Germany) installed at a height of approximately 2 m above the litter tray. Video observations of hens in the litter trays were conducted at sample period 1 (both trials) and sample period 3 (trial 2) to ensure a minimum dustbath duration of 10 min. The beginning of dustbathing behavior was defined when a first vertical wing shake occurred. Single dustbaths ended when a hen exhibited a body wing shake. The durations of all different behavioral patterns within one or more single dustbaths per hen; for example, scratching, bill raking, sitting in substrate, head or side rubbing, and laying on the side were considered to determine the minimum dustbath duration of 10 min within the given time period of 2.5 h. Video recordings were analyzed using Windows Media Player, version 9.0 (Microsoft Corp, Redmond, WA).
Statistical Analysis Data on FLC were analyzed separately for each sampling period. Generalized mixed models (SAS Institute Inc., Cary, NC) were used for sampling period 1 and 2. The initial model included trial (1, 2), substrate (F, L), layer line (Lohmann Selected Leghorn, Lohmann Brown), and all interactions between these variables as fixed factors and compartment and hen within compartment as random factors. Insignificant interactions and variables were removed to produce the final model. Data for sampling periods 1 and 2 were transformed (x0.2) and residuals were tested and found to be normally distributed (Kolmogorov-Smirnoff test). For sampling period 3 (trial 2), the difference between FLC immediately before and after a dustbath in F or L was tested using a paired-difference signed rank test (univariate procedure). Data for sampling period 3 were not transformed.
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
In trial 1, feather sampling was conducted from wk 23 to 25 (sampling period 1) and at wk 36 (sampling period 2). Throughout sampling period 1, samples from all 84 hens were drawn. On 2 consecutive days each, 2 randomly chosen hens of the same compartment were simultaneously placed in the particular litter tray for approximately 2.5 h and allowed to dustbathe. Hens that showed dustbathing behavior for at least 10 min within the time frame given were sampled immediately after they were taken out of the trays before being returned to their conspecifics. The remaining 3 hens per compartment were sampled without prior access to a dustbathing tray. Thus, 4 hens of each compartment were sampled after they had dustbathed and 3 hens per compartment were sampled without access to the litter tray (control). Hens that refused to dustbathe or dustbathed less than 10 min after they were placed in the dustbathing tray were given a second try on the consecutive day and could be sampled then. A hen was sampled once only. After sampling period 1 had been completed, all dustbathing trays were opened for 10 wk. At wk 36, all 84 hens were randomly sampled again (21 hens per day), independent from preceding dustbathing behavior (sampling period 2). In trial 2, with new hens, sampling period 1 was completed during wk 24 to 28. The longer duration of this sampling period compared with trial 1 was due to a higher number of hens refusing to dustbathe on the first day after they had been placed in a particular dustbathing tray. These hens were given a maximum of 2 tries on 2 consecutive days before they were excluded from feather sampling (72 samples in total were taken). Identical to trial 1, 10 wk after hens were given free access to dustbathing trays, sampling period 2 was conducted in wk 39. In contrast to trial 1, at this sampling period, only 76 samples were collected because some hens showed signs of feather loss and were therefore excluded from feather lipid analyses. In trial 2, an additional sampling period 3 (wk 46 to 52) was added. After completion of sampling period 2, all dustbathing trays were closed with a grating again for 6.5 wk. In sampling period 3, 2 feather samples were taken from each of 6 hens per compartment, 2 per day on 3 consecutive days. Only 6 hens per compartment were chosen to slightly reduce the number of samples because each hen was supposed to be sampled twice. One sample was collected immediately before hens were placed in the litter tray; 2 hens were then given simultaneous access to the dustbathing tray for approximately 2.5 h and allowed to dustbathe. Feather samples of the same hens were taken again after they had shown dustbathing behavior for at least 10 min within the given time frame. Thus, feather samples were drawn immediately before and after hens had shown dustbathing behavior. Hens refusing to dustbathe or dustbathing for less than 10 min were not given a second try on the consecutive day but were excluded from feather sampling analysis. Fur-
1879
1880
1.18a 0.01 2.28 1.08b 0.01 1.45 — — — 1.13c 0.01 1.84
1.15deh 0.01 2.01 1.05bg 0.02 1.26 1.05bc 0.01 1.25
RESULTS
1.04d 0.01 1.24
1.21a 0.03 2.59 1.13df 0.03 1.82
F2 L2 C2 F1 L1
Substrate × trial
Figure 1. Mean percentage of particles in different size fractions (mm) for food particles (Food) and lignocellulose (Ligno) in trial 1 and 2.
1.08b 0.01 1.47
1.13a 0.01 1.83
1.17a 0.02 2.19
1.09b 0.01 1.52
— — —
2
1.17ae 0.02 2.22
Results on litter particle size analyses are shown in Figure 1. The F had higher proportions of finer grained particle sizes compared with L. Grain sizes of 1.6 mm accounted for the highest proportion of grain sizes within F, whereas particle sizes of 3.2 mm took on the highest proportion within L. Both F and L appeared to have higher proportions of larger grained particles in trial 1 compared with trial 2.
1LSM
letters in a row-category indicate significant difference (P < 0.05). are based on transformed data and additionally given in back transformed means (btM). 2C = control group; L = lignocellulose; F = food particles. 3LB = Lohmann Brown; LSL = Lohmann Selected Leghorn.
Sampling Period 1
a–hDifferent
1.11 0.01 1.66 1.16a 0.01 2.06 1.06b 0.01 1.34 — — —
1.11 0.01 1.67
1.13 0.02 1.81 1.12 0.01 1.80 1.18a 0.02 2.28 1.09b 0.02 1.52 1.11b 0.01 1.68
Sampling period 1 LSM SE of LSM btM Sampling period 2 LSM SE of LSM btM
LSL LB L C Feather lipid concentration
Substrate2
F
Genotype3
Trial 1
C1
Litter Particle Sizes
In sampling period 1, when hens were sampled after dustbathing behavior in a particular tray or without access to substrate, FLC was influenced by substrate (F2,149 = 19.02, P < 0.001), trial (F1,33 = 20.27, P < 0.001), the interaction of trial × substrate (F2,148 = 4.43, P = 0.014), and trial × layer line (F1,73 = 8.00, P = 0.006, data not shown; Table 1). The FLC was higher in hens with access to F compared with control hens (P < 0.001) and hens on L (P < 0.001), whereas L hens tended to have lower FLC compared with control (P = 0.103). The FLC was higher in trial 1 compared with trial 2 (P < 0.001). When analyzing substrate effects within trials it was evident that hens tended to have (trial 1) or had (trial 2) higher FLC after dustbaths in F compared with the control hens (P = 0.055 and P < 0.001, respectively). Hens with access to L had lower FLC compared with the control group (P = 0.015) in trial 1, whereas no difference was found in trial 2 (P = 0.925; Figure 2). In both trials, hens had higher FLC after dustbathing in F compared with L (P = 0.001 in trial 1 and P < 0.001 in trial 2). No difference in FLC was found between genotypes (F1,65 = 0.01, P = 0.923).
Sampling Period 2 No model reduction was done for sampling period 2 because the 3-way interaction was significant. In sampling period 2, when free access to the trays had been
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
Table 1. Least squares means (LSM) of feather lipid concentration (percentage of dried feather weight) as affected by substrate, genotype, trial, and interaction between substrate and trial of sampling periods 1 and 21
Scholz et al.
FEATHER LIPID CONCENTRATION IN LAYING HENS
provided for 10 wk, FLC was affected by substrate (F1,77 = 101.64, P < 0.001), trial (F1,145 = 43.74, P < 0.001; Table 1), and interactions trial × layer line (F1,49 = 12.82, P = 0.001), substrate × layer line (F1,129 = 5.69, P = 0.019), and substrate × trial × layer line (F1,130 = 19.46, P < 0.001; data not shown). Hens with free access to F as litter substrate had higher FLC compared with those having L (P < 0.001). The FLC was lower in trial 1 compared with trial 2 (P < 0.001).
Sampling Period 3 Hens had higher FLC immediately after, compared with before, a dustbath in F [median 1.26 %, Q1 (25th percentile) = 1.04 %, Q3 (75th percentile) = 1.58 %, S = 263, P < 0.001]. Hens with access to L had lower FLC immediately after a dustbath compared with before (median −0.11 %, Q1 = −0.22 %, Q3 = 0.01 %, S = −170, P = 0.006).
DISCUSSION In sampling period 1, hens were sampled in close temporal proximity to dustbathing behavior in either F or L, thus illustrating short-term effects of the particular substrates on their FLC. The FLC was significantly higher in F compared with L and the control group, whereas FLC only tended to be lower in L compared with the control. Analysis of the interaction between substrate and trial revealed that FLC was in general lower in the second trial compared with the first, so L might have had a less decisive effect because there was less FLC to be removed. Also, the analysis of particle sizes indicates a coarser structure of L in the second trial, possibly reducing its capacity for absorbing and removing lipids.
In sampling period 2, free access to either F or L was provided over a time period of 10 wk, thus giving information on long-term effects of the particular substrates on FLC. Also, without a direct temporal connection to preceding dustbathing behavior, hens on F had higher FLC compared with hens with access to L. When food particles are used as litter substrate in cage systems, they might have a negative long-term effect on plumage quality and thermo-insulatory property, thus resulting in poorer welfare. In sampling period 3, a more precise method to determine short-term differences in FLC was used. The FLC status was measured before and after a dustbath for each hen, thus accounting for individual variation between hens. Results clearly suggest that F is not a suitable dustbathing substrate because FLC was higher immediately after a dustbath in F compared with the status before. In contrast to F, L reduced lipid concentration, thus supporting our hypotheses. According to Van Liere and Bokma (1987) and Van Liere (1992a), dustbathing not only regulates the level of FLC but also affects the quality of hens’ down feathers, and interestingly, even very small differences in feather lipid content were found to have biological impacts on the hen. Van Liere and Bokma (1987) found that marginal differences between feather lipid contents [10.3 mg per gram of feathers (1.03%) versus 14.5 mg per gram of feathers (1.45%)] significantly reduced the fluffiness of back feather down. Furthermore, Van Liere (1992a) could show that hens on wood shavings had higher FLC and decreased fluffiness of down feathers compared with hens on sand or peat. Van Liere (1992a) also reported an increase in plumage surface temperature in hens kept on wood shavings, thus suggesting that impaired quality of down feathers may also lead to higher energy loss due to inferior thermo-insulation. In the present study, differences in lipid concentrations also appeared to be minimal although statistically significant. However, against the background of the findings mentioned above, these differences can be assumed to have a biological effect on the hens. The FLC of hens dustbathing in F was higher (2.28% at sample period 1, 2.06% at sample period 2, back transformed means) than the FLC (1.45%) reported by Van Liere and Bokma (1987), thus suggesting a negative effect on plumage condition. Additionally, Zeman (1988) found that lipid components from preen glands together with natural extract of surface lipids can serve as a stimulant for the red mite. Likewise, Van Liere (1992a) stated that hydrophobic metabolites resulting either from vertebrate skin or microflora of the skin can attract ectoparasites. Martin and Mullens (2012) tested the effect of sand enriched with naturally available dust materials as litter substrate on the prevalence of northern fowl mite infestation in laying hens. Hens dustbathing in sand had lower levels of mite infestation compared with nondustbathing hens and hens deprived of sand. However,
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
Figure 2. Back transformed means of plumage lipid concentration (percent of dried feather weight) including both layer lines (LSL, LB) after dustbathing in lignocellulose, food particles, and without dustbathing (control). Different letters (a,b) within a trial indicate significant differences (P < 0.05). LSL: Lohmann Selected Leghorn; LB: Lohmann Brown.
1881
1882
Scholz et al.
ACKNOWLEDGMENTS The authors thank Ute Lenert and Karl-Heinz Kiemann (Institute of Animal Welfare and Animal Hus-
bandry, Friedrich-Loeffler-Institut, Celle, Germany) for their support.
REFERENCES Berk, J. 2010. Effect of litter type on health, performance and air quality in a forced ventilated turkey house. Page 43 in Proc. 8th International Symposium on Turkey Diseases, Berlin, Germany. World Veterinary Poultry Association and Institute of Poultry Diseases, Free University, Berlin, Germany. Borchelt, P. L., J. Eyer, and D. S. McHenry Jr. 1973. Dustbathing in Bobwhite quail (Colinus virginianus) as a function of dust deprivation. Behav. Biol. 8:109–114. European Union. 1999. Council Directive 1999/74/EC. Laying down minimum standards for the protection of laying hens. Off. J. L 203:54–55. Martin, C. D., and B. A. Mullens. 2012. Housing and dustbathing effects on northern fowl mites (Ornithonyssus sylviarum) and chicken body lice (Menacanthus stramineus) on hens. Med. Vet. Entomol. 26:323–333. Olsson, I. A. S., and L. J. Keeling. 2005. Why in earth? Dustbathing in jungle and domestic fowl reviewed from a Tinbergian and animal welfare perspective. Appl. Anim. Behav. Sci. 93:259–282. Sandilands, V., K. Powell, L. Keeling, and C. J. Savory. 2004a. Preen gland function in layer fowls: Factors affecting preen oil fatty acid composition. Br. Poult. Sci. 45:109–115. Sandilands, V., J. Savory, and K. Powell. 2004b. Preen gland function in layer fowls: Factors affecting morphology and feather lipid levels. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 137:217–225. Scholz, B., J. B. Kjaer, S. Urselmans, and L. Schrader. 2011. Litter lipid content affects dustbathing behaviour in laying hens. Poult. Sci. 90:2433–2439. Scholz, B., S. Urselmans, J. B. Kjaer, and L. Schrader. 2010. Food, wood or plastic as substrates for dustbathing in laying hens: A preference test. Poult. Sci. 89:1584–1589. Sparagano, O., A. Pavlićević, T. Murano, A. Camarda, H. Sahibi, O. Kilpinen, M. Mul, R. Van Emous, S. Le Bouquin, K. Hoel, and M. A. Cafiereo. 2009. Prevalence and key figures for the poultry red mite Dermanyssus gallinae infections in poultry farm systems. Exp. Appl. Acarol. 48:3–10. Van Liere, D. W. 1992a. The significance of fowls’ bathing in dust. Anim. Welf. 1:187–202. Van Liere, D. W. 1992b. Dustbathing as related to proximal and distal feather lipids in laying hens. Behav. Process. 26:177–188. Van Liere, D. W., S. E. Aggrey, F. M. R. Brouns, and P. R. Wiepkema. 1991. Oiling behaviour and the effect of lipids on dustbathing behaviour in laying hens (Gallus gallus domesticus). Behav. Processes 24:71–81. Van Liere, D. W., and S. Bokma. 1987. Short-term feather maintenance as a function of dust-bathing in laying hens. Appl. Anim. Behav. Sci. 18:197–204. Van Liere, D. W., J. Kooijman, and P. R. Wiepkema. 1990. Dustbathing behaviour of laying hens as related to quality of dustbathing material. Appl. Anim. Behav. Sci. 26:127–141. Zeman, P. 1988. Surface skin lipids of birds—A proper host kairomone and feeding inducer in the poultry red mite, Dermanyssus gallinae. Exp. Appl. Acarol. 5:163–173.
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
dustbathing in the litter, which became a mixture of straw particles mixed with manure and feed, had no effect on ectoparasite infestation and mite populations were similar to the numbers of mites in hens without a dustbox. Consequently, the effect of feed on mite infestation within this “litter substrate mixture” remains unclear. Food particles do not remove lipids from the plumage and may even stimulate red-mite infestation, thus causing poorer welfare. Furthermore, red mite infestation has been described to cause economic losses (Sparagano et al., 2009). The relatively small amounts of F provided as litter in practice may not necessarily lead to further plumage lipid accumulation but may simply not affect FLC because they are quickly used up. In the present study, both substrates were available in sufficient amounts as the floor of each tray was fully covered when offered for dustbathing. Only a small amount of substrate could be removed from the trays by dustbathing or scratching behavior. Further research is needed to determine the minimum amount of suitable dustbathing material needed to enable the removal of stale lipids from the plumage. Food particles did not serve to remove feather lipids, whereas L was found to reduce FLC in the amount given here. Using L as litter material in cage systems might induce technical problems. The coarser structure of L compared with F may be problematic for an automatic supply of substrate due to flow characteristics. Also, coarser particles may easily block the passage holes of the supply pipe, thus preventing particles from dropping onto the scratching mats. Finer grinding, however, may increase dust emission. In conclusion, the present study showed that dustbathing in F increased feather lipid content, whereas dustbathing in L decreased lipid content. Therefore, feed is not a suitable dustbathing material for laying hens. Because F are widely used in cage housing systems for laying hens, these results are quite alarming and priority should be given to the development of automatic litter dispensing systems for litter types more suitable as substrates for dustbathing.
Key words: dustbathing, food particle, laying hen, feather lipid concentration, welfare 2014 Poultry Science 93:1877–1882 http://dx.doi.org/10.3382/ps.2013-03231
INTRODUCTION The provision of dustbathing material has been recognized as important in laying hen husbandry systems. Since 2012, dustbathing substrate has become compulsory in all types of housing systems for layers within the European Union to allow hens the ability to “satisfy their ethological needs” (Article 2, § 2b, European Union, 1999). The purpose of dustbathing behavior has been examined in a variety of studies. Apart from possibly serving to remove skin parasites (Borchelt et al., 1973), dustbathing in laying hens was found to regulate the amount of feather lipids (Van Liere et al., 1990; Van Liere, 1992a,b), and importantly, to maintain plumage condition (Van Liere and Bokma, 1987; Olsson and Keeling, 2005). It is well known that the uropygial gland is responsible for the production of ©2014 Poultry Science Association Inc. Received April 9, 2013. Accepted April 14, 2014. 1 Corresponding author: [email protected]
preen oil, which is distributed onto the plumage during preening. Additionally, sebaceous secretions from the skin most likely add to lipid contents of the feathers (Sandilands et al., 2004a). Plumage lipid concentration was found to be strongly influenced by the presence or absence of litter material with hens housed on wire having higher feather lipid concentration (FLC) compared with hens kept on litter (Sandilands et al., 2004b). In a study by Van Liere and Bokma (1987), hens prevented from dustbathing showed an increase in lipids on their back feathers, whereas the original level of lipids was restored after they were given access to river sand as litter material. Particularly the removal of stale lipids was found to play a central role in dustbathing duration. Dustbath duration increased after stale lipids had been artificially distributed on breast feathers. This increase was due to higher numbers of side-lying and side-rubbing bouts, which are functionally crucial to remove lipids (Van Liere et al., 1991). To serve the function of lipid removal adequately, the type of dustbathing litter is important. Van Liere
1877
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
ABSTRACT Within the European Union, dustbathing material in cage-housing systems for laying hens became compulsory in 2012. In practice, most producers use food particles as litter substrate. The feed is dropped in small amounts on scratching mats by an automatic transporting system. However, because dustbathing behavior is meant to remove stale lipids from hens’ plumage, food particles may not be a suitable substrate due to their fat content. This study analyzes feather lipid concentration (FLC) of laying hens with access to food particles (F) or lignocellulose (L) as litter substrates. In each of 2 identical trials, 84 laying hens of 2 genotypes (Lohmann Selected Leghorn, Lohmann Brown) were kept in 12 compartments (7 hens each). Compartments were equipped with a grid floor and additionally contained a closed dustbathing tray holding F or L. Feather samples (150 feathers) were taken 2 times throughout the experiment. At 23 wk of age, 4 hens per compartment were sampled after they were
1878
Scholz et al.
MATERIALS AND METHODS Experimental Design In each of 2 experimental trials, 84 laying hens (42 Lohmann Selected Leghorn and 42 Lohmann Brown) were used. Hens were reared in an aviary system (trial 1, access to litter material (chopped straw) from wk 6 to 10; no access to any other litter outside of these times) or cages (trial 2, without access to litter substrate). Hens were transferred to the experimental stable at the age of 19 wk and individually marked with wing badges. In the experimental stable, hens were kept in 12 compartments, which were arranged in a single row.
Within each compartment, hens were kept in groups of 7, randomly selected from the same genotype (6 compartments per genotype). Genotypes were arranged in a random order. Compartments (floor space: 2.0 × 3.0 m, height: 2.0 m) provided 0.86 m2 space per hen. They contained a plastic grid floor, a circular feed trough, a water dispenser with drinking nipples, a perch (100 cm length), and a nest box without litter filling. Hens were offered food (trial 1: 11.0 MJ of ME, 3.40% crude lipid, 16.2% CP, 3.89% Ca, 0.66% P; trial 2: 11.0 MJ of ME, 3.40% crude lipid, 17.0% CP, 4.35% Ca, 0.50% P) and water ad libitum. From wk 22 (trial 1) and wk 24 (trial 2), the light period was 14 h and was set from 0330 to 1730 h. Each compartment contained a litter tray (750 × 800 mm), which was either supplied with food particles (F) or lignocellulose (L; SoftCell, Agromed Austria GmbH, Kremsmünster, Austria). Substrates were randomly assigned to compartments balanced for genotype, and 3 compartments of each genotype received each substrate. The F were identical to the food provided at the feed troughs. An analysis of litter crude lipid content was conducted before the experiment using fresh substrate. A crude lipid content of 3.40% was measured for F and 0.50% for L. Litter substrates were provided at approximately 15 mm height and fully covered the floor of each litter tray. In both trials, the proportions of litter particle sizes of approximately 110 to 120 g of litter material each were examined with help of a particle sieve (according to DIN 4188) at 3 times throughout each trial and data on particle sizes for each substrate were averaged across all 3 samples of the same trial. The first sample per tray was taken immediately before substrates were distributed on the trays (12 samples). Throughout the experiment, a second sample of each tray was taken on the first day after hens were given free access to the litter trays (12 samples) and a third sample immediately before the second feather sampling period (12 samples, 36 samples per trial in total). The density of substrates was estimated by weighing 1 L of fresh particles per substrate (trial 1: L: 516 g/L, F: 695 g/L; trial 2: L: 534 g/L, F: 734 g/L). Throughout the experiment, the filling level of each tray was controlled each time before hens were placed into the particular trays or on a daily basis when free access to substrate was provided. Substrate was added when necessary and soiled litter substrate was removed.
Feather Sampling and Feather Lipid Analysis Before feather sampling was conducted, hens were given a habituation period of approximately 4 wk in which they could adapt to the experimental compartments. Throughout this period, litter trays were closed with a grating, but litter could be seen from the experimental compartments.
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
(1992a) found an effect of litter type on FLC. In his study, hens were offered wood-shavings, sand, or peat as litter substrate and hens on wood-shavings showed higher lipid concentrations compared with those on sand or peat. Wood shavings may not have had a fine enough grain to infiltrate layers’ plumage and absorb lipids. In practice, however, the type of dustbathing litter in cage-systems is often determined by the technical properties of the system. Due to frequent coupling of food chain and litter supply pipe, hens in enriched cages are predominantly offered food as litter substrate, which is automatically supplied in small amounts on scratching mats up to a few times per day. Hens will preferentially dustbathe on substrates other than food particles, even if the alternatives are of a coarser grain (Scholz et al., 2010). It was assumed that the fat content of food may possibly inhibit lipid removal from hens’ plumage and may therefore lead to an avoidance of food as a litter source for dustbathing. When food particles of different experimentally varied lipid contents were provided as litter material, hens most frequently chose the feed with the lowest lipid content for dustbathing behavior (Scholz et al., 2011). Although food particles are already a widely used litter material in cage systems and have gained even more importance from 2012 due to the legal requirements mentioned above, the adequacy of food particles as dustbathing material related to their effect on FLC in laying hens has not been investigated. The aim of our study was to measure, for the first time, the effect of food particles on FLC in laying hens either having access to food particles or lignocellulose as dustbathing material. Lignocellulose is a woodderived material of pelleted lignin, hemicellulose, and cellulose, which is naturally almost fat-free in its composition and is highly preferred for dustbathing behavior compared with wood shavings and food (Scholz et al., 2010). Furthermore, it is used as litter for turkeys (Berk, 2010). We hypothesized that hens had increased FLC when food particles were provided as litter substrate and decreased FLC when they were given access to lignocellulose compared with nondustbathing control hens.
FEATHER LIPID CONCENTRATION IN LAYING HENS
thermore, hens with intact feather cover were excluded from feather lipid analyses. A total of 68 hens [n = 32 (F), n = 36 (L)] were sampled. At each sampling, 150 feathers per hen were taken from 3 defined locations. Fifty feathers were taken from the breast region and 50 feathers were taken from each of both sides lateral to the body. Only apparently clean feathers with a maximum length of 7 cm were carefully cut off at their base and filled in plastic bags. Feather samples were dried in an oven for 15 h at 40°C. Feather lipids were then extracted by the Soxhlet extraction method. After extracting lipids with hexane for at least 6 h (69°C), the solvent was removed with a rotary evaporator. The lipid residue was dried for 6 h (40°C) and put in an exsiccator for a minimum time of 3 h. Lipid residues were weighed and FLC were given as a percentage of dried feather weight.
Video Analysis Each compartment was equipped with a video camera (S/W-CCD-Mini-Camera C3172, ELV Elektronik AG, Leer, Germany) installed at a height of approximately 2 m above the litter tray. Video observations of hens in the litter trays were conducted at sample period 1 (both trials) and sample period 3 (trial 2) to ensure a minimum dustbath duration of 10 min. The beginning of dustbathing behavior was defined when a first vertical wing shake occurred. Single dustbaths ended when a hen exhibited a body wing shake. The durations of all different behavioral patterns within one or more single dustbaths per hen; for example, scratching, bill raking, sitting in substrate, head or side rubbing, and laying on the side were considered to determine the minimum dustbath duration of 10 min within the given time period of 2.5 h. Video recordings were analyzed using Windows Media Player, version 9.0 (Microsoft Corp, Redmond, WA).
Statistical Analysis Data on FLC were analyzed separately for each sampling period. Generalized mixed models (SAS Institute Inc., Cary, NC) were used for sampling period 1 and 2. The initial model included trial (1, 2), substrate (F, L), layer line (Lohmann Selected Leghorn, Lohmann Brown), and all interactions between these variables as fixed factors and compartment and hen within compartment as random factors. Insignificant interactions and variables were removed to produce the final model. Data for sampling periods 1 and 2 were transformed (x0.2) and residuals were tested and found to be normally distributed (Kolmogorov-Smirnoff test). For sampling period 3 (trial 2), the difference between FLC immediately before and after a dustbath in F or L was tested using a paired-difference signed rank test (univariate procedure). Data for sampling period 3 were not transformed.
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
In trial 1, feather sampling was conducted from wk 23 to 25 (sampling period 1) and at wk 36 (sampling period 2). Throughout sampling period 1, samples from all 84 hens were drawn. On 2 consecutive days each, 2 randomly chosen hens of the same compartment were simultaneously placed in the particular litter tray for approximately 2.5 h and allowed to dustbathe. Hens that showed dustbathing behavior for at least 10 min within the time frame given were sampled immediately after they were taken out of the trays before being returned to their conspecifics. The remaining 3 hens per compartment were sampled without prior access to a dustbathing tray. Thus, 4 hens of each compartment were sampled after they had dustbathed and 3 hens per compartment were sampled without access to the litter tray (control). Hens that refused to dustbathe or dustbathed less than 10 min after they were placed in the dustbathing tray were given a second try on the consecutive day and could be sampled then. A hen was sampled once only. After sampling period 1 had been completed, all dustbathing trays were opened for 10 wk. At wk 36, all 84 hens were randomly sampled again (21 hens per day), independent from preceding dustbathing behavior (sampling period 2). In trial 2, with new hens, sampling period 1 was completed during wk 24 to 28. The longer duration of this sampling period compared with trial 1 was due to a higher number of hens refusing to dustbathe on the first day after they had been placed in a particular dustbathing tray. These hens were given a maximum of 2 tries on 2 consecutive days before they were excluded from feather sampling (72 samples in total were taken). Identical to trial 1, 10 wk after hens were given free access to dustbathing trays, sampling period 2 was conducted in wk 39. In contrast to trial 1, at this sampling period, only 76 samples were collected because some hens showed signs of feather loss and were therefore excluded from feather lipid analyses. In trial 2, an additional sampling period 3 (wk 46 to 52) was added. After completion of sampling period 2, all dustbathing trays were closed with a grating again for 6.5 wk. In sampling period 3, 2 feather samples were taken from each of 6 hens per compartment, 2 per day on 3 consecutive days. Only 6 hens per compartment were chosen to slightly reduce the number of samples because each hen was supposed to be sampled twice. One sample was collected immediately before hens were placed in the litter tray; 2 hens were then given simultaneous access to the dustbathing tray for approximately 2.5 h and allowed to dustbathe. Feather samples of the same hens were taken again after they had shown dustbathing behavior for at least 10 min within the given time frame. Thus, feather samples were drawn immediately before and after hens had shown dustbathing behavior. Hens refusing to dustbathe or dustbathing for less than 10 min were not given a second try on the consecutive day but were excluded from feather sampling analysis. Fur-
1879
1880
1.18a 0.01 2.28 1.08b 0.01 1.45 — — — 1.13c 0.01 1.84
1.15deh 0.01 2.01 1.05bg 0.02 1.26 1.05bc 0.01 1.25
RESULTS
1.04d 0.01 1.24
1.21a 0.03 2.59 1.13df 0.03 1.82
F2 L2 C2 F1 L1
Substrate × trial
Figure 1. Mean percentage of particles in different size fractions (mm) for food particles (Food) and lignocellulose (Ligno) in trial 1 and 2.
1.08b 0.01 1.47
1.13a 0.01 1.83
1.17a 0.02 2.19
1.09b 0.01 1.52
— — —
2
1.17ae 0.02 2.22
Results on litter particle size analyses are shown in Figure 1. The F had higher proportions of finer grained particle sizes compared with L. Grain sizes of 1.6 mm accounted for the highest proportion of grain sizes within F, whereas particle sizes of 3.2 mm took on the highest proportion within L. Both F and L appeared to have higher proportions of larger grained particles in trial 1 compared with trial 2.
1LSM
letters in a row-category indicate significant difference (P < 0.05). are based on transformed data and additionally given in back transformed means (btM). 2C = control group; L = lignocellulose; F = food particles. 3LB = Lohmann Brown; LSL = Lohmann Selected Leghorn.
Sampling Period 1
a–hDifferent
1.11 0.01 1.66 1.16a 0.01 2.06 1.06b 0.01 1.34 — — —
1.11 0.01 1.67
1.13 0.02 1.81 1.12 0.01 1.80 1.18a 0.02 2.28 1.09b 0.02 1.52 1.11b 0.01 1.68
Sampling period 1 LSM SE of LSM btM Sampling period 2 LSM SE of LSM btM
LSL LB L C Feather lipid concentration
Substrate2
F
Genotype3
Trial 1
C1
Litter Particle Sizes
In sampling period 1, when hens were sampled after dustbathing behavior in a particular tray or without access to substrate, FLC was influenced by substrate (F2,149 = 19.02, P < 0.001), trial (F1,33 = 20.27, P < 0.001), the interaction of trial × substrate (F2,148 = 4.43, P = 0.014), and trial × layer line (F1,73 = 8.00, P = 0.006, data not shown; Table 1). The FLC was higher in hens with access to F compared with control hens (P < 0.001) and hens on L (P < 0.001), whereas L hens tended to have lower FLC compared with control (P = 0.103). The FLC was higher in trial 1 compared with trial 2 (P < 0.001). When analyzing substrate effects within trials it was evident that hens tended to have (trial 1) or had (trial 2) higher FLC after dustbaths in F compared with the control hens (P = 0.055 and P < 0.001, respectively). Hens with access to L had lower FLC compared with the control group (P = 0.015) in trial 1, whereas no difference was found in trial 2 (P = 0.925; Figure 2). In both trials, hens had higher FLC after dustbathing in F compared with L (P = 0.001 in trial 1 and P < 0.001 in trial 2). No difference in FLC was found between genotypes (F1,65 = 0.01, P = 0.923).
Sampling Period 2 No model reduction was done for sampling period 2 because the 3-way interaction was significant. In sampling period 2, when free access to the trays had been
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
Table 1. Least squares means (LSM) of feather lipid concentration (percentage of dried feather weight) as affected by substrate, genotype, trial, and interaction between substrate and trial of sampling periods 1 and 21
Scholz et al.
FEATHER LIPID CONCENTRATION IN LAYING HENS
provided for 10 wk, FLC was affected by substrate (F1,77 = 101.64, P < 0.001), trial (F1,145 = 43.74, P < 0.001; Table 1), and interactions trial × layer line (F1,49 = 12.82, P = 0.001), substrate × layer line (F1,129 = 5.69, P = 0.019), and substrate × trial × layer line (F1,130 = 19.46, P < 0.001; data not shown). Hens with free access to F as litter substrate had higher FLC compared with those having L (P < 0.001). The FLC was lower in trial 1 compared with trial 2 (P < 0.001).
Sampling Period 3 Hens had higher FLC immediately after, compared with before, a dustbath in F [median 1.26 %, Q1 (25th percentile) = 1.04 %, Q3 (75th percentile) = 1.58 %, S = 263, P < 0.001]. Hens with access to L had lower FLC immediately after a dustbath compared with before (median −0.11 %, Q1 = −0.22 %, Q3 = 0.01 %, S = −170, P = 0.006).
DISCUSSION In sampling period 1, hens were sampled in close temporal proximity to dustbathing behavior in either F or L, thus illustrating short-term effects of the particular substrates on their FLC. The FLC was significantly higher in F compared with L and the control group, whereas FLC only tended to be lower in L compared with the control. Analysis of the interaction between substrate and trial revealed that FLC was in general lower in the second trial compared with the first, so L might have had a less decisive effect because there was less FLC to be removed. Also, the analysis of particle sizes indicates a coarser structure of L in the second trial, possibly reducing its capacity for absorbing and removing lipids.
In sampling period 2, free access to either F or L was provided over a time period of 10 wk, thus giving information on long-term effects of the particular substrates on FLC. Also, without a direct temporal connection to preceding dustbathing behavior, hens on F had higher FLC compared with hens with access to L. When food particles are used as litter substrate in cage systems, they might have a negative long-term effect on plumage quality and thermo-insulatory property, thus resulting in poorer welfare. In sampling period 3, a more precise method to determine short-term differences in FLC was used. The FLC status was measured before and after a dustbath for each hen, thus accounting for individual variation between hens. Results clearly suggest that F is not a suitable dustbathing substrate because FLC was higher immediately after a dustbath in F compared with the status before. In contrast to F, L reduced lipid concentration, thus supporting our hypotheses. According to Van Liere and Bokma (1987) and Van Liere (1992a), dustbathing not only regulates the level of FLC but also affects the quality of hens’ down feathers, and interestingly, even very small differences in feather lipid content were found to have biological impacts on the hen. Van Liere and Bokma (1987) found that marginal differences between feather lipid contents [10.3 mg per gram of feathers (1.03%) versus 14.5 mg per gram of feathers (1.45%)] significantly reduced the fluffiness of back feather down. Furthermore, Van Liere (1992a) could show that hens on wood shavings had higher FLC and decreased fluffiness of down feathers compared with hens on sand or peat. Van Liere (1992a) also reported an increase in plumage surface temperature in hens kept on wood shavings, thus suggesting that impaired quality of down feathers may also lead to higher energy loss due to inferior thermo-insulation. In the present study, differences in lipid concentrations also appeared to be minimal although statistically significant. However, against the background of the findings mentioned above, these differences can be assumed to have a biological effect on the hens. The FLC of hens dustbathing in F was higher (2.28% at sample period 1, 2.06% at sample period 2, back transformed means) than the FLC (1.45%) reported by Van Liere and Bokma (1987), thus suggesting a negative effect on plumage condition. Additionally, Zeman (1988) found that lipid components from preen glands together with natural extract of surface lipids can serve as a stimulant for the red mite. Likewise, Van Liere (1992a) stated that hydrophobic metabolites resulting either from vertebrate skin or microflora of the skin can attract ectoparasites. Martin and Mullens (2012) tested the effect of sand enriched with naturally available dust materials as litter substrate on the prevalence of northern fowl mite infestation in laying hens. Hens dustbathing in sand had lower levels of mite infestation compared with nondustbathing hens and hens deprived of sand. However,
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
Figure 2. Back transformed means of plumage lipid concentration (percent of dried feather weight) including both layer lines (LSL, LB) after dustbathing in lignocellulose, food particles, and without dustbathing (control). Different letters (a,b) within a trial indicate significant differences (P < 0.05). LSL: Lohmann Selected Leghorn; LB: Lohmann Brown.
1881
1882
Scholz et al.
ACKNOWLEDGMENTS The authors thank Ute Lenert and Karl-Heinz Kiemann (Institute of Animal Welfare and Animal Hus-
bandry, Friedrich-Loeffler-Institut, Celle, Germany) for their support.
REFERENCES Berk, J. 2010. Effect of litter type on health, performance and air quality in a forced ventilated turkey house. Page 43 in Proc. 8th International Symposium on Turkey Diseases, Berlin, Germany. World Veterinary Poultry Association and Institute of Poultry Diseases, Free University, Berlin, Germany. Borchelt, P. L., J. Eyer, and D. S. McHenry Jr. 1973. Dustbathing in Bobwhite quail (Colinus virginianus) as a function of dust deprivation. Behav. Biol. 8:109–114. European Union. 1999. Council Directive 1999/74/EC. Laying down minimum standards for the protection of laying hens. Off. J. L 203:54–55. Martin, C. D., and B. A. Mullens. 2012. Housing and dustbathing effects on northern fowl mites (Ornithonyssus sylviarum) and chicken body lice (Menacanthus stramineus) on hens. Med. Vet. Entomol. 26:323–333. Olsson, I. A. S., and L. J. Keeling. 2005. Why in earth? Dustbathing in jungle and domestic fowl reviewed from a Tinbergian and animal welfare perspective. Appl. Anim. Behav. Sci. 93:259–282. Sandilands, V., K. Powell, L. Keeling, and C. J. Savory. 2004a. Preen gland function in layer fowls: Factors affecting preen oil fatty acid composition. Br. Poult. Sci. 45:109–115. Sandilands, V., J. Savory, and K. Powell. 2004b. Preen gland function in layer fowls: Factors affecting morphology and feather lipid levels. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 137:217–225. Scholz, B., J. B. Kjaer, S. Urselmans, and L. Schrader. 2011. Litter lipid content affects dustbathing behaviour in laying hens. Poult. Sci. 90:2433–2439. Scholz, B., S. Urselmans, J. B. Kjaer, and L. Schrader. 2010. Food, wood or plastic as substrates for dustbathing in laying hens: A preference test. Poult. Sci. 89:1584–1589. Sparagano, O., A. Pavlićević, T. Murano, A. Camarda, H. Sahibi, O. Kilpinen, M. Mul, R. Van Emous, S. Le Bouquin, K. Hoel, and M. A. Cafiereo. 2009. Prevalence and key figures for the poultry red mite Dermanyssus gallinae infections in poultry farm systems. Exp. Appl. Acarol. 48:3–10. Van Liere, D. W. 1992a. The significance of fowls’ bathing in dust. Anim. Welf. 1:187–202. Van Liere, D. W. 1992b. Dustbathing as related to proximal and distal feather lipids in laying hens. Behav. Process. 26:177–188. Van Liere, D. W., S. E. Aggrey, F. M. R. Brouns, and P. R. Wiepkema. 1991. Oiling behaviour and the effect of lipids on dustbathing behaviour in laying hens (Gallus gallus domesticus). Behav. Processes 24:71–81. Van Liere, D. W., and S. Bokma. 1987. Short-term feather maintenance as a function of dust-bathing in laying hens. Appl. Anim. Behav. Sci. 18:197–204. Van Liere, D. W., J. Kooijman, and P. R. Wiepkema. 1990. Dustbathing behaviour of laying hens as related to quality of dustbathing material. Appl. Anim. Behav. Sci. 26:127–141. Zeman, P. 1988. Surface skin lipids of birds—A proper host kairomone and feeding inducer in the poultry red mite, Dermanyssus gallinae. Exp. Appl. Acarol. 5:163–173.
Downloaded from http://ps.oxfordjournals.org/ at University of Southern Queensland on June 20, 2015
dustbathing in the litter, which became a mixture of straw particles mixed with manure and feed, had no effect on ectoparasite infestation and mite populations were similar to the numbers of mites in hens without a dustbox. Consequently, the effect of feed on mite infestation within this “litter substrate mixture” remains unclear. Food particles do not remove lipids from the plumage and may even stimulate red-mite infestation, thus causing poorer welfare. Furthermore, red mite infestation has been described to cause economic losses (Sparagano et al., 2009). The relatively small amounts of F provided as litter in practice may not necessarily lead to further plumage lipid accumulation but may simply not affect FLC because they are quickly used up. In the present study, both substrates were available in sufficient amounts as the floor of each tray was fully covered when offered for dustbathing. Only a small amount of substrate could be removed from the trays by dustbathing or scratching behavior. Further research is needed to determine the minimum amount of suitable dustbathing material needed to enable the removal of stale lipids from the plumage. Food particles did not serve to remove feather lipids, whereas L was found to reduce FLC in the amount given here. Using L as litter material in cage systems might induce technical problems. The coarser structure of L compared with F may be problematic for an automatic supply of substrate due to flow characteristics. Also, coarser particles may easily block the passage holes of the supply pipe, thus preventing particles from dropping onto the scratching mats. Finer grinding, however, may increase dust emission. In conclusion, the present study showed that dustbathing in F increased feather lipid content, whereas dustbathing in L decreased lipid content. Therefore, feed is not a suitable dustbathing material for laying hens. Because F are widely used in cage housing systems for laying hens, these results are quite alarming and priority should be given to the development of automatic litter dispensing systems for litter types more suitable as substrates for dustbathing.
