Effect of Egg Weight and Position Relative to Incubator Fan
on Broiler Hatchability and Chick Quality1
O. Elibol* and J. Brake†2
*Department of Animal Science, Faculty of Agriculture, University of Ankara, Ankara 06110, Turkey;
and †Department of Poultry Science, North Carolina State University, Raleigh 27695-7608
ABSTRACT Two experiments, which included 3 incu-
bators, were carried out to investigate the effects of egg
weight and position relative to incubator (setter) fan on
embryonic mortality, second quality chicks, and fertile
hatchability of broiler eggs. Three egg weight groups
termed small (~62.4 g), average (~65.4 g), and large
(~68.9 g) were set in either the incubator trolley most
distant from the fan (FAR) or in the incubator trolley
nearest the fan (NEAR) as would be the case during
single-stage operation in this type of incubator. Fertile
hatchability decreased in the large egg weight group
due to increased percentage late embryonic mortality in
experiment 1, and both percentage early and late em-
bryonic mortality in experiment 2. Percentage late em-
bryonic mortality and second quality chicks increased
Key words: egg weight, incubator position, hatchability, chick quality, broiler hatching egg
INTRODUCTION
Long-term genetic selection has resulted in an enor-
mous increase in the growth rate of broilers and evi-
dently embryonic metabolism as Hulet and Meijerhof
(2001) reported that heat production of eggs from
modern broilers was substantially higher than that
reported in 1960. Furthermore, previous studies have
shown that large eggs did not hatch as well as small
eggs (Landauer, 1967; Ogunshile and Sparks, 1995;
French, 1997). Although there could be maternal ef-
fects and fertility could differ between large and small
eggs, French (1997) found that as egg mass increased
thermal conductance did not increase proportionally, so
larger eggs would be expected to have greater difficulty
losing embryonic metabolic heat as well as greater dif-
ficulty gaining heat during the initiation of incubation
©2008 Poultry Science Association Inc.
Received January 9, 2008.
Accepted May 19, 2008.
1
The use of trade names in this publication does not imply en-
dorsement of the products mentioned nor criticism of similar prod-
ucts not mentioned.
2
Corresponding author: jbrake@ncsu.edu
1913
and percentage fertile hatchability decreased for eggs
in the FAR position in experiment 1 only. A significant
interaction of incubator position × egg weight group
for late embryonic mortality, second quality chicks,
and fertile hatchability was found in experiment 1, but
only late embryonic mortality was so affected in experi-
ment 2. Experiment 2 was conducted so that eggshell
temperatures could be measured. Large eggs in the
FAR position at transfer time (E 18) exhibited signifi-
cantly higher eggshell temperatures than did the other
groups probably because air velocity or air distribution
was modified in the FAR position of the incubator and
large eggs were most negatively influenced in the trol-
ley in this position.
2008 Poultry Science 87:1913–1918
doi:10.3382/ps.2008-00014
(Lourens et al., 2005). When large and small eggs were
incubated under similar conditions, large eggs exhib-
ited higher temperatures during later incubation (Mei-
jerhof and van Beek, 1993; Meijerhof, 2002). French
(1994) found turkey hatchability to progressively de-
crease with increasing egg size at high air temperature
(38.5°C) but that large eggs exhibited improved hatch-
ability when incubated at a reduced air temperature
(36.5°C) during the second half of incubation mainly
due to a decrease in late embryo mortality. Lourens et
al. (2006) also observed that a higher heat production
required a lower air temperature for large eggs from E
15 onward, which suggested that embryonic growth in
large eggs increased at an increasing rate from E 15
onward. Adjusting incubator temperature avoided the
adverse effects of high egg temperature during the last
week of incubation on embryo development (Lourens
et al., 2005).
Measurement of air temperature around eggs within
incubators has shown that, depending on the design
of the incubator, air temperatures can differ between
0.4 and 3.0°C from the setpoint temperature (Kaltofen,
1969; Mauldin and Buhr, 1995; French, 1997). French
(2001) concluded from his review of previous work that
1914 Elibol and Brake
there was a strong correlation between the estimated
total metabolic heat production of the eggs within the
incubator and the air temperature around the eggs.
The total metabolic heat production of eggs was depen-
dent on the stage of embryo development, size of the
eggs, and fertility of the eggs (number of live embryos)
so that when either egg mass or fertility was increased
during the latter stage of incubation so did the air tem-
perature within the incubator (French, 2002).
The effectiveness of heat transfer from eggs to the
surrounding incubator air and uniformity of egg tem-
perature has been demonstrated to be mainly deter-
mined by the rate of air flow over the eggs as well as
difference between egg and air temperatures (Sother-
land et al., 1987; Owen, 1991; French, 1997). French
(2001) found variations of up to 1.2°C within an incu-
bator and that reducing this air temperature variation
required a uniform air flow throughout the incubator.
Preliminary observations of distinct differences in
fertile hatchability and chick quality between eggs
placed in trolleys that were positioned close to and
most distant from the incubator fan in a commercial
hatchery created an interest in characterization of the
details of this practical problem. The most obvious fac-
tor to examine in detail was egg weight (size). There-
fore, the present study was conducted to evaluate effect
of egg weight and position (location) within an incuba-
tor (setter) during incubation on embryonic mortality,
second quality chicks, and apparent hatchability of fer-
tile broiler eggs.
MATERIALS AND METHODS
Broiler hatching eggs were produced from 2 multi-
house commercial flocks of Ross 308 feather-sexable
strain females mated to Ross 344 males. Males and fe-
males had been grown sex-separate in light-controlled
facilities on an 8-h photoperiod and photostimulated at
21 wk of age. The feeding and BW programs were as
generally described by Ross (1998). Experimental eggs
were collected 4 times daily and stored for 2 d at 18°C
and 75% RH before setting. All remaining space in the
incubator was filled with eggs from the same flock that
had been stored from 2 to 5 d before setting.
In experiment 1, hatching eggs were obtained from
a flock at 51 wk of age. A large number of eggs were
weighed individually and divided into 3 egg weight
groups termed small, average, and large. Mean egg
weights were 62.5 ± 0.12, 65.6 ± 0.13, and 69.0 ± 0.17
g for the 3 groups, respectively. Each egg weight group
was randomly divided into 2 groups that were set in
either the wheeled incubator trolley most distant from
the fan (FAR) or in the trolley nearest the fan (NEAR)
as would be the case for single-stage operation (Figure
1). In experiment 1, two setters were used as machine
replicates. An incubation tray of 150 eggs constituted
an experimental replicate. There were a total of 60
trays and all egg weight-position combinations were
represented by 10 trays each (5 trays per incubator) for
a total of 9,000 eggs.
Experiment 2 was conducted as experiment 1 except
that the eggs came from a second multihouse flock at 58
wk of age and mean egg weight was 62.4 ± 0.16, 65.3 ±
0.14, and 68.8 ± 0.18 g for the small, average, and large
groups, respectively. In experiment 2, only one setter
was used and there were a total of 30 trays, thus, each
egg weight-position combination was represented by 5
replicate trays each for a total of 4,500 eggs. Eggshell
temperature was measured at E 18 of experiment 2
from 10 fertile eggs in each egg weight-position combi-
nation group in the middle of both the NEAR and FAR
dollies using an infrared ear thermometer (Braun Ear
Thermometer Type 6013, The Gillette Company, Bos-
ton, MA) by placing the sensor tip of the instrument,
previously allowed to equilibrate with incubator tem-
perature for 15 min, in direct contact with the equator
of the egg (Leksrisompong et al., 2007). The middle dol-
lies, which were full of eggs, were removed from the in-
cubators immediately before the eggshell temperature
measurements to make access to the eggs possible. The
eggs in the FAR trolley were measured first.
The machines used were a Petersime Model 576 set-
ter and a Model 192 hatcher. The setter air tempera-
ture set points were 37.4 ± 0.2°C dry bulb and 28.9 ±
0.2°C wet bulb. The hatcher air temperature set points
were 37.2 ± 0.2°C dry bulb and 30 ± 0.2°C wet bulb. In-
cubators were monitored remotely by computer 6 times
daily for proper operation. All experimental groups
were placed in a single hatcher at the time of trans-
fer on E 18 in both experiments, but relative positions
within the machines were maintained. The general air
flow and temperature patterns of this type of machine
have been described (Van Brecht et al., 2003).
At the time of removing the chicks from the hatch-
ers, all unhatched eggs were opened and examined
macroscopically by a single experienced individual to
determine percentage fertility and percentage embry-
onic mortality [early (E 0 to 6), middle (E 7 to 17), late
(E 18 to 21 plus pipped)]. Determination of fertility at
hatching has a small margin of error, which determi-
nation at any other time will also experience, but such
small errors should be randomly distributed and not
significantly affect the current results, based upon the
experience of the authors. Percentage fertile hatchabil-
ity was calculated as the number of first quality chicks
hatched per 100 fertile eggs set. Percentage second
quality chicks was calculated as the number that were
not able to stand properly or chicks that showed visible
signs of poor incubation conditions, such as improper-
ly healed navels, per 100 fertile eggs. Eggs that were
cracked were excluded from the analysis. The inci-
dence of contaminated eggs was less than 1% (data not
shown). The results were analyzed by ANOVA with the
GLM procedure of SAS (SAS Institute Inc., 1990). The
data of experiment 1 were initially analyzed as a 2 × 3
factorial with incubator as a block whereas egg weight
 EGG WEIGHT AND INCUBATOR POSITION
Figure 1. Design of the experiment showing the NEAR and FAR trolley relative to fan position and air flow. Within the trolley at each posi-
tion, eggs were placed in the 3 egg weight groups (large, average, and small) to complete a 2 × 3 design.
group and incubator trolley position were the main
effects in a randomized complete block design. There
were no effects found due to incubator, so the data from
the 2 incubators were combined and data of both exper-
iments 1 and 2 were analyzed as a completely random-
ized 2 × 3 factorial with egg weight group and incuba-
tor position as the main effects. Between-tray variation
(residual) was the source of the error term.
RESULTS AND DISCUSSION
Fertile hatchability, percentage second quality
chicks, and percentage embryonic mortality as affected
by egg weight and position relative to incubator fan in
experiment 1 are shown in Table 1. Fertility averaged
94.1% and did not differ due to treatment (data not
shown). Fertile hatchability decreased in the large egg
weight group primarily due to an increased percentage
late dead. Further, percentage late dead decreased as
egg weight decreased across all 3 egg weight classes.
Fertile hatchability was higher from the eggs in the
NEAR position as compared with the FAR position
largely due to significantly lower percentage late em-
bryonic mortality and percentage second quality chicks.
There were significant interactions of egg weight group
× position for percentage fertile hatchability, percent-
age second quality chicks, and percentage late dead
embryos.
Fertile hatchability, percentage second quality
chicks, and percentage embryonic mortality as affected
1915
by egg weight and position relative to incubator fan
in experiment 2 are shown in Table 2. Fertility aver-
aged 80.6% in this older flock but did not differ due to
treatment (data not shown). Fertile hatchability was
decreased significantly in the large egg weight group
due to increased percentage early and late embryonic
mortality. Fertile hatchability was only affected nu-
merically by incubator position in experiment 2. There
was also a significant interaction of incubator position
× egg weight for percentage late dead as in experiment
1. It was thought that the higher early deads in the
large egg weight group was due to a slower initial rise
in egg temperature (Lourens et al., 2005).
Hatchability of fertile eggs decreased with increas-
ing egg weight as percentage late embryonic mortal-
ity increased, as expected within a single batch of eggs
from one maternal source, in both experiments (Tables
1 and 2). This result was consistent with previous re-
ports (Landauer, 1967; Ogunshile and Sparks, 1995;
French, 1997). An explanation for increased late em-
bryonic mortality due to increasing egg size was that
larger eggs would be expected to have greater difficulty
initially achieving adequate embryonic temperature
and then losing embryonic metabolic heat during later
incubation (Lourens et al., 2005). The higher heat pro-
duction and increased difficulty of heat dissipation in
large eggs has been found to result in higher embryo
temperatures in large eggs (Meijerhof and van Beek,
1993; Meijerhof, 2002). A 70-g egg has a 27% larger
embryo but 8% less surface area for gaseous exchange
1916 Elibol and Brake
Table 1. The effect of egg weight and position relative to incubator fan on fertile hatchability of first quality chicks, second quality chicks, and
embryonic mortality in experiment 1

Position Egg weight1

relative Position,
Variable to fan Large Average Small mean ± SE

(%)
Fertile hatchability2 NEAR 91.00 ± 1.033 88.90 ± 1.033 90.56 ± 1.033 90.15 ± 0.60a
FAR 81.69 ± 1.033 87.68 ± 1.033 88.31 ± 1.033 85.89 ± 0.60b
Mean 86.34 ± 0.73b 88.29 ± 0.73a 89.44 ± 0.73a
Second quality chicks2 NEAR 1.15 ± 0.533 1.91 ± 0.533 1.55 ± 0.533 1.53 ± 0.29b
FAR 4.58 ± 0.533 2.36 ± 0.533 2.51 ± 0.533 3.15 ± 0.29a
Mean 2.86 ± 0.36 2.13 ± 0.36 2.03 ± 0.36
Early dead NEAR 3.93 ± 0.54 4.24 ± 0.54 4.52 ± 0.54 4.23 ± 0.31
FAR 5.37 ± 0.54 5.10 ± 0.54 4.35 ± 0.54 4.94 ± 0.31
Mean 4.65 ± 0.38 4.67 ± 0.38 4.43 ± 0.38
Mid dead NEAR 0.22 ± 0.15 0.14 ± 0.15 0.49 ± 0.15 0.28 ± 0.08
FAR 0.57 ± 0.15 0.43 ± 0.15 0.28 ± 0.15 0.43 ± 0.08
Mean 0.40 ± 0.10 0.29 ± 0.10 0.38 ± 0.10
Late dead NEAR 3.00 ± 0.523 3.96 ± 0.523 2.25 ± 0.523 3.07 ± 0.30b
FAR 6.87 ± 0.523 3.79 ± 0.523 3.57 ± 0.523 4.74 ± 0.30a
Mean 4.94 ± 0.37a 3.87 ± 0.37b 2.91 ± 0.37c
a,b
Main effect means ± SE for egg weight and position effects that possess different superscripts differ significantly (P < 0.05). There were 10
replicate trays of 150 eggs each used for each interaction mean divided equally among 2 similar incubators.
1
Mean egg weights from 51-wk-old flock were 62.5, 65.6, and 69.0 g for the small, average, and large groups, respectively.
2
Calculated based upon the number of first quality or second quality chicks hatched per 100 fertile eggs set.
3
There was a significant (P ≤ 0.05) interaction of egg weight and position effects as shown by the respective interaction means.

relative to a 55-g egg (Peebles and Brake, 1987) and
evidently for heat gain or dissipation as well because
thermal conductance does not change proportionally to
egg mass (French, 1997).
The eggshell temperature data of experiment 2 in
Table 3 suggested an interaction between increasing
egg weight and incubator location with respect to in-
creasing egg temperature at E 18. This would some-
what explain the position × egg weight interaction for
percentage late deads in experiment 2. Given the con-
sistency of these particular incubators (Van Brecht et
al., 2003), the significantly higher late dead embryos
found in experiment 1, and the effects of increased fer-
tility on egg temperature (French, 2002), it would be
highly probable that the effects on egg temperature
with respect to position and egg weight also occurred
in experiment 1. Difference in air velocity was a prob-
able explanation because it has been found to play an
important role in heat transfer from eggs to their en-
vironment. With greater air velocity, more heat will

Table 2. The effect of egg weight and position relative to incubator fan on fertile hatchability of first quality chicks, second quality chicks, and
embryonic mortality in experiment 2

Position Egg weight1

relative Position,
Variable to fan Large Average Small mean ± SE

(%)
Fertile hatchability2 NEAR 81.73 ± 1.45 86.06 ± 1.45 87.09 ± 1.45 84.96 ± 0.84
FAR 78.82 ± 1.45 84.54 ± 1.45 86.43 ± 1.45 83.26 ± 0.84
Mean 80.27 ± 1.02b 85.30 ± 1.02a 86.76 ± 1.02a
Second quality chicks2 NEAR 1.04 ± 0.32 0.51 ± 0.32 0.94 ± 0.32 0.83 ± 0.18
FAR 1.50 ± 0.32 0.99 ± 0.32 0.80 ± 0.32 1.10 ± 0.18
Mean 1.27 ± 0.22 0.75 ± 0.22 0.87 ± 0.22
Early dead NEAR 9.54 ± 1.00 6.70 ± 1.00 5.81 ± 1.00 7.35 ± 0.57
FAR 10.41 ± 1.00 7.32 ± 1.00 7.66 ± 1.00 8.46 ± 0.57
Mean 9.98 ± 0.70a 7.01 ± 0.70b 6.74 ± 0.70b
Mid dead NEAR 1.26 ± 0.21 0.99 ± 0.21 0.32 ± 0.21 0.86 ± 0.12
FAR 0.83 ± 0.21 0.17 ± 0.21 0.78 ± 0.21 0.59 ± 0.12
Mean 1.05 ± 0.15 0.58 ± 0.15 0.55 ± 0.15
Late dead NEAR 5.93 ± 0.753 5.21 ± 0.753 4.88 ± 0.753 5.34 ± 0.43
FAR 8.10 ± 0.753 5.98 ± 0.753 3.21 ± 0.753 5.76 ± 0.43
Mean 7.01 ± 0.52a 5.59 ± 0.52b 4.05 ± 0.52b
a,b
Main effect means ± SE for egg weight and position effects that possess different superscripts differ significantly (P < 0.05). There were 5
replicate trays of 150 eggs each used for each interaction mean set in a single incubator.
1
Mean egg weights from 58-wk-old flock were 62.4, 65.3, and 68.8 g for the small, average, and large groups, respectively.
2
Calculated based upon the number of first quality or second quality chicks hatched per 100 fertile eggs set.
3
There was a significant (P ≤ 0.05) interaction of egg weight and position effects as shown by the respective interaction means.
 EGG WEIGHT AND INCUBATOR POSITION
Table 3. Effect of egg weight and egg position relative to incubator
fan on eggshell temperature at transfer (°C) on E 18 in experiment
21
Position relative to fan
Egg weight NEAR FAR
Large 38.7 ± 0.1b
39.4 ± 0.1a
Average 38.8 ± 0.1b 38.8 ± 0.1b
Small 38.6 ± 0.1b 38.8 ± 0.1b
a,b
Means ± SE that possess different superscripts differ signifi-
cantly (P < 0.05).
1
There were 10 fertile eggs from each of 5 trays in each cell mea-
sured.
be removed from the eggshell during late incubation
or accumulated by the egg during early incubation. At
low air velocity, egg temperature has been found to in-
crease. Egg temperature increased with increasing egg
weight in the study of Meijerhof and Lourens (1999).
This could explain the position × egg weight interac-
tions for percentage late deads as well as percentage
fertile hatchability and second quality chicks because
the more evident effects in experiment 1 were consis-
tent with greater fertility (embryo heat) in the presence
of similar egg mass. There have also been problems re-
ported with chick quality from large eggs in the pres-
ence of a lower air flow in an incubator, which could be
consistent with the data of experiment 1. This means
that problems with embryo temperature may not only
be reflected in a reduced fertile hatchability but may
also influence chick quality and posthatch growth and
performance. Data from a recent study by Hulet et al.
(2007) found that chicks hatched from eggs with a high
eggshell temperature during the last 3 d of incubation
exhibited a lower BW at 44 d of age than chicks hatched
from eggs with lower eggshell temperature.
The present data support the suggestion of Meijer-
hof (2002) that the design of incubators should focus
primarily on providing a uniform embryo temperature
for all eggs within an incubator with a maximum dif-
ference in embryo temperature of 0.3°C between indi-
vidual eggs in the incubator as a goal. The results of
the present research can extend the conclusion of Mei-
jerhof (2002) to include more consistent temperatures
for larger eggs in areas of lower air flow to improve
heat gain or dissipation, fertile hatchability, and chick
quality and decreasing late dead embryos. However, in
practice this may be quite difficult because different air
temperatures have been shown to be required to main-
tain eggshell temperatures constant at 37.8°C in small
and large egg weight classes to obtain no significant
differences in embryonic mortality and fertile hatch-
ability (Lourens et al., 2006). One solution that has
been explored would be to use these single-stage in-
cubators on a multistage basis, placing the freshly set
eggs in the hot locations in the incubator and using the
cool locations for the eggs that were near the end of the
incubation process. Improved fertile hatchability re-
1917
sulted with larger old flock eggs but not smaller young
flock eggs in such an experiment (Elibol and Türkoğlu,
2001), suggesting a discrete egg size threshold, consis-
tent with the present and previous reports.
REFERENCES
Elibol, O., and M. Türkoğlu. 2001. Effect of single and multi-
stage incubation systems on hatching performance of broil-
er breeders eggs. Turk. J. Vet. Anim. Sci. 25:335–340.
French, N. A. 1994. Do incubation temperature requirements
vary between eggs? Proc. 9th Eur. Poult. Congr., Glasgow,
UK. Vol. II:395–398.
French, N. A. 1997. Modeling incubation temperature; the ef-
fects of incubator design, embryonic development and egg
size. Poult. Sci. 76:124–133.
French, N. A. 2001. Temperatures in a tunnel incubator—A
long term study. Avian Poult. Biol. Rev. 12:184–186.
French, N. A. 2002. The critical importance of incubation
temperature. Pages 17–20 in Practical Aspects of Com-
mercial Incubation. D. C. Deeming, ed. Ratite Conference
Books, Lincolnshire, UK.
Hulet, R., G. Gladys, D. Hill, R. Meijerhof, and T. El-Shiekh.
2007. Influence of egg shell embryonic incubation temper-
ature and broiler breeder flock age on posthatch growth
performance and carcass characteristics. Poult. Sci.
86:408–412.
Hulet, R. M., and R. Meijerhof. 2001. Real time incubation
temperature control and heat production of broiler eggs.
Poult. Sci. 80(Suppl. 1):128. (Abstr.)
Kaltofen, K. S. 1969. The effect of air movements on hatch-
ability and weight loss of chicken eggs during artificial in-
cubation. Pages 177–190 in The Fertility and Hatchability
of the Hen’s Egg. T. C. Carter and B. M. Freeman, ed.
Oliver and Boyd, Edinburgh, UK.
Landauer, W. 1967. The hatchability of chicken eggs as in-
fluenced by environment and heredity. Monograph 1 (Re-
vised), Univ. Conn. Agric. Exp. Stn., Storrs, CT.
Leksrisompong, N., H. Romero-Sanchez, P. W. Plumstead,
K. E. Brannan, and J. Brake. 2007. Broiler Incubation. 1.
Effect of elevated temperature during late incubation on
body weight and organs of chicks. Poult. Sci. 86:2685–
2691.
Lourens, A., R. Molenaar, H. van den Brand, M. J. W. Heet-
kamp, R. Meijerhof, and B. Kemp. 2006. Effect of egg size
on heat production and the transition of energy from egg
to hatchling. Poult. Sci. 85:770–776.
Lourens, A., H. van den Brand, R. Meijerhof, and B. Kemp.
2005. Effect of eggshell temperature during incubation on
embryo development, hatchability and post-hatch devel-
opment. Poult. Sci. 84:914–920.
Mauldin, J. M., and R. J. Buhr. 1995. What is really happen-
ing in your incubator? Int. Hatchery Pract. 9:19–22.
Meijerhof, R. 2002. Design and operation of commercial in-
cubators. Pages 41–46 in Practical Aspects of Commercial
Incubation. D. C. Deeming, ed. Ratite Conference Books,
Lincolnshire, UK.
Meijerhof, R., and S. Lourens. 1999. Embryo temperature
is the key factor in incubation. World Poultry-Elsevier
15:42–43.
Meijerhof, R., and G. van Beek. 1993. Mathematical model-
ing of temperature and moisture loss of hatching eggs. J.
Theor. Biol. 165:27–41.
Ogunshile, G., and N. Sparks. 1995. Effect of broiler egg
weight on hatchability. Br. Poult. Sci. 36:861–862.
Owen, J. 1991. Principles and problems of incubator design.
Pages 205–224 in Avian Incubation. S. G. Tullett, ed. But-
terworth Heinemann, London, UK.
1918 Elibol and Brake

Peebles, E. D., and J. Brake. 1987. Eggshell quality and hatch-
ability in broiler breeder eggs. Poult. Sci. 66:596–604.
Ross Breeders. 1998. Ross 308 Parent Stock Management
Manual. Ross Breeders Limited, Newbridge, Midlothian,
Scotland, UK.
SAS Institute Inc. 1990. SAS User’s Guide: Statistics. Ver-
sion 6 ed. SAS Institute Inc., Cary, NC.
Sotherland, P. R., J. R. Spotila, and C. V. Paganelli. 1987.
Avian eggs: Barriers to the exchange of heat and mass. J.
Exp. Zool. Suppl. 1:81–86.
Van Brecht, A., J. M. Aerts, P. Degraeve, and D. Berckmans.
2003. Quantification and control of the spatiotemporal
gradients of air speed and air temperature in an incuba-
tor. Poult. Sci. 82:1677–1687.