Agriculture Reference
In-Depth Information
broiler breeders must be capable of predict-
ing voluntary food intake. Where this vari-
able is an input to the model, as is most
often the case, it is naive to believe that
feeding programmes can be successfully op-
timized, when the composition of the food
offered has such important effects on volun-
tary food intake. Food intake must therefore
be an output from, and not an input to, a
model. A reproducing animal needs to be
supplied with nutrients in order to meet the
requirements for maintenance of the body
and for reproduction. The theory of food in-
take and growth proposed by Emmans
(1981, 1989) is based on the premise that
birds attempt to grow at their genetic poten-
tial, which implies that they attempt to eat
as much of a given feed as would be neces-
sary to grow and reproduce at that rate. The
same principle can be applied to laying
hens (Emmans and Fisher, 1986). To calcu-
late the daily energy and nutrient require-
ments of a laying hen, her protein weight
(for maintenance) and potential protein and
lipid output (in eggs) needs to be known.
By comparing these requirements with the
content of nutrients in the feed, the 'de-
sired' feed intake can be determined: this is
the amount of feed that would be needed to
meet the requirement for the first limiting
nutrient in the feed (Emmans, 1981). The
bird may not be capable of consuming this
amount of feed, its intake possibly being
constrained by either the bulkiness of the
feed or the inability to lose sufficient of the
heat generated to the environment. In this
case feed intake will be less than desired
and performance would be compromised.
This theory has been shown to predict
food intake and hence growth and carcass
composition with considerable accuracy
(Ferguson and Gous, 1997, 2002; Ferguson
et al
.,
1997; Wellock
et al
.,
2004). Burnham
et al
. (1992) and Gous
et al
. (1987), among
many others, have shown that broilers and
laying hens increase food intake as the
limiting nutrient in the feed is reduced, at-
tempting thereby to obtain more of the
limiting nutrient, until a dietary concentra-
tion is reached where performance is so
constrained that food intake falls. The com-
mon misconception that 'birds eat to satisfy
their energy requirements' is clearly naive
and of no value in predicting voluntary
food intake.
The critical features of a model to pre-
dict food intake in hens would be predic-
tions of the body protein weight of the bird
and its potential egg output on each day,
from which nutrient requirements for main-
tenance and output may be calculated; a de-
scription of the nutrient content of the feed
on offer; and a description of the effective
temperature of the environment in which
the bird is housed. Although the principle
of predicting food intake is the same for
growing and reproducing birds, the descrip-
tion of potential growth and of egg output
differs markedly between the two.
The situation with broiler breeder hens
differs from that of full-fed laying hens in
that a daily allowance of feed is allocated,
this being less than would normally be con-
sumed if the birds were given
ad libitum
access to feed. Yet the principles applied
to voluntary intake prediction, described
above, remain: the difference is that the de-
sired food intake of the birds may not al-
ways be achieved, thus the actual food in-
take would be that constrained by the farm
manager. Consequently, egg output will be a
function of the amount of limiting nutrient
remaining after the maintenance require-
ment of the hen has been met. Whether the
consequences of underfeeding are more
likely to be evident with broiler breeders
than with commercial laying hens, given
that laying hens are fed
ad libitum
, would
depend on the daily amount of food allo-
cated to the breeders in relation to their po-
tential egg output, and on the density of the
feed allocated to the laying hens and the en-
vironmental temperature to which they are
subjected.
Not all hens in a broiler breeder flock
will consume exactly the same amount of
feed, some birds being more aggressive than
others. These differences in intake may be
accommodated in a simulation model by al-
locating an aggressiveness factor to each
hen based on a range of deviations from a
mean of zero (Wellock
et al
.,
2004). The cor-
relation between aggressiveness and poten-
tial egg output is not known, but could be
