Published November 20, 2014
Effects of ambient temperature, feather cover, and housing
system on energy partitioning and performance in laying hens1
M. M. van Krimpen,*2 G. P. Binnendijk,* I. van den Anker,†
M. J. W. Heetkamp,† R. P. Kwakkel,‡ and H. van den Brand†
*Wageningen UR Livestock Research, PO Box 65, NL-8200 AB Lelystad,
The Netherlands; †Adaptation Physiology Group, Department of Animal Sciences, Wageningen
University, PO Box 338, NL-6700 AH Wageningen, The Netherlands; and ‡Animal Nutrition Group,
Department of Animal Sciences, Wageningen University, PO Box 338, NL-6700 AH Wageningen, The Netherlands
ABSTRACT: Environmental factors, such as ambient
temperature (T), feather cover (FC), and housing system (HS), probably affect energy requirements of laying
hens. Using a 3 × 2 × 2 factorial arrangement, interaction effects of T (11, 16, and 21°C), FC (100 and 50%),
and HS (cage and floor housing) on energy partitioning
and performance of laying hens were investigated. Six
batches of 70 H&N Brown Nick laying hens, divided
over 2 respiration chambers, were exposed to the T levels
in three 2-wk periods. Heat production (HP) was determined by indirect calorimetry. The ME intake was calculated by subtracting energy in manure/litter from that in
feed and wood shavings. The NE was calculated by subtracting HP from ME. The ME intake increased by 1%
for each degree reduction in T. In hens with intact plum-
age, HP was not affected by T, whereas at decreasing T,
HP increased in hens with 50% FC (P < 0.01). At 21°C,
HP was not affected by HS, whereas in the floor system,
HP at 16 and 11°C was 5.8 and 3.0% higher, respectively, than in cages (P < 0.05). The NE for production was
25.7% higher in cages compared to the floor system (P <
0.05). In cages, 24.7% of NE for production was spent on
body fat deposition, whereas in the floor system, 9.0% of
NE for production was released from body fat reserves.
The ME intake was predicted by the equation (R2 = 0.74)
ME intake (kJ/d) = 612 BW0.75 – (8.54 × T) + (28.36 ×
ADG) + (10.43 × egg mass) – (0.972 × FC). Hen performances were not affected by treatments, indicating the
adaptive capacity of young laying hens to a broad range
of environmental conditions.
Key words: ambient temperature, energy partitioning, housing system, laying hens, performance
© 2014 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2014.92:5019–5031
doi:10.2527/jas2014-7627
INTRODUCTION
Environmental factors, such as ambient temperature,
season, and housing system, affect feed intake of laying
hens (Chwalibog and Baldwin, 1995). Current equations
to predict feed intake are largely based on cage-housed
hens (Herremans et al., 1989; NRC, 1994). However, in
Europe, conventional cages were phased out, resulting
in a shift to noncage housing systems. Moreover, beak
1The authors gratefully acknowledge funding of the Dutch Product
Board for Livestock and Eggs, the Dutch Product Board for Animal
Nutrition, and the Dutch Ministry of Economic Affairs. Furthermore,
we wish to thank the students C. van der Pol, V. Nortey, P. de Lely,
T. Verhoeven, and the animal care takers of the experimental facility
“De Haar” for their enormous help in performing all observations.
2Corresponding author: marinus.vankrimpen@wur.nl
Received January 18, 2014.
Accepted September 20, 2014.
trimming will be limited or prohibited in the coming
years (Drake et al., 2010). Reduced plumage conditions
in flocks with intact beaks as a consequence of feather
pecking (Blokhuis and Van der Haar, 1989) might affect the energy requirement of the hens. Outdoor access and noncage housing systems might stimulate
physical activity of the hen, resulting in an increased
maintenance requirement (Luiting, 1990). Moreover,
hens in the outdoor area are exposed to more variable
and—especially in European autumns and winters—reduced ambient temperatures compared to indoor-housed
hens. A decrease in ambient temperature might result in
an increased maintenance requirement of laying hens
(Emmans, 1974; Al-Saffar and Rose, 2002), suggesting
a higher feed intake in noncage housing systems.
Current equations to predict ME intake (NRC, 1994;
Sakomura, 2004) do not take the plumage condition and
5019
5020
van Krimpen et al.
the physical activity level of the bird related to housing system into account. According to Herremans et al.
(1989), classical models based on weight, weight change,
and egg production are inadequate in predicting ME intake, especially in situations of elevated heat production,
for example, lower temperature or poor feathering.
Several studies investigated individual effects of
housing system (Ketelaars et al., 1985), ambient temperature, and feather cover (Herremans et al., 1989; Peguri
and Coon, 1993) on energy partitioning. The aim of the
current study, however, was to study the interaction effects of ambient temperature, plumage condition, and
housing system on energy partitioning and performance
of laying hens and to develop an equation that predicts
ME intake based on the results of this experiment.
MATERIALS AND METHODS
Experimental Design
In 6 subsequent batches, effects of ambient temperature (T), plumage condition, and housing system (HS)
were assessed, using a 3 × 2 × 2 factorial arrangement
with the following factors: T (regular [21°C], average
[16°C], and low [11°C]), feather cover (FC; 100 vs.
50%), and HS (floor vs. cage housing).
Housing systems were chosen to provide a low (cage)
and high (floor) level of physical activity. Housing system
was allotted to batch number. Within a batch, plumage
condition was allotted to 1 of 2 chambers, whereas T levels were allotted to subsequent periods within each batch
and chamber. In total, this experiment comprised 12 treatments with 3 replicates per treatment (36 observations: 6
batches × 2 chambers × 3 periods/chamber). Before each
measuring period, animals were habituated to the HS
during a pre-experimental period of 4 wk. The Animal
Use and Care Committee of Wageningen UR Livestock
Research approved the experimental protocols.
Housing, Hens, and Management during
the Pre-experimental Period (17 to 20 wk of Age)
Divided over 6 subsequent batches of 90 hens each,
a total of 540 17-wk-old H&N Brown Nick layer hens,
obtained from 35- to 50-wk-old breeder flocks, were
used in this study. All hens within 1 batch originated
from the same breeding flock, but flocks differed among
batches. During the rearing period, hens were treated
with routine vaccinations. Hens were wing numbered
for individual identification. From 17 wk onward, hens
were fed a standard maize–wheat–soybean meal layer
mash diet (with 11.8 MJ of ME/kg, 6.5 g of digestible
Lys/kg, 5.8 g of digestible Met + Cys/kg, and 38 g of
Ca/kg). All the feed for the experiment was produced in
1 batch by RDS (Wijk bij Duurstede, the Netherlands).
Feed and water were provided ad libitum.
During the pre-experimental period (17 to 20 wk of
age), the hens were kept in a climate controlled room
at the experimental facility of Wageningen University
(Wageningen, The Netherlands) to adapt to their new environment before the start of the actual study. The light
schedule was gradually extended by 1 h/wk from a 12:12
h light:dark scheme at 17 wk of age to a 16:8 h light:dark
scheme at 21 wk of age. Light intensity was 20 lux.
During this period, half of the available hens were
defeathered by 50%. For this, all feathers on the right
side of an imaginary sagittal line at the back of the hens,
except for the neck, head, and the downy feathers in the
cloaca region, were manually cut using carpet scissors.
To compensate for those uncut areas, a small area at the
left side of the back of the hens was defeathered as well
to realize a total of 50% plumage loss. Feathers in the
cloaca region were not cut, to prevent an increased attractiveness for feather pecking at this area.
In 3 of the 6 batches, the hens were cage housed
(measuring 75 by 60 by 50 cm; 6 hens/cage; 750 cm2/
hen). In the other batches, the hens were housed in a 6
m2 floor pen (3.0 by 2.0 m), holding 9 hens/m2 (1,111
cm2/hen) plus an added 1 m2 for 2 feed bins. The cages
were equipped with 2 drinking nipples, a feeding trough
at the front side, and a 75 cm long and 5 cm high perch
that was placed on the bottom of each cage. The floor
pen was equipped with 10 drinking nipples, a feeding
trough, and 2 laying nests. The activity of the floorhoused hens was stimulated by placing perches 30 cm
above the floor and by use of saw dust as litter material.
Housing, Hens, and Management during
the Experimental Period (21 to 26 wk of Age)
At the age of 21 wk, per batch, 70 healthy hens out
of 90 were placed in 1 of 2 identical respiration chambers
(Verstegen et al., 1987). Total weight of the animals per
chamber was standardized by reducing variation in mean
BW by removing the lightest and heaviest hens. Batches
were alternately assigned to cage housing or floor housing.
Each batch of 6 wk was subdivided in 3 periods of 2 wk.
In each period, 1 of the 3 ambient T levels was applied.
The first week of each period was used for adaptation of
the hens to the new environment. During the second week
of each period, observations were performed.
The chambers measured 3.7 by 1.47 m (5.4 m2),
each holding a total of 35 hens. When the chambers
were equipped as a cage house, 7 cages (5 hens/cage)
identical to the ones used in the pre-experimental period
were placed in the chamber. The cages were placed 75
cm from the floor. Cages were covered with a net to allow the Doppler radar system(Wageningen University,
Energy partitioning in laying hens
Wageningen, The Netherlands), which registers physical activity, to work properly. When the chambers were
equipped as a floor house, the floor was covered with
a 2 cm deep layer of wood shavings. A feed bin covered 0.396 m2 of the surface area. Two laying nests
with flat roofs on which the hens could sit were used,
meaning that no surface area was lost by placing laying
nests. The space beneath the laying nests was sealed off.
Furthermore, the chamber held a total length of 6.7 m of
perches (0.19 m/hen) and 10 drinking nipples.
Differences in ambient T were realized at a constant
level of relative humidity (60%) and air circulation rate
(0.1 to 0.2 m/s). The ambient T was measured on 4 different spots at animal level by Pt100 4-wired resistant
sensors type 1/3 DInB (accuracy ± 0.05°C). Temperature
at animal level was controlled in a master–slave setup
were the slave set point is automatically adapted every
10 min. In this way, the air blown from the climate control unit in the animal area, mixed with the present air,
causes a constant ambient T at animal level throughout
the day. All animals received ad libitum water and standard commercial diet.
Measurements
Hens were habituated to the climate respiration
chambers for 7 d before measurements started. Thereafter,
energy and N balances were assessed per chamber over
a 7-d measuring period. Feed intake (g/d) was monitored
per chamber and both fresh feed and left over feed were
weighed weekly. All individual animals were weighed
before the start and at the end of the measuring period,
whereafter BW gain (BWG; g/d) was calculated. Daily,
eggs were collected from each respiration chamber. At
the end of each measuring period, eggs were counted
and weighed, to calculate egg mass production (g/d). Per
period, 30 eggs were randomly selected, whereafter egg
content was separated from shells. Eggshells were dried
in a stove at 70°C for 16 h and then weighed and sampled
only during the first batch as energy and nitrogen content
was expected to be comparable for all samples. Egg content per measuring period was homogenized and sampled.
Exchange of O2, CO2, and CH4 was measured in
9-min intervals, as described by Verstegen et al. (1987).
Total heat production (HPtot; expressed as kJ·kg–
0.75·d–1) during the last 6 d of the experimental period
was calculated according to the equation of Romijn and
Lokhorst (1966):
HPtot = 16.20 × O2 + 5.00 × CO2,
in which HPtot is expressed in kilojoules and O2 and
CO2 are expressed in liters.
5021
Physical activity was monitored continuously by 2
Doppler radar devices per chamber according to the method
used by Wenk and van Es (1976). Changes in the frequency
of the reflected radar waves, due to movement of the hens
(Doppler effect), were converted into electrical pulses. The
principle of this method is that every change of the surface
of animals due to movements results in a change in frequency of the reflected ultrasound waves emitted by the
meters. Per day, the 9-min data on HPtot were related to
activity according to the following equation:
HPtot:ij = m + Di + β1 × X1j + β2 × X2j + eij,
in which HPtot:ij = heat production (HP) during day period i and 9-min period j; m = overall mean; Di = fixed effect of day period i (i = 1, 2); X1j and X2j = activity counts
during 9-min period j of the Doppler device of chamber 1
and 2, respectively; β1 and β2 = regression coefficient of
HP on activity counts of Doppler devices 1 and 2, respectively; and eij = error term. Heat production and physical
activity exhibit circadian rhythms (Aschoff et al., 1974).
Physical activity only partially accounts for the circadian
rhythm in HPtot. Therefore, a fixed effect of day period
with 2 levels was included in the equation. The day was
divided in a light period from 0100 to 1900 h and a dark
period from 1900 to 0100 h. The HP related to activity
(HPact) was calculated for each 9-min period as follows:
HPact:j = β1 × X1j + β2 × X2j,
in which HPact:j = activity-related HP during 9-min period j; X1j and X2j = activity counts during 9-min period j of the burglar device of pens 1 and 2, respectively;
and β1 and β2 = the estimated regression coefficients
of HPtot on activity from the first equation. The HP not
related to physical activity (HPcor) was derived by subtracting HPact from HPtot. Both HPact and HPcor were
calculated for each 9-min period.
Manure and, in the floor chambers, soiled litter was
collected quantitatively, homogenized, and sampled after
the 7-d measuring period. Dry matter content in feed, egg
content, egg shell, dust, feathers, and manure/soiled litter was determined according to ISO 6496 (ISO, 1998b).
Kjeldahl N content was determined according to ISO 5983
(ISO, 1997) in feed, dust, feathers, egg content, egg shell,
manure/soiled litter, air, and condensed water. Aerial NH3
was collected in a 25% H2SO4 solution in a wash bottle,
through which samples of the total outgoing airflow were
directed, and NH4+ was collected in water that condensed
on the heat exchanger. Gross energy content in feed, egg
content, dust, feathers, fresh wood shavings, and manure/
soiled litter was analyzed using adiabatic bomb calorimetry
(IKA-calorimeter C7000; IKA Works GmbH & Co. KG
, Staufen, Germany) according to ISO 9831 (ISO, 1998a).
5022
van Krimpen et al.
The ME intake (expressed in kJ·kg–0.75·d–1) was
calculated by subtracting the energy in manure/litter
from that in feed plus fresh wood shavings. The ME:GE
ratio was calculated as ME/GE × 100. The ME for maintenance (MEm; kJ·kg–0.75·d–1) was calculated as ME
intake – NE for protein deposition (kJ)/0.54 – NE for fat
deposition (kJ)/0.74 (Romijn and Lokhorst, 1966).
The NE (kJ·kg–0.75·d–1) was calculated by subtracting HP from ME. Retention of N (NR; g/d) was estimated from N in feed, wood shavings, manure/litter, and
dust as well as from aerial NH3 and NH4+ of water that
condensed on the heat exchanger. Net energy as protein
(NEp; kJ·kg–0.75·d–1) was calculated as 23.8 × 6.25 ×
NR, in which 23.8 kJ/g was used as the energy content
of protein (van Es, 1980). Net energy as fat (NEf; kJ·kg–
0.75·d–1) was calculated by subtraction of NE from NE.
p
Based on the amount of protein and fat deposited in eggs,
NEp could be subdivided in NEp in BW gain and in eggs.
Likewise, NEf could be subdivided as energy retention
as fat in BW gain and in eggs.
Statistical Analysis
Three out of the 36 balance studies were considered as
outliers and therefore excluded from the dataset. All outliers concerned floor housing at 11°C, 2 with 100% FC and 1
with 50% FC. In 2 of these balance studies, an outbreak of
feather pecking occurred, resulting in high levels of mortality and, as a consequence, a disturbed energy balance. In
the third balance study, the energy balance (very low level
of NE for production) was not in line with the observed
hen performance. This resulted in a nonorthogonal dataset,
and therefore, restricted maximum likelihood (Genstat 8
Reference Manual; Genstat 8 Committee, 2002) was used
to assess the effects of T, FC, and HS and their interaction
on energy partitioning, egg performances, and behavior.
The applied split-plot model was
Yijk = µ + Ti + FCj + HSk + T × FC + T × HS +
FC × HS + T × FC × HS + eijk,
in which Yijk = dependent variable, µ = overall mean, Ti =
fixed effect of T level i (i = 3; 11, 16, and 21°C), FCj =
fixed effect of FC j (j = 2; 100 and 50%), and HSk = fixed
effect of HS k (k = 2; cage and floor housing). Chamber
number within batches and period number within batches were used as random effects. The threshold for significance was set at P < 0.05.
Linear regression was applied to predict the ME intake, thereby using hen characteristics (e.g., BW, rate of lay,
egg weight, feed to gain [F:G] ratio) and housing conditions (ambient T and HS) as explanatory factors. First, only
1 factor (BW0.75) was involved in the model to predict
ME intake. Second, the RSEARCH procedure (Genstat 8
Reference Manual; Release 3; Genstat 8 Committee, 2002)
was applied to extend the model with the other possible
model parameters. This procedure provided the P-values
of the included model parameters as well as the adjusted R2
and the Mallow’s Cp (a measure of mean squared error of
prediction) of all possible subset combinations. The criteria
for the selected model were that 1) all model parameters
contributed to the model, 2) adjusted R2 was as high as possible, and 3) Cp was as low as possible.
RESULTS
Interaction Effects of Ambient Temperature,
Feather Cover, and Housing System
No interaction between HS, T, and FC were found on
feed intake, rate of lay, egg weight, egg mass, ADG, and
F:G ratio (Table 1). No interaction effects between T, FC,
and HS on energy partitioning were found, except for HP,
in which an interaction was found between T × FC and T
× HS, as shown in Fig. 1 and 2, respectively. In hens with
an intact plumage, total HP was not affected by T, whereas HP linearly increased in hens with a 50% FC, from
637.6 kJ·kg–0.75·d–1 at 21°C to 691.0 kJ·kg–0.75·d–1 at
11°C (P < 0.01; Fig. 1). At 21°C, HP was not affected by
HS, whereas HP in the floor system was increased by 5.8
and 3.0% at 16 and 11°C, respectively, compared to the
cage system (P < 0.05; Fig. 2). An interaction between T
and HS occurred for NE used for body protein deposition
(P < 0.05). In cage-housed hens, T did not affect NE for
body protein deposition, whereas in the floor-housed hens,
NE used for body protein deposition showed a 2-fold increase at 11 and 16°C (on average 54.1 kJ·kg–0.75·d–1)
compared to 21°C (26.6 kJ·kg–0.75·d–1).
Main Effects of Ambient Temperature
Ambient T did not affect any performance parameter
(Table 1), although feed intake tended to increase with decreasing T (P < 0.10) from 114.5 g/d at 21°C to 119.4 g/d
at 11°C. Temperature did not affect GE intake of the hens.
The ME intake tended (P < 0.10) to increase by 9.9%
from 858.1 kJ·kg–0.75·d–1 at 21°C to 942.8 kJ·kg–0.75·d–1
at 11°C (Table 2). At T = 11 and 16°C, the ME:GE ratio,
based on GE intake from feed only, was 3 units higher
than the ME:GE ratio at T = 21°C (P < 0.01). The HPcor
increased by 9.3% from 578.9 kJ·k–0.75·d–1 at 21°C to
633 kJ·kg–0.75·d–1 at 11°C (P < 0.05). The MEm was similar at 21 and 16°C (479 kJ·kg–0.75·d–1) but tended (P <
0.10) to increase by 3.4% at 11°C (495.5 kJ·kg–0.75·d–1).
The NE for BWG tended (P < 0.10) to be higher in hens
housed at 11 and 16°C (on average 66.3 kJ·kg–0.75·d–1)
compared to hens housed at 21°C (15.4 kJ·kg–0.75·d–1),
whereas NE for BWG was not affected by FC and HS
5023
Energy partitioning in laying hens
Table 1. Effects of ambient temperature (T), feather cover (FC), and housing system (HS) and their interaction on
feed intake, rate of lay, egg weight, egg mass, feed:gain (F:G) ratio, ADG, and BW of 21- to 26-wk-old H&N Brown
Nick laying hens
Trait
Feed intake, g/d
Rate of lay, %
Egg weight, g
Egg mass, g/d
F:G ratio, kg/kg
ADG, g/d
BW,2 kg
Treatment1
Cage housing
100% FC
T = 11°C
T = 16°C
T = 21°C
50% FC
T = 11°C
T = 16°C
T = 21°C
Floor housing
100% FC
T = 11°C
T = 16°C
T = 21°C
50% FC
T = 11°C
T = 16°C
T = 21°C
SEM
Main effect T
T = 11°C
T = 16°C
T = 21°C
SEM
Main effect FC
100%
50%
SEM
Main effect HS
Cage
Floor
SEM
P-values
T
FC
HS
T × FC
T × HS
FC × HS
T × FC × HS
118.2
115.7
113.8
98.1
96.5
96.2
58.2
58.1
57.0
57.1
56.0
54.8
2.07
2.07
2.08
2.4
2.0
1.4
1.85
1.86
1.84
126.9
121.9
119.7
97.5
98.2
96.2
59.4
58.7
57.8
57.9
57.7
55.6
2.20
2.12
2.16
2.9
2.8
2.2
1.85
1.85
1.84
107.0
116.4
110.3
97.5
97.3
98.0
56.7
58.4
57.3
55.1
56.8
56.2
2.08
2.05
1.97
2.8
2.0
0.4
1.79
1.80
1.80
125.1
120.8
114.2
3.95
97.7
96.2
95.1
1.90
56.9
59.1
58.0
2.19
56.0
56.9
55.3
2.60
2.21
2.13
2.07
0.059
2.5
2.2
1.5
0.82
1.78
1.83
1.80
0.037
119.3
118.7
114.5
2.14
97.7
97.0
96.4
0.98
57.8
58.6
57.5
1.52
56.5
56.8
55.5
1.74
2.21
2.13
2.07
0.040
2.6
2.2
1.4
0.55
1.82
1.83
1.82
0.020
113.6b
121.4a
1.49
97.3
96.8
1.17
57.6
58.3
1.09
56.0
56.6
1.34
2.05b
2.15a
0.036
1.8b
2.3a
0.38
1.82
1.82
0.021
119.4
115.6
1.83
97.1
97.0
1.18
58.2
57.7
1.52
56.5
56.1
1.77
2.11
2.09
0.049
2.3
1.9
0.53
1.85
1.80
0.020
0.158
0.013
0.689
0.300
0.269
0.619
0.957
0.253
0.028
0.578
0.583
0.855
0.397
0.608
0.780
0.964
0.183
0.813
0.832
0.820
0.595
0.074
0.010
0.309
0.459
0.630
0.742
0.582
0.308
0.722
0.973
0.432
0.832
0.551
0.502
0.845
0.125
0.876
0.932
0.774
0.875
0.768
0.742
0.661
0.923
0.719
0.857
0.533
0.826
a,bMean
values within treatments lacking a common superscript were different (P ≤ 0.05).
general, results of individual treatments were based on 3 observations, results of the main effect of T were based on 12 observations, and results of the
main effects of FC and HS were based on 18 observations. Because 3 out of the 36 observations were considered as outlier, some treatments had a reduced
number of observations.
2BW is determined at the end of the period.
1In
(Table 3). Temperature did not affect NE partitioning. A
schematic overview of the effect of T (11, 16, and 21°C)
on energy partitioning as percentage of GE intake in the
present study is shown in Fig. 3. The MEm relative to GE
tended (P < 0.10) to be lower at 16°C compared to 21°C,
whereas the MEm to GE ratio at 16°C was in between.
Main Effects of Feather Cover
Hens with 50% FC consumed 8 g/d more feed (P =
0.010), had a 5% higher FC (P < 0.05), and gained 0.5 g/d
more BW (P < 0.05) than hens with 100% FC (Table 1).
Daily GE intake of hens with 50% FC was 64 kJ·kg–0.75
5024
van Krimpen et al.
Figure 1. Hourly means of total heat production (HPtot; SEM = 11.2)
of hens with 50% feather cover (FC; dotted lines and open symbols) or 100%
FC (solid lines and closed symbols) at ambient temperature (T) levels of 11
(squares), 16 (triangles), or 21°C (circles). The dark period is indicated by
a shaded background. This figure shows an interaction effect of FC and T
(P = 0.005), indicating that HPtot was not affected by T in hens with an intact
plumage, whereas in hens with a 50% FC, heat production increased if T
decreased. OC = Degree Celcius (°C).
Figure 2. Hourly means of total heat production (HPtot; SEM = 13.7)
of hens housed in cages (dotted lines and open symbols) or in a floor system (solid lines and closed symbols) at ambient temperature (T) levels of 11
(squares), 16 (triangles), or 21°C (circles). The dark period is indicated by a
shaded background. This figure shows an interaction effect of housing system
(HS) and T (P = 0.040), indicating that at 21°C, heat production (HP) was
not affected by HS, whereas HP in the floor system was increased at 16 and
11°C, respectively, compared to the cage system. OC = Degree Celcius (°C).
Modeling Expected ME Intake
higher than of hens with the 100% FC (1,340 vs. 1,404
kJ; P < 0.05; Table 2). Removing 50% of feathers resulted in a 55.6 kJ (6.3%) higher ME intake (881.6 vs.
937.2 kJ·kg–0.75·d–1; P < 0.05) and a 5% higher HPcor of
29.9 kJ·kg–0.75·d–1 compared to the 100% FC treatment
(589.6 vs. 619.5 kJ·kg–0.75·d–1; P < 0.05). The MEm was
8% higher in 50% defeathered hens (37.5 kJ·kg–0.75·d–1;
P < 0.05). Feather cover did not affect NE partitioning.
Main Effects of Housing System
Daily GE intake of the floor-housed hens was
380 kJ·kg–0.75 (32%) higher than of cage-housed hens
(1,182 vs. 1,562 kJ; P < 0.001) as shown in Table
2. Housing system did not affect feed and ME intake.
The ME:GE ratio of cage-housed hens was 1.6 units
higher than of floor-housed hens (76.6 vs. 78.2; P <
0.05). Housing hens in the floor system resulted in a
7.4% higher HPcor of 39.0 kJ·kg–0.75·d–1 compared to
hens housed in the cage system (585.1 vs. 624.1 kJ·kg–
0.75·d–1; P < 0.05). The ratio between NE and ME was
higher in cage-housed hens than in floor hens (31.9
vs. 26.2%; P < 0.05; Table 3). The NE for production
was 60.7 kJ·kg–0.75·d–1 (25.7%) higher in hens housed
in cages compared to hens housed in the floor system
(236.2 vs. 296.9 kJ·kg–0.75·d–1; P < 0.05). Hens housed
in cages spent 72.3 kJ·kg–0.75·d–1 on NE for body fat
deposition, whereas hens housed in the floor system lost
21.2 kJ·kg–0.75·d–1 from their body reserves (P < 0.05).
Egg performance of the hens was not affected by HS, as
shown in Table 1.
Based on the variables of this experiment, the following equation was developed to estimate expected
ME intake of laying hens:
ME = 612 BW0.75 – (8.54 × T) + (28.36 × ADG) +
(10.43 × egg mass) – (0.972 × FC),
in which ME is expressed in kilojoules per day, BW
is expressed in kilograms, T is expressed in degrees
Celsius, ADG and egg mass are expressed in grams per
day, and FC is expressed in percent. The R2 of the model
was 0.74, where SE of the model parameters were 45.4,
3.0, 6.0, and 0.45 for BW0.75, T, ADG, egg mass, and
FC, respectively.
The factor “housing system” did not contribute to
the prediction of the ME intake.
DISCUSSION
Effect of Ambient Temperature
The conversion from GE to ME was more efficient
under conditions of reduced T. The ME content of the
diet was assumed to be 11.8 MJ/kg, but these findings
indicated that the dietary ME content was T dependent.
The recalculated ME content of the diet was 11.74, 12.29,
and 12.38 MJ/kg at 21, 16, and 11°C, respectively. The
higher ME intake at 11°C compared to 21°C was partly
used for higher HP (30 kJ·kg–0.75·d–1) and partly for
higher BWG (72 kJ·kg–0.75·d–1). Hence, ME intake increased by 0.90% (8.5 kJ·kg–0.75·d–1; 13.3 kJ/hen·d–1)
for each degree reduction in T.
5025
Energy partitioning in laying hens
Table 2. Effects of ambient temperature (T), feather cover (FC), and housing system (HS) and their interaction on
GE and ME partitioning, total heat production (HPtot), heat production related to activity (HPact), heat production not
related to physical activity (HPcor), ME for maintenance (MEm), and ME for production (MEprod; (in kJ·kg–0.75·d–1)
of 21- to 26-wk-old H&N Brown Nick laying hens
Trait
Treatment2
Cage housing
100% FC
T = 11°C
T = 16°C
T = 21°C
50% FC
T = 11°C
T = 16°C
T = 21°C
Floor housing
100% FC
T = 11°C
T = 16°C
T = 21°C
50% FC
T = 11°C
T = 16°C
T = 21°C
SEM
Main effect T
T = 11°C
T = 16°C
T = 21°C
SEM
Main effect FC
100%
50%
SEM
Main effect HS
Cage
Floor
SEM
P-values
T
FC
HS
T × FC
T × HS
FC × HS
T × FC × HS
GE intake feed/litter
ME intake
ME:GE1 ratio
HPtot
HPact
HPcor
MEm
MEprod
1,174
1,150
1,115
922.9
912.7
864.8
78.4
79.2
77.4
612.1
602.5
607.8
41.3
43.3
60.1
587.4
559.9
548.3
463.3
447.0
459.1
457.6
463.8
403.4
1,265
1,217
1,174
1,018.1
952.9
893.1
80.3
78.1
75.9
679.1
644.6
638.2
49.9
48.4
45.8
633.4
594.4
587.3
510.9
489.3
495.5
505.0
461.5
395.4
1,535
1,563
1,503
874.0
896.1
819.3
77.0
76.8
73.8
626.6
649.8
618.1
19.9
39.3
38.4
632.8
612.4
597.2
470.8
481.3
473.5
385.0
413.3
343.7
1,639
1,585
1,543
41.96
956.3
947.5
855.3
37.5
77.8
79.7
74.7
1.09
702.9
670.2
637.1
14.06
30.9
33.0
57.4
13.31
678.6
641.0
582.6
23.79
537.0
498.3
488.7
11.71
417.9
447.3
364.6
37.85
1,404
1,379
1,334
28.39
942.8
927.3
858.1
24.0
78.4a
78.5a
75.4b
0.48
655.2
641.8
625.3
7.20
35.5
41.0
50.4
6.72
633.0a
601.9b
578.9c
11.96
495.5
479.0
479.2
6.27
441.4
446.5
376.8
24.13
1,340b
1,404a
17.76
881.6b
937.2a
15.5
77.1
77.6
0.36
619.5
662.0
8.15
40.4
44.2
5.51
589.6b
619.5a
9.81
465.8b
503.3a
6.25
411.1
431.9
15.09
1,182b
1,562a
22.99
927.4
891.4
19.2
78.2a
76.6b
0.35
630.7
650.8
8.15
48.1
36.5
5.96
585.1b
624.1a
9.81
477.5
491.6
6.57
447.8
395.3
19.60
0.496
0.015
<0.001
0.102
0.978
0.437
0.389
0.054
0.023
0.294
0.226
0.791
0.873
0.936
0.002
0.425
0.032
0.701
0.357
0.122
0.465
<0.001
<0.001
0.045
0.005
0.040
0.632
0.636
0.364
0.809
0.206
0.882
0.703
0.608
0.446
0.014
0.033
0.011
0.634
0.590
0.430
0.678
0.063
0.016
0.164
0.165
0.382
0.431
0.280
0.092
0.178
0.149
0.668
0.792
0.386
0.790
a–cMean
values within treatments lacking a common superscript were different (P ≤ 0.05).
on GE intake of only feed, whereas GE intake of litter is neglected.
2In general, results of individual treatments were based on 3 observations, results of the main effect of T were based on 12 observations, and results of the
main effects of FC and HS were based on 18 observations. Because 3 out of the 36 observations were considered as outlier, some treatments had a reduced
number of observations.
1Based
Because hens converted GE to ME at lower T more
efficiently, feed intake in the current experiment increased only by 0.42% for each degree reduction in T.
This value is considerably lower compared to the 0.95%
reported by Al-Saffar and Rose (2002), who based this
value on a literature review to the effects of T on egg laying characteristics. Peguri and Coon (1993), who used
laying hens with 100 and 50% FC, reported in the range
of 23.9 to 12.8°C an increase in feed intake across both
FC of 0.75% for each 1°C reduction in T. In the pres-
5026
van Krimpen et al.
Table 3. Effects of ambient temperature (T), feather cover (FC), and housing system (HS) and their interaction on
NE partitioning of production (NE Prod.), NE for BW gain (NE_BWG), NE for fat retention in BW gain (NE_BWG
Fat), NE for protein retention in BW gain (NE_BWG protein), NE retained in egg (NE_Egg), NE for fat retention
in egg (NE_Egg Fat), and NE for protein retention in egg (NE_Egg Protein; in kJ·kg–0.75·d–1) of 21- to 26-wk-old
H&N Brown Nick laying hens
Trait
Treatment1
Cage housing
100% FC
T = 11°C
T = 16°C
T = 21°C
50% FC
T = 11°C
T = 16°C
T = 21°C
Floor housing
100% FC
T = 11°C
T = 16°C
T = 21°C
50% FC
T = 11°C
T = 16°C
T = 21°C
SEM
Main effect T
T = 11°C
T = 16°C
T = 21°C
SEM
Main effect FC
100%
50%
SEM
Main effect HS
Cage
Floor
SEM
P-values
T
FC
HS
T × FC
T × HS
FC × HS
T × FC × HS
NE:ME ratio
NE Prod.
NE_BWG
33.5
33.9
29.7
309.4
309.2
256.7
92.2
83.4
34.7
99.9
78.1
22.4
–7.7
5.4
12.3
217.2
225.8
222.1
114.6
123.8
120.6
102.6
102.0
101.5
33.7
32.6
28.4
341.6
310.2
254.3
123.3
80.9
25.1
117.8
89.3
26.1
5.5
–8.5
–1.1
218.2
229.4
229.2
112.7
123.3
126.2
105.5
106.0
103.0
24.7
27.8
24.3
223.4
248.4
201.8
11.9
32.4
–5.7
–52.6
–15.5
–29.2
63.6
47.9
23.5
203.7
216.1
207.4
104.2
110.7
103.3
99.0
105.3
104.1
26.4
29.0
25.1
2.24
249.5
276.2
218.1
29.16
35.9
65.5
7.6
29.37
–22.5
14.7
–22.0
27.91
54.3
50.8
29.6
9.46
208.6
210.7
210.5
11.7
106.0
106.1
109.1
8.79
103.4
104.6
101.4
3.62
29.6
30.8
26.9
1.42
281.0
286.0
232.7
18.68
65.8
65.5
15.4
18.83
35.7
41.7
–0.7
17.97
28.9
23.9
16.1
5.94
211.9
220.5
217.3
8.11
109.4
116.0
114.8
6.07
102.6
104.5
102.5
2.43
29.0
29.2
0.84
258.2
275.0
10.91
41.5
56.4
13.38
17.2
33.9
10.68
24.2
21.8
5.12
215.4
217.8
6.41
112.9
113.9
4.86
102.4
104.0
1.96
31.9
26.2
1.11
296.9a
236.2b
14.02
73.3
24.6
17.94
72.3a
–21.2b
14.11
1.0
44.9
6.94
223.6
209.5
8.96
120.2
106.6
6.83
103.4
103.0
2.71
0.135
0.780
0.027
0.990
0.654
0.149
0.972
0.097
0.165
0.043
0.759
0.708
0.396
0.848
NE_BWG Fat NE_BWG Protein
0.067
0.191
0.154
0.732
0.533
0.264
0.768
0.125
0.134
0.013
0.810
0.268
0.471
0.950
0.316
0.657
0.012
0.467
0.026
0.208
0.140
NE_Egg
0.687
0.306
0.326
0.443
0.952
0.303
0.513
NE_Egg Fat NE_Egg Protein
0.589
0.385
0.219
0.179
0.788
0.785
0.700
0.717
0.146
0.972
0.268
0.727
0.124
0.499
a,bMean
values within treatments lacking a common superscript were different (P ≤ 0.05).
general, results of individual treatments were based on 3 observations, results of the main effect of T were based on 12 observations, and results of the
main effects of FC and HS were based on 18 observations. Because 3 out of the 36 observations were considered as outlier, some treatments had a reduced
number of observations.
1In
ent experiment, hen performances were not affected by
ambient T. Based on their review of literature, Al-Saffar
and Rose (2002), however, reported an average increase
in 2.5% egg mass and of 11% in F:G ratio if T decreased
from 21 to 11°C. According to these authors, the addi-
tional feed intake at lower T due to thermoregulation
concomitantly increased intake of other nutrients, thereby explaining the small increase in egg mass output. The
effect of ambient T on feed intake of hens in the current
study, however, was rather low, and therefore limited
Energy partitioning in laying hens
Figure 3. Schematic representation of the effect of temperature (T;
11/16/21°C) on energy partitioning as percentage of GE intake in the present
study. NE eggP = NE for egg production as protein; NE eggF = NE for egg
production as fat; NE bwgP = NE for BW gain as protein; NE bwgF = NE for
BW gain as fat. *P ≤ 0.05; **P ≤ 0.01; #P ≤ 0.10. 1)Interaction effect between
T × feather cover (FC; P = 0.026). 2)Interaction effect between T × FC (P =
0.005) and T × housing system (P = 0.040).
amounts of additional other nutrients were consumed at
lower T, which might explain the absence of significant
effects of T on hen performance and F:G ratio.
The MEm in the present study was similar at 21 and
16°C but increased by 3.5% if T further decreased to
11°C. Peguri and Coon (1993) observed across 100 and
50% FC even a 19.5% increase in MEm requirements
if T decreased from 23.9 to 12.8°C. The reason for the
large difference between both studies is not clear. Peguri
and Coon (1993) used hens with a higher age (59 to 65
wk of age). Moreover, they did not use respiration chambers. The MEm in that study was estimated by subtracting the calculated energy deposition in eggs and BWG
from the calculated ME intake, which is less accurate
compared to the method used in the current experiment.
Results of the present experiment showed that laying hens with an intact FC had an improved insulation
capacity under conditions of reduced T, thereby maintaining HP on a constant level, whereas HP in hens with
50% FC increased with decreasing T. Similar interactions
between T and FC were observed in earlier experiments
(Richards, 1977; Peguri and Coon, 1993), although differences between treatments were more expressed in
older studies. Richards (1977), for instance, observed at
20°C an increase in HP of 59% in poorly feathered hens
(65% feather removal) compared to well-feathered hens,
whereas in the present experiment at 21°C this difference was only 4%. It is not fully clear whether this dissimilarity is related to external conditions (e.g., relative
humidity, air velocity) or hen conditions (e.g., insulation
capacity of the remaining plumage).
Effect of Feather Cover
A schematic overview of the effect of FC (100 and
50%) on energy partitioning as percentage of GE intake
in the present study is shown in Fig. 4. In the context of
5027
Figure 4. Schematic representation of the effect of feather cover (FC;
100/50%) on energy partitioning as percentage of GE intake in the present study.
NE eggP = NE for egg production as protein; NE eggF = NE for egg production
as fat; NE bwgP = NE for BW gain as protein; NE bwgF = NE for BW gain as fat.
*P ≤ 0.05. 1)Interaction effect between temperature × FC (P = 0.005)
thermoregulation, plumage plays an important role due
to its isolating capacities (Herremans et al., 1989; Peguri
and Coon, 1993). Adaptive behavioral patterns involving preening, feather ruffling, dust bathing, and conductive heat loss via sitting are used by the bird to expose
or close thermal windows in the plumage with flexibility
(Gerken et al., 2006). Therefore, damage of the plumage reduces the possibility of a bird to prevent heat loss
from the body. In the present experiment, removing 50%
of feathers resulted in a 5.4% higher GE intake, and ME
intake similarly increased. In the experiment of Peguri
and Coon (1993), removing 50% of feathers resulted in
a 12.7% higher ME intake (across the ambient T values
of 12.8 and 23.9°C), which is much higher than the effect
found in the present experiment. The difference between
both experiments occurred mainly at low T. In the present
experiment, defeathering increased ME intake at 11°C by
9.9%, whereas defeathering in the experiment of Peguri
and Coon (1993) at 12.8°C resulted in an 18% higher feed
intake. In both experiments, feathers were removed at the
same body parts. It is supposed that differences might
be explained by other factors, such as breed (Lohmann
Brown Lite vs. DeKalb XL White Leghorn), age of hens
(21 vs. 59 wk), air velocity, relative humidity, BW, or insulation capacity of the remaining plumage.
Interestingly, Peguri and Coon (1993) also tested
the effect of 100% removing of feathers. Across T treatments, feed intake increased with 13.5 g/d after removing of the first 50% of feathers, and feed intake increased
with another 17.5 g/d after removing of the second 50%
of feathers, indicating a nearly linear relationship between the level of feather removal and energy intake.
In the present experiment, feed efficiency decreased by
4.9% (F:G ratio = 2.15 vs. 2.05) as a result of removing
50% of feathers, whereas it decreased by 12% (0.156 vs.
0.139 g egg mass/kcal) in the study of Peguri and Coon
(1993), probably due to the same factors as mentioned regarding the differences between both studies in ME intake.
5028
van Krimpen et al.
Figure 5. Schematic representation of the effect of housing system (HS;
cage/floor) on energy partitioning as percentage of GE intake in the present
study. NE eggP = NE for egg production as protein; NE eggF = NE for egg
production as fat; NE bwgP = NE for BW gain as protein; NE bwgF = NE for
BW gain as fat. *P ≤ 0.05; ***P ≤ 0.001. 1)Interaction effect between temperature (T) × HS (P = 0.026). 2)Interaction effect between T × HS (P = 0.040).
Effect of Housing System
A schematic overview of the effect of HS (cage and
floor housing) on energy partitioning as percentage of GE
intake in the present study is shown in Fig. 5. Hens housed
in cages were more efficient in converting GE to ME, and
HP was reduced as well compared to floor-housed hens.
Housing system was added to the experiment to realize
differences in physical activity of the hens. It was expected
that the activity level in the floor system would be higher
than in the cages. In the floor system, hens had more space
compared to cages (1,430 vs. 750 cm2/hen), allowing them
to walk and forage, and elevated perches were available on
which they could jump. On average, activity related HP
accounted only for 4.7% of total ME intake. The HPact
did not differ between systems, although HS largely affected behavioral patterns. Based on video recordings, we
observed that cage-housed hens spent more time with their
head in the feeding trough (24.9 vs. 16.3% of observation
period), and sitting behavior was increased as well (37.5 vs.
23.4%; data not shown). Overall, HPact and MEm did not
differ between HS in the present experiment. In their study,
Ketelaars et al. (1985) found in 1 of the 2 performed experiments a reduced MEm in cage-housed hens compared
to hens free housed on a slatted floor. They assumed that
this difference was related to the higher activity of the floor
hens, which is not in line with our findings. In the present
study, wood shavings were used as litter, which might prevent heat loss if hens are sitting on the floor, whereas no
insulation material was available in the study of Ketelaars
et al. (1985), which might explain the differences in results
between both experiments. The radar system was not able
to discriminate between types of activities, for example,
head shaking or jumping on a perch, which could be a possible weakness of the determination of HPact in the present
study. Moreover, HPact might be slightly underestimated
in the floor system. When the hens were in the nest boxes,
the radar system could not detect their activities. Based on
our video observations (data not shown), we found that
the actual number of observed hens was on average 16%
lower than the real number of hens available, indicating
that on average 5 to 6 hens should be stayed into the nest
boxes. Even if HPact of the floor treatment was increased
by 16%, this level remains still lower than the HPact of the
cage treatment.
Floor-housed hens retained more energy as protein
compared to cage-housed hens (44.9 vs. 1.0 kJ·kg–0.75·d–1,
which is equivalent with 1.9 vs. 0.04 g protein·kg–0.75·d–1).
Floor-housed hens mobilized energy from body fat depots,
whereas cage-housed hens retained energy as fat (–21.2 vs.
72.3 kJ·kg–0.75·d–1, which is equivalent with –0.5 vs. 1.8 g
fat·kg–0.75·d–1). These results are in line with findings of
Bolhuis et al. (2008), who observed higher levels of energy
retained as protein and lower levels of energy retained as
fat in pigs housed on straw bedding compared to pigs in a
barren environment without straw. These authors hypothesized that these differences might be related to higher postprandial activity levels of pigs housed on straw bedding,
which could blunt postprandial plasma glucose increase
and fat synthesis. In the present experiment, however,
HPact did not differ between HS, which suggests that HS
did not affect the amount of energy hens spent on physical
activity. The radar system, however, could not discriminate
between different types of movements. Behavioral observations showed that hens in the cage system had a lot of
head movements, whereas hens in floor system spent a
relatively large part of their time budget on foraging. Both
type of movements delivered pulses for the radar system,
although it is expected that the energy costs of head movements will be lower compared to those of foraging, which
is stressed by the differences in protein and fat deposition
between both HS, as shown in Table 3.
Prediction of Energy Intake
Based on the results of this experiment, an equation
to estimate daily ME intake of laying hens was calculated.
Several other authors also developed comparable equations (NRC, 1994; Sakomura, 2004). Although these are
empirical equations, it is possible to speculate on the meanings of the regression coefficients of the prediction. For
example, based on the equation of the present study, the
tabular value for a 1.0-kg hen at 21°C in weight balance
and laying no eggs was estimated to be 419 kJ/d. The filling of the same values in the equations of the NRC (1994)
and Sakomura (2004) resulted in higher ME requirements,
namely 553 and 484 kJ/d, respectively. Data from indirect
calorimetry, however, consistently indicate maintenance
requirements for layers close to 400 kJ·kg–0.75·d–1. Pesti
et al. (1990) calculated 396 and 436 kJ·kg–0.75·d–1 for
maintenance requirements of black and red hens, respectively. Macleod and Jewitt (1988) estimated values of 420
Energy partitioning in laying hens
and 435 kJ·kg–0.75·d–1, whereas Balnave et al. (1978) reported 388 kJ·kg–0.75·d–1 for ovariectomized laying hens.
Therefore, the coefficient for maintenance requirements in
the equations of the NRC (1994) and Sakomura (2004)
seem to overestimate the biological value.
In the equation of the present study, the coefficient
for BWG was estimated as 28.4 kJ/g. This value is in
line with 23.0 kJ/g in the function of the NRC (1994)
and 27.9 kJ/g according to Sakomura (2004).
In the equation of the present experiment, the coefficient for egg mass production was estimated to be
10.43 kJ/g, whereas the NRC (1994) and Sakomura
(2004) used coefficients of 8.66 and 10.03 kJ/g, respectively. The NRC (1994) seems to underestimate the energy costs for egg mass production. The energy content
of eggs was found to range from 7.1 kJ/g in hens at 27
wk of age to 7.5 kJ/g at 48 wk of age (Chwalibog, 1992).
This variation was related to a higher fat to protein ratio/
kilogram egg with increasing age of the hens. According
to Chwalibog (1992), efficiency of energy conversion
into protein and fat is 0.50 and 0.79, respectively, whereas an averaged efficiency for combined protein and fat
deposition of 0.625 was assumed. Considering these efficiency factors, the calculated costs of energy for synthesizing 1 g of egg varied from 11.38 (27 wk) to 12.02
(48 wk) kJ/g. The estimated coefficient of 10.43 kJ/g in
the present experiment approached the physiological
based value calculated for hens at early lay.
Feather cover was not taken into account in the equations of the NRC (1994) and Sakomura (2004). According
to Herremans et al. (1989), prediction of feed intake is
poor when data resulted from a broad range of T or FC,
due to large variation in HP. According to these authors,
defeathering accounted for 48% of the variation in heat
production in their dataset. In the present study, each percent less FC increased daily energy requirement by 0.972
kJ. This implies that energy intake in fully naked hens
should increase by 97.2 kJ/d, which agrees with 8.2 g feed
intake (if energy content = 11.8 kJ/g). It should be considered, however, that only 2 levels of FC (100 and 50%
FC) were included in the present experiment. Based on
these levels, energy costs of other levels were estimated
on a linear scale. This assumption, however, is in line
with findings of Peguri and Coon (1993), who observed a
nearly linear relationship between FC and energy intake
in their study, in which FC ranged from 0 to 100%.
In the present experiment, the estimated daily energy
requirement directly related to HS did not contribute to
the prediction of energy intake. In meta-analysis studies,
in which performance levels of flocks in cage vs. noncage
HS are compared, feed intake level of floor hens ranged
from comparable (Van Horne, 1996) to 3% higher (Aerni
et al., 2005) than cage-housed hens. These results, however, could be confounded by the fact that the surveyed
5029
flocks differed in factors such as breed, performance levels, diet composition, etc. No data from studies based on
indirect calorimetry are available in literature to discuss
the impact of HS on energy requirement. It was expected that, due to a higher level of locomotion, feed intake
level of the floor-housed hens should be increased compared to cage-housed hens, but this was not the case. It
is hypothesized that the absence of an increased energy
factor for floor housing could be explained by a considerable higher apparent fecal nitrogen digestibility of floorhoused hens compared to the cage-housed hens (38.5 vs.
57.0%; Table 4). This might result in an improved supply
of digestible protein/AA on gut level, consequently saving feed intake. The higher nitrogen digestibility of floorhoused hens might be the result of consuming fibers from
the litter, while litter of course was absent in the cages.
Moderate amounts of fiber might improve the development of organs, enzyme production, and nutrient digestibility in poultry (Hetland et al., 2004; Mateos et al., 2012).
Some of these effects are a consequence of better gizzard
function, with an increase in gastroduodenal refluxes that
facilitate the contact between nutrients and digestive enzymes. These effects often result in improved performance
and animal health, but the potential benefits depend to a
great extent on the physicochemical characteristics of the
fiber source. Especially coarse, insoluble fiber sources are
able to improve the functioning of the gut (Van Krimpen
et al., 2007; Mateos et al., 2012), and the used litter source,
wood shavings, satisfies these properties.
In the equation of the present experiment, the SE of
observations amounted 64.6 kJ·kg–0.75·d–1, indicating
that 7.2% of ME intake remained unexplained by this
equation. This unexplained fraction is called residual
feed consumption (RFC). By including only BW0.75,
BWG, and egg mass as explanatory factors in the linear
regression of ME intake, RFC in literature ranges between 47 and 180 kJ·kg–0.75·d–1, which equals a variation coefficient of 4 to 12% (Luiting, 1990). Including
only BW0.75, BWG, and egg mass as explanatory factors
to the equation of the present study resulted in a RFC of
74 kJ·kg–0.75·d–1. Therefore, it appears that the factors
T and FC reduced RFC by only 9.4 kJ·kg–0.75·d–1. The
main component of RFC variation seems to be the variation in MEm, due to differences in physical activity, FC,
basal metabolic rate, body temperature, and body composition (Luiting, 1990; Gabarrou et al., 1997; Bordas
and Minvielle, 1999; Van Eerden et al., 2004). Each of
these differences might be subject to genetic variation,
accumulating into genetic variation in RFC. Bentsen
(1983) found a SD of 70 to 80 kJ·kg–0.75·d–1 over the
entire laying period, with a correlation of 0.5 between
adjacent 4-wk periods. Physical activity is reported to
cause 9 to 25% of total HP (40 to 117 kJ·kg–0.75·d–1)
in ad libitum fed laying hens (Boshouwers and Nicaise,
5030
van Krimpen et al.
Table 4. Main effects of ambient temperature (T), feather cover (FC), and housing system (HS) on N partitioning, N
retained in egg (N_egg), and N retained in BW gain (N_BWG; in g/kg–0.75·d–1), and apparent fecal N digestibility
(N digestibility; %) of 21- to 26-wk-old H&N Brown Nick laying hens
Trait
Treatment1
Main effect T
T = 11°C
T = 16°C
T = 21°C
SEM
Main effect FC
100%
50%
SEM
Main effect HS
Cage
Floor
SEM
P-values
T
FC
HS
T × FC
T × HS
FC × HS
T × FC × HS
a,bMean
N intake, g/d
N_egg, g/d
N_BWG, g/d
N emission, g/d
N in dust filter, g/d
N in feces, g/d
N digestibility, %
1.90
1.98
2.00
0.04
0.69
0.71
0.69
0.02
0.11
0.16
0.20
0.04
0.05
0.07
0.06
0.01
0.022
0.028
0.033
0.006
1.01
1.01
1.04
0.03
46.8
48.6
47.8
1.82
1.91b
2.01a
0.03
0.69
0.70
0.01
0.16
0.15
0.03
0.06
0.06
0.01
0.026
0.030
0.005
0.97b
1.07a
0.02
49.0
46.4
1.50
1.94
1.98
0.04
0.70
0.70
0.02
0.01
0.30
0.05
0.03
0.09
0.01
0.017
0.039
0.007
1.20a
0.85b
0.02
38.5
57.0
1.94
0.147
0.011
0.460
0.076
0.601
0.339
0.370
0.717
0.146
1.000
0.255
0.777
0.124
0.499
0.318
0.601
0.013
0.570
0.024
0.208
0.140
0.005
0.560
<0.001
0.702
<0.001
0.594
0.335
0.510
0.150
0.091
0.604
0.981
0.422
0.309
0.071
0.024
<0.001
0.959
0.113
0.294
0.295
0.343
0.133
0.003
0.638
0.036
0.644
0.338
values within treatments lacking a common superscript were different (P ≤ 0.05).
1In general, results of the main effect of T were based on 12 observations and results of the main effects of FC and HS were based on 18 observations. Because
3 out of the 36 observations were considered as outlier, some treatments had a reduced number of observations.
1985; Macleod et al., 1988). In the present experiment,
HPact accounted only for 6.6% (42 kJ·kg–0.75·d–1) of
total HP. Surprisingly, RFC was not reduced after including HPact as explanatory factor to our regression
equation. The HPact in the present experiment was determined by radar devices. The estimated energy costs of
standing contributes for 2 to 44 kJ·kg–0.75·d–1 (Luiting,
1990). Standing, however, is not detected as an activity by the radar, and therefore HPact could be underestimated in the present experiment. Moreover, HPact may
be divided into a muscular energy fraction associated
with the work involved in movement and a physical heat
loss fraction associated with breaking of the insulation
layer. Therefore, the effect of activity level on HP could
be confounded with effects of variation in FC (Luiting,
1990). Feather cover contributes to the prediction of ME
intake in the present study, but despite this, RFC is still
6.9%. Part of this RFC might be explained by bird to
bird differences in insulation capacity of the FC and in
the surface of nude body areas, such as comb, wattles,
and legs (Luiting, 1990; Gabarrou et al., 1997; Bordas
and Minvielle, 1999; Van Eerden et al., 2004).
Conclusions
The results of the present experiment indicate the
importance of maintaining FC for laying hens, especially in cold conditions, to prevent heat loss. Despite rather
considerable differences between treatments, rate of lay,
egg weight, and egg mass were not affected, indicating
the adaptive capacity of young laying hens to a broad
range of environmental conditions.
LITERATURE CITED
Aerni, V., M. W. G. Brinkhof, B. Wechsler, H. Oester, and E. Frohlich.
2005. Productivity and mortality of laying hens in aviaries: A
systematic review. Worlds Poult. Sci. J. 61:130–142.
Al-Saffar, A. A., and S. P. Rose. 2002. Ambient temperature and the
egg laying characteristics of laying fowl. Worlds Poult. Sci. J.
58:317–331.
Aschoff, J., H. Biebach, A. Heise, and T. Schmidt. 1974. Day-night
variation in heat balance. In: J. L. Monteith and L. E. Mount,
editors, Heat loss from animals and man. Butterworths, London.
p. 147- 172.
Balnave, D., D. J. Farrell, and R. B. Cumming. 1978. Minimum metabolizable energy requirement of laying hens. Worlds Poult.
Sci. J. 34:149–154.
Bentsen, H. B. 1983. Genetic variation in feed efficiency of laying
hens at constant body weight and egg production. 1. Efficiency
measured as a deviation between observed and expected feed
consumption. Acta Agric. Scand. 33:289–304.
Energy partitioning in laying hens
Blokhuis, H. J., and J. W. Van der Haar. 1989. Effects of floor type during rearing and of beak trimming on ground pecking and feather
pecking in laying hens. Appl. Anim. Behav. Sci. 22:359–369.
Bolhuis, J. E., H. van den Brand, S. T. Staals, T. Zandstra, S. J.
Alferink, M. J. Heetkamp, and W. J. Gerrits. 2008. Effects of
fermentable starch and straw-enriched housing on energy partitioning of growing pigs. Animal 2:1028–1036.
Bordas, A., and F. Minvielle. 1999. Patterns of growth and feed intake in divergent lines of laying domestic fowl selected for residual feed consumption. Poult. Sci. 78:317–323.
Boshouwers, F. M. G., and E. Nicaise. 1985. Automatic gravimetric
calorimeter with simultaneous recording of physical activity for
poultry. Br. Poult. Sci. 26:531–541.
Chwalibog, A. 1992. Factorial estimation of energy requirement for
egg production. Poult. Sci. 71:509–515.
Chwalibog, A., and R. L. Baldwin. 1995. Systems to predict the energy and protein requirements of laying fowl. Worlds Poult. Sci.
J. 51:187–196.
Drake, K. A., C. A. Donnelly, and M. S. Dawkins. 2010. Influence of
rearing and lay risk factors on propensity for feather damage in
laying hens. Br. Poult. Sci. 51:725–733.
Emmans, G. C. 1974. The effect of temperature on the performance of
laying hens. In: T. R. Morris and B. M. Freeman, editors, Energy
requirements of poultry. British Poultry Science LTD, Edinburgh.
p. 79–90.
Gabarrou, J. F., P. A. Géraert, M. Picard, and A. Bordas. 1997. Dietinduced thermogenesis in cockerels is modulated by genetic selection for high or low residual feed intake. J. Nutr. 127:2371–2376.
Gerken, M., R. Afnan, and J. Dori. 2006. Adaptive behaviour in chickens in relation to thermoregulation. Arch. Geflugelkd. 70:199–207.
Herremans, M., E. Decuypere, and O. Siau. 1989. Effects of feather
wear and temperature on prediction of food intake and residual
food consumption. Br. Poult. Sci. 30:15–22.
Hetland, H., M. Choct, and B. Svihus. 2004. Role of insoluble nonstarch polysaccharides in poultry nutrition. Worlds Poult. Sci. J.
60:415–422.
International Organization for Standardization (ISO). 1997. Animal
feeding stuffs. Determination of nitrogen content and calculation of crude protein content. Kjeldahl method. ISO 5983. ISO,
Geneva, Switzerland.
International Organization for Standardization (ISO). 1998a. Animal
feeding stuffs. Determination of gross calorific value. ISO 9831.
ISO, Geneva, Switzerland.
International Organization for Standardization (ISO). 1998b. Animal
feeding stuffs. Determination of moisture and other volatile
matter content. ISO 6496. ISO, Geneva, Switzerland.
Ketelaars, E. H., A. Arets, W. Van der Hel, A. J. Wilbrink, and M. W.
A. Verstegen. 1985. Effect of housing systems on the energy
balance of laying hens. Neth. J. Agric. Sci. 33:35–43.
Luiting, P. 1990. Genetic variation of energy partitioning in laying hens – Causes of variation in residual feed consumption.
Worlds Poult. Sci. J. 46:133–152.
5031
Macleod, M. G., and T. R. Jewitt. 1988. Maintenance energy requirements of laying hens: A comparison of measurements made by 2
methods based on indirect calorimetry. Br. Poult. Sci. 29:63–74.
Macleod, M. G., T. R. Jewitt, and J. E. M. Anderson. 1988. Energy
expenditure and physical activity in domestic fowl kept on standard and interrupted lighting patterns. Br. Poult. Sci. 29:231–244.
Mateos, G. G., E. Jimenez-Moreno, M. P. Serrano, and R. P. Lazaro.
2012. Poultry response to high levels of dietary fiber sources
varying in physical and chemical characteristics. J. Appl. Poult.
Res. 21:156–174.
NRC. 1994. Nutrient requirements of poultry. National Research
Council, 9th rev. ed. Natl. Acad. Press, Washington, DC.
Peguri, A., and C. Coon. 1993. Effect of feather coverage and temperature on layer performance. Poult. Sci. 72:1318–1329.
Pesti, G. M., E. Thomson, and D. J. Farrell. 1990. Energy exchange of
2 breeds of hens in respiration chambers. Poult. Sci. 69:98–104.
Richards, S. A. 1977. Influence of loss of plumage on temperature
regulation in laying hens. J. Agric. Sci. 89:393–398.
Romijn, C., and W. Lokhorst. 1966. Heat regulation and energy metabolism in the domestic fowl. In: C. Horton-Smith and E. C.
Amoroso, editors, Physiology of the domestic fowl. Oliver and
Boyd, Edinburgh. p. 211–227.
Sakomura, N. 2004. Modeling energy utilization in broiler breeders, laying hens and broilers. Brazilian J. Poultry Sci. (Revista
Brasileira de Ciência Avícola). V6:1–11.
Van Eerden, E., H. Van Den Brand, H. K. Parmentier, M. C. M. De
Jong, and B. Kemp. 2004. Phenotypic selection for residual feed
intake and its effect on humoral immune responses in growing
layer hens. Poult. Sci. 83:1602–1609.
van Es, A. J. H. 1980. Energy costs of protein deposition in ruminants. In: P. J. Buttery and D. B. Lindsay, editors, Protein deposition in animals. Butterworths, London. p. 215–224.
Van Horne, P. L. M. 1996. Production and economic results of commercial flocks with white layers in aviary systems and battery
cages. Br. Poult. Sci. 37:255–261.
Van Krimpen, M. M., R. P. Kwakkel, G. André, C. M. van der PeetSchwering, L. A. den Hartog, and M. W. Verstegen. 2007. Effect
of nutrient dilution on feed intake, eating time and performance
of hens in early lay. Br. Poult. Sci. 48:389–398.
Verstegen, M. W. A., W. Van der Hel, H. A. Brandsma, A. M. Henken,
and A. M. Bransen. 1987. The Wageningen respiration unit
for animal production: A description of the equipment and its
possibilities. In: M. W. A. Verstegen and A. M. Henken, editors, Energy metabolism in farm animals. Martinus Nijhoff,
Dordrecht, The Netherlands. p. 21–48.
Wenk, C., and A. J. H. van Es. 1976. Eine Methode zur Bestimmung
des Energieaufwandes für die körperliche Aktivität von wachsenden Küken. (In German.) Schweiz. Landwirtsch. Monatsh.
54:232–236.