Sunteți pe pagina 1din 196

UNIVERSITÉ FRANÇOIS – RABELAIS DE TOURS

ÉCOLE DOCTORALE SSBCV


ÉQUIPE Dynamiques Nutritionnelles

THÈSE présentée par :

Maxime TRAINEAU
Soutenance prévue le : 29 AVRIL 2014

pour obtenir le grade de : Docteur de l’université François – Rabelais de Tours


Discipline/ Spécialité : Science de la vie et de la santé

Détermination du champ d’application du


système d’alimentation séquentielle chez les
poules pondeuses
THÈSE dirigée par :
M LESCOAT Philippe Pr, Agroparistech, Paris

CO ENCADREMENT par :
Mme BOUVAREL Isabelle Dr, ITAVI, Nouzilly

RAPPORTEURS :
M BESSEI Werner Pr, Université de Hohenheim
M DOURMAD Jean-Yves HDR, INRA UMR PEGASE Rennes

JURY :
M BESSEI Werner Pr, Université de Hohenheim
M BRESSAC Christophe MC-HDR, Université François Rabelais Tours
M DOURMAD Jean-Yves HDR, INRA UMR PEGASE Rennes
Mme LAUNAY Claire IA, INZO° Chierry
M LESCOAT Philippe Pr, Agroparistech, Paris
M NYS Yves DR, INRA – URA Nouzilly

1
2
Remerciements
Et c’est ainsi que cette aventure s’achève !!

Ces quelques lignes vont me permettre de remercier les personnes qui m’ont soutenu, aidé, et
épaulé pendant ces trois ans (et quelques mois) de thèse. Ce ne sera pas forcément très
conventionnel comme remerciements mais bon on m’a tellement répété que « les
remerciements c’est là où tu peux te lâcher ! » et bien allons y !

Je tiens tout d’abord à remercier les membres du jury Werner Bessei et Jean-Yves Dourmad
en qualité de rapporteurs ainsi que Christophe Bressac, Claire Launay et Yves Nys en qualité
d’examinateurs.

Je remercie toutes les personnes qui m’ont accordé leur confiance pour mener à bien ce projet
de thèse : Philippe Lescoat, Lucien Roffidal et Isabelle Bouvarel. Je me rappelle encore notre
première rencontre lors de l’entretien pour la thèse ; Pour l’anecdote c’est à cette occasion que
j’ai coupé mes cheveux !
Je remercie Lucien ainsi que toute l’équipe de INZO (Claire Launay, Caroline Mulsant et
Jean-Chales Bethuel) pour avoir soutenu ce projet, pour avoir réfléchi avec moi aux
différentes expériences à mener, d’avoir eu un œil critique sur mon travail et m’avoir permis
de présenter mes résultats à de nombreuses reprises.
Je remercie également Philippe Lescoat avec qui j’ai réellement aimé travailler. Je le
remercie de sa confiance, de son implication dans mon travail et de ses encouragements.
Merci également à Isabelle qui m’a refait confiance après mon stage de M2 et qui m’a apporté
de nombreux et précieux conseils durant ces trois années !

Je voulais également remercier celle qui m’a suivi depuis mes premiers pas de jeune stagiaire
M1 à l’INRA : Christine Leterrier ! Merci d’avoir toujours gardé un œil presque maternel sur
moi et même si tu aimais bien mes cheveux longs, le conseil de les couper venait de toi  !

Je tiens également à remercier chaleureusement Michel Couty et Philippe Didier sans qui je
n’aurais pas pu mener à bien ces expériences ! Au cours de ces 3 ans nous avons tout de
même pesé environ 10038 Kg d’aliment et déplacé 79968 gamelles… voyez bien que sans
eux ce n’était pas possible !! Je voulais également remercier l’ensemble des animaliers qui
sont venus le week-end échanger les gamelles et qui ont aidé lors des chantiers d’abattages.

3
Merci à Dolores Batonon avec qui j’ai vraiment aimé travailler, ensemble nous fait du bon
travail pour la 2ème expérience soldée par une publication en co-premiers auteurs !! Sans
oublier non plus nos sorties soirées passées ensemble.

Merci également à toutes les personnes de l’équipe avec qui j’ai vraiment passé de bons
moments à l’INRA ! Merci aux « filles de l’ITAVI » : Angélique, Marie, Laure, Amandine !
Un grand merci également aux membres de l’équipe DN : merci Michel Lessire et Jean Marc
pour m’avoir aidé quand j’en avais besoin, Merci à Nathalie Meme pour son aide précieuse
pour les dosage et les chantiers, merci Pauline Fraysse (attention la prochaine sur la liste ça
sera toi ! ), merci à Bertrand Meda le volleyeur cassé, pour ces conseils et son humour parfois
incompris, merci à Bastien Viot et Arnaud Lapébie les 2 bêtes sauvages avec qui on a bien
déliré ( surtout avec les écureuils ninja), merci à Juliette Lairez, Emilie, Marion, Justine.
Merci à Adeline Guillon et Marjorie Bouchier avec qui j’ai passé de très bons moments et de
très bonnes soirées (la fameuse balade du brame du cerf), merci à Maryse Lecomte de la part
de ton chouchou et merci à Xavière Rousseau pour les bonnes rigolades durant notre
colocation.

Je voulais aussi faire un petit big up à tous les copains du volley Flo, Julien, Papa Donnat,
Cyril, Claude, Céline, Massimo, Steph et tous les autres !!! Vous avez grandement contribué à
ma santé mentale en tant que soupape de sécurité !

Merci aussi à tous les autres copains de soirée, de vacances, de délire ! Merci à Aléma pour sa
participation à la thèse entre 2 verres de vin au Lioran, Merci à Cissou mon vieux pote depuis
maintenant 10 ans , Vincent Puard le fêtard invétéré qui m’a accompagné dans des bonnes
soirées et qui m’a cependant bien aidé pour ma thèse, merci à Tarik la grosse masse (AH
BOOONNN) présent si tu as besoin de te changer la tête, merci à Carolus la maman du volley
toujours là pour moi et qui nous a permis de suivre une véritable série TV tellement il y avait
d’histoires croustillantes ! Merci à Pauline Jenny pour sa bonne humeur perpétuelle et pour
son rire communicatif !

Enfin je tenais à remercier très profondément les personnes les plus proches à savoir mes
parents qui m’ont permis d’arriver jusqu’ici sans embuche et qui m’ont toujours soutenu dans
mes décisions, mon frère avec qui j’ai toujours tout partagé et à qui je souhaite de réussir ses
projets et enfin à Viviane qui a partagé au quotidien les hauts et les bas et qui m’a
énormément aidé et soutenu depuis maintenant plusieurs années.

4
5
6
Résumé
L’alimentation séquentielle (AS) consiste à distribuer de manière cyclique deux aliments
différents dont la combinaison permet des apports nutritionnels adaptés sur toute la journée.
Les premiers travaux donnaient des résultats très encourageants surtout en termes d’épargne
d’aliment Au cours de ces expériences, les poules recevaient le matin un aliment composé
uniquement de blé entier, riche en énergie et pauvre en protéines et l’après midi un aliment
complémentaire riche en protéines et calcium et pauvre en énergie. L’objectif de la thèse a été
d’évaluer les limites de ce système d’AS afin de pouvoir utiliser ce système de distribution à
échelle commerciale. Trois expérimentations ont été réalisées en station expérimentale mais
proches de conditions d’élevage. La première question portait sur la variation de la
composition en matières premières des régimes et la possibilité de substituer au blé entier une
fraction également riche en énergie et qui permettait les mêmes gains induits par la graine
entière (développement du gésier) (Expérience 1). La seconde question portait sur la capacité
d’adaptation des poules à des situations de déséquilibre alimentaire et de stress thermique
(Exp. 2) et enfin la dernière question portait sur la modulation des apports en énergie et en
protéines au cours de la journée (Exp. 3). Un travail de méta-analyse mené simultanément a
permis de quantifier les réponses sur la régulation de l’ingestion chez les poules pondeuses
ainsi que sur l’impact de la nature des ingérés sur les performances de ponte et de croissance.
Les expériences ont mis en évidence la possibilité de distribuer d’autres céréales (blé et/ou
maïs broyé, fibres insolubles) à la place du blé entier dans la fraction du matin avec un gain
inférieur au blé entier (IC) mais supérieur à l’alimentation continue et avec néanmoins un
développement des organes digestifs équivalent au blé entier (développement identique des
gésiers). L’ingestion totale inférieure en AS par rapport à une alimentation continue (Exp. 1 et
2) et un niveau de production équivalent permettent d’améliorer l’IC des poules en AS par
rapport à des poules alimentées de façon classique. Ce niveau d’ingestion inférieur provoque
cependant un poids vif inférieur pour les poules en AS et une composition corporelle
différente, les poules en AS étant moins grasses (Exp 1, 2). Cependant, concernant
l’ingestion, les poules semblent capables d’adapter leur ingestion sur le niveau d’énergie de
l’aliment mais pas sur son niveau de protéines (méta-analyse, Exp. 2). Pour la temporalité des
apports nutritionnels en AS, les offres en énergie doivent se concentrer dans la fraction du
matin afin d’obtenir les meilleures performances tandis que les apports protéiques peuvent
être soit concentrés dans la fraction de l’après midi soit continus au cours de la journée (Exp.
3). Ces connaissances sur l’AS doivent encore être complétées afin de pouvoir utiliser ce

7
système d’alimentation à grande échelle. Néanmoins l’AS est clairement prometteur en
termes d’économies de ressources.

8
Abstract
Sequential feeding (SF) consists in splitting the nutrients into two different diets
whose combination allows adequate nutrient supply throughout the day. In previous works
hens received whole wheat in the morning and in the afternoon a balancer diet (rich in protein
and calcium and low in energy). This early work gave encouraging results, especially for
cereal sparing with an improvement of FCR compared with continuous feeding. The goal of
this thesis was to explore the limits of this feeding system in order to use it on a commercial
scale. Three experiments were conducted in experimental station but close to on farm rearing
conditions. The first question was the utilzation of different feedstuffs and the possibility of
substituting whole wheat in the morning fraction by a cereal combination, rich in energy and
that gave the same gain as whole seeds (Experiment 1). The second question was about the
adaptability of hens to unbalanced dietary and heat stress (Exp. 2.). The last question was the
modulation of energy and protein supplies throughout the day in SF (Exp. 3). A meta-
analysis was done simultaneously to evaluate the regulation of feed intake in laying hens and
the impact of nutrient intake on laying and growth performances.
Experiments highlighted the possibility of distributing other cereals or feedstuffs
instead of whole wheat. This combination of cereal gave better performances than continuous
diet but lower than whole wheat. Nevertheless these combinations developed digestive organs
like whole wheat. The lower total feed intake in SF compared to a continuous distribution
(Exp. 1 and 2) and equivalent laying performances improved FCR for hens with SF. However,
the low level of feed intake induced a lower body weight and a different body composition
between SF hens and conventional feeding birds (SF hens being thinner) (Exp 1, 2).
Moreover, for feed intake, hens were able to modulate their intake on the energy level in the
diet but not on the level of protein (meta analysis, Exp . 2). Concerning the time of nutrient
distribution in SF, in order to get the best laying performances, energy supplies had to be
given in the morning fraction while protein could be given in the afternoon or throughout the
day (Exp. 3 ) . This knowledge is SF needs to be completed in order to use this feeding system
in commercial farms. SF is a very promising feeding system to spare natural resources.

9
10
Table des matières
Remerciements ........................................................................................................................... 3
Et c’est ainsi que cette aventure s’achève !!............................................................................... 3
Abstract ...................................................................................................................................... 9
Table des matières .................................................................................................................... 11
Liste des tableaux ..................................................................................................................... 13
Introduction .............................................................................................................................. 17
Chapitre 1 : Etude bibliographique ......................................................................................... 21
I Panorama de la production et de la consommation d’œuf au niveau mondial .................. 23
II Etat de la filière française ................................................................................................. 25
III Fonction de ponte ............................................................................................................ 28
III-1 Cycle de formation d’un œuf ................................................................................... 28
III-2 Description de la courbe de ponte ............................................................................ 32
III -3 Facteurs de variations de la courbe de ponte .......................................................... 33
IV Ingestion chez les volailles ............................................................................................. 35
IV-1 Cycle d’ingestion et évolution de la cinétique au cours de la journée ..................... 35
IV-2 Effet des modalités d’apport et de la forme de l’aliment sur l’ingestion :............... 36
IV-3 Impact de la composition des régimes sur l’ingestion ............................................. 38
V Alimentation séquentielle, système de distribution innovant ........................................... 41
Conclusion ............................................................................................................................ 45
Chapitre 2 : Relations entre composition des aliments, niveaux d’ingestion et performances
des poules pondeuses : approche par méta-analyse.................................................................. 47
Résumé du travail de méta-analyse ...................................................................................... 49
I Objectifs et questions de recherche .................................................................................... 51
II Construction de la base de données .................................................................................. 51
III Analyse des différentes bases de données ....................................................................... 52
III-1 Analyse de la base initiale........................................................................................ 53
III-2 Etablissement de lois de réponses : approches par étapes ....................................... 58
III-3 Analyses par GLM de la base initiale ...................................................................... 64
IV « Modèle relatif » énergie / protéines ............................................................................. 69
IV-1 Modèles GLM, modulation d’apports énergétiques (Tab 8) : ................................. 77
IV-2 Modèles GLM, modulation d’apports protéiques (Tab 9): ...................................... 77
V Analyse de la base en modèle relatif Lysine / Méthionine............................................... 78
11
V-1 Analyses par GLM base Lys et Met .......................................................................... 82
VI Conclusion et perspectives .............................................................................................. 87
VII Références de la méta-analyse ....................................................................................... 91
Chapitre 3 : Effet du blé broyé avec ou sans fibres insolubles et/ou blé entier sur les
performances des poules pondeuses en alimentation séquentielle ........................................... 97
Résumé Publication 1 : ......................................................................................................... 99
Chapitre 4 : Capacité des poules pondeuses à ajuster leur consommation d’aliment après s’être
vu offrir précédemment un régime nutritionnellement déséquilibré ...................................... 113
Résumé Publication 2 : ....................................................................................................... 115
Chapitre 5 : Modulation des apports énergétiques et protéiques en alimentation séquentielle
chez les poules pondeuses ...................................................................................................... 145
Résumé Publication 3 : ....................................................................................................... 147
Chapitre 6 Discussion Générale ............................................................................................. 177
Valorisation et Perspectives ............................................................................................... 183
Conclusion générale ........................................................................................................... 186
Références .............................................................................................................................. 189

12
Liste des tableaux
Tableau 1 Informations incluses dans la base ingestion des poules pondeuses et performances
.......................................................................................................................................... 54
Tableau 2 Description de la base initiale variables nominales et continues ............................ 56
Tableau 3 Influence de la nature des régimes et de l’ingestion sur les performances (ponte et
croissance) et sur la quantité ingérée en fonction de la classe de lignée et la classe d’âge.
Chaque ligne correspond à un modèle statistique comprenant deux facteurs et une ou
plusieurs variables explicatives continues. ....................................................................... 67
Tableau 4 Influence de la nature des ingérés (énergie, protéines, Lys, Met) sur les
performances (ponte et croissance) en fonction de la classe de lignée et la classe d’âge
Chaque ligne correspond à un modèle statistique comprenant deux facteurs et une ou
plusieurs variables explicatives continues. ....................................................................... 68
Tableau 5 Description de la base modèle relatif effet énergie / protéines variables nominales
.......................................................................................................................................... 70
Tableau 6 Description de la base modèle relatif effet énergie ................................................. 70
Tableau 7 Description de la base modèle relatif effet protéines .............................................. 70
Tableau 8 Influence de la quantité d’énergie dans les régimes de l’ingestion d’aliment et de
celle d’énergie sur les performances (ponte et poids vif ) et sur la quantité ingérée en
fonction des lignées et de ’âge. Chaque ligne correspond à un modèle statistique
comprenant deux facteurs et une ou plusieurs variables explicatives continues. ............. 75
Tableau 9 Influence de la quantité de protéines dans les régimes, de l’ingestion d’aliment et
de celle de protéines sur les performances (ponte, poids d’œuf et poids vif) et sur la
quantité ingérée en fonction de la classe de lignée et la classe d’âge. Chaque ligne
correspond à un modèle statistique comprenant deux facteurs et une ou plusieurs
variables explicatives continues ....................................................................................... 76
Tableau 10 Description de la base modèle relatif Méthionine / Lysine variables nominales .. 79
Tableau 11 Description de la base modèle relatif effet Méthionine ........................................ 79
Tableau 12 Description de la base modèle relatif effet Lysine ................................................ 79
Tableau 13 Influence de la quantité de Lys et Met dans les des régimes et de l’ingestion
d’aliment sur les performances (ponte et poids vif) et sur la quantité ingérée en fonction
de la classe de lignée et la classe d’âge. Chaque ligne correspond à un modèle statistique
comprenant deux facteurs et une ou plusieurs variables explicatives continues .............. 84

13
Tableau 14 Influence de la quantité de Lys et Met dans ingérée sur les performances (ponte et
poids vi) en fonction de la classe de lignée et la classe d’âge Chaque ligne correspond à
un modèle statistique comprenant deux facteurs et une ou plusieurs variables explicatives
continues........................................................................................................................... 85

14
Liste des figures

Figure 1 Fonctionnement de la filière pondeuse en France (Elaboration ITAVI, 2011) Erreur !


Signet non défini.
Figure 2 Cycle de formation d'un œuf (Adapté Sauveur, 1988) .............................................. 29
Figure 3 Evolution du taux de ponte et du poids des œufs en fonction de l'âge (Courbe de
ponte théorique ISA BROWN, 2013) .............................................................................. 33
Figure 4 Effets opposés de la longueur des cycles lumineux sur le taux de ponte et le poids
d’œuf (d’après Shanawany, 1982).................................................................................... 34
Figure 5 Evolution de l'ingestion au cours d'une journée de distribution (nuit de 22h à 6h)
(d'après Chah et Moran, (1985)) ...................................................................................... 35
Figure 6 Adaptation de la consommation d'aliment en fonction de la concentration en énergie
du régime (Bouvarel et al., 2010) .................................................................................... 39
Figure 7 Poids moyen de l'œuf en fonction de la teneur en méthionine totale de l'aliment
(Bouvarel et al., 2010)...................................................................................................... 40
Figure 8 Composition des régimes : EMAn, protéines brute, acides aminés ........................... 55
Figure 9 Performances de ponte : taux de ponte, poids des œufs et masse d’œuf ................... 55
Figure 10 Ingestion : aliment, énergie, protéines, Indice de Consommation et poids ............. 55
Figure 11 Caractéristiques des régimes : représentation partielle du plan d’expérience méta-
analytique (Chaque couleur correspond à un essai) ......................................................... 57
Figure 12 Influence de la composition des régimes sur l’ingestion ......................................... 58
Figure 13 Influence l’âge et de la souche sur l’ingestion des poules ....................................... 59
Figure 14 Influence de la composition nutritionnelle des régimes sur l’ingestion en fonction
de la souche et de l’âge des poules ................................................................................... 59
Figure 15 Influence de la souche et de l’âge sur les performances de ponte ........................... 60
Figure 16 Influence de la quantité d’aliment ingéré sur la production d’œuf .......................... 61
Figure 17 Influence de la nature et du niveau de l’ingéré sur la production d’œuf (énergie et
protéines) .......................................................................................................................... 62
Figure 18 Influence de la nature et du niveau de l’ingéré sur la production d’œuf (Lysine et
Méthionine) ...................................................................................................................... 63
Figure 19 Modèle d’évolution de l’ingestion ou des performances de ponte pour des poules
brunes, de classe d’âge 2M avec trois niveaux de PB ...................................................... 66
Figure 20 Influence de la composition nutritionnelle des régimes sur l’ingestion .................. 71

15
Figure 21 Influence de la nature et du niveau de l’ingéré sur la production d’œuf ................. 72
Figure 22 Influence de la nature et du niveau de l’ingéré sur le poids vif et l’Indice de
Consommation ................................................................................................................. 73
Figure 23 Modèle d’évolution des performances de ponte (taux de ponte et poids des œufs)
pour des poules brunes, de classe d’âge 2M avec trois niveaux de PB ............................ 78
Figure 24 Influence de la composition des régimes sur les performances de ponte ................ 80
Figure 25 Influence de la nature et du niveau de l’ingéré sur la production d’œuf ................. 81
Figure 26 Modèle d’évolution de la masse d’œuf en fonction de la composition en Lys ou Met
pour des poules brunes, de classe d’âge 2M avec trois niveaux d’ingestions .................. 83

16
Introduction

17
18
Introduction générale

Dans la filière pondeuse, l’aliment a une place extrêmement importante car le coût de
celui-ci représente les 2/3 du prix de revient de l’œuf (ITAVI, 2013). Toute amélioration des
systèmes d’alimentation permettrait des économies financières importantes. Les principales
réflexions afin d’optimiser la valorisation des aliments tournent autour de systèmes
d’alimentation innovants, d’utilisation plus variée de céréales (afin d’être moins soumis aux
spéculations du marché) mais également d’une utilisation de matières premières alternatives
et de coproduits issus de processus de transformation de céréales (Blesbois, 2010).
L’alimentation des poules est un des enjeux majeurs pour la filière, la production d’un œuf
par la poule étant très fortement dépendante de la quantité et de la qualité de l’ingéré. La mise
en place de systèmes de distribution innovant est une des solutions qui permettront
d’optimiser l’utilisation de l’aliment dans la filière pondeuse. La gestion d’un régime
alimentaire complexe (différent d’un aliment un unique distribué sur la journée) est un
système qui reste aujourd’hui peu utilisé à échelle commerciale. Plusieurs systèmes
d’alimentation permettent de distribuer non pas un aliment unique au cours de la journée mais
des aliments différents, mimant ainsi partiellement des conditions d’alimentation plus variées.
Le mélange de céréales entières avec un complémentaire ou la distribution séparée mais
simultanée de plusieurs fractions nutritionnellement différents, sont deux type de distribution
qui permettent une alternative aux distributions classiques continues. Ces types de
distributions permettent aux animaux d’exprimer leurs choix et leurs préférences alimentaires
ce qui n’aboutit pas forcément à un équilibre optimal des besoins quotidien de la poule.
Une autre technique est la distribution séquentielle de l’aliment. Le principe étant de
distribuer de façon cyclique deux aliment différents dont la combinaison permet des apports
équilibrés au cours de la journée Les différents travaux concernant ce type de distributions
sont détaillés plus précisément dans le chapitre suivant. Les principaux résultats montrent une
amélioration de l’indice de consommation par l’utilisation de ce type de distribution
séquentielle. Lors de ces précédents travaux (Umar Faruk et al ., 2010) la fraction du matin
contenait uniquement du blé entier ( riche en énergie et pauvre en protéines) et la fraction de
l’après midi un complémentaire (riche en protéine / calcium et pauvre en énergie).

L’objectif de la thèse a été d’étudier les champs d’application du système


d’alimentation séquentielle chez la poule pondeuse et donc de mieux comprendre et évaluer
19
les limites de ce système d’alimentation. Le travail portait sur la possibilité de diversifier les
apports dans les deux fractions proposées aux poules en alimentation séquentielle.

Dans ce document, une première partie bibliographique a permis de faire le point sur
la fonction de ponte chez la poule, puis les facteurs de variation de l’ingestion chez la poule et
enfin un panorama des différents travaux portant sur l’alimentation séquentielle.
Le chapitre 2 a permis d’approfondir, avec une méta-analyse les relations entre
composition des aliments, niveaux d’ingestion et performances des poules pondeuses. Ce
travail de méta analyse a permis dans un premier temps de définir quels étaient les facteurs
nutritionnels qui intervenaient sur les variations de l’ingestion puis de déterminer quel était
l’impact de la nature des ingérés sur la production chez la poule.
Trois publications relatives aux trois expérimentations réalisées en station
expérimentale sont ensuite présentées. La première publication, réalisées avec des poules
élevées en cage collective portait sur la possibilité de substituer le blé entier de la fraction du
matin par une combinaison d’autres céréales tout en conservant l’effet bénéfique de la graine
entière dans le cadre d’une distribution séquentielle. Cependant ces travaux ont mis en
évidence une fragilité des poules placées en AS lorsque celles-ci sont soumises à une forte
élévation de température mais également à la difficulté de faire ingéré aux poules la fraction
du matin riche en énergie. La seconde expérimentation a porté sur la capacité des poules
pondeuses à ajuster leur consommation d’aliment après s’être vu offrir précédemment un
régime nutritionnellement déséquilibré ainsi que l’impact de températures élevées sur la
distribution séquentielle de l’aliment. L’apport de régimes déséquilibrés en énergie et en
protéines a cependant posé la question de la temporalité de ces apports nutritionnels sur la
journée. La troisième expérimentation portait sur la variation des apports énergétiques et
protéiques en alimentation séquentielle chez les poules pondeuses. L’objectif était donc de
pouvoir évaluer le moment de la journée le plus opportun pour les apports en énergie et en
protéine.
Le dernier chapitre de discussion est une synthèse de l’ensemble des travaux, avec une
confrontation aux résultats de la littérature et une évaluation des perspectives restant sur les
futurs travaux en alimentation séquentielle.

20
Chapitre 1 :
Etude bibliographique

21
22
I Panorama de la production et de la consommation d’œuf au niveau
mondial
La production mondiale d’œufs est en croissance depuis 2000 et a atteint 65 millions
de tonnes aux derniers recensements (FAO, 2013). Sur cette dernière décennie, la production
est très dynamique avec une croissance annuelle moyenne de 2,2%. Le principal producteur
d’œufs est aujourd’hui la Chine (37% de la production mondiale) suivi de l’UE (10%) puis les
Etats Unis et l’Inde). Cette forte croissance est en grande partie due à la forte valeur nutritive
de l’œuf qui est la protéine animale consommable la moins chère (rapport entre la teneur en
protéine de l’œuf et son prix) par rapport aux autres sources de protéines animales comme la
viande. Le commerce mondial d’œufs et d’ovoproduits reste principalement dominé par les
échanges intra-communautaires. La majeure partie des échanges d’œufs coquille ou
d’ovoproduits liquides se déroulent entre les pays de l’UE. En 2010, les échanges
internationaux ont porté sur 2 millions de tonnes d’œufs coquille, 55800 tonnes de poudre
d’œufs et 272 tonnes d’ovoproduits liquides. Les échanges d’œufs coquille représentent
environ 78% des échanges d’œufs et d’ovoproduits (Magdelaine et al., 2010). Néanmoins, les
échanges d’ovoproduits ont connu une très forte augmentation pendant cette dernière
décennie (+120% pour les œufs séchés et +90% pour les œufs liquides). Il existe trois types
de pays producteurs d’œufs dans le monde. Tout d’abord, il s’agit des pays dans lesquels la
production est très développée (la Chine, les pays de l’Union Européenne et les Etats-Unis). Il
y a ensuite des zones de productions émergentes qui prennent une place de plus en plus
importante sur le marché mondial avec des pays tels que l’Inde, le Japon le Mexique et le
Brésil.
La Chine est aujourd’hui le plus gros producteur mondial d’œufs dans le monde (37 %
de la production mondiale). De 1991 à 2000, la production chinoise s’est beaucoup
développée avec une augmentation d’environ 8 % par an pour passer de 9.46 millions de
tonnes en 1991 à 22.21 millions de tonnes en 2000. Ce développement est en fait surtout dû à
l’émergence de petits producteurs dans les régions du Nord de la Chine (dix provinces
assurent 80% de la production chinoise). Mises à part deux provinces (Shandong et Hebei)
composées d’élevages très industrialisés fournissant la grande distribution, la production
chinoise est essentiellement tournée vers les marché locaux avec des petites unités de
production (un million d’éleveurs ont moins de 200 poules) (Magdelaine et Braine, 2010).

23
Toutefois, depuis 2001, la production chinoise connait un relatif mouvement de concentration
des unités de production, privilégiant ainsi les grandes exploitations (actuellement en Chine,
cinq fermes de pondeuses possèdent une capacité de production de plus d’un million de
poules). Cette industrialisation de la production permet à la Chine de se tourner peu à peu vers
l’exportation d’œufs coquille. Enfin, que ce soit dans les grandes ou les petites unités de
production, les poules pondeuses élevées dans le pays sont élevées en cage (Bell and Weaver
2002).
La production d’œufs de l’UE à 27 représente 10% de la production mondiale. Les
principaux pays producteurs d’œufs sont la France, l’Allemagne, l’Italie, l’Espagne, les Pays-
Bas et la Pologne. Leur production est relativement stable depuis la dernière décennie même
si l’on observe une baisse d’environ 2,5 % entre 2011 et 2012 (FAO, 2013). Cette baisse peut
être principalement imputée à la mise aux normes des cages bien-être. En effet, les systèmes
de production européens ont dû s’adapter aux exigences des consommateurs mais également à
l’évolution de la législation liée au bien être des poules en cage (2012, directive européenne
1999/74/CE). Cependant, au début de l’année 2013, les nombreux élevages ayant déjà été mis
aux normes ont permis à la production d’œufs de retrouver son niveau de 2011. La spécificité
de la production de l’UE à 27 est que 32 % de la production dépendait, en 2008, de systèmes
alternatifs à la cage (plein air, sol, biologique), le pourcentage grimpant à 42% en 2011. La
consommation des œufs au sein de l’UE est d’environ 6,1 millions de tonnes soit environ 200
par habitant. Il existe cependant de fortes différences de consommation entre les différents
pays européens (180 œufs/habitant en Espagne et aux Pays Bas contre presque 300 pour le
Danemark). C’est donc un ensemble de facteurs qui vont influencer le devenir de la
production européenne d’œufs : l’évolution de la règlementation (bien être, environnement),
la volatilité du prix des matières premières, les volontés des consommateurs et surtout la
possible ouverture du marché européen à de nouveaux concurrents comme la Chine ou le
Brésil.
Les Etats-Unis produisent environ 9 % de la production totale d’œuf, se plaçant ainsi
comme le 3éme producteur d’œufs au Monde. La croissance de la production reste positive
depuis ces 10 dernières années (+ 1.2%/an) avec néanmoins une croissance plus faible que la
Chine (+2.6% /an) mais supérieure à la croissance européenne (-0.3%/an). La production nord
américaine est très concentrée. En effet, 255 entreprises américaines possèdent 95 % du
cheptel des pondeuses. Actuellement, il existe 61 entreprises productrices d'œufs ayant plus
d’un million de poules, l’ensemble représentant environ 87% de la production totale. Parmi

24
elles, 16 entreprises possèdent un cheptel de plus de 5 millions de poules. Cette importante
capacité de production permet aux Etats-Unis d’exporter de nombreux produits. Tout d’abord,
les œufs-coquille sont largement exportés, principalement vers Hong Kong, le Mexique, les
Emirats Arabes Unis, l'UE-27 et le Canada, ces cinq marchés représentant 86,4% des
exportations nord-américaines, soit 17,69 millions de dollars (US Department of Agriculture,
Egg Board américain et USAPEEC). Ensuite, les exportations d’ovoproduits transformés, qui
sont majoritairement à destination du Japon, représentent 33,5% du total des exportations des
ovoproduits dans le monde. Pour finir, la consommation intérieure d’œufs aux Etats-Unis est
supérieure à la consommation moyenne de l’UE avec 248 œufs par habitant (FAO, 2013).

Depuis 1998, des nouveaux pays producteurs d’œuf ont fait leur apparition sur le
marché mondial. L’Inde, le Mexique et le Brésil sont les principaux nouveaux pays qui sont
entrés dans les échanges internationaux. En effet entre 1998 et 2008 le Brésil à augmenté sa
production de 2,8% / an, le Mexique de 4,8% / an et l’Inde de 5.4% / an (Magdelaine et
Braine, 2010). La production de ces pays est inférieure aux trois pôles principaux de
production (UE, USA et Chine) mais sont sur une dynamique de développement très rapide et
seront dans la prochaine décennie des pays qui pèseront sur les échanges internationaux
d’œufs.

II Etat de la filière française


Tout d’abord, la production française a connu une période de croissance dans les
années 1990, avec près de 16 milliards d'œufs produits en 1999. Depuis les 15 dernières
années cependant, elle est globalement en légère baisse. Ainsi, en 2012, la France aurait
produit 12.5 milliards d'œufs (soit 761 000 tonnes), en repli de 2.9 % par rapport à 2011
(ITAVI). En fait, si on observe plus attentivement, on remarque que depuis 2002, le secteur
œufs et ovoproduits est déficitaire en volume mais excédentaire en valeur, avec des évolutions
divergentes. En volume, la France est importatrice nette d'œufs en coquille et exportatrice
nette d'ovoproduits. En valeur, la France est en général excédentaire et dégage un bénéfice
commercial compris en 10 et 40 millions d'euros selon les années. Néanmoins, l’année 2012
se caractérise par un retournement de la tendance puisque le secteur œufs et ovoproduits
apparait déficitaire de 48.5 M€ contre un excédent de 4.7 M€ en 2011, ce qui participe au
déficit global des échanges d'œufs en coquille et d’ovoproduits. Ainsi, le taux d'autosuffisance

25
global est estimé à 93 % en 2012. En ce qui concerne la consommation, celle-ci est en

Figure 1 Fonctionnement de la filière pondeuse en France (Elaboration ITAVI, 2011

26
France d’environ 209 œufs par an et par habitant (123 œufs coquille et 86 sous forme
d’ovoproduits). En outre, la part allouée à la consommation d’œufs produits dans des
systèmes de production alternatifs tend à augmenter. Afin de répondre à cette évolution de la
demande et aux différentes législations, la production française, provenant pour 43% de
Bretagne, tend alors à se diversifier puisque en 2008 18.7% des poules françaises sont élevées
dans des systèmes alternatifs (biologique, label rouge, plein air et au sol) (Magdelaine et al.,
2011). La caractéristique principale de ces systèmes alternatifs français est que la majeure
partie des poules ont accès à un parcours extérieur. Seulement 15% des systèmes alternatifs
sont des systèmes de volières en France. Cependant, si l’on compare aux autres producteurs
européens, la France utilise proportionnellement moins de systèmes alternatifs à la cage : 20%
des élevages français de pondeuses n’utilisent pas de cage, contre 50% au Royaume-Uni, 57%
aux Pays Bas, et quasiment 90% en Allemagne (Magdelaine et al., 2011). En raison de cette
diversité des systèmes d’élevage français, la taille des cheptels dans les exploitations varie
énormément. En effet, la capacité moyenne de 24 000 pondeuses en 2004 cache une grande
diversité de situations, les productions en cage ayant une moyenne de 46 000 individus, et les
productions plein air ayant une moyenne de 7000 individus seulement. Globalement, en 2010,
la filière pondeuse comprenait 7 couvoirs, 470 élevages de poulettes et 1700 élevages de
pondeuses alimentant 65 centres d’emballage et 56 fabricants d’ovoproduits. (Figure 1).

27
III Fonction de ponte

III-1 Cycle de formation d’un œuf

III-1-1 Formation d’un œuf


La fonction de ponte chez la poule est une production cyclique qui apparait à partir de
19-20 semaines d’âge. La formation d’un œuf a bien été étudiée et décrite par Sauveur en
1988. D’un point de vue physiologique, l’appareil reproducteur de la poule se divise en deux
parties importante : les ovaires, essentiellement constituées de tissu conjonctif et de cellules
interstitielles servant à la synthèse des hormones stéroïdiennes et l’oviducte qui capte le jaune
et le conduit lors de la formation du blanc et de la coquille jusqu’au cloaque (Sauveur 1988).
L’ovaire comprend dès la naissance tous les follicules qui donneront plus tard les jaunes mais
seul l’ovaire gauche va se développer et devenir fonctionnel pour la reproduction future. Sous
le contrôle des hormones gonadotropes, l’ovaire des oiseux sécrète trois principales familles
de stéroïdes sexuels : œstrogène, androgènes et progestérone. Les œstrogènes (synthétisés par
les cellules interstitielles) servent à la croissance de l’oviducte, à la synthèse par le foie des
protéines et des lipides du jaune, au transport sanguin des lipoprotéines et du calcium et à la
synthèse des protéines du blanc. En outre, les oestrogènes sont essentiels dans le contrôle du
comportement d’oviposition. Les androgènes sont des stéroïdes sexuels typiquement mâles,
ils n’ont donc chez la femelle que peu d’action. Ils participent néanmoins à la stimulation du
développement de la crête et, en synergie avec les oestrogènes, au développement de
l’oviducte. Enfin, la progestérone sert au contrôle des activités cellulaires impliquées dans la
croissance de l’oviducte et de certaines protéines du blanc. Elle contrôle également les
rythmes d’ovulation et d’oviposition en agissant sur la libération de LH-RH par
l’hypothalamus (Nys et al., 2011). L’oviducte gauche, qui se situe à proximité de l’ovaire
gauche, est formé de six parties distinctes : l’infundibulum (qui capte l’ovule et qui permet la
fécondation en présence de spermatozoïdes), le magnum (longue partie de l’oviducte où
s’effectue la synthèse et le dépôt du blanc), l’isthme de l’utérus (où les deux membranes
coquillères sont synthétisées et déposées), l’utérus (où se déroule la calcification de la
coquille), le vagin (où la coquille est terminée et colorée) et enfin le cloaque (qui permet
l’expulsion de l’œuf formé) (Nys et al., 2011). La production presque quasi journalière d'un
œuf par la poule est rendue possible par le développement simultané des follicules ovariens,
conduisant à la présence journalière d'un seul follicule prêt à ovuler. En moyenne, une poule
pond un œuf tous les jours pendant 3-6 jours consécutifs puis s'arrête pour une journée ou
28
SOXV /RUV G¶XQ F\FOH FKDTXH RYLSRVLWLRQ HVW VXLYLH  PLQXWHV SOXV WDUG SDU XQH QRXYHOOH
RYXODWLRQ/DIRUPDWLRQGHO
°XIHVWGRQFSULQFLSDOHPHQWUpJLHSDUODVpFUpWLRQGHVKRUPRQHV
VWpURwGHVGHO
RYDLUHPDLVpJDOHPHQWFRQWU{OpHHWV\QFKURQLVpHSDUODGXUpHG¶pFODLUHPHQW(Q
HIIHW XQH DXJPHQWDWLRQ GH OD ORQJXHXU GX MRXU SURYRTXH XQH DXJPHQWDWLRQ GH O
DFWLYLWp
RYDULHQQH/DIRUPDWLRQG¶XQ°XIHVWGRQFXQSURFHVVXVFRPSOH[HUpJLjODIRLVSDUO¶DQLPDO
OXLPrPHPDLVpJDOHPHQWSDUOHVFRQWUDLQWHVHQYLURQQHPHQWDOHV

 


)LJXUH&\FOHGHIRUPDWLRQG
XQ°XI $GDSWp6DXYHXU 

/¶LQJHVWLRQG¶DOLPHQWVHVWIRUWHPHQWOLpHjODIRUPDWLRQGHO¶°XIHWjO¶RYLSRVLWLRQ(Q
HIIHW :RRG*XVKHW +RUQH  HW 7D\ORU   RQW ELHQPLVHQpYLGHQFHXQHLQJHVWLRQ
G¶DOLPHQWV VXSpULHXUH ORUV G¶XQH RYXODWLRQ HW G¶XQH RYLSRVLWLRQ SDU UDSSRUW j XQ MRXU VDQV


Cette augmentation d’ingestion est de l’ordre de 13 à 14% pour des poules blanches un jour
de ponte par rapport à un jour sans formation d’oeuf. En effet, selon Morris et Taylor (1967),
la consommation d’aliment le matin par des poules lors de la sécrétion d’albumen (28,1 g / p)
n'est pas comparable à celle sans sécrétion et donc formation d’œuf (19,7 g / p). Les variations
de la consommation répondent aux variations des besoins en acides aminés pour la synthèse
des protéines de l'œuf. L'étude de Shevchenk et Sherapanov. (1986) montre qu'environ 70%
des protéines du blanc sont synthétisées dans l’oviducte. Ils ont donc suggéré que la poule a
besoin de plus de protéines au cours de la matinée (moment où le jaune se trouve dans le
magnum), ce qui correspond au moment de la synthèse des protéines du blanc. Au cours de
cette période, un déficit de synthèse des protéines dans le magnum peut entraîner une
réduction de blanc d'œuf et donc du poids des œufs. Chah et Moran (1985), ont également
montré que lorsque la poule a la possibilité de choisir ses aliments (un aliment riche en
énergie, un aliment riche en protéines, et un dernier riche en calcium), celles-ci préfèrent
consommer l’aliment riche en protéines au moment du dépôt d’albumen. L'augmentation
simultanée de l'apport en protéines avec une ovule en transit à travers le magnum, semble être
une réponse à une synthèse accrue d'albumine sous l’influence des stéroïdes ovariens (Edwin
et Moran, 1986). Cependant ces auteurs concluent que la synthèse d’albumen étant continue
au cours de la journée, les apports en protéines doivent donc être continus au cours de la
journée. Il parait donc difficile de définir avec certitude le moment d’apport optimal d’une
grande quantité de protéines dans la journée, (apport le matin pour Shevchenko et al. (1986),
apport continu pour Edwin et Moran (1986)).
Un des éléments essentiels pour la fabrication de la coquille est le calcium. Là aussi la
consommation de celui-ci est strictement liée à la formation de l’œuf. Plusieurs auteurs
indiquent que les pics de consommation de calcium dans la journée sont liés à la position de
l’œuf dans l'oviducte. Mongin et Sauveur (1974) ont montré qu’en apportant le calcium sous
forme particulaire séparée de la ration d’aliment, les poules montraient un appétit spécifique
pour le calcium au moment de la formation de la coquille. Lorsque le calcium est mélangé
dans les granulés, l’augmentation de la consommation associée à la formation d’un œuf
s’expliquerait donc par cet appétit spécifique pour le calcium.

III-1-2 Synthèse des constituants de l’œuf


Le jaune :
La vitellogénèse (accumulation de jaune à l’intérieur d’un follicule ovarien) est un
processus long qui commence chez la jeune poulette puis se termine juste avant l’ovulation.
30
Cette phase fait uniquement appel à des constituants transportés par voie sanguine. Lorsque le
follicule est « sélectionné » dans le pool des follicules indifférenciés, celui-ci grossit pendant
60 jours (dépôts de protéines et de lipides). Huit jours avant ovulation, la masse de ce
follicule augmente rapidement jusqu’ à se détacher de la grappe ovarienne et passer dans
l’oviducte. Les constituants de ce jaune sont : des livétines (protéines du plasma qui
s’accumulent dans le jaune et qui sont synthétisées à 520% par le foie), des phosvitines
(protéines spécifiques du jaune synthétisées uniquement par le foie) et des lipoprotéines
(également synthétisées à 100% pour le foie uniquement pour servir de constituant au jaune)
(Sauveur 1988). Ces protéines et lipides sont synthétisés par le foie (jusqu’à 2,5g par jour)
sous le contrôle des œstrogènes. Ainsi, lorsque la poule est en ponte, le taux de synthèse par le
foie est constant au cours de la journée, ce qui correspond à un dépôt lui aussi en continu.
Le blanc :
La sécrétion et le dépôt des protéines du blanc se font dans le magnum (partie
supérieure de l’oviducte). Le blanc de l’œuf est une solution aqueuse (88% d’eau) avec des
protéines (synthétisées et sécrétées uniquement par le magnum) et des minéraux. La
production de protéines est à peu près continue chez une poule en ponte même si cette
synthèse est plus active juste après le passage d’un jaune. Cette augmentation de synthèse
protéique est due à une augmentation de l’activité métabolique induite par le passage d’un
jaune. Les principales protéines sécrétées sont l’avidine, l’ovomucine (toutes deux sécrétées
par les cellules calciformes de l’epithélium), l’ovalbumine et le lyzozome (tous deux sécrétés
par les cellules des glandes tubulaires) (Nys et al., 2011). Ces protéines ainsi produites se
déposent sur le jaune, pénétrant dans le magnum. Ce dépôt dure environ 3h30. Au fur et à
mesure de son transit, les protéines du blanc accumulées dans les cellules épithéliales, sont
libérées dans la lumière du magnum. Cette libération est provoquée par la distension tissulaire
induite par le passage du jaune. Cette arrivée massive de protéines est ainsi issue d’une
stimulation mécanique et ne demande donc par d’énergie spécifique.
La coquille :
L’œuf qui pénètre dans l’utérus pour la formation de la coquille, est déjà recouvert de
deux membranes coquillières formées de fibres protéiques déposées dans l’isthme. La
première action que subit l’œuf avant formation de la coquille dans l’utérus est une
hydratation du blanc (action qui va doubler le poids de l’œuf). Ce transfert d’eau est
accompagné de sécrétion de sodium, potassium et bicarbonate. Une expulsion de l’œuf à ce
stade donnera des œufs « mous », caractéristique que l’on rencontre essentiellement en début

31
de ponte. La composition de la coquille (qui pèse environ 6g) est constituée majoritairement
de cristaux de carbonate de calcium (CaCo3) (Nys et al., 2011). Dix heures après ovulation, la
croissance et le dépôt des cristaux de carbonate de calcium durent 12h. La vitesse de dépôt de
ces cristaux est de 0,30 à 0,35 g/heure. La sécrétion de calcium dans le liquide utérin est
continue et dépend d’un phénomène de stockage relargage. Le calcium pour la formation de la
coquille provient en fait uniquement des apports du sang. Dans le sang, le calcium est présent
2+
sous deux formes : une forme diffusable (Ca lié à des petites molécules de citrate) et une
forme non diffusable (calcium lié à des protéines) servant au transport entre le foie et les
ovaires par exemple. Cependant, si le calcium servant à la fabrication de la coquille provient
du sang à court terme, celui-ci est disponible pour l’animal grâce à l’alimentation. Cette phase
de calcification de la coquille est la phase la plus longue mais est également la plus
importante dans la mesure où c’est la calcification qui va donner la solidité de l’œuf, facteur
essentiel pour pouvoir vendre les œufs. En 2008, les pertes mondiales dues à des coquilles
fragiles représentaient 500 millions de dollars US (Etches, 2008).

III-2 Description de la courbe de ponte


Entre 1 et 20 semaines, la poule est immature et n’a pas encore débuté son cycle de
ponte. Elle dispose cependant déjà de l’ensemble de ses ovules qui serviront ensuite à la
production des jaunes d’œufs. A partir de la semaine 19, les poules entrent en ponte
progressivement en raison de leur stade de maturité sexuelle différent, ce qui explique
l’ascendance de la première partie de la courbe. Cependant, cette période d’entrée en ponte est
caractérisée par l’observation de quelques malformations des œufs : ponte de deux œufs par
jour dont un est mal formé, production d’œufs sans coquille dits « mous », plusieurs jours
sans ponte ou même des œufs avec deux jaunes. Ces anomalies disparaissent rapidement
après quelques jours de ponte. Le pic de ponte ou « cloche » est obtenu 4 à 8 semaines après
l’entrée en ponte. Il sera obtenu d’autant plus rapidement que le troupeau est homogène. Sa
valeur caractérise la productivité de l’élevage et sa conduite. Après le pic, l’intensité de ponte
décroît linéairement (1%/semaine) en fonction du temps (avec l’âge). Pour des considérations
commerciales, la production n’est plus souvent rentable lorsqu’elle devient inférieure à 60 ou
65 % soit à l’âge de 70 à 72 semaines (Sauveur, 1988)

32
100 70
90
Taux de ponte (%)

Poids des œufs (g)


65
80
70
60
60
50 55
40
50
30
taux ponte
20
45
10
0 40 Poids des
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 œufs

Age en semaines

Figure 3 Evolution du taux de ponte et du poids des œufs en fonction de l'âge (Courbe
de ponte théorique ISA BROWN, 2013)

III -3 Facteurs de variations de la courbe de ponte


Le taux de ponte moyen dans un élevage est un des critères essentiels et un indicateur
fort de production que l’éleveur contrôle au quotidien. Chez des poules pondeuses qui
produisent des œufs destinés à la consommation et qui sont placées dans un environnement
constant, la baisse du taux de ponte qui peut être observée est généralement due à une baisse
de la ponte de tous les animaux plutôt qu’à un arrêt total de ponte de certains d’entre eux
(Sauveur, 1988). Le premier facteur qui induit une diminution de la productivité des animaux
est l’âge de ceux-ci. En effet, la baisse de ponte s’explique par un ralentissement de l’activité
folliculaire puisque la phase d’accroissement du jaune de l’œuf dure plus longtemps au fur et
à mesure que la poule avance dans l’âge. De plus, la souche à laquelle appartient la poule
affecte également le taux de ponte de celle-ci. Un autre facteur commun à l’ensemble des
poules, âge et souche confondu, est la longueur de la photopériode. Lorsque la poule est
soumise à une photopériode quotidienne constante, la longueur de cette photopériode agit peu
sur l’intensité de ponte a partir d’un seuil de 9h de lumière par jour. A l’opposé, des variations
de photopériode affectent fortement la ponte (Sauveur, 1988). Toute décroissance de la durée
33
G¶pFODLUHPHQWHQWUDLQHXQHGLPLQXWLRQGHSRQWH /¶XWLOLVDWLRQGHF\FOHVXSpULHXUjKSHXW
QpDQPRLQVrWUHEpQpILTXHVXUODUpGXFWLRQGHMRXUVRODSRXOHQHSRQGSDV K HWSHUPHW
O¶DXJPHQWDWLRQGXSRLGVGHO¶°XI DXJPHQWDWLRQGXSRLGVGXMDXQHHWGXEODQF &HWWHILJXUH
)LJ LOOXVWUHELHQO¶HIIHWGXF\FOHOXPLQHX[VXUOHWDX[GHSRQWHHWVXUOHSRLGVGHV°XIV
 Pourcentage de la valeur obtenue en nycthémère de 24h














)LJXUH(IIHWVRSSRVpVGHODORQJXHXUGHVF\FOHVOXPLQHX[VXUOHWDX[GHSRQWHHWOHSRLGV

G¶°XI G¶DSUqV6KDQDZDQ\ 


'HV GpILFLWV DOLPHQWDLUHV SHXYHQW HQVXLWH DIIHFWHU OH WDX[ GH SRQWH $LQVL .HVKDYDU]
 LQGLTXHTXHODSURGXFWLRQQHSHXWrWUHPDLQWHQXHjXQQLYHDXFRQVWDQWVLRQQ¶DSSRUWH
SDVXQQLYHDXDGpTXDWGHSURWpLQHVDXFRXUVGHODMRXUQpH HQYLURQ (QILQOHVFRQGLWLRQV
HQYLURQQHPHQWDOHV MRXHQW pJDOHPHQW XQ U{OH LPSRUWDQW VXU OD FRXUEH GH SRQWH 8QH
WHPSpUDWXUH DPELDQWH pOHYpH LQGXLW XQH GLPLQXWLRQ GH OD SURGXFWLRQ G
°XIV  $ULPD HW DO
 'HVPDODGLHVRXXQHFRQFHQWUDWLRQGHSDUDVLWHVSHXWDYRLUpJDOHPHQWXQHIIHWQpIDVWH
VXU OD SURGXFWLRQ G¶°XIV $LQVL XQH PXOWLSOLFDWLRQ GH SRX[ URXJHV LQGXLW XQH SUHVVLRQ
VDQLWDLUH LPSRUWDQWH TXL YD UpGXLWH OD TXDOLWp HW OD SURGXFWLRQ G¶°XIV FKH] OHV SRXOHV
SRQGHXVHV &KDXYHHWDO/HVQDHWDO 



,9,QJHVWLRQFKH]OHVYRODLOOHV

,9&\FOHG¶LQJHVWLRQHWpYROXWLRQGHODFLQpWLTXHDXFRXUVGHODMRXUQpH
 /¶LQJHVWLRQ GHV SRXOHV SRQGHXVHV HVW XQ pOpPHQW HVVHQWLHO SRXU OD FRQGXLWH RSWLPDOH
G¶XQpOHYDJHGHSRXOHHWODPDvWULVHGHVIDFWHXUVGHYDULDWLRQVHQHVWODSULQFLSDOHFOp8QGHV
SUHPLHUVpOpPHQWVjSUHQGUHHQFRPSWHHVWOHFDUDFWqUHF\FOLTXHGHO¶LQJHVWLRQ(QHIIHWOHV
SRXOHVRQWXQU\WKPHG¶LQJHVWLRQTXLYDULHVHORQOHVGLIIpUHQWVPRPHQWVGHODMRXUQpH&KDKHW
0RUDQ  RQWPLVHQpYLGHQFHTXHODFRQVRPPDWLRQGHVSRXOHVpWDLWU\WKPpHVHORQGHX[
SLFV G¶LQJHVWLRQ GDQV OD MRXUQpH /HV SRXOHV SODFpHV HQ DOLPHQWDWLRQ OLEUH HW VpSDUpH
VpSDUDWLRQGHVIUDFWLRQVDSSRUWpHVHQPrPHWHPSVDYHFXQHIUDFWLRQULFKHHQpQHUJLHXQH
IUDFWLRQ ULFKH HQ SURWpLQHV HW XQH IUDFWLRQ FDOFLTXH  RQ PRQWUp XQ SLF G¶LQJHVWLRQ TXDWUH
KHXUHV DSUqV O¶DOOXPDJH SXLV XQ VHFRQG SLF GHX[ KHXUHV DYDQW O¶H[WLQFWLRQ GH OD OXPLqUH
)LJXUH 










)LJXUH(YROXWLRQGHO
LQJHVWLRQDXFRXUVG
XQHMRXUQpHGHGLVWULEXWLRQ QXLW

GHKjK  G
DSUqV&KDKHW0RUDQ  

/HSUHPLHUSLFG¶LQJHVWLRQFRUUHVSRQGjXQHLQJHVWLRQLQGXLWHSDUODSpULRGHGHMH€QH
IRUFpSHQGDQWODSpULRGHGHQXLW SDVGHFRQVRPPDWLRQG¶DOLPHQWHQSpULRGHGHQXLW5LFKDUG
HW DO   /H VHFRQG SLF G¶LQJHVWLRQ TXHOTXHV KHXUHV DYDQW O¶H[WLQFWLRQ GH OD OXPLqUH
FRUUHVSRQGUDLWjXQHFRQVRPPDWLRQSUpYHQWLYHGHVSRXOHVDILQG¶DYRLUOHVUpVHUYHVQpFHVVDLUHV
SRXU SDVVHU OD QXLW 6DYRU\   3OXV OD SpULRGH G¶pFODLUHPHQW HVW ORQJXH SOXV OHV SRXOHV
YRQWFRQVRPPHUG¶DOLPHQWTXHFHVRLWSRXUGHVSRXOHVGHjVHPDLQHVG¶kJHVRXSRXUGHV
SRXOHV GH  j  VHPDLQHV G¶kJH 0RUULV HW DO /HZLV HW DO   (QILQ O¶DVSHFW
VRFLDO YD LQIOXHQFHU OH F\FOH G¶LQJHVWLRQ GHV SRXOHV (Q HIIHW GHV SRXOHV SODFpHV HQ JURXSH



vont interagir régulièrement avec leurs congénères et donc modifier leur comportement en
fonction du groupe. Collins et Stumper (2007) ont montré que la facilitation sociale conduisait
à un effet de masse très important initié par certains individus qui entrainent par la suite la
même action de la part de leurs congénères. Ils ont montré en effet que plus le nombre de
poulets en train de manger était important, plus la probabilité de voir un autre poulet arriver à
la mangeoire pour consommer était importante. Cette notion de patterns d’excitation
synchronisés et d’ingestion de groupe est un élément important à prendre en compte car
aujourd’hui les poules pondeuses en cages sont placées obligatoirement dans des cages
collectives (de 5 à 60 poules par cage). Enfin, plusieurs autres facteurs environnementaux
interviennent dans la régulation de l’ingestion. En effet, de fortes températures dans le
bâtiment vont induire une diminution de la consommation (Syafwan et al., 2012) due au
maintien de l’homéothermie et afin d’éviter une augmentation de la température de l’animal
induite par la production de chaleur lors du métabolisme (Zhou and Yamamoto., 1997). La
thermolyse, qui est le besoin de dissiper la chaleur produite par la prise alimentaire et le
métabolisme est un régulateur majeur : selon le « bilan thermique » de l’animal, celui-ci va
augmenter ou réduire son ingestion.

IV-2 Effet des modalités d’apport et de la forme de l’aliment sur l’ingestion :


Les modalités d’apports et la forme des aliments sont des éléments importants qui vont
influencer considérablement l’ingestion des poules. L’aliment distribué sous forme de granulé
est systématiquement plus consommé qu’un aliment distribué sous forme de farine (Yo et al.,
1997). Ces auteurs on également monté que si le maïs était apporté sous forme de graines
entières, celui-ci était plus consommé que lorsqu’il était présenté sous forme concassé et
encore moins consommé lorsqu’il était présenté sous forme de farine fine. Ainsi, lorsque
l’aliment est présenté sous forme de farine avec des particules de différentes tailles, les
volailles choisissent toujours de consommer en premier les particules de plus grosse taille
(adaptées à la taille de leur bec) quelle que soit la composition en matières premières
(Wauters et al., 1997, Safaa et al., 2009). Chez les poules pondeuses, l’ingestion de particules
grossières permet le développement du gésier, ce qui engendre une amélioration significative
du broyage mécanique des aliments (rendant plus disponible les nutriments pour la digestion
et donc l’assimilation). Cependant, ce tri particulaire peut avoir des conséquences
nutritionnelles importantes en fonction des régimes alimentaires. En effet, Wauters et al.,
(1997) ont mis en évidence que selon la composition du régime (Maïs/Soja, Blé/Soja ou

36
Maïs/Tournesol) les écarts nutritionnels dus au tri particulaire peuvent être de l’ordre de -2.3 à
+0.2% de protéines brutes ingérées. Picard et al., (2000) ont également montré que la vitesse
d’ingestion, c'est-à-dire le temps mis par le poulet pour ingérer sa ration alimentaire, était
également influencé par la taille des particules et par la dureté de l’aliment. En effet, un
aliment avec une dureté importante et des grosses particules est ingéré beaucoup plus
rapidement qu’un aliment avec beaucoup de fines et tendre. Safaa et al., (2009) ont montré
que les farines de tailles grossières chez les poules pondeuses amélioraient significativement
l’ingestion mais également induisaient un développement plus important du gésier, permettant
ainsi un broyage mécanique des aliments plus efficient (rendant plus disponible les nutriments
pour la digestion et donc l’assimilation). Chez la poule il n’y a pas d’ingestion minimum
(l’ingestion peut être nulle sur la journée si l’aliment est refusé) mais les poulets de chairs
consomment l’aliment jusqu’à leur maximum physique d’ingestion (Bokkers et Koen, 2003).
Les tailles de particules ingérées vont donc induire également des phénomènes de régulation
de l’ingestion au travers d’une satiété physique déclenchée par le remplissage du jabot. La
dilatation de cet organe en raison de l’ingestion d’aliment induit très rapidement une
diminution de la consommation de celui-ci (Richardson, 1970). Il parait donc raisonnable de
penser que l’ingestion rapide de grosses particules ou de graines entières, provoquera
rapidement une forte distension du jabot et donc un arrêt de la consommation.
L’encombrement physique des organes supérieurs du tractus digestif est donc un des éléments
déterminant de la régulation de l’ingestion et dépend fortement de la qualité des aliments
proposés. Certains auteurs ont mis en évidence la possibilité d’influencer le temps de transit
dans la partie supérieure du tube digestif, non plus grâce à la forme de l’aliment, mais
également pas la composition de celui-ci. Ainsi, Shelton et al., (2005) ont montré qu’une
faible densité volumique de l’aliment induisait une plus faible consommation. Van Krimpen
et al., (2011) ont également mis en évidence que l’incorporation de fibres insolubles dans
l’aliment augmentait le temps de rétention de l’aliment dans le jabot, le gésier et l’intestin,
ralentissant ainsi le rythme d’ingestion et la quantité ingérée.

L’ingestion n’est pas constante et uniforme au cours de la journée mais est dépendante
d’un rythme propre à l’animal. En effet, il existe chez les poules un rythme d’ingestion propre
à chaque individu. Duncan et al., (1970) ont montré que pour les poules pondeuses l’ingestion
est ponctuée par des phases de repas qui varient pour chaque individu en durée et en quantité
d’aliment consommé. C’est pourquoi Clifton (1979) affirme qu’il est impossible de proposer

37
un modèle unique décrivant les rythmes de consommation des poules puisque la rythmicité de
la phase alimentaire varie en fonction des conditions environnementales, de la forme des
aliments et est propre à chaque individu. L’existence naturelle de ces repas permet de moduler
les apports des aliments et donc d’optimiser et de réguler l’ingestion chez les volailles et
notamment chez les poules pondeuses. Dans une situation d’alimentation en libre choix, les
poules ont été nourries avec trois aliments différents offerts en même temps dans trois
contenants séparés. Les poules recevaient un aliment concentré en protéines sous forme de
granulés, du maïs entier et du carbonate de calcium. Au cours de la période de ponte, les
poules en libre choix ont pondu des œufs plus lourds et eu de meilleurs indices de
consommation que celles nourries avec un aliment continu (Olver et Malan, 2000). Aucune
différence n’a été observée pour le taux de ponte et la quantité d’aliment ingéré. Cependant si
l’aliment est offert ad libitum, certains auteurs ont observé des dérives dans la consommation
d’une des fractions qui semble la plus appétante pour des poules placées en situation de libre
choix (Picard et al., 1993). Ces travaux montrent bien la difficulté à piloter l’ingestion des
poules poue optimiser la production. L’alimentation séquentielle permet néanmoins de mieux
contrôler la distribution et l’ingestion d’aliment et permet de diminuer l’ingestion tout en
maintenant un niveau de production comparable à l’alimentation continue (Umar Faruk et al.,
2010). Cette modalité d’apport en séquentielle sera plus détaillée dans le chapitre suivant.

IV-3 Impact de la composition des régimes sur l’ingestion


La composition des aliments est également un des éléments majeurs de la régulation
de l’ingestion. Chez les poules pondeuses, la composition du régime en protéines, en énergie
et en calcium est un régulateur fort de l’ingestion. Chah et Moran (1985) ont mis en évidence
que des poules placées en situation de choix alimentaire avec un aliment riche en énergie, un
aliment riche en protéines et du carbonate de calcium avaient des rythmes d’ingestion
différents en fonction des aliments. En effet, l’aliment riche en énergie était consommé
principalement 2 heures après l’allumage puis pendant deux heures avant l’extinction de la
lumière. Un rythme quasiment identique est observé par ces mêmes auteurs pour l’aliment
riche en protéines. Bouvarel et al., (2010) ont mis en évidence au travers d’un travail de méta-
analyse, que l’ingestion d’aliments était très fortement corrélée au niveau d’énergie
métabolisable dans le régime. En effet, plus le régime est concentré en énergie plus le niveau
d’ingestion diminue (Figure 6).

38
Figure 6 Adaptation de la consommation d'aliment en fonction de la concentration en énergie
du régime (Bouvarel et al., 2010)

Les auteurs ont toutefois mis en évidence que lors d’une forte teneur en énergie dans le
régime ou d’une très forte dilution, la régulation de la poule régulait moins bien son ingestion.
D’autres auteurs confirment bien cette régulation sur le niveau d’énergie en ajoutant qu’il n’y
a pas d’effet sur le poids des œufs (Summers et Leeson 1993, Van Krimpen et al., 2007). La
régulation de l’ingestion en fonction de l’énergie ingérée est induite par des régulations
métaboliques intervenant à moyen terme. Le contrôle de l’homéostasie énergétique est
considéré par les nutritionnistes comme le principal facteur de la régulation du comportement
alimentaire. La régulation coordonnée de l’ingestion et des dépenses énergétiques répond à
des signaux environnementaux externes (composition de l’aliment, disponibilité de l’aliment,
photopériode, température,…) et à des signaux physiologiques internes (stockage d’énergie,
hormones et niveaux en nutriments et métabolites). La région hypothalamique représente le
point de convergence de nombreuses informations périphériques et participe au contrôle de
l’équilibre énergétique corporel de la prise alimentaire. Elle contient de multiples voies
nerveuses peptidergiques divisées en deux catégories élémentaires : les voies anaboliques et
les voies cataboliques, impliquées dans la régulation de l’ingestion et l’homéostasie
énergétique (Woods et al., 1998). La voie anabolique entraîne une augmentation nette de

39
l’énergie ingérée et du stockage, tandis que la voie catabolique résulte en une baisse nette de
l’énergie ingérée et du stockage.
Le niveau de protéines dans l’aliment n’est pas un élément majeur permettant aux
poules de réguler leur niveau d’ingestion. En effet, Summers et Leeson (1993) n’on pas mis
en évidence de régulation de l’ingestion sur le niveau de protéines (pas de différences
d’ingestion entre un régime à 17% de PB et un régime à 13% de PB). Ces auteurs ont
cependant mis en évidence que les poules qui consommaient le régime plus riche en protéines
avaient un poids d’œufs exporté plus important. Néanmoins plus récemment, certains auteurs
confirment bien l’absence de régulation de l’ingestion sur le niveau de PB de l’aliment
(14.5% PB contre 17.4% PB) et sur le niveau d’acides aminés (Lys 6,5 g/Kg Met 3,2 g/Kg
contre Lys 7,8 g/Kg, Met 3,9 g/Kg) (Mohiti-Asli et al., 2012). Ces auteurs ont mis en
évidence que le taux de protéines et d’acides aminés ne semblant pas avoir d’influence sur le
poids des poules et le poids des œufs. Mohiti-Asli et al., (2012) ont toutefois mis en évidence
qu’une alimentation plus pauvre en protéines favorisait le dépôt de gras abdominal (sans
pourtant provoquer de différences de poids). Concernant les acides aminés, à 8 jours d’âge,
des poulets recevant un aliment déficient en lysine, méthionine et tryptophane (- 45%)
réduisent leur ingestion de 60%, 6 h après le changement d’aliment (Picard et al., 1993). Lors
d’alternance d’aliments à teneur réduite et normale en lysine, dans la journée, les poulets de
chair apprennent à reconnaître les aliments et adaptent ainsi leur ingestion (Bizeray et al.,
2002). Il existe un niveau optimal à atteindre afin de répondre correctement aux besoins des
animaux et atteindre les attentes sur le poids des œufs (Figure 7). On voit bien sur la figure
une relation curvilinéaire entre le poids des œufs et la teneur en méthionine dans l’aliment
(Bouvarel et al., 2010), avec une asymptote à 0.36-0.38% de Met dans l’aliment,

acide aminé n’as pas valorisé et donc perdu.

Figure 7 Poids moyen de l'œuf en fonction de la teneur en méthionine totale de l'aliment 40


(Bouvarel et al., 2010)
La régulation de l’ingestion sur le niveau de calcium dans l’aliment est bien
documentée. En effet, la littérature met bien en évidence un besoin calcique spécifique en fin
de journée. Le calcium dans une situation de libre choix est uniquement consommé l’après
midi avec une consommation quasiment nulle durant les 10 premières heures après l’allumage
(Chah et Moran, 1985). Cette consommation de calcium spécifique dans l’après midi
correspond en fait au moment de la formation de la coquille. La forme du calcium ne semble
cependant pas impacter la consommation à partir du moment où les particules de carbonate de
calcium sont suffisamment grosse pour pouvoir être spécifiquement choisie par la poule
(Anchieta de Araujo et al., 2011) La taille grossière des particules permettait à la poule de
choisir spécifiquement le moment de la journée où consommer le calcium et donc éviter toute
consommation d’aliment induite par un calcium qui serait broyé trop finement ou incorporé
dans un granulé. Anchieta de Araujo et al., (2011) ont bien mis en évidence que l’ingestion
totale d’aliments était la même mais avec des rythmes différents au cours de la journée

V Alimentation séquentielle, système de distribution innovant


Plusieurs systèmes d’alimentation alternatifs à l’alimentation classique continue ont
été étudiés afin d’essayer d’optimiser la distribution d’aliment tout en conservant les
performances de ponte. L’alimentation mélangée ou l’alimentation en libre choix sont des
systèmes qui ont chacun des avantages et des inconvénients. L’objet de ce chapitre porte sur
l’alimentation séquentielle, système de distribution choisi pour ces travaux de thèse car ne
présente pas de contraintes de distribution comme l’alimentation séparée (plusieurs
mangeoires en même temps) et permet d’éviter aux poules de trier et de consommer l’aliment
qu’elles préfèrent dans le cadre d’une alimentation mélangée.
La majorité des travaux sur l’alimentation séquentielle ont été réalisés sur des poulets
de chairs et non sur des poules pondeuses. Le format et l’âge des animaux n’est absolument
pas le même, ces deux animaux sont néanmoins de la même espèce (Gallus gallus).Les
travaux en AS sur les poulets de chairs peuvent donc apporter des informations qui pourraient
être utiles aux les réflexions sur les travaux chez les poules pondeuses. Le concept
d’alimentation séquentielle est apparu pour la première fois en 1975 avec les travaux de Gous
et Du Preez qui alternaient dans la journée deux aliments avec des teneurs différentes en
acides aminés. Chez le poulet de chair, ce système d’alimentation séquentielle a été largement
étudié et appliqué avec succès avec par exemple du blé entier qui était donné en alternance
avec un régime riche en protéines (Noirot et al., 1998). Pour des cycles de 12h comprenant

41
une alternance entre un régime riche et un régime pauvre en acides aminés, les poulets ont été
capable de réguler leur ingestion afin d’avoir une croissance identique aux poulets placés en
alimentation classique continue (Gous et Du Preez, 1975). En revanche, lorsque ces auteurs
proposaient une alimentation séquentielle avec des cycles de 24h, les poulets placés dans ce
système de distribution ont eu une croissance inférieure (bien que le niveau d’ingestion soit
resté le même que celui des poulets en alimentation continue). Lorsque l’alimentation
séquentielle est proposée avec des cycles de 48h, avec deux aliments à teneur en protéines
différente, des baisses de croissance ont été observées (Rosebrough et al., 1989). Rose et al.
(1995) affirment que des séquences supérieures à 12h en alimentation séquentielle chez le
poulet peuvent placer celui-ci dans des situations de carence nutritionnelle. Plus récemment,
Bouvarel et al. (2004) ont distribué deux aliments variant par leur teneur en énergie et en
protéines (l’un riche en énergie et pauvre en protéines et l’autre pauvre en protéines et riche
en énergie). Lorsque les deux aliments étaient offerts par séquence de 12h, l’ingestion était
fortement diminuée par rapport à un aliment offert de façon continue, ce qui réduisait la
croissance des poulets. Les séquences étant trop rapprochées, l’hypothèse était que les poulets
attendaient l’aliment suivant riche en protéines qui devait probablement leur paraitre plus
appétant. Cependant, des cycles de 48h induisaient une plus faible diminution de
consommation que pour les poulets placés en alimentation continue avec des performances de
croissance équivalentes. Or, dans le contexte de l’élevage de poules pondeuses, des cycles de
12, 24 ou 48h ne sont pas envisageables car la poule à des besoins spécifiques en énergie,
protéine et calcium pour la formation de l’œuf sur une journée.
Plusieurs études ont été réalisées sur la distribution séquentielle de l’aliment chez les
poules pondeuses, cependant toutes ne s’accordent pas forcément sur des résultats
systématiquement positifs en alimentation séquentielle. Blair et al. (1973) ont observé une
augmentation de l’ingestion quand les poules pondeuses étaient nourries avec un mélange de
blé entier, orge et maïs concassé dans la matinée suivi d'un concentré de protéines dans
l'après-midi. Cependant, la production d'œufs et le poids des œufs étaient similaires par
rapport à l’alimentation complète unique. Robinson (1985) qui offrait le matin un aliment
concentré en protéines et l’après-midi de l’avoine avec du carbonate de calcium, a observé
une ingestion réduite, conduisant à une réduction de la production d'œufs et du poids d’œuf.
La même tendance a été rapportée par Reichmann et Connor (1979) qui avaient offert le matin
un aliment riche en énergie et l’après midi un aliment riche en protéines et calcium. Plus
récemment, Umar Faruk et al. (2010) offraient le matin du blé entier et l’après midi un

42
aliment complémentaire riche en énergie et calcium. Ces auteurs on montré que les poules
alimentées de façon séquentielle, avaient une consommation totale inférieure mais
conservaient un niveau de production équivalent à celui des poules nourries avec un aliment
classique continu ce qui à permis d’améliorer significativement l’indice de consommation.
Umar Faruk et al. (2011) qui apportaient également des céréales le matin (donc un aliment
riche en énergie et pauvre en protéines le matin) et un complémentaire l’après midi (riche en
protéines, calcium et pauvre en énergie) on également mis en évidence une amélioration
significative de l’IC. L’apport de blé entier dans la fraction du matin donnait les meilleurs
résultats (IC) par rapport à un apport le matin avec du blé broyé même grossièrement.
L’alimentation séquentielle entrainant néanmoins un gain de poids inférieur par rapport à une
alimentation continue (Umar Faruk et al., 2010, 2011 ;), les poules pourraient être toutefois
moins robustes vis-à-vis des contraintes environnementales négatives (chaleur, infection de
poux rouge, ...).

43
44
Conclusion
L’organisation de la filière pondeuse dans le monde et en France dépend de nombreux
facteurs, répondant tous au besoin de produire de façon optimale et raisonnable. Les
contraintes pesant sur les filières (importance du coût alimentaire, volatilité du prix des
matières premières, concurrence entre les pays, …), la « cyclicité » des phénomènes liés à la
ponte avec les besoins qui y sont associés et les connaissances sur les régulations d’ingestion
ont permis d’évaluer la place importante de l’aliment dans la filière pondeuses. Cependant il
reste encore à quantifier les réponses des poules pondeuses en fonction des aliments apportés
En effet la littérature est unanime quant à la capacité de modulation de l’ingestion des poules
sur le niveau énergétique de l’aliment. Cependant la littérature ne donne que peu
d’informations sur la variation d’ingestion d’aliment en fonction du niveau de protéines ou de
certains acides aminés essentiels de celui –ci. L’ingestion de protéines et de certains acides
aminés ayant pourtant un fort impact sur la production d’œufs chez la poule pondeuse, la
littérature manque également d’informations concernant l’impact de la nature des ingérés sur
les performances de ponte. Afin de pouvoir évaluer les besoins de la poule et apporter les
nutriments nécessaires au meilleur moment de la journée dans le cadre d’une distribution
séquentielle, il est impératif de pouvoir définir clairement les différents facteurs susceptibles
de moduler l’ingestion mais également l’impact de l’ingestion des nutriments sur la ponte. Un
travail de méta-analyse permettra d’apporter ces connaissances manquantes et ainsi apporter
des pistes quant à une utilisation optimale de l’alimentation séquentielle.

45
46
Chapitre 2 :
Relations entre composition des aliments,
niveaux d’ingestion et performances des
poules pondeuses : approche par méta-analyse

47
48
Résumé du travail de méta-analyse
Les recherches en sciences animales et plus particulièrement en nutrition s’appuient
sur des traitements de bases de données de plus en plus importantes. En effet, pour un sujet,
les nombres de publications et de résultats par publication se sont largement accrus.
L’utilisation de nouveaux moyens de traitement de l’information par méta-analyse permet
d’apporter de nouvelles informations et de nouveaux résultats. Il est donc nécessaire de
pouvoir extraire des relations quantitatives des réponses des animaux à partir des publications.
En conséquence les analyses par méta-analyses des bases de données construites à partir de
données provenant d’expérimentations diversifiées deviennent essentielles. Les méta-analyses
suivent une démarche en plusieurs phases. La première consiste à définir les objectifs
permettant de sélectionner les publications à inclure dans la base de méta analyse. Dans notre
étude, les deux objectifs concernaient tout d’abord l’influence de la nature des régimes sur
l’ingestion et ensuite la quantité et qualité des ingérés influençant les performances des
animaux. Les étapes suivantes conduisent à la construction de la base et sa description,
permettant d’écarter certaines données qui sont trop extrêmes par rapport aux autres données
et de comprendre la structure du méta dispositif expérimental testé. La base comprenanit 40
publications publiées entre 1998 et 2013 Enfin la dernière étape consiste en une analyse par
Modèle Linéaire General., proposant des modèles statistiques simples, lois de réponse des
poules pondeuses à leur aliment.
Concernant les principaux résultats, l’ingestion est fortement influencée par la souche
des poules et l’âge de celles-ci (plus importante pour des poules brunes que pour des poules
blanches et ingestion qui augmente avec l’âge). Concernant les variations de l’ingestion, la
concentration en énergie dans le régime était l’élément le plus important. Le niveau de
protéines influence peu l’ingestion alors que les teneurs en Lys et en Met n’ont pas d’effet
significatif. Pour les performances de ponte, le taux de ponte est amélioré avec l’ingestion de
protéines, sans effet de la teneur en énergie ou de la quantité d’énergie ingérée. L’élévation du
poids des œufs est liée à l’augmentation de l’ingestion d’énergie et de protéines. Les teneurs
en Lys et Met dans le régime et des quantités ingérées améliorent systématiquement le taux de
ponte, le poids des œufs et la masse d’œuf. Le poids des poules est quant à lui influencé par la
souche (poules brunes plus lourdes) et par l’âge mais pas par la nature des ingérés.
Ces travaux de méta analyse permettent donc de répondre à des interrogations portant
sur l’ajustement de la consommation des poules et le lien entre quantité et qualité des ingérés
sur les performances. Ces travaux mériteraient néanmoins d’être validés par une
49
expérimentation et complétés surtout sur la partie des acides aminés (la base n’étant pas crée
pour répondre assez précisément sur l’effet de ces acides aminés sur l’ingestion et les
performances des poules).

50
I Objectifs et questions de recherche

L’ingestion chez les poules pondeuses est principalement déterminée par la concentration en
énergie et dans une moindre mesure par l’apport en acides aminés essentiels contenus dans
l’aliment. En effet lorsque l’aliment est dilué ou enrichi en énergie la variation de l’ingéré est
une fonction quasi-linéaire de la concentration énergétique de l’aliment (Bouvarel et al,
2010). Cependant si cette régulation sur l’énergie semble claire, la poule est capable d’autres
types de régulations impactant sa consommation d’aliment. La première question de la méta-
analyse est de voir comment la teneur de l’aliment en différents éléments (énergie, protéines,
acides aminés) influence l’ingestion de la poule pondeuse en fonction du type de poules
(blanche ou brune) et de leur classe d’âge (début, milieu ou fin de ponte). La réponse au
calcium n’est pas étudiée, son influence à court terme étant bien documentée en lien avec la
formation quotidienne de la coquille. Par ailleurs, les facteurs environnementaux
(température, pression sanitaire, etc…) n’ont pas été testés statistiquement, les données
n’étant pas disponibles ou suffisamment fiables dans la base collectée. Leurinfluence est
cependant importante. La seconde question consiste à quantifier l’influence de la valeur
nutritionnelle (énergie, protéines et acides aminés) des aliments et de la quantité des
nutriments ingérés sur les performances (taux de ponte, poids des œufs, masse d’œuf et indice
de consommation, croissance) en fonction de leur type génétique et de leur âge au cours du
cycle de production.

II Construction de la base de données


La base de données répertorie des publications en revue à comité de lecture sur
l’alimentation des poules pondeuses, publiées entre 1998 et 2013. Cette plage de temps a été
choisie au regard des progrès génétiques et des modes d’élevages, susceptibles d’influencer
les paramètres des modèles obtenus. Seules les publications qui donnaient au minimum, la
composition nutritionnelle du régime (Teneur en énergie métabolisable à bilan azoté nul
(EMAn), Protéines Brutes (PB), acides aminés, ...) et les performances des poules (taux de
ponte, poids d’œuf, masse d’œuf et ingestion) ont été conservées. La base inclue plusieurs
types d’informations reprises dans le tableau 1: identification unique de chaque observation,
informations générales sur la publication (nom du premier auteur, année de publication), des
informations qualitatives ou semi-quantitatives (type des poules, âge, température, durée
d’éclairage, …), puis les compositions des régimes (proportions des différentes matières

51
premières, EMAn, PB, acides aminés, P et Ca). Enfin les données relatives aux performances
des poules (poids des poules, ingestion, taux de ponte, poids des œufs, masse d’œuf,
composition des œufs). Chaque traitement de chaque expérimentation s’est vu attribuer un
code unique permettant d’identifier facilement chaque ligne dans la base de données et ainsi
de porter un regard critique sur les graphiques et par conséquent sur les équations obtenues.
En effet, pour chaque traitement, la nature de l’effet testé était notée et a permis de relativiser
ou de renforcer les résultats observés. Ainsi, il a été fait le choix d’écarter quelques
publications de la base de données car les poules présentaient des performances
anormalement élevées ou faibles par rapport aux autres poules en lien avec les facteurs testés
dans un nombre limité d’expérimentations (lignées expérimentales, températures élevées).
Pour comparer les différentes expérimentations entre elles, certaines variables ont été
recalculées à partir des autres informations disponibles afin d’avoir une approche homogène.
Par exemple, l’ingestion d’énergie a été recalculée grâce au niveau énergétique des régimes
multiplié par l’ingestion d’aliment. Les indices de consommation ont tous été recalculés grâce
à la masse d’œuf et l’ingestion totale d’aliment. La base finale comprend 40 publications
représentant 76 essais avec un total de 332 traitements.

III Analyse des différentes bases de données


L’analyse de la base de données s’est déroulée en plusieurs étapes.
La première étape définit la répartition des données les unes par rapport aux autres. Des
histogrammes visualisent l’ensemble des essais pour une variable donnée. Cela permet
d’isoler l d’eventuelles données aberrantes. Les graphiques bivariées (Fig 13) montrent des
données pour une variable définie en fonction d’un facteur qualitatif, ici le type de poule et
leur classe d’âge.
La seconde étape représente et compare les liens entre les différentes variables explicatives.
Une représentation sous forme d’un graphique bivariée avec en abscisse une variable
explicative et en ordonnée une autre variable explicative, relie les observations d’un même
essai et visualise le lien global entre les deux variables concernées. Ces représentations
mettnte en évidence les relations entre les variables explicatives et visualisent le plan
d’expérience méta-analytique.
La troisième étape met en relation graphique plusieurs variables en plaçant en abscisse
un variable explicative et en ordonnée une variable expliquée. Ces représentations graphiques

52
permettent de quantifier le lien global entre les deux variables et de commencer à représenter
une possible relation statistique entre les variables.
Enfin pour finir l’analyse de la base de données, une analyse par Modèle Linéaire General a
été effectuée:

Variable expliquée i jk = Ordonnée à l’origine + Effet Lignée i + Effet âge j + α * VE1 ijk + β* VE2 ijk + Ɛijk

VE: Variable Explicative

Ces analyses par GLM permettent donc d’obtenir des propositions de modèles statistiques à
interpréter en fonction des paramètres obtenus.

III-1 Analyse de la base initiale


L’analyse de cette base commence par sa description. Les intitulés de chaque colonne
relevés dans les différentes publications scientifiques sont décrits dans le tableau 1. Chaque
variable est décrite graphiquement par des histogrammes (Fig 8, 9, 10).et numériquement en
donnant le nombre de variables, le minimum, le maximum, la médiane et l’écart -type (Tab
2).

53
Tableau 1 Informations incluses dans la base ingestion des poules pondeuses et performances

ID Numéro spécifique à chaque ligne TSAA Méthionine + Cystéine dans les régimes (%)
Nom du premier auteur et année de
Publication Thr Thréonine dans les régimes (%)
publication de l'article
Numéro de l'essai (une publication peut
Essai Ca Calcium dans les régimes (%)
comprendre plusieurs essais)
Nom Nom commercial de la lignée P Phosphore total (%)
Taux de
Lignée Catégorie de lignées (brunes ou blanches) Production d’œuf (%)
ponte
température ou variation de température au
Température Poids d’œuf Poids des œufs (g)
cours de l'essai (°C)
Masse
Début Age au début de l'essai (semaine) Poids d'œuf * Taux de ponte / 100 (g/poule/j)
d’œuf
Fin Age en fin d'essai (semaine) Ingestion Ingestion d'aliment (g/j)
Indice de consommation (ingestion / masse
Durée durée de l'essai (j) IC
d'œuf produite)
Age moyen des poules au cours de l'essai
Age EM ingérée énergie métabolisable ingérée (kcal/poule/j)
(semaine)
Classes d'âges par rapport à l'âge moyen (20
à 35 semaines: début de ponte (1D), 35,1 à 60 Protéines
Classe d’âge Protéines brutes ingérées (g/poule/j)
: milieu de courbe de ponte (2M) et 50,1 à ingérées
100 : fin de courbe de ponte (3F))
PV Poids moyen au cours de l'essai (g) Lys ingérée Lysine ingérée (mg/poule/j)
Caractérisation qualitative du traitement
Facteur Met ingérée Méthionine ingérée (mg/poule/j)
expérimental
Valeur énergétique en EMAn des régimes
EMA n % jaune Proportion de jaune dans l'œuf (%)
offerts (Kcal/Kg)
Valeur en protéines brutes des régimes
PB % blanc Proportion de blanc dans l'œuf (%)
offerts (%)
Met Méthionine dans les régimes (%) % coquille Proportion de coquille dans l'œuf (%)

54
140 140 60

120 120 50

100 100
40
80 80
30
60 60
20
40 40

20 10
20

0 0 0
2100 2300 2500 2700 2900 3100 8 10 12 14 16 18 20 22 24 ,4 ,5 ,6 ,7 ,8 ,9 1 1,1
70 CP (%) Lys (%)
AME n (Kcal/Kg)
EMAn (Kcal / Kg) PB (%)
60

50

40

30

20

10

0
,15 ,2 ,25 ,3 ,35 ,4 ,45 ,5 ,55 ,6 ,65
Met (%)

Figure 8 Composition des régimes : EMAn, protéines brute, acides aminés

70 70 120

60 60
100
50 50
80
40 40
60
30 30

20 40
20

10 10 20

0 0 0
50 55 60 65 70 75 80 85 90 95 100 45 47,5 50 52,5 55 57,5 60 62,5 65 67,5 70 25 30 35 40 45 50 55 60 65 70
Taux de ponte (%) Poids des œufs (g) Masse d'oeuf (g/poule/j)

Figure 9 Performances de ponte : taux de ponte, poids des œufs et masse d’œuf

80
120 140
70
100 120
60
100
50 80
80
40 60
60
30
40
20 40

10 20 20

0 0 0
70 80 90 100 110 120 130 140 150 175 200 225 250 275 300 325 350 375 400 425 450 6 8 10 12 14 16 18 20 22 24 26
Ingestion d'aliment (g/j) Ingestion énergie(kcal/poule/j) Ingestion protéines (g/poule/j)

90
35
80
30
70
60 25

50 20
40
15
30
10
20
10 5

0 0
1,4 1,6 1,8 2 2,2 2,4 2,6 2,8 3 3,2 3,4 1200 1300 1400 1500 1600 1700 1800 1900 2000
IC Poids Vif (g)

Figure 10 Ingestion : aliment, énergie, protéines, Indice de Consommation et poids 55


Variable Nb
Nominales valeurs Publications Essais Lignées Classes d’âge
Brunes Blanches 1D 2M 3F
N 332 40 76
136 171 121 134 77

Variable Continues Nb Min Max Moyenne Médiane ETM


valeurs
Température (°C) 218 18 28 21 ,8 21 0 ,2
Début (Semaines) 332 18 72 37 ,6 32 1
Fin (Semaines) 332 24 81 51 ,5 50 0 ,8
Durée (Jours) 332 28 329 10 ,7 84 3,7
Age (Semaines) 332 22,5 75,5 44,5 40 0,9
PV initial (g) 156 1207 1987 1613 1598 16,4
PV final (g) 168 1375 2091 1731,3 1724 11,8
PV (g) 132 1303 1930 1679,5 1685 12,1
EMA n (Kcal/Kg) 332 2200 3130 2766,8 2758 7,9
PB (%) 332 9,4 21,9 16,5 16,9 0,1
Lys (g) 205 0,5 1,1 0,8 0,8 0,007
Met (g) 216 0,2 0,6 0,4 0,4 0,005
TSAA (g) 89 0,6 3,2 0,8 0,7 0,42
Thr (g) 56 0,4 0 ,8 0,6 0,6 0,009
Ca (g) 275 2,5 4,4 3,7 3,8 0,02
P (g) 164 0,2 1,2 0,5 0,4 0,01
Taux de ponte (%) 318 52,8 98,9 84 85 0,5
Poids d’œuf (g) 314 47,4 69,4 61,2 61,8 0,3
Masse d’œuf (g) 328 29,1 65,2 51,6 52,4 0,3
Ingestion (g) 332 73,0 146,1 108,6 108,8 0,7
IC 328 1,6 3,3 2,1 2,1 0 ,01
EM ingérée (Kcal) 332 204,0 424,0 299,9 299,5 1 ,86
Protéines ingérées (g) 332 7,2 24,3 17,9 17,9 0 ,1
Lys ingéré (mg) 205 379,0 1160,0 866,6 867 8 ,5
Met ingéré (mg) 216 166 940 412,4 405 6 ,4
% jaune (%) 107 21,2 30,8 26,23 26,0 0 ,2
% blanc (%) 122 54,4 68,3 62,1 62,8 0 ,3
% coquille (%) 107 8,6 11,3 11 10,3 0 ,2
Tableau 2 Description de la base initiale variables nominales et continues
56
24

22

20

18

PB (%)
16

14

12

10

8
2100 2300 2500 2700 2900 3100
EMA n (Kcal/Kg)
PB (%) = 9,478 + ,003 * EMA n (Kcal/Kg); R^2 = ,05
1,1 ,65
,6
1
,55

Met (%)
,9 ,5
,45
Lys (%)

,8
,4
,7 ,35

,6 ,3
,25
,5
,2
,4 ,15
8 10 12 14 16 18 20 22 24 8 10 12 14 16 18 20 22 24
PB (%) PB (%)
Lys (%) = ,136 + ,041 * PB (%); R^2 = ,501 Met (%) = ,019 + ,022 * PB (%); R^2 = ,279

Figure 11 Caractéristiques des régimes : représentation partielle du plan d’expérience méta-


analytique (Chaque couleur correspond à un essai)

Ces différents graphiques décrivent les aliments expérimentaux. Tout d’abord, une
grande diversité de teneurs en énergie métabolisable, en protéines et acides aminés est
présente dans cette base, comme les histogrammes (Fig 8,9,10) et le tableau 2 l’illustrent.
Cette variation continue a conduit à conserver l’ensemble des données, hormis quelques unes,
en nombre très limitées qui étaient extrêmisées, en lien avec le facteur expérimental étudié. Le
niveau d’énergie métabolisable n’est pas directement lié à la teneur en protéines comme cela
apparait sur le graphique (Fig 11) : Pour un niveau d’énergie, les teneurs en protéines des
aliments diffèrent beaucoup. Pour certains régimes, l’augmentation de la concentration en
énergie apparait corrélée avec les protéines, mais des situations inverses existent. Il y a donc
une grande variabilité dans les niveaux nutritionnels des régimes proposés mais également des
schémas expérimentaux très différents entre les essais. Ainsi certaines publications testaient
l’influence de différents niveaux d’énergie ou de protéines sans faire varier les autres
caractéristiques nutritionnelles (Fig 11). A l’inverse les niveaux d’acides aminés, ici Lys et
Met, dans l’aliment sont corrélés positivement aux niveaux de protéines (Fig 11). Par
57
conséquent, les lois de réponses obtenues entre PB, Lys et Met, seront à discuter de façon
prudente, les 3 variables étant fortement liées dans le plan d’expérience complexe de la base
de données.

III-2 Etablissement de lois de réponses : approches par étapes

Des approches par régressions ont été effectuées pour analyser les liens entre variables
influençant l’ingestion chez la poule, Les premières tentent d’évaluer si la composition
nutritionnelle du régime influence la prise alimentaire (Fig 12).
150
150
140
140
130

Ingestion d'aliment (g/j)


Ingestion d'aliment (g/j)

130
120
120
110
110
100
100
90
90
80
80
70
70 8 10 12 14 16 18 20 22 24
2100 2300 2500 2700 2900 3100
PB (%)
EMA n (Kcal/Kg)
Ingestion d'aliment (g/j) = 146,107 - 2,272 * PB (%); R^2 = ,083
Ingestion d'aliment (g/j) = 195,356 - ,031 * EMA n (Kcal/Kg); R^2 = ,123

150 140

140 130
Ingestion d'aliment (g/j)
Ingestion d'aliment (g/j)

130
120
120
110
110
100
100
90
90

80 80

70 70
,4 ,5 ,6 ,7 ,8 ,9 1 1,1 ,15 ,2 ,25 ,3 ,35 ,4 ,45 ,5 ,55 ,6 ,65
Lys (%) Met (%)
Ingestion d'aliment (g/j) = 129,81 - 27,495 * Lys (%); R^2 = ,043 Ingestion d'aliment (g/j) = 116,97 - 23,985 * Met (%) ; R^2 = ,019

Figure 12 Influence de la composition des régimes sur l’ingestion

Ces graphiques montrent l’influence de la composition nutritionnelle des régimes sur


l’ingestion. Il semble, comme cela est attendu, que plus la concentration en énergie dans le
régime augmente, plus l’ingestion diminue. Cependant dans certains essais, à un niveau
d’énergie similaire, des ingestions différentes sont observées. Les teneurs en protéines et en
acides aminés de l’aliment ne semblent pas avoir d’impact évident sur l’ingestion. Deux

58
IDFWHXUV TXDOLWDWLIV FROOHFWpV GDQV OD EDVH VHPEOHQW rWUH UHOLpV j O¶LQJHVWLRQ LO V¶DJLW GHV
OLJQpHVHWGHVFODVVHVG¶kJH
  

 

  

 


 

 
 



 %ODQFKHV 






%UXQHV

 


 
 
,QJHVWLRQG
DOLPHQW JM ,QJHVWLRQG
DOLPHQW JM


 )LJXUH,QIOXHQFHO¶kJHHWGHODVRXFKHVXUO¶LQJHVWLRQGHVSRXOHV

/¶LQJHVWLRQ DSSDUDLW VXSpULHXUH SRXU OHV SRXOHV EUXQHV FRPSDUDWLYHPHQW DX[ SRXOHV
EODQFKHV HW O¶LQJHVWLRQ DXJPHQWH DYHF O¶kJH )LJ   /HV UHSUpVHQWDWLRQV JUDSKLTXHV
VXLYDQWHVFRPELQDQWFODVVHG¶kJHOLJQpHVHWOHVYDULDEOHVTXDQWLWDWLYHVFRQFHUQDQWO¶pQHUJLHHW
OHV SURWpLQHV QH VHPEOHQW SDV IDLUH DSSDUDvWUH G¶LQWHUDFWLRQV PDMHXUHV HQWUH FHV GLIIpUHQWV
SDUDPqWUHV )LJ 

 

 
,QJHVWLRQG
DOLPHQW JM

 
,QJHVWLRQG
DOLPHQW JM




  




  




  





  

 %ODQFKHV

 %UXQHV
 
     
      
(0$Q .FDO.J (0$Q .FDO.J
  

 

,QJHVWLRQG
DOLPHQW JM
,QJHVWLRQG
DOLPHQW JM

 
   



  









 
  
%ODQFKHV
  %UXQHV 

 
                  

 3%  3% 


)LJXUH,QIOXHQFHGHODFRPSRVLWLRQQXWULWLRQQHOOHGHVUpJLPHVVXUO¶LQJHVWLRQHQIRQFWLRQGH
ODVRXFKHHWGHO¶kJHGHVSRXOHV 

100 100
95 95
90 90
85 85
80 80
75 75
70 70
65 65 1D

60 Blanches 60 2M

55 Brunes 55 3F
50 50
Taux de ponte (%) Taux de ponte (%)

70
70
67,5
67,5
65
65
62,5
62,5
60
60
57,5
57,5
55
55
52,5 1D
52,5
50 Blanches
50 2M
47,5 Brunes
47,5 3F
45
Poids des œufs (g) 45
Poids des œufs (g)

70 70
65 65

60 60
55 55
50 50
Unités

45 45
40 40 1D
35 Blanches 35 2M

30 Brunes 30 3F

25 25
Masse d'oeuf (g/poule/j) Masse d'oeuf (g/poule/j)

Figure 15 Influence de la souche et de l’âge sur les performances de ponte

La classe de lignées ne semble pas influencer sur le taux de ponte (Fig 15). Cependant
les poules blanches semblent pondre des œufs plus petits que les poules brunes, induisant
également une masse d’œuf inférieure pour les poules blanches. Selon l’âge des poules, le
poids des œufs s’accroit mais le taux de ponte diminuant, la masse d’œuf est globalement

60
stable. Cette dernière observation qualitative doit néanmoins être pondérée par le fait que les
poules concernées ne sont pas les mêmes dans les différentes classes d’âge (la proportion de
poule balvhe et brune pour chaque classe d’âge n’étant pas la même).
.

100
95
90
Taux de ponte (%)

85
80
75
70
65
60
55
50
70 80 90 100 110 120 130 140 150
Ingestion d'aliment (g/j)
Taux de ponte (%) = 72,567 + ,105 * Ingestion d'aliment (g/j); R^2 = ,023
70
70
67,5
65
Masse d'oeuf (g/poule/j)

65
60
Poids des œufs (g)

62,5
55
60
50
57,5
55 45

52,5 40

50 35

47,5 30
45 25
70 80 90 100 110 120 130 140 150 70 80 90 100 110 120 130 140 150
Ingestion d'aliment (g/j) Ingestion d'aliment (g/j)
Poids des œufs (g) = 35,475 + ,238 * Ingestion d'aliment (g/j); R^2 = ,445
Masse d'oeuf (g/poule/j) = 24,495 + ,249 * Ingestion d'aliment (g/j); R^2 = ,3

Figure 16 Influence de la quantité d’aliment ingéré sur la production d’œuf

Les graphiques ci dessus mettent en avant une absence de corrélation entre le taux de
ponte et la quantité d’aliment ingéré hormis dans le cas d’ingestions particulièrement faibles
où le taux de ponte semble diminuer (Fig 16). Cependant, il semble visuellement, et les
régressions linéaires simples appuient cette impression, que le poids des œufs et la masse
d’œufs augmentent avec l’ingestion d’aliment.

61
100
95
90

Taux de ponte (%)


100 85
95 80
Taux de ponte (%)

90 75
85 70
80 65
75 60
70 55
65 50
60 6 8 10 12 14 16 18 20 22 24 26
55 Ingestion protéines (g/poule/j)
50 Taux de ponte (%) = 55,442 + 1,595 * Ingestion protéines (g/poule/j); R^2 = ,169
175 200 225 250 275 300 325 350 375 400 425 450
Ingestion énergie(kcal/poule/j)
Taux de ponte (%) = 69,578 + ,048 * Ingestion énergie(kcal/poule/j); R^2 = ,034

70
67,5
70 65
67,5 Poids des œufs (g) 62,5
65 60
62,5 57,5
Poids des œufs (g)

60 55
57,5 52,5
55
50
52,5
47,5
50
45
47,5 6 8 10
12 14 16 18 20 22 24 26
45 Ingestion protéines (g/poule/j)
175 200 225 250 275 300 325 350 375 400 425 450 Poids des œufs (g) = 47,621 + ,759 * Ingestion protéines (g/poule/j); R^2 = ,14
Ingestion énergie(kcal/poule/j)
Poids des œufs (g) = 36,679 + ,082 * Ingestion énergie(kcal/poule/j); R^2 = ,389

70
65
70
60
Masse d'oeuf (g/poule/j)

65
55
Masse d'oeuf (g/poule/j)

60
50
55
45
50
40
45
35
40
35 30

30 25
6 8 10 12 14 16 18 20 22 24 26
25
Ingestion protéines (g/poule/j)
175 200 225 250 275 300 325 350 375 400 425 450
Masse d'oeuf (g/poule/j) = 23,876 + 1,551 * Ingestion protéines (g/poule/j); R^2 = ,36
Ingestion énergie(kcal/poule/j)
Masse d'oeuf (g/poule/j) = 22,427 + ,097 * Ingestion énergie(kcal/poule/j); R^2 = ,32

Figure 17 Influence de la nature et du niveau de l’ingéré sur la production d’œuf (énergie et


protéines)

62
100
95
90

Taux de ponte (%)


85
80
100
75
95
70
Taux de ponte (%)

90
65
85
60
80
55
75
50
70 100 200 300 400 500 600 700 800 900 1000
65
Met ingérée (mg/poule/j)
60
55 Taux de ponte (%) = 67,737 + ,034 * Met ingérée (mg/poule/j); R^2 = ,128
50
300 400 500
600 700 800 900 1000 1100 1200
Lys ingérée (mg/poule/j)
Taux de ponte (%) = 64,497 + ,023 * Lys ingérée (mg/poule/j); R^2 = ,083
70
Poids des œufs (g) 67,5
65
62,5
60
70 57,5
67,5 55
65 52,5
Poids des œufs (g)

62,5 50
60 47,5
57,5 45
55 100 200 300 400 500 600 700 800 900 1000
52,5
Met ingérée (mg/poule/j)
50
Poids des œufs (g) = 57,739 + ,008 * Met ingérée (mg/poule/j); R^2 = ,027
47,5
45
300 400 500 600 700 800 900 1000 1100 1200
Lys ingérée (mg/poule/j)
Poids des œufs (g) = 48,989 + ,013 * Lys ingérée (mg/poule/j); R^2 = ,13
60

55
Masse d'oeuf (g/poule/j)

50
70
45
65
40
Masse d'oeuf (g/poule/j)

60
55 35
50
30
45
25
40
100 200 300 400 500 600 700 800 900 1000
35
Met ingérée (mg/poule/j)
30
Masse d'oeuf (g/poule/j) = 39,219 + ,026 * Met ingérée (mg/poule/j); R^2 = ,18
25
300 400 500 600 700 800 900 1000 1100 1200
Lys ingérée (mg/poule/j)
Masse d'oeuf (g/poule/j) = 31,112 + ,023 * Lys ingérée (mg/poule/j); R^2 = ,199

Figure 18 Influence de la nature et du niveau de l’ingéré sur la production d’œuf (Lysine et


Méthionine) 63
Les approches suivantes se sont axées sur le lien possible entre le niveau et la nature
des ingérés et la production d’œufs. Concernant les ingérés d’énergie ou de protéines, le taux
de ponte ne semble pas affecté (Fig 17). Il semblerait néanmoins que l’augmentation
d’énergie ingérée mais également de protéines induisent une augmentation du calibre des
œufs ainsi que de la masse d’œuf exportée, même si les très grandes variabilités inter
expérimentales atténuent la mise en avant de ces effets d’ingestion. Pour la Méthionine, il
semble difficile d’établir une relation claire entre ingestion et performances de ponte. Pour la
Lysine, il semblerait que plus l’ingestion de Lysine augmente plus le calibre et la masse d’œuf
augmente (Fig 18).
Ces régressions linéaires simples sont plutôt à classer dans des analyses graphiques que
statistiques. Il est plus pertinent de déterminer si l’ingestion est influencée par des
combinaisons de facteurs qualitatifs (ici classe d’âge et lignée) et de variables quantitatives
caractérisant les aliments, donc de dissocier les différentes sources de variation liées à la
poule (lignée, âge) de celle induite par la composition de l’aliment ou son niveau d’apport
journalier . Cette discrimination entre facteurs de variation isole leur influence respective et
réduit la variance résiduelle donc améliore l’analyse statistique de chaque facteur par rapport
à l’approche globale.

III-3 Analyses par GLM de la base initiale

Après les analyses par Modèle Linéaire Général, il est possible de déterminer quels
sont les facteurs qui semblent influencer statistiquement l’ingestion. Celle-ci semble
dépendante de plusieurs facteurs comme cela apparait dans le tableau 3. Tout d’abord la
souche des poules impacte la prise alimentaire avec une consommation supérieure pour les
poules brunes par rapport aux poules blanches. Ensuite l’âge influence également l’ingestion
avec une augmentation de la consommation. La composition nutritionnelle du régime a un
impact sur l’ingestion avec une relation inverse entre concentration énergétique du régime et
consommation. La concentration du régime en protéines tend également impacter l’ingestion
mais dans une moindre mesure. Les teneurs en Lys et en Met n’impactent pas l’ingestion.
L’ingestion dépend donc principalement de l’âge et de la souche des poules mais également
de la concentration en énergie du régime.

Les analyses par GLM quantifient l’impact des facteurs sur la production d’œuf, le
poids de poules et l’indice de consommation. Tout d’abord, l’âge est corrélé à la production

64
d’œufs (taux de ponte, poids des œufs et masse d’œuf). En effet le taux de ponte diminue avec
l’âge alors que le poids des œufs et la masse d’œuf augmentent. La souche des poules semble
n’impacter que le taux de ponte avec un taux de ponte légèrement inférieur pour les poules
blanches par rapport aux poules brunes mais n’influencer ni le poids ni la masse d’œuf. La
nature et les quantités ingérées des différents nutriments impactent également les
performances de production. L’ingestion d’aliment (Tab 3) améliore significativement
l’ensemble des performances. Le taux de ponte est augmenté avec l’ingestion de protéines
(Tab 4) mais pas avec l’ingestion d’énergie. Pour le poids des œufs et la masse d’œuf, les
ingérés énergétique et protéique impactent positivement le poids des œufs. La quantité de Lys
et de Met sont liées positivement aux performances de ponte. Le poids des poules est corrélé à
l’ingestion d’aliment mais également à la nature de ces ingérés. En effet les ingestions
d’énergie, de protéines et de lysine ont un effet positif sur le poids vif des animaux an
contraire de la méthionine. Pour l’IC, une ingestion plus élevée d’énergie le diminuer alors
qu’ un surplus de protéines augmente l’IC. Les niveaux d’ingestion de Lys et de Met
n’influencent pas l’IC. Les performances des poules (ponte et croissance) dépendent donc de
plusieurs facteurs comme la souche des animaux, l’âge des poules, mais également de la
quantité et la qualité de l’aliment ingéré.

Les analyses par GLM décrivent des modèles évaluant l’ingestion ou les performances
de ponte. Ci-dessous trois exemples incluant trois niveaux de protéines illustrent l’impact
conjoint des teneurs en énergie ou de l’ingestion d’aliment sur les performances des poules.
Ces valeurs de 13, 15 et 17 % de protéines ont été choisies car elles couvrent bien l’ensemble
des valeurs qui ont pu être rencontrées dans les publications (Fig 19).

65
140

Ingestion (g/p) 130

120

110
PB 13%
100 PB 15%
PB 17%
90
2 200 2 400 2 600 2 800 3 000
85
EMAn (kcal/Kg)
68,0

80

Taux de ponte (%)


65,4
poids oeufs (g)

75
62,8

70
60,2

65
57,6

60
90 100 110 120 130
55,0 ingestion (g/p)
90 100 110 120 130
ingestion (g/p)

Figure 19 Modèle d’évolution de l’ingestion ou des performances de ponte pour des


poules brunes, de classe d’âge 2M avec trois niveaux de PB

Cependant, les facteurs étant liés les uns aux autres, il est difficile de définir l’effet à
affecter à chaque facteur sur les performances des poules. De plus cette façon de traiter les
différents effets ne prend pas en compte la diversité entre les essais. L’analyse des essais
comparés à un témoin spécifique défini pour chaque essai permettra de prendre en compte
partiellement cette variabilité due aux nombreux traitements et essais (traitement des données
sur un modèle relatif).

66
7DEOHDX,QIOXHQFHGHODQDWXUHGHVUpJLPHVHWGHO¶LQJHVWLRQVXUOHVSHUIRUPDQFHV SRQWHHWFURLVVDQFH HWVXUODTXDQWLWpLQJpUpHHQIRQFWLRQ
GH OD FODVVH GH OLJQpH HW OD FODVVH G¶kJH &KDTXH OLJQH FRUUHVSRQG j XQ PRGqOH VWDWLVWLTXH FRPSUHQDQW GHX[ IDFWHXUV HW XQH RX SOXVLHXUV
YDULDEOHVH[SOLFDWLYHVFRQWLQXHV


7DEOHDX,QIOXHQFHGHODQDWXUHGHVLQJpUpV pQHUJLHSURWpLQHV/\V0HW VXUOHVSHUIRUPDQFHV SRQWHHWFURLVVDQFH HQIRQFWLRQGHODFODVVHGH
OLJQpHHW ODFODVVHG¶kJH &KDTXHOLJQHFRUUHVSRQGjXQPRGqOHVWDWLVWLTXHFRPSUHQDQW GHX[IDFWHXUVHW XQHRXSOXVLHXUV YDULDEOHVH[SOLFDWLYHV
FRQWLQXHV

IV « Modèle relatif » énergie / protéines
Pour établir les lois de réponses sur ce modèle, l’ensemble des performances mais
aussi des paramètres qui caractérisent les régimes vont être comparés au témoin de chaque
essai. Ces comparaisons par rapport à ces témoins spécifiques permettent d’évaluer l’effet des
régimes testés par rapport à un « niveau de base » fixé au sein de chaque essai. Pour chaque
essai, un témoin est défini, et une nouvelle base de données est crée en prenant ces valeurs
relatives. Les essais sélectionnés pour la conception de cette base présentent des différences
soit pour la composition en énergie métabolisable (base spécifique pour tester l’effet énergie),
soit pour la composition en protéine (base spécifique pour tester l’effet niveau de protéine).
Les essais qui ne présentaient pas, ou de très faibles différences dans la composition pour ces
deux paramètres, ne sont pas pris en compte pour ces nouvelles bases de données.
Ces bases pour traiter les données par un modèle relatif comprennent 22 publications
et 34 essais pour la base énergie et 27 publications et 50 essais pour la base protéine. Les
valeurs présentées dans les tableaux sont les différences au témoin, exprimées dans l’unité de
la variable considérée (Tab 5, 6, 7).

69
Tableau 5 Description de la base modèle relatif effet énergie / protéines variables nominales

.
Nb valeurs Publication Essai Lignée Classe d’âge

Brunes Blanches 1D 2M 3F
Energie 132 22 34
88 44 57 45 30
Protéines 193 27 50 84 109 84 61 48

Tableau 6 Description de la base modèle relatif effet énergie

Nb valeurs Min Max Moyenne SEM


PV (g) 57 -104 38,5 -10,9 3,98
EMA n (Kcal/Kg) 132 -580 400 -16,5 13,93
Taux de ponte (%) 132 -8 6,4 -0,06 0,18
Poids d’œuf (g) 128 -2,2 2,6 -0,02 0,06
Masse d’œuf (g/j) 128 -4,62 3,358 -0,04 0,12
Ingestion (g/j) 132 -12,6 29,8 1,01 0,54
IC 128 -0,39 0,777 0,018 0,01
EM ingérée (Kcal/j) 132 -23,55 31,262 0,47 0,83

Tableau 7 Description de la base modèle relatif effet protéines

Nb valeurs Min Max Moyenne SEM


PV (g) 45 -104 38,5 -12,5 4,9
PB (%) 193 -3,5 3 -0,13 0,08
Taux de ponte (%) 193 -12,3 6,4 -0,56 0,22
Poids d’œuf (g) 193 -2,97 2,6 -0,04 0,06
Masse d’œuf (g/j) 193 -9,6 4,3 -0,27 0,15
Ingestion (g/j) 193 -13,2 29,8 0,43 0,4
IC 193 -0,39 0,78 0,02 0,01
Protéines ingérées (g/j) 193 -4,3 3,59 -0,08 0,08

70
35
30
25

Delta Ingestion aliment (g) 20


15
10
5
0
-5
-10
-15
-600 -400 -200 0 200 400 600
Delta Energie (Kcal)
Delta Ingestion aliment (g) = ,471 - ,033 * Delta Energie (Kcal); R^2 = ,707

35
30
Delta Ingestion aliment (g)

25
20
15
10
5
0
-5
-10
-15
-4 -3 -2 -1 0 1 2 3 4
Delta Protéines (g)
Delta Ingestion aliment (g) = ,232 - 1,465 * Delta Protéines (g); R^2 = ,085

Figure 20 Influence de la composition nutritionnelle des régimes sur l’ingestion

Ces graphiques montrent la relation forte déjà observée précédemment entre la


concentration en énergie du régime et l’ingestion. La concentration en protéines ne semble
pas avoir d’effet clair sur cette ingestion (Fig 20). Ces relations sont à confirmer par des
GLM.

71
8 8
6

Delta Taux de ponte (%)


6
4
Delta Taux de ponte (%)

4
2
2 0
0 -2
-2 -4
-6
-4
-8
-6
-10
-8 -12
-10 -14
-10 0-30 10-20 20 30 40 -2 -1 -5 0-4 -3
1 2 3 4
Delta Energie ingérée (Kcal) Delta Protéines ingérées (g)
Delta Taux de ponte (%) = -,024 + ,047 * Delta Energie ingérée (Kcal); R^2 = ,048 Delta Taux de ponte (%) = -,406 + 1,339 * Delta Protéines ingérées (g); R^2 = ,271

3
3
2,5 2
2
Delta Poids des œufs (g)
Delta Poids des œufs (g)

1
1,5
1 0
,5
0 -1

-,5 -2
-1
-1,5 -3
-2
-4
-2,5 -5 -4 -3 -2 -1 0 1 2 3 4
-10 0 -30 -20
10 20 30 40 Delta Protéines ingérées (g)
Delta Energie ingérée (Kcal)
Delta Poids des œufs (g) = -,012 + ,022 * Delta Energie ingérée (Kcal); R^2 = ,1 Delta Poids des œufs (g) = -,002 + ,341 * Delta Protéines ingérées (g); R^2 = ,215

6
4
3 4
Delta Masse œufs (g)

2 2
Delta Masse d'œuf (g)

1 0
0 -2
-1
-4
-2
-6
-3
-8
-4
-10
-5
-5 -4 -3 -2 -1 0 1 2 3 4
-10 -30
0 10-20 20 30 40
Delta Energie ingérée (Kcal) Delta Protéines ingérées (g)
Delta Masse d'œuf (g) = -,007 + ,043 * Delta Energie ingérée (Kcal); R^2 = ,097 Delta Masse œufs (g) = -,149 + 1,068 * Delta Protéines ingérées (g); R^2 = ,381

Figure 21 Influence de la nature et du niveau de l’ingéré sur la production d’œuf

72
Ces représentations graphiques, qui mettent en relation l’ingestion soit d’énergie soit
de protéines avec les performances de ponte (taux ponte, poids des œufs et masse d’œufs)
montrent l’absence de réponse majeure à des variations d’apport énergétique au sein des
essais concernés (Fig 21). Il semble néanmoins exister des effets de l’ingestion de protéines
mais qui ne sont pas bien mis en évidence. Des analyses statistiques complémentaires par
GLM quantifieront ces possibles relations.

60
40
60
20

Delta Poids vif (g)


40
0
20
-20
Delta Poids vif (g)

0
-40
-20
-60
-40
-80
-60
-100
-80
-120
-100
-1,5 -1 -,5 0 ,5 1 1,5 2 2,5 3
-120 Delta Protéines ingérées (g)
-30 -20 -10 0 10 20 30 40 Delta Poids vif (g) = -9,439 + 5,034 * Delta Protéines ingérées (g); R^2 = ,016
Delta Energie ingérée (Kcal)
Delta Poids vif (g) = -6,144 + 1,105 * Delta Energie ingérée (Kcal); R^2 = ,103

,5
,4
,5 ,3
,4 ,2
,3
Delta IC

,1
,2 0
,1 -,1
Delta IC

0 -,2
-,1 -,3
-,2
-,4
-,3 -,5
-,4 -5 -4 -3 -2 -1 0 1 2 3 4
-,5 Delta Protéines ingérées (g)
-30 -20 -10 0 10 20 30 40 Delta IC = ,011 - ,011 * Delta Protéines ingérées (g); R^2 = ,012
Delta Energie ingérée (Kcal)
Delta IC = ,012 - ,003 * Delta Energie ingérée (Kcal); R^2 = ,043

Figure 22 Influence de la nature et du niveau de l’ingéré sur le poids vif et l’Indice de


Consommation

73
Il n’y a pas de lien visible entre les apports en énergie et protéines et les poids des
poules et indice de consommation (Fig 22).

Les analyses par GLM ont été effectuées pour le modèle relatif, sur des essais où il y
avait des variations dans les niveaux d’énergie et de protéines des régimes. Pour faciliter la
lecture, les résultats sont présentés sous la forme de deux tableaux (Tab 8, 9), le premier
portant sur des essais où le niveau d’énergie de régimes variaient et le deuxième où le niveau
de protéine des régimes étaient différents selon les traitements.

74
7DEOHDX,QIOXHQFHGHODTXDQWLWpG¶pQHUJLHGDQVOHVUpJLPHVGHO¶LQJHVWLRQG¶DOLPHQWHWGHFHOOHG¶pQHUJLHVXUOHVSHUIRUPDQFHV SRQWHHWSRLGV
YLI HWVXUODTXDQWLWpLQJpUpHHQIRQFWLRQGHVOLJQpHVHWGH¶kJH&KDTXHOLJQHFRUUHVSRQGjXQPRGqOHVWDWLVWLTXHFRPSUHQDQWGHX[IDFWHXUVHW
XQHRXSOXVLHXUVYDULDEOHVH[SOLFDWLYHVFRQWLQXHV
















7DEOHDX  ,QIOXHQFH GH OD TXDQWLWp GH SURWpLQHV GDQV OHV UpJLPHV GH O¶LQJHVWLRQ G¶DOLPHQW HW GH FHOOH  GH SURWpLQHV VXU OHV
SHUIRUPDQFHV SRQWHSRLGVG¶°XIHWSRLGVYLI HWVXUODTXDQWLWpLQJpUpHHQIRQFWLRQGHODFODVVHGHOLJQpHHWODFODVVHG¶kJH&KDTXH
OLJQHFRUUHVSRQGjXQPRGqOHVWDWLVWLTXHFRPSUHQDQWGHX[IDFWHXUVHWXQHRXSOXVLHXUVYDULDEOHVH[SOLFDWLYHVFRQWLQXHV


















Ces analyses statistiques mettent en évidence, tout en tenant compte de l’effet
« essais » que le niveau d’énergie dans le régime a bien un fort impact sur la consommation
d’aliment (Tab 8) sans effet de la souche et de l’âge, alors que le niveau de protéines dans la
ration n’impacte pas l’ingestion et cela pour tous les âges (Tab 9).

IV-1 Modèles GLM, modulation d’apports énergétiques (Tab 8) :


Pour les essais sélectionnés, les effets des variables sont différents selon les
performances. Pour le taux de ponte, seule la souche des animaux semble avoir un effet
(supérieur pour les poules blanches) mais pas d’impact de l’âge, de l’ingéré énergétique et de
l’ingestion d’aliment sur la production d’œuf. Le niveau d’énergie ingérée influence
cependant fortement le poids des œufs ainsi que la masse d’œuf produite. La nature de
l’ingéré, plus ou moins riche en énergie, semble également impacter positivement
l’augmentation du poids vif alors que l’ingestion seule d’aliment ne montre pas d’effet ; c’est
donc bien le niveau d’énergie de l’aliment qui impacte le poids vif et non simplement la prise
alimentaire. Cependant l’indice de consommation n’est pas du tout lié au niveau d’énergie
ingéré mais est très fortement lié au niveau d’ingestion d’aliment, avec une augmentation de
l’ingestion qui dégrade l’IC.

IV-2 Modèles GLM, modulation d’apports protéiques (Tab 9):

Pour l’ensemble des performances relevées (ponte etpoids vif), la souche ne semble
pas avoir d’impact. Pour les performances de ponte (taux de ponte, poids des œufs et masse
d’œuf), l’ingestion d’aliment et spécifiquement de protéines augmentent ces paramètres.
L’effet croisé ingestion d’aliment x teneur en protéines dans le régime montre encore une fois
que l’ingestion et le taux de protéines dans le régime permettent d’améliorer les performances
de ponte. La classe d’âge influence également le taux de ponte et la masse d’œuf avec une
diminution transitoire du taux de ponte en milieu de cycle, ce qu’il faut considérer avec
prudence étant donnée le jeu de valeurs utilisée et l’approche relative. Aucun facteur ne
semble affecter le poids vif des animaux . L’IC quant à lui semble être influencé par la classe
d’âge avec un IC variable, mais cela peut être relié à la nature des données. Les courbes de
modèles linéaires réalisées à partir des coefficients des tableaux d’analyse GLM illustrent
l’effet positif du niveau d’ingestion sur le taux de ponte et le poids des œufs en fonction de
trois taux de protéines définis. Ces valeurs de -2, 0 et 2 % ont été choisie car elles couvrent

77
ELHQO¶HQVHPEOHGHVYDOHXUVTXLRQWSXrWUHUHQFRQWUpHVGDQVOHVSXEOLFDWLRQVHWGDQVODEDVH
GHGRQQpHVUHODWLYH )LJ 






GHOWDSRLGVGHV°XIV J
GHOWDWDX[SRQWH 

 













    
    
GHOWDLQJHVWLRQ J  GHOWDLQJHVWLRQ J 



)LJXUH0RGqOHG¶pYROXWLRQGHVSHUIRUPDQFHVGHSRQWH WDX[GHSRQWHHWSRLGVGHV°XIV 
SRXUGHVSRXOHVEUXQHVGHFODVVHG¶kJH0DYHFWURLVQLYHDX[GH3%


9$QDO\VHGHODEDVHHQPRGqOHUHODWLI/\VLQH0pWKLRQLQH

 3RXU pWDEOLU OHV ORLV GH UpSRQVHV VXU FHV PRGqOHV OHV HVVDLV VpOHFWLRQQpV SRXU OD
FRQFHSWLRQGHFHWWHEDVHSUpVHQWHQWGHVGLIIpUHQFHVVLJQLILFDWLYHVVRLWSRXUODFRPSRVLWLRQHQ
0pWKLRQLQH VRLW SRXU OD FRPSRVLWLRQ HQ /\VLQH &HWWH EDVH GH GRQQpHV FRPSUHQG 
SXEOLFDWLRQV HW  HVVDLV SRXU OD EDVH /\VLQH HW  SXEOLFDWLRQV HW  HVVDLV SRXU OD EDVH
SURWpLQHV 7DE 



Tableau 10 Description de la base modèle relatif Méthionine / Lysine variables nominales

Nb
Publication Essais Lignée Classe d’âge
valeurs

Brunes Blanches 1D 2M 3F
Met 115 16 16
56 59 42 33 40

Lys 97 15 15 44 52 46 38 12

Tableau 11 Description de la base modèle relatif effet Méthionine

Nb valeurs Min Max Moyenne SEM


Met (g) 115 -0,1 0,3 0,0 0,01
Taux de ponte (%) 115 -35,4 7,0 -0,7 0,42
Poids d’œuf (g) 111 -5,8 5,3 0,3 0,14
Masse d’œuf (g) 111 -24,3 5,0 -0,3 0,30
Ingestion (g) 115 -19,8 29,8 0,3 0,49
IC 115 -0,39 0,89 0,01 0,02
Met ingérée (mg) 115 -162,1 542,1 19,0 9,15
Poids Vif (g) 19 -58,5 10,0 -11,7 4,49

Tableau 12 Description de la base modèle relatif effet Lysine

Nb valeurs Min Max Moyenne SEM


Lys (g) 97 -0,4 0,2 0,0 0,01
Taux de ponte (%) 97 -40,6 7,0 -1,4 0,66
Poids d’œuf (g) 93 -5,8 5,3 0,1 0,16
Masse d’œuf (g) 93 -27,7 5,0 -0,8 0,48
Ingestion (g) 97 -30,7 29,8 0,1 0,68
IC 97 -0,39 0,89 0,03 0,02
Lys ingérée (mg) 97 -514,7 157,3 -14,2 10,56
Poids Vif (g) 23 -58,5 10,0 -10,1 3,79

79
10
10
5
5

delta Taux de ponte (%)


0
PONTE(%)
(%)

0
-5 -5
de ponte

-10 -10
Taux TAUX

-15 -15
-20
DeltaDELTA

-20
-25
-25
-30
-30 -35
-35 -40
-40 -45
-,15 -,1 -,05 0 ,05 ,1 ,15 ,2 ,25 ,3 -,5 -,4 -,3 -,2 -,1 0 ,1 ,2
DELTA MET delata Lys (%)
Delta Met (%) Delta Lys (%)
6
6
4
Delta poids des oeufs (g)
Delta poids des œufs (g)

4
2
Delta poids des oeufs (g)

0
0

-2
-2

-4
-4

-6 -6

-8 -8
-,15 -,1 -,05 0 ,05 ,1 ,15 ,2 ,25 ,3 -,5 -,4 -,3 -,2 -,1 0 ,1 ,2
DELTA MET
Delta Met (%)
delata
Delta LysLys
(%)(%)

10 10
Masse (g/poule/j)

5 5
delta Masse d'oeuf (g/poule/j)
d'oeuf (g)

0
0

-5
-5
d’œuf

-10
-10
Delta

-15
Delta masse

-15
-20
-20
-25
-25
-30
-,15 -,1 -,05 0 ,05 ,1 ,15 ,2 ,25 ,3
-,5 -,4 -,3 -,2 -,1 0 ,1 ,2
DELTA(%)
Delta Met MET
delataLys
Delta Lys(%)
(%)

Figure 24 Influence de la composition des régimes sur les performances de ponte

80
Ces graphiques montrent que les performances des poules dépendent de la
composition en acides aminés des régimes (Fig 24). En effet lorsque le taux de Met ou de Lys
dans le régime diminue fortement, les performances sont rapidement dégradées. Si le niveau
de ces acides aminés est proche du témoin, les performances ne différent pas de celui -ci.
Enfin si le niveau de ces acides aminés dans le régimes et le niveau ingéré sont fortement
augmentés, les performances de ponte semblent arriver à un « plateau ». Cela suggère qu’il
existe un niveau optimal d’apport de Lysine et de Méthionine (Fig 25).
10 10
5 5
0 0

delta Taux de ponte (%)


(%)
ponte (%)

-5 -5
ponte

-10 -10
-15
taux

-15
Deltade

-20
-20
Delta Taux

-25
-25
-30
-30 -35
-35 -40
-40 -45
-200 -100 0 100 200 300 400 500 600 -600 -500 -400 -300 -200 -100 0 100 200
6
6 Delta Lys ingérée (mg)
DeltaDELTA
MetMET
ingérée (mg)
INGEREE
(g) (g)
œufs

4 4
Delta poids des oeufs (g)
oeufs

2
desdes

2
poids

0 0
Delta poids

-2
Delta

-2

-4 -4

-6
-6

-8 -8
-200 -100 0 100 200 300 400 500 600 -600 -500 -400 -300 -200 -100 0 100 200
Delta
DELTAMet ingérée (mg)
MET INGEREE
10 10 Delta Lys ingérée (mg)
5
5
Masse (g/poule/j)

delta Masse d'oeuf (g/poule/j)

0
d'oeuf (g)

0
-5
-5
d’œuf

-10
-10
-15
Delta
Delta masse

-15
-20

-20 -25

-25 -30
-200 -100 0 100 200 300 400 500 600 -600 -500 -400 -300 -200 -100 0 100 200

DELTA MET INGEREE Delta Lys ingérée (mg)


Delta Met ingérée (mg)

Figure 25 Influence de la nature et du niveau de l’ingéré sur la production d’œuf 81


V-1 Analyses par GLM base Lys et Met

Ces bases en relatif mettent en évidence plusieurs effets sur les performances de ponte
ou le poids des poules. En effet ni la composition en Met ou en Lys n’influencent l’ingestion
globale d’aliment ce qui suggère que les poules ne régulent pas leur ingestion sur la
concentration de ces acides aminés dans le régime (Tab 13), sauf sans doute en cas de
carences avérées, cas ici non traité. Par ailleurs la concentration de ces deux acides aminés
essentiels dans le régime induit une amélioration des performances de ponte (taux de ponte,
poids des œufs et masse d’œuf). L’indice de consommation est également amélioré avec
l’augmentation de la concentration de Lys ou de Met dans le régime. Cette amélioration de
l’IC est certainement due à l’amélioration des performances de ponte. Cependant, cette
analyse par GLM montre ses limites car elle ne met pas en évidence le « plateau » observé
graphiquement au dessus (Fig 24, 25). Il aurait fallu tester d’autres modèles de types
quadratiques. Comme observé pour la concentration de Lys et de Met dans le régime,
l’ingestion de ces deux acides aminés améliore significativement les performances de ponte
(taux de ponte, poids des œufs et masse d’œuf) (Tab 14). Il n’y a pas d’effet du niveau
d’ingestion de ces deux acides aminés sur le poids des poules. L’IC est quant à lui également
amélioré avec le niveau d’ingestion de Lys et de Met. Ces analyses par GLM permettent donc
d’obtenir des coefficients qui servent à la conception des illustrations proposées ci-dessous
(Fig 26). Ces deux graphiques montrent l’effet positif de la concentration en Lys ou en Met
sur la masse d’œuf avec dans le cas de la Lys un fort effet du niveau d’ingestion sur la masse
d’œuf alors que pour la Met il n’y a pas d’effet ingestion sur la masse d’œuf. . Ces valeurs de
-10, 0 et 10g ont été choisie car elles couvrent bien l’ensemble des valeurs qui ont pu être
rencontrées dans les publications et dans la base relative.

82


 

'HOWDPDVVHRHXI J
'HOWDPDVVHRHXI J

 

 

 

 
        

'HOWD/\V  'HOWD0HW 

)LJXUH0RGqOHG¶pYROXWLRQGHODPDVVHG¶°XIHQIRQFWLRQGHODFRPSRVLWLRQHQ/\V
RX0HWSRXUGHVSRXOHVEUXQHVGHFODVVHG¶kJH0DYHFWURLVQLYHDX[G¶LQJHVWLRQV



7DEOHDX,QIOXHQFHGHODTXDQWLWpGH/\VHW0HWGDQVOHVGHVUpJLPHVHWGHO¶LQJHVWLRQG¶DOLPHQWVXUOHVSHUIRUPDQFHV SRQWHHW

SRLGV YLI  HW VXU OD TXDQWLWp LQJpUpH HQ IRQFWLRQ GH OD FODVVH GH OLJQpH HW OD FODVVH G¶kJH &KDTXH OLJQH FRUUHVSRQG j XQ PRGqOH

VWDWLVWLTXHFRPSUHQDQWGHX[IDFWHXUVHWXQHRXSOXVLHXUVYDULDEOHVH[SOLFDWLYHVFRQWLQXHV























7DEOHDX,QIOXHQFHGHODTXDQWLWpGH/\VHW0HWGDQVLQJpUpHVXUOHVSHUIRUPDQFHV SRQWHHWSRLGVYL HQIRQFWLRQGHOD
FODVVHGHOLJQpHHW ODFODVVHG¶kJH&KDTXHOLJQHFRUUHVSRQGjXQPRGqOH VWDWLVWLTXH FRPSUHQDQW GHX[IDFWHXUVHW XQHRX
SOXVLHXUVYDULDEOHVH[SOLFDWLYHVFRQWLQXHV










86
VI Conclusion et perspectives

Cette méta-analyse regroupe 40 publications. Elle permet de confirmer certaines


connaissances sur l’alimentation des poules pondeuses mais également d’apporter des
informations quantifiées sur l’ingestion chez la poule et ses impacts sur les performances.
Concernant l’ingestion, la souche de poule ainsi que l’âge sont des facteurs qui influent sur
l’ingestion avec une quantité plus importante pour des poules brunes que pour des poules
blanches et une ingestion qui augmente avec l’âge. Concernant les régulations de l’ingestion,
les analyses ont confirmé quantitativement que la concentration en énergie dans le régime
était l’élément le plus important. Le niveau de protéines influence peu l’ingestion alors que les
teneurs en Lys et en Met n’ont pas d’effet significatif. Néanmoins, la constitution de la base
de données ne prend sans doute pas en compte des cas de carences fortes en acides aminés,
provoquant des dégradations des performances et corrélativement de l’ingestion. Ces
observations sont vraies pour l’ensemble de poules, quels que soient les souches ou l’âge,
même si des régulations doivent surement varier selon les souches. Les interactions n’ayant
pas été testées en lien avec la complexité du méta-plan d’expérience, il n’est pas possible de
donner plus d’élément de réponses.

Concernant les performances de ponte, le taux de ponte est amélioré avec l’ingestion
de protéines mais ni par la teneur en énergie du régime ou la quantité d’énergie ingérée. Pour
le poids des œufs, les ingérés énergétique et protéique en accroissent le poids. Les teneurs en
Lys et Met dans le régime, mais également les quantités ingérées de ces acides aminés
améliorent systématiquement le taux de ponte, le poids des œufs et la masse d’œuf. Le poids
des poules est quant à lui influencé par la souche (poules brunes plus lourdes) et par l’âge
(augmentation avec l’âge). Mais les teneurs en énergie, protéines ou Lys n’influencent pas le
poids des poules. Seule la Met dans le régime et la quantité totale d’aliment ingéré est corrélée
au poids. Une nuance peut toutefois être apportée car les mesures de poids ici sont un poids
moyen et donc une mesure figée. Afin de pouvoir notre l’influence d’autres nutriments il
serait pertinent d’évaluer l’évolution de la croissance et donc plutôt évaluer un flux au cours
des expérimentations plutôt qu’un bilan figé.

La base initiale a été établie pour évaluer quantitativement les régulations de


l’ingestion puis les relations entre quantité et qualité de l’ingéré sur les performances des
poules. Il est difficile d’obtenir de modèles très représentatifs car la grande diversité des

87
données incite à la prudence. Cependant cette diversité apporte des éléments comme les effets
induits par les souches des poules et leur âge. Pour améliorer ce travail, il faudrait pour
certaines variables (comme les ingestions de protéines, de Lys et Met) utiliser un modèle
quadratique plutôt qu’un modèle linéaire. En effet le modèle linéaire ne peut pas prendre en
compte un possible plateau dans les performances suivant une augmentation linéaire ou quasi-
linéaire dans le cas d’une carence ou subcarence d’un élément en amont de ce plateau.
Concernant ces acides aminés, pour prendre en compte les systèmes d’alimentation actuels, il
aurait été pertinent de recalculer les concentrations des acides aminées en quantité digestible
pour la poule. Or dans nos analyses, les valeurs de Lys et de Met totales ont été conservées.
Plus globalement, cela encouragerait à recalculer l’ensemble des régimes dans un système
d’alimentation commun de façon à gommer un certain nombre de différences artéfactuelles.

Au cours de l’écriture de la base de données, la composition en matières première de


l’ensemble des régimes avait été relevée. Ces informations n’ont cependant pas été utilisées
pour l’instant. Néanmoins la composition des régimes peut permettre d’évaluer si des matières
premières peuvent influencer l’ingestion d’aliment, notamment en termes d’appétence, ou à
travers des interactions nutritionnelles positives comme négatives sachant que les
conséquences de la nature des matières premières utilisées peuvent jouer sur les performances
des poules. La composition en minéraux des régimes (Calcium et Phosphore) a également été
répertoriée et non utilisée. Or les poules ont un appétit spécifique pour calcium. Ces éléments
pourraient également entrer en compte comme des facteurs qui vont influencer l’ingestion. Le
niveau de calcium dans notre base de donnée n’étant pas un élément qui variait beaucoup, il
n’est pas sur que dans cette base nous ayons pu mettre en évidence un lien entre calcium et
ingestion.

Afin de tester les résultats obtenus au cours de ces travaux de méta-analyse et de


combler des manques (comme sur la nature des régimes), il serait pertinent de réaliser des
expérimentations combinant des niveaux nutritionnels variables associés à des jeux de
matières premières diversifiés. Une approche complémentaire pourrait être de confronter ces
résultats avec des données en élevage. L’utilisation de différentes matières premières et leur
effet sur le niveau d’ingestion sera évalué au cours de la première expérience de la thèse. La
seconde expérience permettait d’évaluer la possibilité de contrôler l’ingestion et d’induire des
différences de croissances chez les poules par la variation des apports en énergie et en
protéines. Cette variation de poids n’ayant pas été mis en évidence dans ce travail de méta

88
analyse il était intéressant de le tester. La temporalité des apports doit probablement jouer un
rôle majeur dans la régulation de l’ingestion chez les poules mais également agir sur les
performances de ponte et de croissance chez les poules. Cet aspect de « moment de
distribution » n’étant pas évalué dans ce travail de méta analyse, l’expérience 3 permettait
d’apporter des informations sur l’effet d’apports nutritionnels variés à différents moments de
la journée.

89
90
VII Références de la méta-analyse
Abd-Elrazig SM and Elzubeir EA 1998. Effects of feeding pearl millet on laying hen
performance and egg quality. Animal Feed Science and Technology 76, 89-94.

Almeida VR, Dias AN, Bueno CFD, Couto FAP, Rodrigues PA, Nogueira WCL and Faria
Filho DE 2012. Crude Protein and Metabolizable Energy Levels for Layers Reared in
Hot Climates. Brazilian Journal of Poultry Science 14, 203-208.

Antruejo A, Azcona JO, Garcia PT, Gallinger C, Rosmini M, Ayerza R, Coates W and Perez
CD 2011. Omega-3 enriched egg production: the effect of alpha-linolenic omega-3 fatty
acid sources on laying hen performance and yolk lipid content and fatty acid
composition. British Poultry Science 52, 750-760.

Bennett CD and Classen HL 2003. Performance of two strains of laying hens fed ground and
whole barley with and without access to insoluble grit. Poultry Science 82, 147-149.

Boling SD, Douglas MW, Johnson ML, Wang X, Parsons CM, Koelkebeck KW and
Zimmerman RA 2000. The effects of dietary available phosphorus levels and phytase
on performance of young and older laying hens. Poultry Science 79, 224-230.

Burley HK, Patterson PH and Elliot MA 2013. Effect of a reduced crude protein, amino acid-
balanced diet on hen performance, production costs, and ammonia emissions in a
commercial laying hen flock. Journal of Applied Poultry Research 22, 217-228.

Gunawardana P, Roland DA, Sr. and Bryant MM 2008. Effect of Energy and Protein on
Performance, Egg Components, Egg Solids, Egg Quality, and Profits in Molted Hy-Line
W-36 Hens. Journal of Applied Poultry Research 17, 432-439.

Harms RH, Russell GB and Sloan DR 2000. Performance of four strains of commercial layers
with major changes in dietary energy. Journal of Applied Poultry Research 9, 535-541.

Hartini S, Choct M, Hinch G, Kocher A and Nolan JV 2002. Effects of light intensity during
rearing and beak trimming and dietary fiber sources on mortality, egg production, and
performance of ISA brown laying hens. Journal of Applied Poultry Research 11, 104-
110.

91
Hetland H, Svihus B and Choct M 2005. Role of insoluble fiber on gizzard activity in layers.
Journal of Applied Poultry Research 14, 38-46.

Hsu JC, Lin CY and Chiou PWS 1998. Effects of ambient temperature and methionine
supplementation of a low protein diet on the performance of laying hens. Animal Feed
Science and Technology 74, 289-299.

Hu CH, Zuo AY, Wang DG, Pan HY, Zheng WB, Qian ZC and Zou XT 2011. Effects of
broccoli stems and leaves meal on production performance and egg quality of laying
hens. Animal Feed Science and Technology 170, 117-121.

Jalal MA, Scheideler SE and Marx D 2006. Effect of bird cage space and dietary
metabolizable energy level on production parameters in laying hens. Poultry Science 85,
306-311.

Jalal MA, Scheideler SE and Pierson EM 2007. Strain response of laying hens to varying
dietary energy levels with and without Avizyme supplementation. Journal of Applied
Poultry Research 16, 289-295.

Jiang W, Zhang L and Shan A 2013. The effect of vitamin E on laying performance and egg
quality in laying hens fed corn dried distillers grains with solubles. Poultry Science 92,
2956-2964.

Keshavarz K 1998. Further investigations on the effect of dietary manipulation of protein,


phosphorus, and calcium for reducing their daily requirement for laying hens. Poultry
Science 77, 1333-1346.

Khatun A, Ali MA and Dingle JG 1999. Comparison of the nutritive value for laying hens of
diets containing atolls (Azolla pinnata) based on formulation using digestible protein
and digestible amino acid versus total protein and total amino acid. Animal Feed
Science and Technology 81, 43-56.

Krimpen vMM, Veldkamp T, Binnendijk GP and de Veer R 2010. Effect of four processed
animal proteins in the diet on digestibility and performance in laying hens. Poultry
Science 89, 2608-2616.

92
Krimpen MMv, Kwakkel RP, Peet-Schwering CMCvd, Hartog LAd and Verstegen MWA
2009. Effects of nutrient dilution and nonstarch polysaccharide concentration in rearing
and laying diets on eating behavior and feather damage of rearing and laying hens.
Poultry Science 88, 759-773.

Panda A K, Rama Rao SV, Raju MVLN, Niranjan M and Reddy MR 2011. Effect of nutrient
density on production performance, egg quality and humoral immune response of brown
laying(Dahlem Red) hens in the tropics. Tropical Animal Health Production 44, 293–
299

Laudadio V and Tufarelli V 2011. Influence of substituting dietary soybean meal for
dehulled-micronized lupin (Lupinus albus cv. Multitalia) on early phase laying hens
production and egg quality. Livestock Science 140, 184-188.

Lazaro R, Garcia M, Aranibar MJ and Mateos GG 2003. Effect of enzyme addition to wheat-,
barley- and rye-based diets on nutrient digestibility and performance of laying hens.
British Poultry Science 44, 256-265.

Leeson S, Summers JD and Caston LJ 2001. Response of layers to low nutrient density diets.
Journal of Applied Poultry Research 10, 46-52.

Lei QB, Shi LX, Zhang KY, Ding XM, Bai SP and Liu YG 2011. Effect of reduced energy,
protein and entire substitution of inorganic phosphorus by phytase on performance and
bone mineralisation of laying hens. British Poultry Science 52, 202-213.

Mirzaie S, Zaghari M, Aminzadeh S, Shivazad M and Mateos GG 2012. Effects of wheat


inclusion and xylanase supplementation of the diet on productive performance, nutrient
retention, and endogenous intestinal enzyme activity of laying hens. Poultry Science 91,
413-425.

Moghaddam NH, Kazemi Fard M, Agah MJ, Hosseini SJ and Mirakzehi MT 2012. Effect of
different levels of methionine, protein and tallow on the productive performance and
egg quality of laying hens in the late-phase production. Revista Brasileira de Ciência
AvÃcola 14, 149-158.

93
Murugesan GR and Persia ME 2013. Validation of the effects of small differences in dietary
metabolizable energy and feed restriction in first-cycle laying hens. Poultry Science 92,
1238-1243.

Novak C, Yakout H and Scheideler S 2004. The combined effects of dietary lysine and total
sulfur amino acid level on egg production parameters and egg components in Dekalb
Delta laying hens. Poultry Science 83, 977-984.

Perez-Bonilla A, Novoa S, Garcia J, Mohiti-Asli M, Frikha M and Mateos GG 2012. Effects


of energy concentration of the diet on productive performance and egg quality of brown
egg-laying hens differing in initial body weight. Poultry Science 91, 3156-3166.

Roberts SA, Xin H, Kerr BJ, Russell JR and Bregendahl K 2007. Effects of dietary fiber and
reduced crude protein on nitrogen balance and egg production in laying hens. Poultry
Science 86, 1716-1725.

Safaa HM, Serrano MP, Valencia DG, Arbe X, Jimenez-Moreno E, Lazaro R and Mateos GG
2008. Effects of the levels of methionine, linoleic acid, and added fat in the diet on
productive performance and egg quality of brown laying hens in the late phase of
production. Poultry Science 87, 1595-1602.

Safaa HM, Serrano MP, Valencia DG, Frikha M, Jimenez-Moreno E and Mateos GG 2008.
Productive performance and egg quality of brown egg-laying hens in the late phase of
production as influenced by level and source of calcium in the diet. Poultry Science 87,
2043-2051.

Safaa HM, Jimenez-Moreno E, Valencia DG, Frikha M, Serrano MP and Mateos GG 2009.
Effect of main cereal of the diet and particle size of the cereal on productive
performance and egg quality of brown egg-laying hens in early phase of production.
Poultry Science 88, 608-614.

Skrivan M, Marounek M, Bubancova I and Podsednicek M 2010. Influence of limestone


particle size on performance and egg quality in laying hens aged 24-36 weeks and 56-68
weeks. Animal Feed Science and Technology 158, 110-114.

Swiatkiewicz S and Koreleski J 2009. Effect of crude glycerin level in the diet of laying hens
on egg performance and nutrient utilization. Poultry Science 88, 615-619.
94
Traineau M, Bouvarel I, Mulsant C, Roffidal L, Launay C and Lescoat P 2013. Effects on
performance of ground wheat with or without insoluble fiber or whole wheat in
sequential feeding for laying hens. Poultry Science 92, 2475-2486

Umar Faruk M, Bouvarel I, Meme N, Rideau N, Roffidal L, Tukur HM, Bastianelli D, Nys Y
and Lescoat P 2010. Sequential feeding using whole wheat and a separate protein-
mineral concentrate improved feed efficiency in laying hens. Poultry Science 89, 785-
796.

Valkonen E, Venalainen E, Rossow L and Valaja J 2008. Effects of dietary energy content on
the performance of laying hens in furnished and conventional cages. Poultry Science 87,
844-852.

Yan L, Meng QW, Ao X, Wang JP, Jang HD and Kim IH 2011. Evaluation of dietary wild-
ginseng adventitious root meal on egg production, egg quality, hematological profiles
and egg yolk fatty acid composition in laying hens. Livestock Science 140, 201-205.

95
96
Chapitre 3 : Effet du blé broyé avec ou sans
fibres insolubles et/ou blé entier sur les
performances des poules pondeuses en
alimentation séquentielle

Traineau M, Bouvarel I, Mulsant C, Roffidal L, Launay C and Lescoat P 2013. Effects on


performance of ground wheat with or without insoluble fiber or whole wheat in sequential
feeding for laying hens. Poultry Science 92, 2475-2486
97
98
Résumé Publication 1 :

Lors des travaux précédents, les poules recevaient en alimentation séquentielle (AS)
du blé entier le matin et un aliment complémentaire l’après midi riche en protéines et en
calcium. Cette distribution permettait de conserver un niveau de production équivalent à une
distribution classique tout en réduisant l’ingestion ce qui permettait d’améliorer l’indice de
consommation. Cependant, afin de ne pas être dépendant de l’utilisation d’une céréale unique,
il parait inévitable de pouvoir substituer le blé entier par une combinaison de céréales qui
permettrait d’obtenir les mêmes résultats et conserver l’effet bénéfique dû à la forme entière
du blé. Notre hypothèse était que l’inclusion de fibres insolubles ainsi que d’une faible
proportion de blé entier devrait pourvoir mimer le même effet positif sur le développement du
gésier qu’un régime uniquement composé de blé entier.

Huit traitements ont été offerts aux 432 Isa Brown qui étaient placées dans des cages
collectives de trois poules de 20 à 35 semaines d’âge. Le premier traitement correspondait à
une alimentation classique et continue. Les sept autres traitements étaient distribués de façon
séquentielle avec le même aliment complémentaire riche en protéines et en calcium apporté
dans la fraction de l’après midi. L’un des traitements en AS offrait du blé entier dans la
fraction du matin tandis que trois autres régimes étaient basés sur l’utilisation de blé broyé
(blé broyé, blé broyé + 5 % de fibres insolubles, blé broyé + 20% de blé entier) et encore trois
autres sur une base de maïs broyé (maïs broyé, maïs broyé + 5% de fibres insolubles, maïs
broyé + 20% de blé entier). La distribution séquentielle avec le blé entier a donné les
meilleurs résultats avec un indice de consommation meilleur que pour le régime continu mais
également meilleur que l’ensemble des autres régimes en AS. Les régimes séquentiels en base
maïs et en base blé donnent des résultats intermédiaires entre l’AS blé entier et l’alimentation
continue. En effet, la taille du gésier des poules avec ces régimes est comparable a la taille de
ceux des poules ayant ingéré du blé entier mais supérieure aux gésiers des poules en
alimentation continue. Les six régimes séquentiels ont également un indice de consommation
amélioré (-3.2%) par rapport au régime continu. Il parait donc possible de substituer le blé
entier par une combinaison de céréales en AS avec toutefois des performances légèrement
inférieures, bien que toujours supérieures à celles de l’alimentation classique.

99
100
PRODUCTION, MODELING, AND EDUCATION

Effects on performance of ground wheat with or without insoluble fiber


or whole wheat in sequential feeding for laying hens

M. Traineau,*† I. Bouvarel,‡ C. Mulsant,† L. Roffidal,† C. Launay,† and P. Lescoat*1

*Institut National de la Recherche Agronomique, Unité de Recherches Avicoles (UR83), F-37380 Nouzilly, France;
†Innovation en Nutrition et Zootechnie (INZO), Service R&D Pondeuses, Rue de l’Eglise, CS 90019,
02400 Chierry, France; and ‡Institut Technique de l’Aviculture (ITAVI), F-37380 Nouzilly, France

ABSTRACT Sequential feeding (SF) is an innovative wheat and 20% whole wheat, and ground wheat with
system for laying hens consisting of nutrients separat- 5% insoluble fiber) and 3 ground corn-based (ground
ing energy, protein, and calcium supplies to fulfill nutri- corn, ground corn and 20% whole wheat, and ground
ent requirements at the relevant time of day. In previ- corn with 5% insoluble fiber) provided in the morning.
ous studies, hens received whole wheat in the morning All sequential regimens received the same balancer diet
and a balancer diet (rich in protein and calcium) in rich in protein and calcium in the afternoon. Whole
the afternoon. To improve SF utilization, the aim was wheat SF gave the best results with an improved FCR
to substitute whole wheat in the morning by an alter- compared with continuous control and all other SF di-
native energy supply: ground wheat and ground corn, ets. Wheat- and corn-based diets showed intermediate
with or without a proportion of whole wheat and in- results between whole wheat SF and continuous feed-
soluble fiber. The goal was to obtain the advantages ing. Gizzard weight was higher and hens were lighter
observed in previous experiments with whole wheat than with conventional continuous feeding, leading to
[bigger gizzard, thinner hens, reduced feed conversion an average FCR improvement of 3.2% compared with
ratio (FCR)]. Four hundred thirty-two ISA Brown hens a continuous control. Thus, it is possible in SF diets to
were housed in collective cages from 20 to 35 wk of substitute, at least partially, whole wheat by ground
age divided into 8 different treatments: a continuous wheat or ground corn with added insoluble fiber or
control diet, a sequential diet with whole wheat in the some whole wheat, allowing more flexibility and eco-
morning, 3 wheat-based diets (ground wheat, ground nomic optimization.
Key words: sequential feeding, laying hen, feedstuff, performance
2013 Poultry Science 92:2475–2486
http://dx.doi.org/10.3382/ps.2012-02871

INTRODUCTION whole grain supply (sorghum, wheat, and corn; Yo,


1996; Noirot et al., 1998) either simultaneously in dif-
In commercial egg production, laying hens are fed ferent containers (choice feeding) or mixed in a single
with a complete diet (i.e., a mix of ground feedstuffs). feeder (loose-mix) or offered at different times of the
This distribution has the advantage of providing a uni- day (sequential feeding). Choice feeding and loose mix
form diet even though it may lack some flexibility re- gave different results depending on the trial and the
garding the instantaneous supply of nutrients to the measured performance indicators: decreased feed intake
bird, as it is well understood for Ca metabolism in con- with a decreased feed conversion ratio (FCR) in loose-
nection with eggshell deposition (Sauveur and Mongin, mix feeding when hens ate a mixture of whole wheat,
1974). Moreover, a uniform diet could be a disadvan- whole barley, and pellet protein concentrate (Blair et
tage regarding the inclusion of some alternative feed- al., 1973). However, the main disadvantage of loose-mix
stuffs and therefore might increase the diet cost (Bou- is that hens sort their food by eating the largest par-
varel et al., 2009). ticles, inducing broad individual variation in terms of
One way to improve efficiency in bird production intake and performance (Picard et al., 1997). Hens were
systems is to implement innovative feeding techniques. fed in a choice-feeding situation with pelleted protein
Different supply systems of feedstuffs have been tested: concentrate, whole yellow maize, and limestone. During
the laying period, hens offered the choice-fed diets laid
heavier eggs and had better food conversion ratios than
©2013 Poultry Science Association Inc.
Received November 2, 2012.
those fed the control diet (Olver and Malan, 2000).
Accepted May 19, 2013. No significant differences between treatments were ob-
1 Corresponding author: philippe.lescoat@agroparistech.fr
served for eggs laid per hen or feed intake (Olver and

2475
2476 TRAINEAU ET AL.

Malan, 2000). The main problem with choice feeding is whole wheat grain restricts the economic flexibility of
the protein/energy actually consumed by the bird. It this system. Consequently, it is necessary to find effi-
may be different compared with a complete feed and cient substitutions for whole wheat. Coarse grinding of
thus could affect the performance of birds (Noirot et corn or wheat may be sufficient to keep the beneficial
al., 1998). Additionally, in current cages, the major effect of whole seed. Another way to mimic the whole
drawback is the requirement of several feeders, decreas- grain effect could be obtained from additional dietary
ing the available space and compatibility with commer- fiber, which could increase gizzard weight (Deaton et
cial feeding systems. al., 1977; Longe, 1984; Hetland et al., 2005). Moreover,
Previous studies on sequential feeding reported an the incorporation of fiber in the diet appeared to play a
increased total intake when hens were fed whole ce- positive role in feather pecking in laying hens. Indeed,
real supply followed by a protein concentrate given ad Van Krimpen et al. (2009) showed that the dilution of
libitum, with the same egg performances (Blair et al., the diet with insoluble nonstarch polysaccharides de-
1973). Sequential feeding has been used in laying hens creased feather pecking and could explain the improved
with a combination of high energy-low protein diet giv- feather condition at 49 wk of age.
en in the morning and a high protein-mineral one in The goal of this study was, therefore, to determine
the afternoon (Leeson and Summers, 1978; Reichmann if whole wheat could be substituted by other feedstuffs
and Connor, 1979; Umar Faruk et al., 2010a). These in sequential feeding, while maintaining the benefits of
authors showed lower feed intakes, with the same egg this technique. The aim was to determine if a combina-
production, leading to an improved fed efficiency com- tion of ground corn or ground wheat with a small pro-
pared with a complete feed diet. However, there is no portion of whole wheat (20% of the morning fraction)
specific research on the feedstuffs and no explanation or the inclusion of insoluble fibers (5% of the morning
of the biological mechanisms that might explain results fraction) could have the same beneficial effect on laying
observed in sequential feeding. Recent studies on se- hens as whole wheat in the morning fraction.
quential feeding showed consistent results: improved
FCR, bigger gizzard, and lighter and thinner hens when
whole wheat was offered in the morning and high pro- MATERIALS AND METHODS
tein and high calcium balancer in the afternoon (Umar
Faruk et al., 2010a, 2011). Change of feed efficiency in Birds and Housing
sequential feeding with whole grain might be connected
to changes in dietary content and particle size of feed- Hendrix ISA Brown hens were delivered to the ex-
stuffs. Sequential feeding also led to greater develop- perimental unit at 15 wk of age. Hens were housed in
ment of the gizzard than occurred with a continuous wire-bottomed cages designed to accommodate 3 hens
diet (Umar Faruk et al., 2010a). Heavier gizzard im- per cage (50 cm width × 48 cm depth × 48 cm height)
proves grinding action, gastric reflex, and surface area equipped with nipple drinker and feeding trough (16.7
of contact between nutrients and digestive enzymes cm of feeder access per hen). Birds were given access to
(Gabriel et al., 2003) and thus might increase digestion a quantity of food fitted to their requirements through-
of food. Moreover, whole grain in the diet enhanced out the habituation period, with a gradual increase in
the development of digestive organs with a modifica- the amount of food given during these 3 wk. Moreover,
tion both of the proventriculus and gizzard and of the they received a whole wheat progressively from wk 16
lower small intestine tract (Nir et al., 1995; Amerah et to 18. The objective was to adapt the bird to whole
al., 2007). Change in gizzard weight might be due to wheat supply before they reached the point of lay be-
an increased frequency of contraction required to re- cause abrupt supply of whole wheat led to birds’ refus-
duce whole grain into finer particles (Hill and Strachan, als as observed by Dezat et al. (2009). Subsequently,
1975; Engberg et al., 2004). Whole wheat in sequen- they were given ad libitum access to food and water
tial feeding also led to an increase in pancreas weight throughout the experimental period. Photoperiod was
compared with a complete diet (Engberg et al., 2004; 10L:14D at wk 16 and reached 16L:8D at wk 18, re-
Gabriel et al., 2008; Umar Faruk et al., 2010a), and it maining at this level until the end of the experiment.
might be assumed that an heavier pancreas could be The dark period lasted from 2000 to 0400 h. The poul-
positively correlated with an increased digestive ability try shed temperature was scheduled to be 20 ± 0.5°C
of the chicken (Péron et al., 2006). but fluctuated between 19 and 27°C due to high en-
Indeed, use of whole wheat could have several ad- vironmental temperatures. The experimental period
vantages: local availability, no grinding processing ran from wk 19 through 35 with 432 birds randomly
cost, and a positive effect on gizzard weight (Nir et selected and divided into 8 groups of 18 cages with 3
al., 1990) even though it might be difficult to expand birds per cage.
this system at the commercial level. Sequential feeding
seems to be an interesting alternative system because Experimental Treatments
it allows control of type and amount of feed offered to
the hens, therefore optimizing nutrient intake at the Eight different diets were tested in the experiment.
relevant time of day. However, compulsory addition of Contents of the diets are given in Table 1. The control
SEQUENTIAL FEEDING IN LAYING HENS 2477
treatment contained 2,650 kcal of ME/kg and 16.9% and shape on the feeding system’s efficiency. Similarly,
CP and was given in 2 equal portions twice a day. For the inclusion of insoluble fibers in the diet is assumed
the 7 sequential combinations, birds received the high to enhance the development of the digestive tract and
energy fraction in the morning and beginning of the particularly the gizzard (Hetland et al., 2005). A small
afternoon (0830 to 1530 h) and the high protein and proportion of fiber was therefore included to 2 sequen-
mineral fraction in the late afternoon and early morn- tial high-energy diets, one with ground wheat and one
ing (1530 to 0830 h). The high protein and mineral with ground corn.
fraction was the same for the 7 sequential diets. These Hens with whole wheat, ground wheat, and ground
were formulated to reach a similar nutritional supply wheat + whole wheat diets received 2,628 kcal of ME/
to the control treatment, assuming a feed intake ratio kg and 17.2% CP; hens with ground wheat + insoluble
of 50/50 between high energy and high protein-mineral fibers received 2,623 kcal of ME/kg and 16.6% CP. Lay-
diet. The control diet was offered as a complete ground ers with ground corn + whole wheat received 2,654 kcal
diet with mostly the same feedstuffs (wheat, corn, soy- of ME/kg and 16.8% CP; layers with ground corn had
bean meal) as the sequential diets to mimic an average 2,662 kcal of ME/kg and 16.9% CP; and those with
supply. ground corn + insoluble fibers received 2,660 kcal of
Because the cereal choice was assumed to influence ME/kg and 16.9% CP. Diets were formulated according
the feeding behavior and therefore the birds’ perfor- to INRA tables (Sauvant et al., 2002).
mance, 3 high-energy diets were formulated with Each bird was offered 125 g of feed distributed twice
mainly wheat, 3 with mainly corn, and 1 only with daily in equal quantities. This quantity represented
whole wheat. Whole wheat was considered as the posi- 105% of the theoretical requirement of 119 g/bird per
tive control due to previous results (Umar Faruk et d (ISA-Hendrix, 2011) and could be considered as ad
al., 2010a). Diets with ground wheat and ground corn libitum supply. In sequential feeding, feed left over from
were tested either with or without the addition of 20% the previous distribution was always removed before
whole wheat to determine the effect of the wheat size the next distribution.

Table 1. Composition of experimental diets given to laying hens between 19 and 35 wk of age with 125 g of control or a combina-
tion of 62.5 g of any wheat- or corn-based diets between 0830 and 1530 h and 62.5 g of balancer between 1530 and 0830 h (no light
between 2000 and 0400 h)1
Wheat-based diet Corn-based diet

Ground Ground Ground


wheat wheat corn Ground
Balancer Sequential + whole wheat Ground + fiber + whole wheat Ground corn + fiber
Item Control diet2 control (SC) (GW+WW) wheat (GW) (GW+F) (GC+WW) corn (GC) (GC+F)
Wheat (%) 33 17 100 100 100 92.7 20
Corn (%) 34.91 16 72.07 89.76 88
Arbocel RC3 (%) 5 5
Soybean oil (%) 0.3 2.6 2.3 0.3
Sunflower meal (%) 6.8 8.5
Rapeseed meal (%) 1.6
Soybean meal (%) 19 43
Corn gluten 60 (%) 2.5 1.1 1.1 1.7 5
Calcium carbonate (%) 7.68 15.04
Dicalcium phosphate 1.31 2.65
(%)
4
Premix (%) 1.004 2.22
dl-Methionine (%) 0.175 0.35
l-Lysine (%) 0.085 0.04
AMEn (kcal/kg) 2,651 2,275 2,980 2,980 2,980 2,972 3,034 3,048 3,046
CP (%) 16.9 23.7 10.7 10.7 10.7 10.0 10.0 10.0 10.1
DM (%) 88.0 89.9 87.4 87.4 87.4 88.0 86.7 86.5 86.8
Ash (%) 11.7 22.1 1.6 1.6 1.6 1.6 1.6 1.6 1.4
Calcium (%) 3.60 7.10 0.040 0.040 0.040 0.067 0.034 0.033 0.043
Available P (%) 0.32 0.55 0.150 0.150 0.150 0.139 0.071 0.051 0.044
Lys (%) 0.719 1.188 0.250 0.250 0.250 0.232 0.253 0.256 0.242
Met (%) 0.419 0.657 0.150 0.150 0.150 0.139 0.185 0.197 0.198
Thr (%) 0.536 0.796 0.270 0.270 0.270 0.250 0.305 0.32 0.323
Trp (%) 0.161 0.251 0.120 0.120 0.120 0.111 0.110 0.112 0.181
1Nutritional values were calculated based on Sauvant et al. (2002).
2Balancer diet: at 27 wk of age, 2% oil was added on top of the balancer diet to boost the birds’ weight until wk 35. This increased AMEn from
2,275 to 2,412 and slightly diluted the other parameters.
3Arbocel RC: pine lignocellulosic insoluble fiber, made by the company J. Rettenmaier and Söhne (Rosenberg, Germany).
4Premix supplied per kilogram of diet: vitamin A, 750,000 IU; vitamin D , 300,000 IU; vitamin E (tocopherol acetate 3a700), 1,000 IU; vitamin
3
K3, 100 mg; zinc, 10,000 mg; manganese, 10,000 mg; copper, 1,000 mg; iron, 4,000 mg; iodine, 150.0 mg; selenium, 30.0 mg; canthaxanthin, 1,950.0
mg; β-apocarotenoid acid, 2,000.0 mg; lutein, 3,560.0 mg; zeaxanthin, 200.0 mg; cryptoxanthin, 80.0 mg; citric acid, 400.00 mg; orthophosphoric acid,
660.00 mg; ethoxyquin, 400 mg; and propyl gallate (E310), 400 mg.
2478 TRAINEAU ET AL.

At 27 wk of age, sequentially fed hens showed a too few hens were laying at 20 wk of age, the first
meaningful weight loss compared with wk 24, whereas period only included 3 wk (21 to 23). To evaluate the
no changes were observed for the controls. This weight differences between diets, several contrasts were tested.
loss was mainly due to a very high temperature for sev- The statistical analysis began by comparing the con-
eral days without decline during periods of night. We trol diet with the sequential whole wheat diet. Second,
added 2% oil on top of the balancer diet to boost the the control diet and the sequential whole wheat were
birds’ weight until wk 35. tested versus the other sequential diets grouped under
corn-based ones (ground corn, ground corn + insoluble
Measurements fibers, ground corn + whole wheat) and wheat-based
ones (ground wheat, ground wheat + insoluble fibers,
For each cage, feed intake was recorded every week. ground wheat + whole wheat). Finally, sequential diets
In sequential feeding, intakes of high energy and high were compared within the same base (corn or wheat).
protein-mineral diets were measured independently. Averaged values from cages were analyzed using Stat-
Egg production was measured by recording the num- View (version 5, SAS Institute Inc., Cary, NC). Data
ber of eggs produced daily and describing their charac- were analyzed based on the 4 following periods: 21 to
teristics (normal, broken, soft, and dirty). Egg weight 23, 24 to 27, 28 to 31, and 32 to 35 wk of age.
was recorded twice weekly. Every 4 wk, egg components Feed and fractional (high energy and high protein –
were measured (weight of yolk, albumen, and eggshell) mineral diets) intakes, laying performances (egg weight,
on all eggs produced on a given day. Eggshells were egg production, egg mass), FCR, BW, and body con-
washed and dried for 12 h in an oven at 70°C, and then tent were tested with a 1-way repeated measures
weighed. ANOVA including the above proposed comparisons. A
Body weight was recorded individually at wk 16, 19, 1-way ANOVA was used to test treatment effect on egg
23, 27, 31, and 35. Body composition was predicted with component weight and digestive organ weight. Tested
a noninvasive method based on bioelectrical impedance factors were considered highly significant if P < 0.01,
analysis at 19, 27, and 35 wk of age. The device uses significant if P < 0.05, could be discussed as a trend if
2 electrodes placed on the bird’s legs and distributes a P < 0.1, and not significant if P > 0.1.
very low intensity current. It is assumed that the elec-
trical impedance of the biological organism is indicative
of the highly conductive fat-free component (Berg and RESULTS
Marchello, 1994). The electrical impedance of a tissue
depends on its fluid and electrolyte content (Rutter et Comparison Between the Control Diet
al., 1998) and therefore on the electrical properties of and the Sequential Whole Wheat
different tissues in connection with their relative pro- (Tables 2 and 3)
portions of muscle, lipid, water, and ash. These propor-
tions could consequently be predicted. This technique The average feed intake tended to be higher for the
has been developed by the company INZO (Chierry control than it was for the whole wheat diet (P < 0.1).
Cedex, France), which calibrated the instrument of im- However, no difference was observed for the laying per-
pedance through chemical analysis of birds to obtain a formance (egg weight, egg production, and egg mass),
consistent prediction of body composition of chickens. whereas FCR was significantly lower (P < 0.05) for
At the end of the trial, 10 birds per treatment, ran- sequential feeding with whole wheat than it was for the
domly selected, were first weighed before being injected control.
with Na pentobarbital solution (1 mL/kg). The abdom- Body weight increased with age, but sequentially
inal cavity was dissected and the weights of pancreas, fed birds were lighter than control ones (−108 g, P <
liver, and spleen were recorded. The gizzard was placed 0.01). According to this comparison between control
into an iced container (−4°C) for 24 h to facilitate the and sequential whole wheat and in connection with the
removal of the surrounding fat before being emptied results of Umar Faruk et al. (2010a, 2011), sequential
and weighed. The small intestine was divided into 3 whole wheat could be considered as a positive control
parts: duodenum (from gizzard to pancreo-biliary in sequential feeding.
duct), jejunum (from pancreo-biliary duct to Meckel’s
diverticulum), and ileum (from Meckel’s diverticulum Feed Intake (Table 2)
to ileo-cecal junction). Each segment was emptied of
its contents and fat surrounding the tube was removed The comparison control versus wheat-based or corn-
before weighing. Hepatic lipid (%/liver) was measured based sequential diets showed a total feed intake that
according to the Folch procedure (Folch et al., 1957). was always higher for the control. No differences be-
tween sequential whole wheat and wheat- or corn-based
Statistical Analysis diets were observed for total feed intake and high en-
ergy intake. High protein-mineral diet intakes tended to
Data collected during the experiment were analyzed be higher (P < 0.1) with wheat-based diets compared
based on 4 different periods that lasted 4 wk. Because with sequential whole wheat, whereas there were no
Table 2. Effects on feed consumption and feed conversion ratio of laying hens from 21 to 35 wk old offered either 125 g of control or a combination of 62.5 g of any wheat- or corn-
based diets between 0830 and 1530 h and 62.5 g of balancer between 1530 and 0830 h (no light between 2000 and 0400 h)1
P-value2 Wheat-based diet P-value Corn-based diet P-value

SC vs. GW vs. Control SC vs. GC vs.


Control Control vs. wheat GW+F vs. vs. corn corn GC+F vs.
Item Control SC vs. SC GW GW+F GW+WW wheat diets diets GW+WW GC GC+F GC+WW SEM diets diets GC+WW

Total feed intake


(g/hen per d)
wk 21 to 23 107.0 100.3 <0.1 103.9 102.8 101.8 <0.05 NS NS 104.5 101.9 102.8 0.46 <0.05 NS <0.05
wk 24 to 27 109.4 107.1 107.9 108.0 107.1 110.7 106.9 108.4 0.49
wk 28 to 31 110.9 109.3 108.5 108.0 107.9 109.5 108.0 109.2 0.34
wk 32 to 35 113.8 111.8 112.3 109.9 109.5 114.0 110.2 111.6 0.42
High protein mineral intake
(g/hen per d)
wk 21 to 23 — 52.0 55.1 55.9 53.7 — <0.1 NS 54.5 56.1 53.3 0.4 — NS <0.01
wk 24 to 27 — 59.3 60.6 60.4 60.8 59.6 60.7 59.1 0.25
wk 28 to 31 — 60.7 60.9 61.3 61.1 59.8 61.4 59.8 0.15
wk 32 to 35 — 61.5 61.2 61.8 61.8 60.8 61.7 60.8 0.15
High energy intake
(g/hen per d)
wk 21 to 23 — 48.2 48.9 46.9 48.1 — NS NS 50.0 45.9 49.5 0.38 — NS <0.01
wk 24 to 27 — 47.8 47.3 47.5 46.3 51.1 46.2 49.2 0.42
wk 28 to 31 — 48.6 47.6 46.6 46.8 49.7 46.6 49.3 0.36
wk 32 to 35 — 50.2 51.1 48.2 47.8 53.2 48.6 50.8 0.46
Feed conversion ratio
SEQUENTIAL FEEDING IN LAYING HENS

wk 21 to 23 2.05 1.95 <0.05 2.02 1.99 2.03 <0.1 <0.05 NS 2.11 2.11 2.00 0.017 <0.1 <0.05 NS
wk 24 to 27 2.00 1.85 1.88 1.95 1.89 1.90 1.88 1.88 0.011
wk 28 to 31 1.94 1.80 1.88 1.85 1.84 1.88 1.81 1.80 0.013
wk 32 to 35 1.97 1.87 1.98 1.88 1.89 1.95 1.90 1.90 0.015
1SC: sequential control; GW+WW: ground wheat + whole wheat; GW: ground wheat; GC+WW: ground corn + whole wheat; GW+F: ground wheat + fiber; GC: ground corn; GC+F: ground corn
+ fiber.
2Values were tested with a 1-way repeated measures ANOVA for the 4 periods: 21 to 23, 24 to 27, 28 to 31, and 32 to 35 wk of age. The P-values represent the whole period effect from wk 21 to 35.
2479
2480

Table 3. Effects on egg production, egg weight, and BW on laying hens from 19 to 35 wk old offered either 125 g of control or a combination of 62.5 g of any wheat- or corn-based
diet between 0830 and 1530 h and 62.5 g of balancer between 1530 and 0830 h (no light between 2000 and 0400 h)1
P-value2 Wheat-based diet P-value Corn-based diet P-value

GW vs. Control SC vs. GC vs.


Control Control vs. SC vs. GW+F vs. vs. Corn Corn GC+F vs.
Item Control SC vs. SC GW GW+F GW+WW wheat diets wheat diets GW+WW GC GC+F GC+WW SEM diets diets GC+WW
Egg production (%)
wk 21 to 23 94.9 92.2 NS 94.2 95.1 91.7 NS NS NS 91.5 91.4 94.8 0.66 NS NS NS
wk 24 to 27 93.6 95.7 96.6 95.8 95.6 97.2 96.5 97.2 0.5
wk 28 to 31 93.9 95.9 94.4 95.4 94.8 94.8 96.3 96.6 0.52
wk 32 to 35 93.8 95.3 92.9 95.5 94.0 94.3 94.8 94.2 0.58
Egg weight (g)
wk 21 to 23 55.8 56.0 <0.1 55.0 54.6 55.0 NS <0.01 NS 54.8 53.7 54.4 0.24 NS <0.05 NS
wk 24 to 27 58.9 60.4 59.5 57.9 59.9 60.1 59.0 59.5 0.18
wk 28 to 31 61.7 63.5 61.7 61.3 62.3 61.9 62.1 62.7 0.2
wk 32 to 35 62.4 63.1 61.9 61.2 61.6 62.3 61.5 62.7 0.2
Egg mass (g/hen per d)
wk 21 to 23 52.9 51.7 NS 51.8 52.0 50.5 NS <0.05 NS 50.2 49.0 51.5 0.43 NS <0.1 NS
wk 24 to 27 55.1 57.8 57.4 55.5 57.2 58.4 56.9 57.8 0.32
TRAINEAU ET AL.

wk 28 to 31 58.0 60.9 58.2 58.5 58.9 58.7 59.8 60.6 0.36


wk 32 to 35 58.6 60.2 57.5 58.4 58.0 58.7 58.2 59.0 0.39
BW (g)
wk 19 1,573 1,535 <0.01 1,560 1,528 1,531 <0.01 NS NS 1,568 1,521 1,531 4.7 <0.01 NS <0.01
wk 23 1,737 1,693 1,723 1,685 1,702 1,733 1,679 1,701 5.8
wk 27 1,735 1,665 1,688 1,664 1,675 1,713 1,643 1,682 7.4
wk 31 1,813 1,712 1,743 1,717 1,732 1,762 1,697 1,723 6.5
wk 35 1,883 1,764 1,790 1,752 1,749 1,836 1,727 1,763 7.6
1SC: sequential control; GW+WW: ground wheat + whole wheat; GW: ground wheat; GW+F: ground wheat + fiber; GC+WW: ground corn + whole wheat; GC: ground corn; GC+F: ground corn
+ fiber.
2Values were tested with a 1-way repeated measures ANOVA for the 4 periods: 21 to 23, 24 to 27, 28 to 31, and 32 to 35 wk of age except for BW values were tested with a 1-way repeated measures
ANOVA for wk 19, 23, 27, 31, and 35. P-values represent the whole period effect from wk 21 to 35.
SEQUENTIAL FEEDING IN LAYING HENS 2481
Table 4. Effects of different diets on weight of digestive organs (% BW) and abdominal fat at 35 wk
of age for laying hens fed between 19 and 35 wk of age with either 125 g of control or a combination
of 62.5 g of any wheat- or corn-based diet between 0830 and 1530 h and 62.5 g of balancer between
1530 and 0830 h (no light between 2000 and 0400 h)1
Small Abdominal
% of BW Gizzard intestine Pancreas Liver fat

Control 1.1 2.0 0.17 2.6 4.0


SC 1.5 2.4 0.22 2.8 3.2
GW 1.4 2.3 0.19 2.9 2.7
GW+F 1.4 2.4 0.19 2.9 2.8
GW+WW 1.5 2.4 0.20 3.0 2.4
GC 1.4 2.2 0.20 2.7 3.6
GC+F 1.6 2.1 0.19 2.9 2.3
GC+WW 1.5 2.2 0.21 3.0 3.1
SEM 0.020 0.030 0.004 0.040 0.13
Control vs. SC2 P < 0.01 P < 0.01 P < 0.05 NS NS
Control vs. wheat diets P < 0.01 P < 0.01 P < 0.01 P < 0.05 P < 0.01
GW vs. GW+F vs. GW+WW NS NS NS NS NS
SC vs. wheat diets NS NS NS NS NS
Control vs. corn diets P < 0.01 P < 0.01 NS P < 0.05 P < 0.05
SC vs. corn diets NS P < 0.05 NS NS NS
GC vs. GC+F vs. GC+WW NS NS NS NS NS
1SC: sequential control; GW+WW: ground wheat + whole wheat; GW: ground wheat; GW+F: ground wheat
+ fiber; GC+WW: ground corn + whole wheat; GC: ground corn; GC+F: ground corn + fiber.
2Values were tested with a 1-way ANOVA at 35 wk of age.

differences with corn-based diets. There were no differ- Body Composition (Figure 1)
ences between the 3 wheat-based diets regarding high
energy and high protein-mineral feed intakes. Converse- Protein content was similar at 19 wk of age for se-
ly, high protein-mineral fraction intake was higher with quential whole wheat and the control. It evolved dif-
ground corn + insoluble fibers compared with ground ferently over time with higher protein levels for hens
corn and ground corn + whole wheat (P < 0.01). Con- with sequential whole wheat compared with the con-
trary to high protein-mineral fraction, birds ate less trol. Conversely, hens were fatter with control diet than
the high energy fraction with ground corn + insoluble with the sequential whole wheat. No differences in pro-
fibers than with ground corn and ground corn + whole tein and fat dynamics between sequential whole wheat
wheat (P < 0.01). and wheat- or corn-based diets were observed. Hens
with ground corn diets were fatter than the other 2
corn-based diets and had a lower protein level. No dif-
Egg Production (Table 4) ferences were observed between wheat-based diets.
Egg production and egg weight increased with age.
Similar egg productions were observed for the 8 diets. FCR (Table 2)
There were no differences between control and corn-
based or wheat-based diets for egg weight and egg The FCR was lower for sequential whole wheat than
mass. Egg weights were significantly higher during the control (P < 0.05). Sequential whole wheat had lower
experiment for sequential whole wheat compared with FCR than wheat- (P < 0.05) or corn-based diets (P <
wheat-based and corn-based diets. A similar trend was 0.05). The FCR of control tended to be higher than
observed for the egg mass. Similar egg weight and egg wheat- or corn-based diets (P < 0.1). No differences
mass were observed among the 3 wheat-based and the in FCR values were observed within corn- and wheat-
3 corn-based diets. based diets.

BW (Table 3) Digestive Organs (Table 4)


Body weight increased with age and hens fed with At 35 wk of age, control feeding resulted in signifi-
the control diet were always heavier than sequential- cantly lighter gizzard and small intestine compared
ly fed ones. No differences between whole wheat and with sequential diets. Hens with the control diet had
wheat- or corn-based diets were found. Similar BW a lighter pancreas than sequential whole wheat. Hens
were observed among the 3 wheat-based diets. The with control diets had a lighter liver than wheat- or
hens with corn-based diets had hierarchical weights corn-based diets and they had higher levels of abdom-
throughout the experiment: hens eating ground corn inal fat. Gizzard, pancreas, liver, and abdominal fat
+ insoluble fibers were the lightest, hens with ground were similar among sequential treatment, except for the
corn + whole wheat had an average weight, whereas small intestine which was heavier with sequential whole
hens with ground corn were the heaviest. wheat than corn-based diets. Livers’ lipid contents
2482 TRAINEAU ET AL.

Figure 1. Effects of different diets on estimated BW composition [protein (%) and fat (%)] at 19, 27, and 35 wk old for laying hens fed between
19 and 35 wk of age with either 125 g of control or a combination of 62.5 g of any wheat- or corn-based diet between 0830 and 1530 h and 62.5 g
of balancer between 1530 and 0830 h (no light between 2000 and 0400 h). SC: sequential control; GW+WW: ground wheat + whole wheat; GW:
ground wheat; GW+F: ground wheat + fiber; GC+WW: ground corn + whole wheat; GC: ground corn; GC+F: ground corn + fiber. Values were
tested with a 1-way repeated measures ANOVA for wk 19, 27, and 35.

were performed on half of the diets (control, whole Egg Contents


wheat, ground wheat + fiber, and ground corn + whole
wheat). Results showed that the livers of control hens Very few differences were observed regarding the egg
were fatter than those of hens with sequential feeding contents. Egg yolks were similar among all treatments
diets (P < 0.01). during the experiment. The only difference was for the
SEQUENTIAL FEEDING IN LAYING HENS 2483
Table 5. Effects of different diets on ME (kcal/g per hen/d) and protein (g/hen per d) intake for laying hens from 21 to 35 wk of age fed between 19 and 35 wk of age with either 125

eggshells, which were always heavier for corn-based di-


g of control or a combination of 62.5 g of any wheat- or corn-based diet between 0830 and 1530 h and 62.5 g of balancer between 1530 and 0830 h (no light between 2000 and 0400 h)1

1SC: sequential control; GW+WW: ground wheat + whole wheat; GW: ground wheat; GW+F: ground wheat + fiber; GC+WW: ground corn + whole wheat; GC: ground corn; GC+F: ground corn
GC+F vs.
GC GC+F GC+WW SEM corn diets corn diets GC+WW
GC vs.

<0.01
<0.01
ets compared with control and sequential whole wheat.

NS
NS
Energy and Protein Intake (Table 5)

<0.01
<0.05
SC vs.
P-value

NS

NS
Energy and protein intakes were computed for 2 dif-
ferent periods: wk 21 to 27 (before supplementation of
oil) and wk 28 to 35 (after oil supplementation). Hens
Control vs.

<0.05
<0.01

offered the control diet ate more energy than hens in


NS

NS

sequential whole wheat with no difference for protein


intake before oil supplementation. After oil supple-
mentation, there was no difference for energy intake,
0.7
0.5
1.2
1.1

but hens with sequential whole wheat ate more pro-


tein than hens receiving the control diet. Layers with
18.3
19.3

the control diet consumed more energy than hens with


278
297
Corn-based diet

wheat and corn-based from wk 21 to 27 but had similar


values from wk 28 to 35. The opposite was observed
18.5
19.4

for protein intakes. Similar levels of energy and protein


273
293

intakes were observed among sequential whole wheat


18.6
19.4

and wheat-based diets. Hens with corn-based diets con-


284
302

sumed more energy and the same quantity of proteins


than sequential whole wheat from wk 21 to 27 with an
GW+F vs.
wheat diets GW+WW
GW vs.

<0.01

inversion from wk 28 to 35.


NS

NS
NS

Hens offered ground wheat diets ingested more energy


than the 2 other wheat-based diets before additional oil
supply, with no difference afterward. Hens with corn-
SC vs.
P-value

based diets had ranked energy intake throughout the


NS
NS
NS
NS

experiment: ground corn + insoluble fibers < ground


corn + whole wheat < ground corn. No differences in
wheat diets

protein intake within wheat- and corn-based diets were


Control vs.

<0.01
<0.01

observed.
NS

NS

2Values were tested with a 1-way ANOVA for the 2 periods 21 to 27 and 27 to 35 wk of age.

DISCUSSION
GW GW+F GW+WW

The sequential whole wheat diet gave the best results


18.6
19.6
Wheat-based diet

289
271

compared with the control diet in our study, consistent


with the findings of Umar Faruk et al. (2010a, 2011).
With the sequential whole wheat diet, hens tended to
18.5
19.3
289
273

consume less food than those offered a control com-


plete diet with the same performance and a lower BW
18.8
19.7

(−4.3%), and therefore their maintenance requirements


294
285

were lower. As an example, according to Sakomura


(2004), when BW requirement for hens with control
P-value2

Control

<0.01
<0.01
vs. SC

NS

NS

diet is 331 kcal of ME/kg between 20 and 35 wk of age,


a decrease of 100 g of BW leads to a lower requirement
of 7 kcal/d. Moreover, hens in sequential feeding had
18.3
19.8
SC

295
270

less fat than hens with a continuous diet. Our results


agreed with Lee and Ohh (2002) with the same effect
Control

18.3
19.0
298
287

of sequential feeding on feed intake and performance of


laying hens. However, in the morning the authors gave
a diet rich in energy and protein and in the afternoon
ME intake (kcal/g per hen/d)

Protein intake (g/hen per d)

a diet with a low level of energy and protein but with


calcium. A second explanation for a lower FCR may be
an improvement in the utilization of nutrients thanks
to a heavier gizzard and pancreas. These might be ex-
wk 28 to 35

wk 21 to 27
wk 28 to 35
wk 21 to 27

plained both by the fractional distribution of the diet


as observed for slow growing chickens (Chagneau et
+ fiber.

al., 2009) and large particle size, the best being whole
Item

grains that stimulate the digestive capacity of hens (Nir


2484 TRAINEAU ET AL.

et al., 1990). Inclusion of large particles in the diet induced an increase of feed intake (McNaughton, 1978).
of laying hens increased the retention time in the giz- However, it could also have negative effect on feed in-
zard (half-emptying time in the gizzard is 6.1 h with take as for Noy and Sklan (2002) where 10% of added
large particles and 3.1 h for small particles (Vergara et cellulose induced a reduction of 5 g of feed intake for
al., 1989), which should enhance the digestion of nutri- chicken. However, they demonstrated that inclusion of
ents as shown for protein by Ouhida et al. (2000) and insoluble fiber sources between 2 and 3% improved the
Rougière (2010). Moreover, relations have been com- growth performance of broilers and the development
puted between the weight of the gizzard and digestive of organs, enzyme production, and nutrient digestibil-
efficiency: a difference of 2.4 mg of gizzard/g of BW in- ity in poultry. However, in sequential feeding, the high
duced a 9% increased AMEn value (Peron et al., 2006). energy fraction contained 5% fiber, which may explain
As the retention time in the gizzard had been shown the negative effect on feed intake through a bulk effect
as linked to the relative weight of this organ (Verdal et of fibers in crop and gizzard. Van Krimpen et al. (2011)
al., 2011), the positive relationship between the gizzard showed that mean retention time in the foregut was
weight and digestive efficiency could be accepted. prolonged when daily insoluble nonstarch polysaccha-
To assess sequential feeding with diverse high en- ride intake increased. This higher mean retention time
ergy/low protein and Ca fractions, they have been in the foregut seemed to indicate a higher level of sa-
compared both with the sequential whole wheat and tiety. Diets with fiber had the same effect on digestive
the control diet. Comparing wheat- or corn-based diets organs as other sequential diets, with a heavier gizzard
with control and sequential whole wheat showed inter- than the control diet.
mediate results. Hens consumed less food in sequen- Questions were raised regarding the sequential feed-
tial diets than continuous feeding with similar laying ing system. Whatever the diet content in sequential
performance. Hens in sequential diets were also lighter feeding, the high energy fractions in the morning were
than the control, leading to a trend toward a better always consumed less than the high protein mineral
FCR. On the other hand, when wheat- or corn-based fractions in the afternoon. This underconsumption of
diets were compared with sequential whole wheat, egg high energy fractions provided a better FCR but could
weights were significantly heavier during the experi- weaken the overall system by a long-term lack of nu-
ment for sequential whole wheat driving to an improved trients. Decrease in consumption could be explained by
FCR with sequential whole wheat compared with the 2 main hypotheses: as dietary energy level increases,
other sequential diets. It might, therefore, be possible chickens satisfy their energy needs through feed intake
to substitute whole wheat by ground wheat, ground reduction (Plavnik et al., 1997; Nahashon et al., 2005)
corn, or both in sequential feeding but with attenuated as observed in our study. The afternoon fraction with
performance. When hens were fed with a classical com- a low energy density increased hens’ consumption due
plete diet, Lázaro et al. (2003) and Safaa et al. (2009) to their simultaneous needs for energy and Ca. The
found that the substitution of corn by wheat did not second hypothesis deals with amino-acid intake: a lack
affect laying performance and feed intake. This is con- might reduce the morning intake while favoring it in
sistent with Umar Faruk et al. (2010b) in hot semi-arid the afternoon. Indeed, chickens react quickly to a defi-
conditions, who substituted corn by millet in sequential ciency in essential amino acids in the diet. As an exam-
feeding for laying hens, obtaining similar results, and ple, at 8 d of age, chickens fed a diet deficient in lysine,
Bouvarel et al. (2009), who used different raw materi- methionine, and tryptophan (45%) reduced their feed
als such as fava bean and rapeseed meal in sequential intake by 60% (Picard et al., 1993).
feeding for chickens with the same performances as the This work showed the possibility of providing more
control diet. However, the use of ground corn alone af- flexibility in a sequential feeding system by substitution
fected the body composition of hens with fatter animals of whole wheat. Indeed, it is possible to use corn or
compared with hens with corn combined with whole wheat coarsely ground, mixed or not with whole grains
wheat or insoluble fiber. and insoluble fiber cellulose. These combinations gave
From this study, limited inclusion of 20% whole intermediate results between control and continuous se-
wheat or supply of 5% cellulose fiber in high energy/ quential feeding with whole wheat. In addition, body
low protein and Ca fraction did not produce similar composition predictions allowed demonstration of the
performances as 100% whole wheat. This supported the existence of different metabolic pathways used by hens
hypothesis of a required minimal level of whole grain in fed sequentially compared with continuously fed ones.
sequential feeding higher than 20% to be efficient. How- However, this study highlighted the fragility of laying
ever, no published study is available on laying hens to hens in sequential feeding, stressing a difficult adapta-
the authors’ knowledge. Regarding insoluble fiber, its tion to heat and a constant low energy fraction intake
negative effect on feed intake of high energy/low pro- in the morning. It would be interesting to successfully
tein and Ca fraction was stronger than expected. Usu- control the intake of hens, thereby driving their weight
ally dietary fiber had been considered as diluents of the gain to increase the robustness of birds against severe
diet (Mateos et al., 2012) and additional fiber supply environmental situations.
SEQUENTIAL FEEDING IN LAYING HENS 2485
ACKNOWLEDGMENTS Longe, O. G. 1984. Effect of increasing the fibre content of a layer
diet. Br. Poult. Sci. 25:187–193.
The authors are grateful to Michel Couty and Mateos, G. G., E. Jimenez-Moreno, M. P. Serrano, and R. P. Laza-
ro. 2012. Poultry response to high levels of dietary fiber sources
Philippe Didier of INRA, UR83 Recherches Avicoles, varying in physical and chemical characteristics. J. Appl. Poult.
Nouzilly, France, for their technical assistance as well as Res. 21:156–174.
their full commitment in the data collection and for the McNaughton, J. L. 1978. Effect of dietary fiber on egg yolk, liver,
care they provided to birds. We thank our experimental and plasma cholesterol concentration of the laying hen. J. Nutr.
108:1842–1848.
unit (Experimental Unit Pôle Expérimental Avicole de Nahashon, S. N., N. Adefope, A. Amenyenu, and D. Wright. 2005.
Tours) for its help in the setup of the experiment. Effect of dietary metabolisable energy and crude protein concen-
trations on growth performance and carcass characteristics of
French guinea fowl broilers. Poult. Sci. J. 84:337–344.
REFERENCES Nir, I., R. Hillel, I. Ptichi, and G. Shefet. 1995. Effect of particle-size
on performance. 3. Grinding pelleting interactions. Poult. Sci.
Amerah, A. M., V. Ravindran, R. G. Lentle, and D. G. Thomas. 74:771–783.
2007. Feed particle size: Implications on the digestion and perfor- Nir, I., J. P. Melcion, and M. Picard. 1990. Effect of particle-size of
mance of poultry. World’s Poult. Sci. J. 63:439–455. sorghum grains on feed-intake and performance of young broilers.
Berg, E. P., and M. J. Marchello. 1994. Bioelectrical-impedance Poult. Sci. 69:2177–2184.
analysis for the prediction of fat-free mass in lambs and lamb Noirot, V., I. Bouvarel, B. Barrier-Guillot, J. Castaing, J. L. Zwick,
carcasses. J. Anim. Sci. 72:322–329. and M. Picard. 1998. Whole cereal grains in broiler nutrition:
Blair, R., W. A. Dewar, and J. N. Downie. 1973. Egg production The come back? Prod. Anim. 11:349–357.
responses of hens given a complete mash or unground grain to- Noy, Y., and D. Sklan. 2002. Nutrient use in chicks during the first
gether with concentrate pellets. Br. Poult. Sci. 14:373–377. week posthatch. Br. Poult. Sci. 81:391–399.
Bouvarel, I., A. M. Chagneau, P. Lescoat, M. Lessire, M. Vilarino, Olver, M. D., and D. D. Malan. 2000. The effect of choice-feeding
K. Crepon, G. Etave, C. Margetyal, and H. Juin. 2009. Use of from 7 weeks of age on the production characteristics of laying
new feed ingredients in sequential feeding. Page 17 in Proc. of the hens. S. Afr. J. Anim. Sci. 30:110–114.
8th Avian French Research Days. St Malo, France 25–26 March, Ouhida, I., J. Perez, J. Piedrafita, and J. Gasa. 2000. The effects of
2009. World Poult. Sci. Assoc. (WPSA). sepiolite in broiler chicken diets of high, medium and low viscos-
Chagneau, A. M., S. Lequelle, P. Lescoat, M. Quentin, J. M. Guil- ity. Productive performance and nutritive value. Anim. Feed Sci.
laumin and I. Bouvarel. 2009. Les poulets à croissance lente valo- Technol. 85:183–194.
risent bien une distribution alimentaire par repas. The 8th Avian Péron, A., J. Gomez, S. Mignon-Grasteau, N. Sellier, J. Besnard, M.
French Research Days, St Malo, France, 25–26 March 2009. Derouet, H. Juin, and B. Carre. 2006. Effects of wheat quality on
Deaton, J. W., L. F. Kubena, F. N. Reece, and B. D. Lott. 1977. digestion differ between the D+ and D− chicken lines selected for
Effect of dietary fibre on the performance of laying hens. Br. divergent digestion capacity. Poult. Sci. 85:462–469.
Poult. Sci. 18:711–714. Picard, M., J. P. Melcion, C. Bouchot, and J. M. Faure. 1997. Peck-
Dezat, E., M. Umar-Faruk, P. Lescoat, L. Roffidal, A. M. Chagneau, ing and prehension of feed particles in domestic fowls. Prod.
and I. Bouvarel. 2009. Short term feed intake of laying hens of- Anim. 10:403–414.
fered loose mix diets including whole wheat and a balancer con- Picard, M. L., G. Uzu, E. A. Dunnington, and P. B. Siegel. 1993.
trasted in particle size. Page 32 in Proc. of the 8th Avian French Food-intake adjustments of chicks—Short-term reactions to de-
Research Days. St Malo, France. 25–26 March 2009. World Poult. ficiencies in lysine, methionine and tryptophan. Br. Poult. Sci.
Sci. Assoc. (WPSA). 34:737–746.
Engberg, R. M., M. S. Hedemann, S. Steenfeldt, and B. B. Jensen. Plavnik, I., E. Wax, D. Sklan, I. Bartov, and S. Hurwitz. 1997. The
2004. Influence of whole wheat and xylanase on broiler perfor- response of broiler chickens and turkey poults to dietary energy
mance and microbial composition and activity in the digestive supplied either by fat or carbohydrates. Poult. Sci. 76:1000–
tract. Poult. Sci. 83:925–938. 1005.
Folch, J., M. Lees, and C. H. S. Solane-Stanley. 1957. A simple Reichmann, K. G., and J. K. Connor. 1979. Effects of meal feeding
method for the isolation and purification of total lipids from ani- of calcium, protein and energy on production and calcium status
mal tissues. J. Biol. Chem. 226:497–509. of laying hens. Br. Poult. Sci. 20:445–452.
Gabriel, I., S. Mallet, and M. Leconte. 2003. Differences in the di- Rougière, N. 2010. Etude compare des parameters digestifs des pou-
gestive tract characteristics of broiler chickens fed on complete lets issus des lignées génétiques D+ et D− sélectionnées pour une
pelleted diet or on whole wheat added to pelleted protein concen- efficacité digestive divergente. Thesis. François Rabelais Univer-
trate. Br. Poult. Sci. 44:283–290. sity, Graduate School SST.
Gabriel, I., S. Mallet, M. Leconte, A. Travel, and J. P. Lalles. 2008. Rutter, K., L. Hennoste, L. C. Ward, B. H. Cornish, and B. J.
Effects of whole wheat feeding on the development of the diges- Thomas. 1998. Bioelectrical impedance analysis for the estima-
tive tract of broiler chickens. Anim. Feed Sci. Technol. 142:144– tion of body composition in rats. Lab. Anim. 32:65–71.
162. Safaa, H. M., E. Jiménez-Moreno, D. G. Valencia, M. Frikha, M. P.
Hetland, H., B. Svihus, and M. Choct. 2005. Role of insoluble fiber Serrano, and G. G. Mateos. 2009. Effedct of main cereal of the
on gizzard activity in layers. J. Appl. Poult. Res. 14:38–46. diet and particle size of the cereal on productive performance and
Hill, K. J., and P. J. Strachan. 1975. Recent advances in digestive egg quality of brown egg-laying hens in early phase of production.
physiology of the fowl. Symp. Zoo. Soc. London 35:1–12. Poult. Sci. 88:608–614.
ISA-Hendrix. 2011. Nutrition Management Guide. 2011 ed. ISA Sakomura, N. K. 2004. Modeling energy utilisation in broiler breed-
Hendrix Genetics, Saint Brieuc, France. ers, laying hens and broilers. Braz. J. Poult. Sci. V6:1–11.
Lázaro, R., M. García, M. J. Arníbar, and G. G. Mateos. 2003. Ef- Sauvant, D., J. M. Perez, and G. Tran. 2002. Table de composition
fect of enzyme addition to wheat-, barley-, and rye-based diets on et de valeur nutritive des matières premières destinées aux ani-
nutrient digestibility and performance of laying hens. Br. Poult. maux d’élevage. Inst. Natl. Rech. Agron., Paris, France.
Sci. 44:256–265. Sauveur, B., and P. Mongin. 1974. Effects of time-limited calcium
Lee, K. H., and Y. S. Ohh. 2002. Effects of nutrient levels and feed- meal upon food and calcium ingestion and egg quality. Br. Poult.
ing regimen of a.m. and p.m. diets on laying hen performances Sci. 15:305–313.
and feed cost. Korean J. Poult. Sci. 29:195–204. Umar Faruk, M., I. Bouvarel, S. Mallet, M. N. Ali, H. M. Tukur, Y.
Leeson, S., and J. D. Summers. 1978. Voluntary food restriction by Nys, and P. Lescoat. 2011. Is sequential feeding of whole wheat
laying hens mediated through dietary self-selection. Br. Poult. more efficient than ground wheat in laying hens? Animal 5:230–
Sci. 19:417–424. 238.
2486 TRAINEAU ET AL.
Umar Faruk, M., I. Bouvarel, N. Meme, N. Rideau, L. Roffidal, H. tration, and particle sizes of nonstarch polysaccharides on digesta
M. Tukur, D. Bastianelli, Y. Nys, and P. Lescoat. 2010a. Sequen- mean retention time and gut development in laying hens. Br.
tial feeding using whole wheat and a separate protein-mineral Poult. Sci. 52:730–741.
concentrate improved feed efficiency in laying hens. Poult. Sci. Verdal, H. D., A. Narcy, D. Bastianelli, H. Chapuis, N. Meme, S.
89:785–796. Urvoix, E. L. Bihan-Duval, and S. Mignon-Grasteau. 2011. Im-
Umar Faruk, M., I. Bouvarel, N. Meme, L. Roffida, H. M. Tukur, Y. proving the efficiency of feed utilization in poultry by selection.
Nys, and P. Lescoat. 2010b. Adaptation of wheat and protein- Genetic parameters of anatomy of the gastro-intestinal tract and
mineral concentrate intakes by individual hens fed ad libitum in digestive efficiency. BMC Gen. 12:(6 July 2011).
sequential or in loose-mix systems. Br. Poult. Sci. 51:811–820. Vergara, P., C. Ferrando, M. Jimenez, E. Fernandez, and E. Gona-
Van Krimpen, M. M., R. P. Kwakkel, C. M. C. van der Peet-Schwer- lons. 1989. Factors determining gastrointestinal transit-time
ing, L. A. den Hartog, and M. W. A. Verstegen. 2009. Effects of of several markers in the domestic-fowl. Q. J. Exp. Physiol.
nutrient dilution and nonstarch polysaccharide concentration in 74:867–874.
rearing and laying diets on eating behavior and feather damage Yo, T. 1996. Adaptation comportementale au choix alimentaire du
of rearing and laying hens. Poult. Sci. 88:759–773. poulet de chair et de la poule pondeuse (Gallus domesticus) en
Van Krimpen, M. M., R. P. Kwakkel, C. M. C. van der Peet-Schwer- milieu tropical. Thesis. Ecole nationale supérieure argonomique
ing, L. A. den Hartog, and M. W. A. Verstegen. 2011. Effects of de Rennes, Rennes, France.
dietary energy concentration, nonstarch polysaccharide concen-
Chapitre 4 : Capacité des poules pondeuses à
ajuster leur consommation d’aliment après
s’être vu offrir précédemment un régime
nutritionnellement déséquilibré

Maxime Traineau, Dolores I. Batonon, Isabelle Bouvarel, Lucien Roffidal and Philippe
Lescoat 2014. Capacity of laying hens in sequential feeding to adjust their feed consumption
when offered previously a nutritionally unbalanced diet. In press.

113
114
Résumé Publication 2 :

En alimentation séquentielle (AS) la distribution d’un régime riche en énergie et


pauvre en protéines le matin et d’un régime riche en protéine et calcium et pauvre en
protéines l’après midi permet d’améliorer significativement l’efficacité alimentaire.
Cependant, les précédents travaux ont mis en évidence l’incapacité des poules à réguler
correctement leur ingestion (particulièrement sur cette fraction du matin), entrainant une
fragilité potentielle des poules, mais aussi à résister à un épisode de fortes chaleurs. Deux
expériences ont été menées en parallèles pour tenter de comprendre et de contrôler l’ingestion
des poules dans un système de distribution séquentielle afin de parvenir à conduire les poules
vers un poids et une composition corporelle optimale. Les niveaux d’ingestion ainsi que les
performances de ponte ont été relevées de 20 à 30 semaines d’âge pour deux ambiances
différentes (20°C et 26°C).
Un régime complet était distribué deux fois par jour avec un niveau constant d’énergie
et de protéines. Les poules alimentées de façon séquentielle recevaient la fraction riche en
énergie (E) le matin et la fraction riche en protéines et calcium (P) l’après midi. Dans la
cellule à température neutre, (Exp.1) les poules recevaient 4 régimes différents des semaines
19 à 23 : un témoin continu, et trois régimes offerts en séquentiel (40% E/60% P; 50% E/50%
P; 60% E/40%P). Les poules dans la cellule en condition de température élevée (Exp.2)
recevaient des semaines 19 à 23soit un aliment témoin continu soit un des deux régimes en
séquentiel (40% E/60% P; 50% E/50% P). Afin d’évaluer la plasticité des poules et de
déterminer leur capacité à converger vers des caractéristiques physiques similaires (poids et
composition corporelles) l’ensemble des poules en AS recevaient le régime 50% E/50% P de
la semaine 24 à 29.
Les résultats ont montré qu’en température neutre, (Exp.1) les poules alimentées en
AS ont un indice de consommation amélioré par rapport à celui des poules alimentées en
continu. De la semaine 24 à 29, l’ensemble de poules en AS ont convergé vers un niveau
d’ingestion équivalent indépendamment du régime reçu précédemment. Ainsi, le poids des
poules et la composition corporelle ne diffèrent pas entre les poules en AS. Les poules en AS
dans des conditions de températures élevées (Exp. 2) ont tendance à consommer moins mais
produisent de la même façon que les poules en alimentation continue. Par conséquent, il peut
en être conclu pour cette expérimentation que les hautes températures diminuent globalement
les performances mais n’empêchent pas l’effet positif de l’AS. Plus globalement, ces
115
expérimentations ont souligné la difficulté de piloter simultanément ingestion, performances
et croissances de poules pondeuses.

116
13-70
Sequential feeding in laying hens

Capacity of laying hens in sequential feeding to adjust their feed consumption


when offered previously a nutritionally unbalanced diet

Dolores I. Batonon1,5*, Maxime Traineau1,3*, Isabelle Bouvarel2, Lucien Roffidal3 and


Philippe Lescoat1,4
1
INRA, UR83 Recherches Avicoles, F-37380 Nouzilly, France
2
ITAVI, Nouzilly, France
3
INZO°, Montgermont, France
4
AgroParisTech, Département Sciences de la Vie et Santé - UFR Développement des Filières Animales, Paris,
France
5
UAC-FSA, Laboratoire de Recherches Avicoles et de Zoo-Economie, Abomey-Calavi, Bénin
Contact: philippe.lescoat@agroparistech.fr
* These authors are considered as the first author

Manuscript received 20 November 2013,

Introduction

The characteristic of sequential feeding is the offering of two dietary components alternately
at different times of the day. This technique was reported as more economical in countries, in
which feedstuffs transport and diet mixing are expensive and where the direct use of locally
produced feedstuffs could be implemented (UMAR FARUK, 2010). Distribution of a high
energy diet in the morning followed by a protein-mineral concentrate in the afternoon resulted
in a reduction in feed energy and protein intakes while maintaining egg production, egg
weight compared to a complete diet given throughout the day (LEESON and SUMMERS,
1978; REICHMANN and CONNOR, 1979). LEE and OHH (2002) also reported a reduction
in feed intake when birds were offered a high energy/protein and low Ca diet in the morning
and a low energy/protein and high Ca in the afternoon.
However, this feeding system impacts birds’ weight. Sequentially fed birds were
lighter when compared with control fed ones (ROBINSON, 1985; UMAR FARUK et al.,
2010a). Low weight could be linked to the lower feed intake, also related to the lower body
fat deposition in sequential feeding (SCANES et al., 1987; UMAR FARUK et al., 2011).
Since laying performances are assumed to depend on bird’s maturity (BOUVAREL et al.,
117
2010; CANET et al., 2012), the question arises whether hens with lower feed intake and
lower body weight compared to complete diet are able to go ahead producing at an equivalent
level of hens fed complete diet for the whole laying cycle.
Moreover, to the authors’ knowledge, there is no specific research on the sensitivity of
laying hens to unbalanced diet both in energy and protein under sequential feeding
management provided at the beginning of the laying period. Most studies on sequential
feeding indicate benefits when birds were offered sequential diets contrasted between
morning and afternoon, but with a daily supply similar to the control diets (BLAIR et al.,
1973; ROBINSON, 1985; UMAR FARUK et al., 2010a). However, KESHAVARZ (1998)
indicated that egg production can’t be maintained by providing an adequate level of good
quality of protein (16%) only half a day, compared to the control group that received a diet
with 16% protein the whole day. The flexibility of hens according to the provided nutrients
has therefore to be studied. This might help to drive sequentially fed birds to a required shape
needed to keep on performing correctly for the whole laying cycle.
Besides, the detrimental effects of high environmental temperature on poultry
performance are commonly observed: decreasing egg production (ARIMA et al., 1976), egg
mass (DEATON et al., 1986) and egg weight (BALNAVE and BRAKE, 2005). Increasing the
dietary level of protein (GUNAWARDANA et al., 2008; ALMEIDA et al., 2012) or allowing
bird to consume adequate nutrients before photostimulation has been suggested to overcome
the negative effect of heat stress (BALNAVE and BRAKE, 2005). Sequential feeding, which
maintains egg production in laying hens with a reduction of feed intake under high
temperatures, might be a solution in this environment (UMAR FARUK, 2010). However, lack
of consistent references regarding sequential feeding under high environmental temperatures
restrains the flexibility of this system. Consequently, further investigations are required on the
ability of laying hens under sequential feeding to control their intakes and performances in
contrasted environments.
Therefore, the present work was designed to investigate the direct impacts of
sequential distribution of unbalanced diets in energy and protein contents on the performances
of laying hens from 20 to 23 weeks-old. A second step was to evaluate the back-effects of 20
to 23 weeks-old unbalanced supplies on performances of laying hens offered balanced
sequential feeding between 24 and 29 week-old. To explore contrasted environments, two
trials were performed: the first with a neutral temperature and the second with a hot one.

118
Materials and Methods

Two different experiments were conducted simultaneously. In both trials, Hendrix Isa Brown
laying hens were housed in a three-tier battery having individual cages (25 x 38cm) equipped
with a nipple drinker. The feeding through was common to 5 cages. Each group of 5
contiguous cages got the same diet. Birds were offered whole wheat in mixed diet or
sequentially from weeks 16 to 18 to implement sequential feeding (UMAR FARUK et al.,
2008).
The experimental period ran from week 19 to 30 of age. Photoperiod was 12L:12D at
week 16, reached 16L:8D at week 18 and remained at this level until the end of the
experiment.

Experimental treatments
Composition and calculated contents of nutrients of diets are described in Table 1. The control
treatment containing 10.9 MJ/kg and 165 g CP/kg was fed ad libitum and was given in equal
portions twice a day between 20 and 29 weeks-old. For the sequential combinations, birds
received the high-energy fraction (E) in the morning and beginning of the afternoon and the
high protein and mineral fraction (P) in the late afternoon and early morning. To test the birds'
flexibility, contrasted sequential diets were provided. Energy fraction (E: 98 g CP/kg, 13.1
MJ/kg) and protein fraction (P: 231.7 g CP/kg, 8.8 MJ/kg) were offered in varying
proportions: 50 g of E between 8:45h and 14:45h and 75 g P between 14:45h and 8:45h for
the E40/P60 diet; 62.5 g of E between 8:45h and 15:45h and 62.5 g of P between 15:45h and
8:45h for the E50/P50 diet; 75 g of E between 8:45h and 16:45h and 50 g of P between
16:45h and 8:45h for E60/P40. Each bird was offered 125 g of feed i.e. 105% of the
theoretical requirement of laying hens (ISA, 2011).
Since corn is richer in energy and lower in protein than wheat (SAUVANT et al.,
2004), the energy fraction was formulated mostly with ground corn and a limited amount of
20% whole wheat to obtain large differences of nutrient supply dynamics between
experimental diets. Inclusion of whole grains in the diet of laying hens was chosen to
stimulate the digestive capacity of hens (NIR et al., 1990) by increasing the gizzard size. Hens
with E40/P60 received 10.5 MJ ME/kg and 178.2 g CP/kg, hens with E50/P50 received 10.9
MJ ME/kg and 164.8 g CP/kg and those in E60/P40 group received 11.4 MJ ME/kg and 151.5

119
g CP/kg. Diets were formulated according to INRA tables (SAUVANT et al., 2004).Water
was given ad libitum throughout the experimental period.
Experiment 1 (Exp. 1)
The objective of this experiment was to evaluate the impacts of unbalanced diets in energy
and protein contents offered sequentially on the performances of laying hens from 20 to 23
weeks old in neutral environmental condition. Experimental design is described Table 2. One
hundred and twenty birds were reared at 20.4°C ± 0.02, kept constant from week 19 to week
30. Birds were divided into 4 groups of 30 birds (6 feeding troughs * 5 contiguous cages).
Four different diets were provided in this experiment between 20 and 23 weeks: control diet
and 3 sequential diets (40E/60P; 50E/50P; 60E/40P). To investigate the ability of laying hens
receiving contrasted diets from 20 to 23 week-old, to converge on similar characteristics at 30
weeks of age, two diets were given between week 24 and 29: the control one to the birds
receiving earlier the same diet and the 50E/50P to all other birds.

Experiment 2 (Exp. 2)
A group of 60 hens was placed in a constantly hot temperature room (25.8°C ± 0.02°C) to
study the impacts of sequential distribution of unbalanced diets from 20 to 23 weeks on bird’s
response in this environment. As in the first experiment (Table 2), birds were divided into 3
groups of 20 birds (4 feeding troughs * 5 contiguous cages). They were habituated to this high
temperature by a gradual increase to avoid acute heat stress (DANIEL and BALNAVE, 1981;
EMERY et al., 1984). An excessive energy supply as provided by 60E/40P could
dramatically decrease the overall feed intake and this diet was therefore omitted. Thus, control
diet and 2 sequential diets (40E/60P; 50E/50P) were provided between 20 and 23 weeks. In
the second period, birds got two diets: the control one and the 50E/50P, as in the first
experiment.

Measurements
Feed intake was recorded weekly for each group of five cages. In the sequentially fed
treatments, intakes of E and P were measured separately. Egg production was measured by
recording the number of eggs produced daily. Egg weight was recorded four times a week.
Weight of egg components (yolk, albumen, shell) was determined on a daily sample every 3
weeks, starting at week 21. Albumen and the chalazae were separated using forceps prior to

120
weighing the yolk. Shells were washed and dried for 12 hours in a drying oven at 70°C and
then weighed.
Birds were weighed individually at 16, 19, 23 and 30 weeks of age. Body composition
was predicted with a noninvasive method based on Bioelectrical Impedance Analysis (BIA) at
week 19, 23 and 30. The technique has been developed by the company INZO°1. It is
assumed that the electrical impedance of the biological organism is indicative of the highly
conductive fat-free component (BERG et al., 1994). The electrical impedance of a tissue
depends on its fluid and electrolyte content (RUTTER et al., 1998) and therefore on their
relative proportions of muscle, lipid, water and ash. Considering that, these proportions could
consequently be predicted. This method allows monitoring the evolution of the body
composition throughout the experimental period.
The weight of the digestive organs was recorded at the end of the trials to assess the
effect of feeding system and temperature conditions. For both experiments, ten birds per
treatment randomly selected were weighed before being injected with sodium Pentobarbital
solution (1 ml/kg body weight). The abdominal cavity was opened, the digestive tract
dissected and separated into gizzard, small intestine, pancreas, liver and spleen. Gizzard was
placed into an iced container (-4°C) for 24h to facilitate the removal of the surrounding fat
before being emptied and weighed.

Statistical analysis
Data from each experiment were analyzed based on 2 periods: from 20 to 23 weeks of age and
from 24 to 29 weeks of age, according to the changing of diets at 24 weeks of age. The
experimental unit for feed consumption (feed intake, fractional intakes, ME and CP intakes)
and feed conversion ratio was the group of 5 contiguous cages due to the common feeding
trough. The analyses on the other variables were based on individual values from cages.
In Exp. 1, control diet was compared to all sequential diets; then, the effects between
the three sequential diets (E40/P60 vs. E50/P50 vs. E60/P40) were tested. A one-way
ANOVA was performed to test the treatment effect on all measured variables. Results were
considered significantly different if P < 0.05, and a Bonferroni-Dunnet pairwise comparison
was used to compare differences between means.
In hot environmental condition (Exp. 2), only the diet effect was tested (Control vs.
E40/P60 vs. E50/P50). Due to the number of replications in this experiment, non-parametric

1
INZO° SAS rue de l’Eglise-BP 50019 02407 Chierry Cedex
121
tests were performed on feed consumption data and on feed conversion ratio. When there are
only two groups (E and P fractions intake), a Mann-Whitney test was performed and Kruskal-
Wallis test was realized for comparison of the 3 diets (total feed intake, ME, CP intake and
FCR). Laying performances and digestive organs weights have been submitted to a one-way
ANOVA. Similarly, results were considered significantly different if P < 0.05, and
Bonferroni-Dunnet pairwise comparison was used to compare differences between means.
Body content was submitted to a one-way repeated measures ANOVA. Statistical
analyses were performed using R version 2.15.2 (R CORE TEAM, 2012).

Results

Effect of the feeding systems in the neutral temperature condition (Exp. 1)


Weeks 20 to 23
From 20 to 23 weeks of age, no difference was observed with E40/P60, E50/P50 and E60/P40
compared to control diet for total feed, ME and CP intakes (Table 3). In sequential feeding,
energy and protein contents did not affect the total feed intake. However, energy fraction (E)
intake was significantly lower with E40/P60 diet compared to the two other sequential diets,
representing respectively about 42%, 51% and 59% of the total feed intake. Regarding protein
fraction (P) intake, the complementary results were observed. ME intake was similar among
the treatments while CP intake was significantly affected by sequential diets. The pairwise
comparison showed that E50/P50 protein intake tended to be different from the E60/P40
whereas both diets are significantly lower than the E40/P60. Egg production was increased
with the sequential diets compared to the control one. Egg weight, egg mass and egg
components proportions were similar between control and sequential diets except for the egg
shell percentage. Feed conversion ratios (FCR) were higher with control group (2.20)
compared to sequential ones (2.04). There was no difference in feed efficiency within the
sequential diets.

Weeks 24 – 29
From weeks 24 to 29, total feed intake tended to be higher for the control than for the
sequential diets (P < 0.10) with no difference between sequential diets (Table 3). All
sequentially fed hens ate a similar quantity of E (around 48% of total feed intake) and P
(around 52% of total feed intake). Layers with control diet tended to eat more ME than hens

122
with sequential diets (P < 0.10) whereas CP intake was similar among all treatments. As
previously observed, egg production was significantly increased with the three sequential
groups compared to the control one. Egg weight, egg mass, egg yolk and albumen proportion
did not present any difference among all diets. Egg shell percentage was found to be lower in
the control compared to the 3 sequential groups. FCR were similar between the four groups
from weeks 24 to 29 of age.
Body weights were similar at week 19 for all diets and they increased with age. Hens
fed sequentially were always significantly lighter than control fed ones. The average body
weight values were 1736 g for hens with control diet and 1638 g for hens with sequential diets
at week 23. At week 30, they were 1813 g and 1659 g respectively for the control and the
sequential ones.
At 30 weeks-of-age, pancreas, liver and spleen proportions were not influenced by the
feeding system (Table 5). Nevertheless, sequential feeding resulted in significantly heavier
gizzard relative to control diet. No difference in the proportion of small intestine was recorded
between all treatments except for E40/P60 diet, which tended to be higher than E50/P50. The
heaviest abdominal fat proportion was obtained with control group, whereas the lowest was
obtained with E40/P60. Figures (1.a) and (1.b) show the time change in fat and protein
contents according to diets from 19 to 30 weeks of age. There was a significant increase in fat
level in all treatments with age, although hens fed sequentially had significantly lower fat
content. Birds in sequential feeding had similar fat content. Hens fed with sequential diets
were significantly higher than the control ones regarding protein proportion while no
difference was observed within sequential diets.
Hens in sequential groups got similar performances at 23 weeks old but they were
significantly different from the control group regarding feed efficiency. The results obtained
from 24 to 29 weeks-of-age indicated that birds with different feed consumption at 23 week-
old converged to the same level of intake when they were given access to the same diet.
Laying performances and body content did not vary within the sequential group with the
change of diets.

Effect of feeding systems on laying hens performances under high temperature (Exp. 2)
Weeks 20 to 23
Total feed intake during the distribution of unbalanced diets (Table 4) tended to be affected by
feeding system: it tended to be lower for E40/P60 than the control while E40/P60 was similar

123
to E50/P50. Birds fed E50/P50 diet consumed more E (around 47% of total feed intake) than
birds in E40/P60 (around 41% of total feed intake) whereas no difference was observed for P
intake. ME intake was significantly reduced with E40/P60 diet compared to E50/P50 and
Control while CP intake was similar among treatments.
Besides, there was no effect of the feeding systems on FCR and laying performances.
The only meaningful difference in terms of egg components weights was observed with egg
albumen proportion, significantly lower in control diet compared to the two sequential ones.
Hens have a similar body weight at week 19 which increased with age. Sequential
feeding reduced body weight, the lower value was recorded for E40/P60 diet at week 23.

Weeks 24 to 29
Feeding systems did not affect total feed intake, E fraction intake, ME and CP intake, from
weeks 24 to 29 (Table 4). The only difference was observed for P fraction intake with the
highest consumption in E50/P50 compared to E40/P60. Moreover, there is no diet effect on
laying performances. Egg weight and egg mass increased with age and there was no
difference between control and sequentially fed birds. Egg albumen tended to be different
according to the diet effect at P < 0.10 while no difference was found for egg yolk and egg
shell proportions. Birds had similar feed conversion ratio: 1.94, 1.93 and 1.90 respectively for
the control, E40P60 and E50/P50 diets. Body weight increased with age and hens in E40/P60
group were lighter than those in control diet at week 30.
No significant difference was observed in gizzard weight between all diets (Table 5).
However, hens fed with control diet have a lighter small intestine than those fed E40/P60.
Pancreas weight tended to be lower for hens with control diet compared to E50/P50 diet
whereas spleen weight tended to be lower for E40/P60 than for the others two diets (P <
0.10). The heaviest abdominal fat weight was obtained with Control.
Figures (2.a) and (2.b) show a change in fat and protein contents from 19 to 30 weeks
of age. There was a significant increase in fat level in all treatments with time with no
difference between diets. As previously observed, protein contents decreased with age but
they were similar between sequential diets and control.
It was pointed out through the results obtained during 20 and 23 weeks-of-age that
laying hens subjected to high temperature tended to eat less and produced as well as the
control. When hens were offered the same diet in sequential feeding from 24 to 29 weeks,
they converged to the same feeding characteristics at week 30.

124
Discussion

Effect of the feeding systems in the neutral temperature room (Exp. 1)


The chosen unbalance between energy and protein within each fraction and the sequential
supply should help to monitor body weight and composition at 30 weeks of age. Regarding
feed intake, the aim was reached since hens in sequential feeding had a daily morning fraction
offer that drove the expected 40%, 50% and 60% intakes of the daily supply in the first
period. Distribution of high-energy fraction followed by a protein-mineral one has improved
feed efficiency for sequential diets compared to the control one in the first period. This
improvement might be due to a slight under-consumption of feed and a higher egg production
for sequentially fed hens compared to control ones as observed previously (UMAR FARUK
et al., 2010a,b; 2011; TRAINEAU et al., 2013). This efficiency may also result of a more
efficient digestion thanks to the fractional distribution (CHAGNEAU et al., 2009).
Besides, during the unbalanced period, similar ME and CP intakes were observed
comparing control to sequential diets, suggesting a regulation of these nutrients by
sequentially fed birds, consistently with BLAIR et al. (1973), MACISAAC and ANDERSON
(2007) and UMAR FARUK et al. (2010b). Hens could fit their feed intake to satisfy their
energy requirements; then, an increase in energy concentration leading to a reduced feed
intake (PEREZ-BONILLA et al., 2012) while increasing dietary protein could increase
(GUNAWARDANA et al., 2008) or have no effect (SUMMERS and LEESON, 1994;
MOHITI-ASLI et al., 2012) on feed intake. In the second period, all hens were offered 50% E
and 50% P. ME intake tended to be lower for sequential diets compared to the control one.
Moreover, hens in sequential feeding were thinner and had lower abdominal fat weight than
those in control continuous diet. As demonstrated by SMITH (1973), higher energy
consumption than it is required for maintenance and production led to fat deposition,
increasing the body weight as observed here with the difference in body weight and fat
content between sequential and control fed birds.
Focusing on sequential diets, differences in CP intake could be observed in the
unbalanced period while no difference in ME intake was observed. Two hypotheses could be
proposed. First of all, the composition of the offered diets (BOUVAREL et al., 2004) leading
to contrasted nutrient intake: the gap between E offered in the morning and P distributed in
the afternoon is more pronounced in crude protein (98.0 g/kg for E vs. 231.7 g/kg for P; 59%
of difference) than in metabolizable energy (13.1 MJ/kg for E vs. 8.8MJ/kg for P; 37.3% of

125
difference). With the same amount of feed, energy and protein unbalance are not symmetrical
which could induce a lack in other nutrients such as amino acids. Regarding two limiting
amino acids, methionine and lysine (NOVAK et al., 2004; MOGHADDAM et al., 2012),
hens in sequential diets did not present any differences in total lysine and methionine intakes
compared to the control diet but within the sequential group, there was a difference (P < 0.01)
in these intakes. As demonstrated by HARMS and RUSSELL (1998), KESHAVARZ (2003)
and MOGHADDAM et al. (2012), simultaneous reduction of dietary protein, methionine and
lysine could reduce laying performances. Consistently, increasing methionine level
significantly increased laying performances (SAKI et al., 2012) up to an asymptotic level
between 0.36 – 0.38% (BOUVAREL et al., 2010). It seems therefore that hens are not able to
finely tune their feed intake on the proportion of protein in the diet as observed by MOHITI-
ASLI et al., (2012) but they are able to adapt their feed intake to the energy in the diet, even
though deviations could be observed (BOUVAREL et al., 2010).

Effect of feeding systems and high temperature on laying hens performances (Exp. 2)
The aim of this experiment was to determine if high environmental temperature was affecting
birds’ performances offered sequential diets. Birds sequentially fed met the expected
consumptions of E and P with the same hierarchy. This feed intake behavior could be
explained by the reduction in appetite when the temperature is above 21°C (SMITH, 1973,
MASHALY et al., 2004). Indeed, at high temperature, bird regulates its heat production by
decreasing global feed intake or specific nutrient yielding heat during processing
(SYAFWAN et al., 2012) which is consistent with our data of a lower E intake for
sequentially fed birds compared to the control diet. Furthermore, the ingestion of excess
carbohydrate produces a heat increment equivalent to 10-15% of the metabolizable energy
while fat may produce 0-5% (KLEYN, 2013). Since E is higher in calculated fat content than
P (46.8 g/kg vs. 34.5 g/kg respectively), this might also contribute to explain the lower E
intake under high temperature. Thus, UMAR FARUK et al., (2010a) hypothesized that the
reduced total feed intake in sequential feeding under hot climate, combined with similar
performance to conventional feeding, could lead to an improved feed utilization. Moreover,
FCR in sequential diets were lower during the first period than in the second one while,
according to the rearing guidelines, FCR decreased with age and temperature (PEGURI and
COON, 1991) and is higher at the onset as observed in neutral temperature.

126
Our results stressed that there was a reduction in body weight and body weight gain
due to treatments: birds fed sequentially had lower body weight compared to conventional
feeding as observed in neutral temperature room (SMITH, 1973; UMAR FARUK et al.,
2010a). Moreover, as it was hypothesized in neutral temperature condition, sequentially fed
birds seem to have a lower fat deposition according to lower maintenance requirements
(UMAR FARUK et al., 2010a). This metabolic mechanism might be similar in high
temperature room since their energy requirements for maintenance decreased with
temperature (DAGHIR, 2008). According to PEGURI and COON (1991) and CHENG et al.
(1997), hot ambient temperatures lead to a decreased growth performance of birds due to a
restriction in nutrients intakes and physiological changes observed in hot environment
(CHENG et al., 1997; MUJAHID, 2011).

Conclusions
These studies indicate that laying hens in sequential feeding were able to modify their feed
intake when diets varied in energy and protein contents. However, they highlighted the
difficulty to increase robustness of animals by controlling weight gain. It would be interesting
in future work to successfully control the performances of sequentially fed birds throughout
the laying period according to their growth performances.
Moreover, these works suggest that high environmental temperature does not disable
sequential feeding benefits. Previous studies have shown that hens supplied with sequential
diet, could be more fragile, which was here observed through a decrease of performances with
temperature. It must therefore be ensured, that hens fed sequentially, under high temperatures,
will not be too thin or deficient in specific nutrient, in order to last a complete commercial
laying period. There is however the need to investigate laying hens’ responses to amino acids
imbalance. Indeed, during the first period, temperature did not affect lysine and methionine
intakes while there was a feeding system effect according to the diets offered. During the
second period, there was a reduction in these two limiting amino acids intakes, suggesting that
the length of exposure to heat may affect the response of birds as observed for protein
(GONZALEZ-ESQUERRA and LEESON, 2005). Exposure to high temperature has been
shown to influence amino acids digestibility (WALLIS and BALNAVE, 1984; BRAKE et al.,
1998; MUJAHID, 2011). Attention should be given to an imbalance that was potentially
induced by sequential feeding of diets varying both in energy and protein content.

127
Acknowledgments
We thank Philippe Didier and Michel Couty (INRA) for their technical assistance. We are
grateful to our experimental unit (UE PEAT) for its help in the set-up of the experiment. The
scientific and financial support of INZO° (Montgermont, France) and the financial help of the
French Embassy in Benin are highly appreciated.

Summary

Sequential feeding consists in splitting the diet into two fractions. Distribution of a high-
energy diet in the morning followed by a diet rich in protein and calcium in the afternoon, can
significantly improve feed efficiency. Nevertheless, previous works showed the difficulty of
hens to regulate their feed intake and to overcome high temperatures when fed sequentially.
Two experiments have been conducted to understand whether, controlling laying hens’ feed
consumption in sequential feeding system, would drive them to an optimal body weight and
body composition. Feed consumption and performances of birds offered sequential diets
unbalanced in energy and protein contents were investigated under two contrasted
temperatures from 20 to 30 weeks of age. A complete diet was fed as control and given in
equal portions twice a day. Birds in sequential diets received the high-energy fraction (E) in
the morning and beginning of the afternoon and the high protein and mineral fraction (P) in
the late afternoon and early morning.
Four different diets were provided between 20 and 23 weeks old for birds in
experiment 1 (Exp. 1): control diet and 3 sequential diets (40% E/60% P; 50% E/50% P; 60%
E/40%P). For birds in high temperature condition in the experiment 2 (Exp. 2), 3 diets were
given: control diet, 2 sequential diets (40% E/60% P; 50% E/50% P). To test the flexibility of
birds to converge to the same characteristics at week 30, birds in control group received the
same diet and all other birds, the 50% E/50% P one from 24 to 29 weeks.
The results showed that under neutral temperature (Exp. 1), hens in sequential groups
were different to the control one between 20 to 23 weeks, regarding feed efficiency. From 24
to 29 weeks, birds with different feed supply at 23 week converged to the same consumption.
Laying performances and body composition did not vary in the sequential groups with the
change of diets. Since birds fed sequentially in experiment 2 (Exp. 2) tended to consume less
and produced as well as control fed birds, it might be concluded that high environmental

128
condition reduced bird’s performances but did not hide the effects of sequential feeding
system.
Key words Laying hens, sequential feeding, unbalanced diets, temperature,
performances

129
References

ALMEIDA, V. R., DIAS, A. N., BUENO, C. F. D., COUTO, F. A. P., RODRIGUES, P. A.,
NOGUEIRA, W. C. L. & FARIA FILHO, D. E. (2012) Crude protein and
metabolizable energy levels for layers reared in hot climates. Braz. J. Poult. Sci., 14:
203-208.
ARIMA, Y., MATHER, F. B. & AHMAD, M. M. (1976) Response of egg-production and
shell quality to increases in environmental-temperature in 2 age-groups of hens. Poult.
Sci., 55: 818-820.
BALNAVE, D. & BRAKE, J. (2005) Nutrition and management of heat-stressed pullets and
laying hens. World’s Poult. Sci. J., 61: 399-406.
BERG, E. P. & MARCHELLO, M. J. (1994) Bioelectrical-Impedance analysis for the
prediction of fat-free mass in lambs and lamb carcasses. J. of Anim. Sci., 72: 322-329.
BLAIR, R., DEWAR, W. A. & DOWNIE, J. N. (1973) Egg production responses of hens
given a complete mash or unground grain together with concentrate pellets. Br. Poult.
Sci., 14: 373-377.
BOUVAREL, I., BARRIER-GUILLOT, B., LARROUDE, P., BOUTTEN, B., LETERRIER,
C., MERLET, F., VILARINO, M., ROFFIDAL, L., TESSERAUD, S., CASTAING, J.
& PICARD, M. (2004) Sequential feeding programs for broiler chickens: Twenty-
four- and forty-eight-hour cycles. Poult. Sci., 83: 49-60.
BOUVAREL, I., NYS, Y., PANHELEUX, M. & LESCOAT, P. (2010) How hen’s diet
influences eggs quality? INRA Prod. Anim., 23(2): 167-182.
BRAKE J. (1998): Optimum dietary arginine: Lysine ratio for broiler chickens is altered
during heat stress in association with changes in intestinal uptake and dietary sodium
chloride. Br. Poult. Sci., 39(5), 639-647.
CANET, Z. E., ROMERA, B. M., FAIN BINDA, V., TERZAGHI, A., DOTTAVIO, A. M. &
MASSO, R. J. D. (2012) Productive indicators at sexual maturity in experimental
populations of free range layers. Rev. Argent. Prod. Anim., 32 : 37-46.
CHAGNEAU, A. M., LECUELLE, S., LESCOAT, P., QUENTIN, M., GUILLAUMIN, J. M.
& BOUVAREL, I. (2009) The slow growing chicken well add value food distribution
by meal. 8èmes Journées de la recherche avicole, St Malo, France, 25-26 March 2009, p
59.
CHENG, T. K., HAMRE, M. L. & COON, C. N. (1997) Effect of environmental temperature,

130
dietary protein, and energy levels on broiler performance. J. Appl. Poult. Res., 6: 1-17.
DAGHIR, N. J. (2008) Nutrient requirements of poultry at high temperatures, in: DAGHIR,
N., J. (Eds) Poultry production in hot climates, 2nd ed, pp. 132-159 (UK, Cromwell
Press).
DANIEL, M. & BALNAVE, D. (1981) Responses of laying hens to gradual and abrupt
increases in ambient-temperature and humidity. Austr. J. Exp. Agric., 21: 189-195.
DEATON, J. W., REECE, F. N. & LOTT, B. D. (1986) Effect of summer cyclic temperatures
versus moderate temperature on laying hen performance. Poult. Sci, 65: 1649-1651.
EMERY, D. A., VOHRA, P., ERNST, R. A. & MORRISON, S. R. (1984) The effect of
cyclic and constant ambient-temperatures on feed consumption, egg-production, egg
weight, and shell thickness of hens. Poult. Sci, 63: 2027-2035.
GONZALEZ-ESQUERRA, R. & LEESON, S. (2005) Effects of acute versus chronic heat
stress on broiler response to dietary protein. Poult. Sci, 84: 1562-1569.
GUNAWARDANA, P., ROLAND, D. A., SR. & BRYANT, M. M. (2008) Effect of Energy
and Protein on Performance, Egg Components, Egg Solids, Egg Quality, and Profits in
Molted Hy-Line W-36 Hens. J. Appl. Poult. Res., 17: 432-439.
HARMS, R. H. & RUSSELL, G. B. (1998) Layer performance when returned to a practical
diet after receiving an amino acid-deficient diet. J. Appl. Poult. Res., 7, 175-179.
ISA. (2011). Nutrition Management Guide, 2011 Edition. ISA Hendrix Genetics, Saint
Brieuc, France
KESHAVARZ, K. (1998) Further investigations on the effect of dietary manipulation of
protein, phosphorus, and calcium for reducing their daily requirement for laying hens.
Poult. Sci, 77: 1333-1346.
KESHAVARZ, K. (2003) Effects of reducing dietary protein, methionine, choline, folic acid,
and vitamin B-12 during the late stages of the egg production cycle on performance
and eggshell quality. Poult. Sci, 82: 1407-1414.
KLEYN, R. (2013). Chicken nutrition. A guide for nutritionists and poultry professionals.
ISBN: 978-1-899043-42-2, 347 pages, Context 2013.
LEE, K. H. & OHH, Y. S. (2002) Effects of nutrient levels and feeding regimen of a.m. and
p.m. diets on laying hen performances and feed cost. Kor. J. Poult. Sci, 29: 195-204.
LEESON, S. & SUMMERS, J. D. (1978) Voluntary food restriction by laying hens mediated
through dietary self-selection. Br. Poult. Sci, 19: 417-424.
MACISAAC, J. L. & ANDERSON, D. M. (2007) Effect of whole wheat, enzyme

131
supplementation and grain texture on the production performance of laying hens. Can.
J. Anim. Sci., 87: 579-589.
MASHALY, M. M., HENDRICKS, G. L. III, KALAMA, M. A., GEHAD, A. E.,
PATTERSON, P. H. & ABBAS, A. O. (2004) Effect of Heat Stress on Production
Parameters and Immune Responses of Commercial Laying Hens. Poult. Sci, 83: 889-
894.
MOGHADDAM, H. N., FARD, M. K., AGAH, M. J., HOSSEINI, S. J. & MIRAKZEHI, M.
T. (2012) Effect of different levels of methionine, protein and tallow on the productive
performance and egg quality of laying hens in the late-phase production. Braz. J.
Poult. Sci, 14: 149-158.
MOHITI-ASLI, M., SHIVAZAD, M., ZAGHARI, M., REZAIAN, M., AMINZADEH, S. &
MATEOS, G. G. (2012) Effects of feeding regimen, fiber inclusion, and crude protein
content of the diet on performance and egg quality and hatchability of eggs of broiler
breeder hens. Poult. Sci 91: 3097-3106
MUJAHID, A. (2011) Nutritional strategies to maintain efficiency and production of chickens
under high environmental temperature. J. Poult. Sci., 48: 145-154.
NIR, I., MELCION, J. P. & PICARD, M. (1990) Effect of particle-size of sorghum grains on
feed-intake and performance of young broilers. Poult. Sci, 69: 2177-2184.
NOVAK, C., YAKOUT, H. & SCHEIDELER, S. (2004) The combined effects of dietary
lysine and total sulfur amino acid level on egg production parameters and egg
components in Dekalb Delta laying hens. Poult. Sci, 83: 977-984.
PEGURI, A. & COON, C. (1991) Effect of temperature and dietary energy on layer
performance. Poult. Sci, 70: 126-138.
PEREZ-BONILLA, A., NOVOA, S., GARCIA, J., MOHITI-ASLI, M., FRIKHA, M. &
MATEOS, G. G. (2012) Effects of energy concentration of the diet on productive
performance and egg quality of brown egg-laying hens differing in initial body weight.
Poult. Sci, 91: 3156-3166.
REICHMANN, K. G. & CONNOR, J. K. (1979) Effects of meal feeding of calcium, protein
and energy on production and calcium status of laying hens. Br. Poult. Sci, 20: 445-
452.
R CORE TEAM (2012) R: A Language and Environment for Statistical Computing. R
Foundation for Statistical Computing.
ROBINSON, D. (1985) Performance of laying hens as affected by split time and split

132
composition dietary regimens using ground and unground cereals. Br. Poult. Sci, 26:
299-309.
RUTTER, K., HENNOSTE, L., WARD, L. C., CORNISH, B. H. & THOMAS, B. J. (1998)
Bioelectrical impedance analysis for the estimation of body composition in rats. Lab.
Anim., 32: 65-71.
SAKI, A. A., HARSINI, R. N., TABATABAEI, M. M., ZAMANI, P. & HAGHIGHT, M.
(2012) Estimates of methionine and sulfur amino acid requirements for laying hens
using different models. Braz. J. Poult. Sci, 14: 209-216.
SAUVANT, D., PEREZ, J.-M. & TRAN, G. (2004) Tables of composition and nutritional
value of feed materials, INRA Editions and AFZ. Wageningen Academic Publishers,
Paris, France.
SCANES, C. G., CAMPBELL, R. & GRIMINGER, P. (1987) Control of energy-balance
during egg-production in the laying hen. J. Nut., 117: 605-611.
SMITH, A. J. (1973) Some effects of high environmental temperatures on the productivity of
laying hens (a review). Trop. Anim. Health Prod., 5: 259-271.
SUMMERS J. D & LEESON, S. (1994) Laying hen performance as influenced by protein
intake to sixteen weeks of age and body weight at point of lay. Poult. Sci, 73: 495-501.
SYAFWAN, S., WERMINK, G. J. D., KWAKKEL, R. P. & VERSTEGEN, M. W. A. (2012)
Dietary self-selection by broilers at normal and high temperature changes feed intake
behavior, nutrient intake, and performance. Poult. Sci, 91: 537-549.
TRAINEAU, M., BOUVAREL, I., MULSANT, C., ROFFIDAL, L., LAUNAY, C. &
LESCOAT, P. (2013) Effects on performance of ground wheat and with or without
insoluble fiber or whole wheat in sequential feeding for laying hens. Poult. Sci., 92:
2475-2486.
UMAR FARUK, M., DEZAT, E. BOUVAREL, I. NYS, Y., & LESCOAT P. (2008) Loose-
mix and sequential feeding of mash diets with whole-wheat: effect on feed intake in
laying hens. Proceedings of the XXIIIth World’s Poultry Congress, Brisbane,
Australia. p. 468.
UMAR FARUK, M. (2010) L’évaluation de l’alimentation mélangée et séquentielle à base de
matières premières localement disponibles sur les performances des poules pondeuses
en France et au Nigéria. Ph. D. Thesis, Université François-Rabelais de Tours.
UMAR FARUK, M., BOUVAREL, I., MEME, N., RIDEAU, N., ROFFIDAL, L., TUKUR,
H. M., BASTIANELLI, D., NYS, Y. & LESCOAT, P. (2010a) Sequential feeding

133
using whole wheat and a separate protein-mineral concentrate improved feed
efficiency in laying hens. Poult. Sci, 89: 785–796.
UMAR FARUK, M., BOUVAREL, I., MÊME, N., ROFFIDAL, L., TUKUR, H. M., NYS,
Y. & LESCOAT, P. (2010b) Adaptation of wheat and protein-mineral concentrate
intakes by individual hens fed ad libitum in sequential or in loose-mix systems. Br.
Poult. Sci, 51: 811-820.
UMAR FARUK, M., BOUVAREL, I., MALLET, S., ALI, M. N., TUKUR, H. M., NYS, Y.
& LESCOAT, P. (2011) Is sequential feeding of whole wheat more efficient than
ground wheat in laying hens? Anim., 5: 230 238.
WALLIS, I. R. & BALNAVE, D. (1984) The influence of environmental-temperature, age
and sex on the digestibility of amino-acids in growing broiler-chickens. Br. Poult. Sci,
25: 401-407.

Corresponding author: P. Lescoat, AgroParisTech, Département Sciences de la Vie et Santé -


UFR Développement des Filières Animales, 16, rue Claude Bernard, 75231 Paris, France; E-
mail: philippe.lescoat@agroparistech.fr

134
Table 1. Composition of experimental diets given to laying hens between 19 and 29 weeks-
old

Ingredient (g/kg) Control diet Energy fraction (E) Protein fraction (P)
Corn 444.5 715.0 170.0
Wheat 160.0 - 125.0
Whole wheat - 200.0 -
Soybean meal 200.0 - 399.0
Calcium carbonate 75.9 149.5
Sunflower meal 78.0 69.0 87.0
Soybean oil 16.0 16.0 17.0
Dicalcium phosphate 13.1 - 2.7
Premix1 10.0 - 20.0
DL-Methionine 1.6 - 3.4
L-Lysine 0.1 - 0.2

Calculated composition (g/kg)


ME (MJ/kg) 2 10.9 13.1 8.8
CP 165.0 98.0 231.7
Analyzed CP (g/kg) 170.2 99.1 231.7
DM 886.5 871.9 901.4
Calcium 36.0 0.4 71.0
Fat 40.2 46.8 34.5
Total P 5.9 2.9 9.0
Available P 3.1 0.7 5.7
Lysine 8.2 3.2 13.2
Methionine 4.5 2.1 7.1
Threonine 6.6 3.6 9.4
Tryptophan 2.0 0.9 3.0
1
Premix: Supplied per kg of diet at Vitamin A 750000 UI, Vitamin D3 300000 UI, Vitamin E (tocopherol
acetate 3a700) 1000 UI, Vitamin K3 100 mg, Zinc 10000 mg, Manganese 10000 mg, Copper 1000 mg, Iron
4000 mg, Iode 150.0 mg, Selenium 30.0 mg ; Canthaxanthin 1950.0 mg, ß-apocarotenoïc acid 2000.0 mg, lutein
3560.0 mg, zeaxanthin 200.0 mg, Cryptoxanthin 80.0 mg, Citric acid 400.00 mg, orthophosphoric acid 660.00
mg, Ethoxyquin 400 mg, Propyl gallate (E310) 400 mg. Nutritional values were calculated according to Sauvant
et al., 2002.
2
Nutritional values were calculated by INZO° adapted from SAUVANT et al., 2002

135
Table 2. Experimental design

Experiment 1 Experiment 2
Environmental condition Neutral temperature High temperature
Control Control
E40/P60 E40/P60
Diets
E50/P50 E50/P50
E60/P40 -
Replicates (group of 5 hens) 6 4
Number of birds 30 20
A Control diet was fed ad libitum between 20 and 29 weeks-old. Between 20 and 23 weeks-
old, 3 sequential diets were provided in the morning for E and in mid-afternoon for P:
E40/P60: 50 g of the energy fraction (E) and 75 g of protein fraction (P); E50/P50: 62.5 g of E
and 62.5 g of P; E60/P40: 75 g of E and 50 g of P. Between 24 and 29 weeks of age, birds
previously with E40/P60, E50/P50 or E60/P40 were offered the same sequential diet,
E50/P50.

136
Table 3. Effect of feeding systems between 20 and 23 weeks-old on performances of laying
hens between 20 and 29 weeks-old in neutral temperature condition (Experiment 1)

E40/P60 vs.
Control Control vs.
E40/P60 E50/P50 E60/P40 SEM 2 E50/P50 vs.
diet1 Sequential
E60/P40
N3 (N = 6) (N = 6) (N = 6) (N = 6)
Total feed intake (g/b/d)
Wk20-23 106.7 105.8 103.6 102.7 1.55 NS NS
Wk24-29 118.8 116.5 115.1 115.3 1.60 < 0.10 NS
Energy fraction intake
(g/b/d)
Wk20-23 44.8a 53.3b 60.3c 0.87 < 0.01
Wk24-29 56.6 55.6 55.3 1.30 NS
Protein fraction intake
(g/b/d)
Wk20-23 60.9a 50.4b 42.4c 1.18 < 0.01
Wk24-29 59.8 59.5 59.9 0.41 NS
ME intake (MJ/b/d)
Wk20-23 1.24 1.19 1.21 1.23 0.02 NS NS
Wk24-29 1.38 1.35 1.33 1.33 0.02 < 0.10 NS
CP intake (g/b/d)
Wk20-23 18.4 19.4a 17.7B 16.4C 0.29 NS < 0.01
Wk24-29 20.5 20.3 20.1 20.2 0.22 NS NS
Feed conversion ratio
Wk20-23 2.20 2.02 2.09 1.99 0.06 < 0.05 NS
Wk24-29 2.05 1.98 1.91 1.90 0.05 NS NS
N4 (N = 30) (N = 30) (N = 30) (N = 29)
Egg production (%)
Wk20-23 82.0 91.2 88.7 93.0 2.92 < 0.05 NS
Wk24-29 92.4 97.7 96.6 98.8 1.48 < 0.05 NS
Egg weight (g)
Wk20-23 56.9 56.8 55.9 55.1 0.67 NS NS
Wk24-29 62.6 61.7 62.3 61.3 0.89 NS NS
Egg mass (g/b/d)
Wk20-23 48.6 52.0 49.7 51.2 1.75 NS NS
Wk24-29 58.9 59.1 60.2 60.6 1.54 NS NS
Egg yolk (%)
Wk20-23 21.6 21.7 21.4 22.0 0.32 NS NS
Wk24-29 23.7 23.8 23.6 23.9 0.34 NS NS
Egg albumen (%)
Wk20-23 68.0 68.0 68.7 68.0 0.40 NS NS
Wk24-29 66.1 65.8 65.9 65.5 0.38 NS NS
Egg shell (%)
Wk20-23 10.44 10.30B 9.82C 9.96BC 0.15 < 0.05 < 0.10
Wk24-29 10.13 10.42 10.40 10.62 0.12 < 0.05 NS
Body weight (g)
Wk 19 1571 1569 1579 1610 22.23 NS NS
Wk 23 1736 1644 1624 1646 24.50 < 0.01 NS
Wk 30 1813 1669 1650 1658 25.53 < 0.01 NS

137
1
A Control diet was fed ad libitum between 20 and 29 weeks-old. Between 20 and 23 weeks-
old, 3 sequential diets were provided in the morning for E and in mid-afternoon for P:
E40/P60: 50 g of the energy fraction (E) and 75 g of protein fraction (P); E50/P50: 62.5 g of E
and 62.5 g of P; E60/P40: 75 g of E and 50 g of P. Between 24 and 29 weeks of age, birds
previously with E40/P60, E50/P50 or E60/P40 were offered the same sequential diet,
E50/P50.
2
Probability mentioned matches the effect of control diet vs. all sequential diets.
3
The experimental unit for the feed ad nutrient consumption and feed conversion ratio was
the group of 5 contiguous cages (N = 6) due to the common feeding trough.
4
The analysis was based on individual values from cages. One bird’s mortality was recorded
in the E60/P40 diet.
a,b,c
: values within the same row not sharing a common superscript letter differ significantly at
P < 0.05. A,B,C: have a trend at P < 0.10, not significantly at P > 0.10.

138
Table 4. Effect of feeding systems between 20 and 23 weeks-old on performances of laying
hens between 20 and 29 weeks-old under hot condition (Experiment 2)

Control vs.
Control
E40/P60 E50/P50 SEM E40/P60 vs.
diet1
E50/P50
N2 (N = 4) (N = 4) (N = 4)
Total feed intake (g/b/d)
Wk20-23 103.7 96.8 98.8 1.92 < 0.10
Wk24-29 111.7 107.4 109.4 1.62 NS
Energy fraction intake (g/b/d)
Wk20-23 39.7 46.6 0.93 < 0.01
Wk24-29 48.2 49.4 0.81 NS
Protein fraction intake (g/b/d)
Wk20-23 57.0 52.1 1.47 NS
Wk24-29 57.1 59.2 0.69 < 0.05
ME intake (MJ/b/d)
Wk20-23 1.20a 1.09b 1.14ab 0.02 < 0.05
Wk24-29 1.19 1.12 1.20 0.07 NS
CP intake (g/b/d)
Wk20-23 17.9 17.9 17.4 0.37 NS
Wk24-29 17.7 17.6 18.8 1.01 NS
Feed conversion ratio
Wk20-23 1.95 1.82 1.87 0.07 NS
Wk24-29 1.94 1.93 1.90 0.06 NS
N3 (N = 20) (N = 20) (N = 20)
Egg production (%)
Wk20-23 96.4 94.5 96.1 2.13 NS
Wk24-29 95.1 92.4 96.3 1.27 NS
Egg weight (g)
Wk20-23 56.3 55.6 54.7 0.85 NS
Wk24-29 61.3 60.3 59.3 0.87 NS
Egg mass (g/b/d)
Wk20-23 54.4 52.7 52.8 1.61 NS
Wk24-29 58.3 55.8 57.2 1.32 NS
Egg yolk (%)
Wk20-23 21.9 21.6 21.0 0.31 NS
Wk24-29 23.4 22.7 22.6 0.34 NS
Egg albumen (%)
Wk20-23 68.2a 68.7ab 69.4b 0.34 < 0.05
Wk24-29 66.8 67.3 67.0 0.37 NS
Egg shell (%)
Wk20-23 9.9 9.7 9.6 0.17 NS
Wk24-29 9.8A 10.1AB 10.3B 0.14 < 0.10
Body weight (g)
Wk 19 1565 1575 1590 22.19 NS
Wk 23 1652a 1550b 1577ab 25.78 < 0.05
Wk 30 1701a 1571b 1624ab 28.89 < 0.01

139
1
A Control diet was fed ad libitum between 20 and 29 weeks-old. Between 20 and 23 weeks-
old, 3 sequential diets were provided in the morning for E and in mid-afternoon for P:
E40/P60: 50 g of the energy fraction (E) and 75 g of protein fraction (P) E50/P50: 62.5 g of
(E) and 62.5 g of (P) ; E60/P40: 75 g of (E) and 50 g of (P). Between 24 and 29 weeks of age,
birds previously with E40/P60, E50/P50 or E60/P40 were offered the same sequential diet:
E50/P50.
2
The experimental unit for the feed consumption and feed conversion ratio was the group of 5
contiguous cages (N = 4) due to the common feed trough.
3
Analysis was based on individual values from cages.
a,b,c
: values within the same row not sharing a common superscript letter differ significantly at
P < 0.05. A,B,C: have a trend at P < 0.10, not significantly at P > 0.10.

140
Table 5. Effects of feeding systems between 20 and 23 weeks-old on weight of digestive
organs and abdominal fat at 30 weeks of age in experiment 1 and 2

Small Abdominal
% BW Gizzard Pancreas Liver Spleen
intestine fat
Experiment 1
Control diet N=10 1.45 1.95 0.19 2.91 0.085 2.7
E40/P60 N=10 1.66 2.15AB 0.20 2.81 0.084 1.4
E50/P50 N=10 1.63 1.95A 0.20 2.89 0.093 1.4
B
E60/P40 N=10 1.63 2.18 0.21 2.79 0.091 1.9

Control vs. Sequential* < 0.01 NS NS NS NS < 0.01


E40/P60 vs. E50/P50 vs. E60/P40 NS < 0.10 NS NS NS NS

Experiment 2
Control diet N=10 1.45 1.70a 0.19A 2.68 0.100 3.1a
E40/P60 N=10 1.67 1.89b 0.20AB 2.80 0.077 1.3b
E50/P50 N=10 1.55 1.88ab 0.2 B 2.82 0.103 1.7b
SEM 0.07 0.06 0.006 0.11 0.007 0.28
Control vs. E40/P60 vs. E50/P50 NS < 0.05 < 0.10 NS < 0.10 < 0.01

A Control diet was fed ad libitum between 20 and 29 weeks-old. Between 20 and 23 weeks-
old, 3 sequential diets were provided in the morning for E and in mid-afternoon for P:
E40/P60: 50 g of the energy fraction (E) and 75 g of protein fraction (P) E50/P50: 62.5 g of
(E) and 62.5 g of (P); E60/P40: 75 g of (E) and 50 g of (P). Between 24 and 29 weeks of age,
birds previously with E40/P60, E50/P50 or E60/P40 were offered the same sequential diet:
E50/P50.
* Probability mentioned matches the effect of control diet vs. all sequential diets.
a,b,c
: values within the same column not sharing a common letter differ significantly at P <
0.05; NS, not significantly at P > 0.10. A,B,C: significantly at P < 0.10.

141
Figure 1. Effects of feeding systems between 20 and 23 weeks-old on estimated body
composition (protein proportion and fat proportion) at 19, 23 and 30 weeks-old for laying
hens fed between 19 and 30 weeks of age with control or a sequential combination of energy
and protein fractions (Exp. 1)

Evolution of fat Evolution of protein

0.18
0.182
0.180

0.17
0.178

(Proportion of BW)
(Proportion of BW)

0.16
0.176

Control
0.174

E40/P60
E50/P50
E60/P40 0.15
0.172

0.14

(a)
0.170

(b)
Diet: p < 0.05 Diet: p < 0.05
Evolution: p < 0.01 Evolution: p < 0.01
0.168

Diet × Evolution: p < 0.01 Diet × Evolution: p < 0.01


0.13

Wk19 Wk23 Wk30 Wk19 Wk23 Wk30

142
Figure 2. Effects of feeding systems between 20 and 23 weeks-old on estimated body
composition (protein proportion and fat proportion) at 19, 23 and 30 weeks-old for laying
hens fed with control or a sequential combination of energy and protein fractions between 19
and 30 weeks of age under high temperature (Exp. 2)

Evolution of fat Evolution of protein

0.182
0.18

(a)
Diet: p > 0.10

0.180
Evolution: p < 0.01
0.17

Diet × Evolution: p < 0.01

0.178
(Proportion of BW)
(Proportion of BW)

0.16

0.176

Control
0.174

E40/P60
0.15

E50/P50
0.172
0.14

(b)
0.170

Diet: p > 0.10


Evolution: p < 0.01
Diet × Evolution: p < 0.01
0.168
0.13

Wk19 Wk23 Wk30 Wk19 Wk23 Wk30

143
144
Chapitre 5 : Modulation des apports
énergétiques et protéiques en alimentation
séquentielle chez les poules pondeuses

M. Traineau, I. Bouvarel, C.Mulsant, L. Roffida3, C.Launay and P. Lescoat Modulation of


energy and protein supplies in sequential feeding in laying hens. Soumis au journal Animal en
janvier 2014

145
146
Résumé Publication 3 :

Lors de l’ensemble des travaux précédents, les poules en alimentation séquentielle


(AS) recevaient le matin un aliment riche en énergie et pauvre en protéines (E) et l’après midi
un aliment riche en protéine/calcium et pauvre en énergie (P). Cependant, il n’y a pas encore
d’évidence claire quant aux cinétiques d’apports des nutriments au cours d’une journée de
ponte (la certitude concerne tout de même l’apport de calcium l’après midi). Afin d’évaluer
l’effet de la modulation des apports au cours de la journée, six régimes ont été offerts : : E0P0
/ E0P0; E0P0/E-P-; E+P+ / E-P-; E+P- / E-P+; E0P+ / E0P- and E+P0 / E-P0 avec E+qui
représente un haut niveau d’énergie, E0 un niveau moyen et E- un niveau bas d’énergie
(même définition pour les protéines P). Pour l’ensemble de ces régimes, le calcium
particulaire était apporté dans la fraction de l’après midi. 168 poules étaient placées dans des
cages individuelles de 20 à 39 semaines d’âge dans deux cellules différentes.
Conformément aux observations des travaux présents, et malgré la modulation des
apports en énergie et en protéines, la consommation de la fraction du matin était
systématiquement inférieure à celle de l’après midi. Les poules qui consommaient le régime.
E0P0 / E-P- n’ont pas suffisamment élevé leur niveau d’ingestion afin de couvrir leur besoins
nutritionnels. Les niveaux d’ingestion d’énergie et de protéines inférieurs n’ont pas permis à
ces poules de conserver un niveau de production équivalent aux poules qui consommaient les
cinq autres régimes. Ces derniers, malgré la diversification des apports nutritionnels, n’ont
présenté aucune différence dans le niveau de nutriments ingérés. Cependant, concernant la
production, les trois régimes E+P- / E-P+; E0P0 / E0P0; E+P0 / E-P0 ont donné les meilleures
performances par rapport aux régimes E+P+ / E-P- et E0P+ / E0P-. Ces résultats suggèrent
donc que l’énergie devrait plutôt être apportée dans la fraction du matin tandis que le niveau
de protéines peut être apporté de façon continue au cours de la journée.

147
148
Modulation of energy and protein supplies in sequential feeding in laying hens

M. Traineau1,3,a, I. Bouvarel2, C.Mulsant3, L. Roffidal3, C.Launay3 and P. Lescoat4

1
INRA, URA (UR83), F-37380 Nouzilly, France
2
ITAVI, F-37380 Nouzilly, France
3
INZO, Service R&D Pondeuses, Rue de l'Eglise, CS 90019, 02400 Chierry, France
4
AgroParisTech, 16 rue Claude Bernard, 75231 Paris Cedex, France

a
Corresponding author: M. Traineau. E-mail: maxime.traineau@gmail.com

Running head: Sequential feeding in laying hens

149
Abstract

Sequential feeding consists in splitting energy (E) and protein / calcium (P) fractions

temporally, improving the feed conversion ratio (FCR) of hens compared to a

continuous distribution during the day. In previous work, the fraction (with a low level

of protein) was provided in the morning while the P fraction (with low level of energy)

was given in the afternoon. However, there is no clear evidence that a requirement in

energy or proteins are connected to these distribution sequences, whereas the

requirement for calcium is known to be required in the afternoon for hens laying in the

afternoon. To evaluate the effects on performances of the modulation of energy and

protein supplies in sequential feeding, six different sequential treatments were

offered: E0P0 / E0P0; E0P0/E-P-; E+P+ / E-P-; E+P- / E-P+; E0P+ / E0P- and E+P0 /

E-P0 where E+ represents a high energy level, E0 a moderate one and E- a low one

(with the same meaning for P regarding protein supply). Afternoon fractions were

provided with particulate calcium. One hundred and sixty-eight Hendrix hens were

housed in individual cages from 20 to 39 weeks of age in two different rooms. Feed

intake in the morning and afternoon fractions, egg production, egg weight, body

weight and weight of digestive organs were recorded. E0P0/E-P- gave the lowest

laying performance accordingly with the lowest level of ingested energy and protein.

All diets have a similar body weight. For the other diets, two of them provided

intermediate laying performances (E+P+ / E-P-; E0P+ / E0P-) whereas three diets

gave the best ones ( E+P- / E-P+; E0P0 / E0P0; E+P0 / E-P0). These results

suggested that supplying protein required for hens continuously throughout the day

or in the afternoon fraction might have a similar impact while supplying energy in the

morning seems to be relevant in sequential feeding.

Keywords Laying hens, sequential feeding, energy, protein, laying performances

150
Implications

Improving laying hen performance may be achieved by maximizing nutrient utilization

from current feedstuffs. Supply of various energy and protein in sequential feeding

could reduce the amount of food ingested while maintaining an equivalent level of

production. Knowledge of the specific needs in energy and protein during the day will

allow optimizing daily intake. This study can increase the robustness of the system to

make it usable on a field scale and to ensure the relevant nutrient supply sequence.

Introduction

Several studies on Sequential Feeding (SF) have yielded good results through a

meaningful improvement of the FCR while maintaining similar performances to hens

fed classically with continuous diets. The morning fractions were rich in energy, while

the afternoon ones were rich in proteins and calcium (Umar Faruk et al., 2010 and

2011; Traineau et al., 2013). However in these SF studies, hens were feeding more

in the morning than in the afternoon, even though they ate the same quantity of

energy than the control on the day-scale. This morning low feed intake might be due

to the regulation of energy concentration (Gunawardana et al., 2009; Pérez-Bonilla et

al., 2012). Increased apparent metabolizable energy (AMEn) content of the diet

reduced feed intake to maintain an optimal level of energy ingested (Valkonen et al.,

2008). Reduced feed intake while maintaining identical laying performance, induced

significant improvement of FCR (Grobas et al., 1999; Valkonen et al., 2008). These

results might support the improvement in FCR for hens feed sequentially with a high

energy fraction in the morning and a high protein fraction in the afternoon. However,

it is difficult to establish that hens need energy mostly in the morning and protein in

the afternoon while a specific appetite for calcium in the afternoon for eggshell

151
fabrication (Mongin and Sauveur, 1974; Chah and Moran, 1987). To improve SF

system robustness and to implement it on a commercial scale, it has to be more

flexible. Traineau et al., (2013) showed that the energy fraction could be formulated

with wheat or corn or with a mix of these two cereals leading to intermediate results

between whole wheat sequential and usual continuous diet. Regarding nutrient

supply, Penz and Jensen, (1991) showed that a feeding timetable could be a solution

in using low-protein diets in layer rations while maintaining similar laying

performances. When hens were fed a “high-protein” diet (16% CP) from 04:00 am to

08:00 am and from 2:00 pm to 8:00 pm and a “low-protein” diet (13% CP) from 08:00

am to 2:00 pm, no difference in performance with control hens (16% CP) were

observed. However, hens fed continuously the “low-protein diet” (13% CP) or fed the

“low-protein” diet from 04:00 am to 08:00 am and from 2:00 pm to 8:00 pm, and the

“high-protein” diet from 8:00 am to 2:00 pm, laid lower egg weights than the control

group. This study supported that a provision of protein at given times influences on

the performances and has strengthened the hypothesis that hens need more protein

after oviposition in order to meet the protein requirements for the synthesis of the

albumen even though secretion might not be fully related to protein synthesis.

However, Keshavarz, (1998) had a similar experimental design (positive control 16%

CP, negative control 13% CP, and two sequential diets combining 13% and 16% CP

fractions), and concluded that optimum performance may be expected as long as the

daily protein intake is adequate and satisfies the daily needs of laying hens.

Therefore, additional experiments are required to further investigate SF. The aim of

this study was to modulate the energy and protein fractions and nutritional contents

in SF for laying hens to determine their impacts on intake, growth and production of

152
hens due to the sequential supply of nutrients. This could help to stress the best

nutrient inputs in the day.

Materials and methods

Birds and housing

A total of 168 Hendrix layer hens were housed in a two-tier battery having

individual cages equipped with a nipple drinker.For the duration of the experiment,

the temperature was maintained between 18 and 22 °C. All hens had individual

feeders and have been randomly assigned in battery cages. The birds were

habituated to SF with a mix of whole wheat and a balancer from weeks 16 to 18. The

photoperiod was 12L: 12D at week 16 and reached 16L: 8D at week 18 and

remained at this level until the end of experiment. Lights were on at week 16 from

05:00h to 17:00h and from 04:00h to 20:00h from weeks 18 to 39.

The experimental period ran from weeks 19 to 39. Birds were housed in

different rooms within the same building and kept in the same cages from week 16 to

39. Birds used in habituation period were divided into 6 groups of 12 birds in the first

room and into 6 groups of 16 birds in the second room. The difference between the

two rooms was mainly related to the size of the cages: in the first room cages were

45cm width × 62cm depth× 80cm height and in the second room 25cm width × 48cm

depth× 48cm height. Moreover, the second room had a meaningful environmental

pressure. This room was progressively infected with red mites for the duration of the

experimental period, which altered whole performance of hens, despite treatment

against mites.

Experimental treatments

153
Diets fed during the experimental period are described in Table 1. Dietary

contents of previous experiments on SF in our lab were based on the utilization of

different raw materials (wheat, maize, …, Traineau et al., 2013) whereas the present

study is focused on SF fractions nutrients contents. Diets were distributed

sequentially. Birds were offered 62.5 g in the morning and beginning of the afternoon

(8:30 am to 3:30 pm) and 62.5 g in the late afternoon and early morning (3:30 pm to

8:00 am and 4:00 am to 8:30 am), 125 g/d being 105% of the recommended diet

supply. For all diets, calcium is provided in coarse form, in the fraction distributed in

the afternoon. No diet was formulated with a low energy level (E-) in the morning

fraction. This fraction was systematically less consumed than the afternoon one for

every previous study, inducing overall shortage in ME intake, since the high energy

fraction was systematically supplied in the morning. To modulate the daily intakes’

dynamic of energy and protein, six sequential diets were offered: E0P0/E0P0 (2601

Kcal/Kg; 16,5 CP/2601 Kcal/Kg; 16,5 CP), E0P0/E-P- (2601 Kcal/Kg; 16.5CP / 2303

Kcal/Kg; 13.5 CP); E+P+ / E-P- (2901 Kcal/Kg; 19.5 CP; 2303 Kcal/Kg; 13.5 CP);

E+P-/E-P+ (2905 Kcal/Kg; 13.5 CP/ 2302 Kcal/Kg; 19.6 CP); E0P+/E0P- (2599

Kcal/Kg; 19.6 CP / 2597 Kcal/Kg; 13.6 CP) and E+P0/E-P0 (2903 Kcal/Kg; 16.4 CP /

2302 Kcal/Kg; 16.5 CP). Diets were formulated according to INRA recommendations

and proposed nutritional contents were based on the hypothesis of 50/50 intakes of

the morning and afternoon fractions (Sauvant et al., 2002). Water was given ad

libitum throughout the experimental period.

Measurements

Particle size profile was measured. Four different sieves were used to

characterize the particle size (2.5 mm, 1.6 mm, 1.18 mm, 1.18 mm). Feed intake was

individually recorded every week for the morning and afternoon fractions. Egg

154
production was recorded on a daily basis. Egg weight was recorded twice a week by

weighing all eggs produced on the measuring day. Weights of egg components (yolk,

albumen, shell) were determined on the daily samples every 4 weeks, starting at

week 22. Shells were washed and dried for 12 hours in a drying oven at 70°C and

then weighed.

Body weight was recorded at 16, 19, 23, 27, 31, 35 and 39 weeks of age.

Body composition was predicted with a noninvasive method based on bioelectrical

impedance analysis (BIA) at 19, 27 and 35 weeks of age. The technique has been

developed by the company INZO°2 and consists in the prediction of body composition

by an instrument of impedance which was calibrated through the chemical analysis of

birds. Impedance measurements of resistance (Rs) and reactance (Xc) are made

using two electrodes placed on the animal’s legs and distributing a very low intensity

current. It is assumed that the electrical impedance of the biological organism is

indicative of the highly conductive fat-free component (Berg et al., 1994). The

electrical impedance of a tissue depends on its fluid and electrolyte content (Rutter et

al., 1998) and therefore on the electrical properties of different tissues in connection

with their relative proportions of muscle, lipid, water and ash. For the duration of a full

day, week 27, feed intake kinetics of hens was measured in the second room. Every

hour the feeders were weighed to set the amount of food ingested by the hens during

this period. This operation was performed for the two fractions (morning and

afternoon) throughout the light period from 04:00 to 20:00. Metabolizable energy

(ME; Kcal/b/d) and protein (g/bird/d) intakes were calculated as a product of feed

intake and ME and protein contents of the experimental diets respectively

2
INZO° SAS rue de l’Eglise-BP 50019 02407 Chierry Cedex
155
At the end of the trial, 10 birds per treatment in the second room were

randomly selected, weighed and injected with Na Pentobarbital solution (1 ml/kg).

The weight of the digestive organs was recorded to assess the effect of feeding

system on these organs. The abdominal cavity was opened and the digestive tract

dissected and separated into gizzard, small intestine, pancreas and liver. Small

intestine was separated into duodenum, jejunum and ileum. Each segment was first

emptied prior to weighing. Gizzard was placed into an iced container (-4°C) for 24h to

facilitate the removal of the surrounding fat before being emptied and weighed.

For the evaluation of the apparent digestibility of the hens, Titanium dioxide was

added at a rate of 5 g/kg as a dietary marker in the two fractions. Chemical analyses

were performed for TiO2 concentrations in each fraction of the diet and in the faeces.

Analysis of TiO2 was performed by colorimetric method. In the presence of sulphuric

acid, TiO2 and H2O2 cause a yellow coloration read by absorbance at 410 nm

(TECAN, Infinite200, Austria). This method was developed by Short et al. (1996).

Apparent digestibility was calculated as follow:

TTAR = 1 – [(Nutrientfaeces/Nutrientdiet) x ( TiO2diet/ TiO2faeces)] x 100

Where TTAR is total tract apparent retention, Nutrientfaeces and TiO2faeces are the

concentrations of dietary components and of TiO2 in the excreta, Nutrientdiet and

TiO2diet represent the concentrations of the same dietary components and the TiO 2 in

the feeds. Concentrations in the sequential diets were obtained by combining the

actual quantities of ingested morning and afternoon fractions with their respective

content of nutrient and Titanium dioxide. At 32 weeks of age, blood sampling in the

wing vein was used to define the level of hematocrit as an indicator of the effect of

red mites on the health of the hens.

Statistical analysis

156
Data collected during the experiment was analyzed based on 2 different periods:

during the increase of lay (from week 20 to week 26) and after the peak (from week

27 to week 39). Averaged values from each period from cages were analyzed using

StatView (version 5, SAS Institute Inc., Cary, NC). A 1-way ANOVA was used to test

treatment and room effect on all measured parameters. If the ANOVA was significant

(p<0.01: **, p<0.05: *, p<0.1: obtained p-value, p>0.1: NS), a post-hoc Bonferroni test

was used, differences between diets on the same line are illustrated with different

letters (a, b, …) if p<0.05 and with capital (A, B, …) if p<0.1

Results

Diet and room effect on egg production, body weight, body composition and FCR

(table 2)

Week 20-26:

No difference due to the diet was observed for the laying performances (egg

weight, egg production and egg mass) and for the FCR. In addition to this, no

difference observed for initial body weight at week 19 and body composition at week

19. Body Weight Gain (BWG) (week 20 to week 27) was significantly affected by the

diets (p=0.02). Hens with E+P+ / E-P- had lower gain than E+P- / E-P+. There was a

difference between the two rooms for egg production (p = 0.02) and egg mass (p =

0.01). An interaction between diet and room was observed (p = 0.04). This interaction

was related to hens consuming E0P0/E0P0 in the second room having higher BWG

(78.18 g) than those in the first room (60.24 g), whereas for all other birds, hens in

the second room had lower BWG than those of the first one.

Week 27-39:

157
Egg production, egg weight, egg mass, body weight (BW) (week 27 and 39),

BWG and FCR were reduced for hens in the second room compared to hens in the

first one. Overall the diet effect was observed on egg performance, FCR, BWG and

BW at 39 weeks of age. However few pairwise comparisons were significant:

statistical analyzes can show degraded performance of hens that consumed the diet

E0P0 / E-P- : lower egg weight than E+P0 / E-P0 (no difference with the four other

diets) ; lower egg mass than E0P0 / E0P0 (no difference with the four other diets);

highest FCR compared to all other diets (except E+P+ / E-P-). There was no

difference for body weight at 27 weeks of age but hens with E+P0 / E-P0 had lower

body weight gain than hens with E+P- / E-P+ (no difference with the four other diets).

There was no interaction between room and diet for all measured parameters. There

was neither a diet nor room effect for the egg components. For body proportion in

protein and fat, an effect of age was observed through a repeated 1-way ANOVA.

There was no diet effect on fat and protein compositions on weeks 19, 27 and 35.

After hens’ slaughtering at 39 weeks of age, main digestive organs and

abdominal fat were weighed. However, there were no significant differences between

diets on any tested values. There was no diet effect on the apparent digestibility.

However the values of apparent digestibility performed on sequentially fed hens,

were similar to values obtained on continuously fed hens. The measure of hematocrit

on week 32 underlined a difference between the two rooms: 30.5 % in the first room

compared to 25.4 % in the second one.

Diet and room effects on feed intake, ME intake, CP intake (table 3)

Week 20-26:

158
All parameters were affected by the room effect and degraded in the second

room. There was no diet effect on total and afternoon feed intakes. However, diet

affected morning feed intakes with a lower level for E0P+ / E0P0- and E0P0 / E0P0

compared to E+P- / E-P+. Diets affected ME intake. For total ME intake, hens with

E+P0 / E-P0 ate more energy than hens with E0P0 / E-P-. For the morning ME

intake, all birds with high energy in the morning (E+P+ / E-P-, E+P- / E-P+ and E+P0

/ E-P0) ate more energy compared to birds with lower levels on energy in the

morning (E0P0 / E-P-, E0P+/E0P- and E0P0 / E0P0). Concerning the afternoon ME

intake, hens with an E- fraction in the afternoon ate less energy than hens with the

two other diets (E0P+ / E0P- and E0P0 / E0P0). For total CP intake, hens with E0P0

/ E-P- ate less protein than all other hens. For the morning CP intake, three different

groups were observed. Hens with P+ in the fraction of the morning (E+P+ / E-P- and

E0P+/E0P-) had the highest level of CP intake, hens with P0 in these fractions in the

morning (E0P0 / E-P-, E0P0 / E0P0 and E+P0 / E-P0) have an intermediate level of

CP intake and hens with P- in the morning (E+P- / E-P+) had the lowest level of CP

intake.

Week 27-39:

All parameters were affected by the room effect and degraded in the second

room. There was no diet effect on feed intakes. For the total ME intake, hens with

E0P0/E-P- ate less energy than all other birds. For ME intake in the morning, all hens

with high energy in the morning (E+P+ / E-P-, E+P- / E-P+ and E+P0 / E-P0) ate

more energy compared to birds with lower level on energy in the morning (E0P0 / E-

P-, E0P+ / E0P- and E0P0 / E0P0). For the afternoon ME intake, all birds with an E-

fraction in the afternoon (E+P+ / E-P-, E+P- / E-P+, E+P0/E-P0 and E0P0/E-P) ate

less energy than hens with the two other diets (E0P+ / E0P- and E0P0 / E0P0).

159
Concerning CP intake, for total CP intake all hens ate more protein than hens with

E0P0/E-P-. For the morning CP intake, three different levels were observed: a high

level of protein intake for E+P+ / E-P- E0P+ / E0P- and E0P0 / E0P0, an intermediate

level for E+P0/E-P0 and E0P0/E-P- and the lowest level for E+P- / E-P+. For CP

afternoon intake, three levels were still reported: the highest for E+P- / E-P+, an

intermediate one for E+P0 / E-P0 and a low level for E+P+ / E-P-, E0P0 / E0P0,

E0P0 / E-P- and E0P+ / E0P-. Concerning feed intake dynamics (figure 1), there was

no diet effect on the rate of eating. However this figure allows us to highlight the daily

shape of ingestion induced by SF: feed intake was induced by the diet distribution

and hens were expecting the next diet due to the learning of the timetable of the feed

supply.

Discussion

The main objective of this work was to study the influence of different kinetics

of nutrient supply in SF on performances of laying hens. Only sequential diets were

given. One diet brought E0P0 in the morning and E0P0 with Ca in the afternoon

which may mimic a continuous diet regarding energy and protein supply. However,

the two fractions were contrasted in their raw materials content and their particle size.

These differences might explain that for the 20-26 week period, hens with E0P0 in

the morning (19% of large particles and 49% fines) consumed less than hens that

had E0P0 + Ca at the same time (29% of large particles and 31% fine). Indeed,

coarse particle induces an increase in intake compared to a thinner diet (Nir et al.,

1990; Safaa et al., 2009). For the first room, with a low environmental pressure, rate

of lay and egg mass were similar to those observed in SF in previous studies (Umar

Faruk et al., 2010; Traineau et al., 2013). However, FCR (+ 5.1%) and egg weight (-

160
5.3%) were different than those observed by Traineau et al., (2013) may be due to a

difference in total protein intake (17.5 in our trial compare to 19.04 for the study of

Traineau et al., (2013)). Summer and Lesson, (1985) showed that egg size could be

altered by 1.3 g when the protein intake varied between 17 and 18.9 g.

For the duration of the experiment, environmental pressure increased in the

second room. From weeks 20 to 26, feed intake levels were lower in the second

room compared to the first one; while egg weight, weight gain and FCR were not

impacted. From weeks 27 to 39, difference between the two rooms strongly

increased. Red mites’ pressure can lead to decreases in feed consumption and

reduce eggs quality and production (Chauve et al., 1998; Lesna et al., 2009). To

implement the evidence of the negative effect of these red mites, measure of

hematocrit on week 32 showed a strong difference between the two rooms (30.5 % in

the first room compared to 25.4 % in the second one). However, the ranking between

the diets was similar in the two rooms, indicating that sequential diets were impacted

in the same order of magnitude whatever their contents.

One question was to determine the impact of nutrients distribution on hens’

responses. First of all, regardless of the protein or energy timetable supply, a higher

intake was observed for the afternoon fraction compared to the morning one as

obtained by Umar Faruk et al., (2010) and Traineau et al., (2013). Despite contrasted

energy and protein supply timetables, hens were not affected on the timing of lay.

However, energy and protein intakes’ dynamics are different during the day between

diets. The afternoon fraction wasn’t energy-rich (no E +). However, as dietary energy

level increases, chickens satisfy their energy needs through feed intake reduction

(Plavnik et al., 1997; Nahashon et al., 2005). The low or medium energy level in the

afternoon fractions could therefore increase hens’ consumption due to their need for

161
energy. The difference in fractions intakes might also take into account the specific

and requirement for Ca in the afternoon for eggshell formation (Chah and Moran,

1985).

A first conclusion is that a diet too diluted (E0P0 / E-P-) does not yield good

results. Hens do not manage to increase their intake to express their production

potential: E0P0 / E-P- has the lowest level of ME and CP content or intake. This low

level induced deficiencies in laying hens and supported their poor performance.

Indeed, Shim et al. (2013) showed that the protein level does not seem to affect feed

intake between weeks 19-34, which is consistent to our trial. It led to a similar feed

intake for all diets and therefore a lack of protein for E0P0 / E-P-, implying low laying

rate, egg weight and egg mass as observed by Shim et al., (2013). However, no

difference for body weight and body composition was observed between all diets.

One hypothesis was that the nutrients are used in priority for growth and

maintenance and only after for egg production. Heywan, (1940) highlight that a low

restriction of diet intake of 87% of the diet offered induced a decrease in the total

average number of eggs produced but did not affect body weight. Nevertheless, a

high feed restriction to 70% of ad libitum reduced egg production, egg weight and

body weight (Jackson, 1970; Benyi et al., 1981).

The other diets could help to determine the best supply time to bring energy

and protein fractions. First on ME and CP intakes, there were no significant

differences between all diets except E0P0/E-P-. However, hens have not eaten the

energy and protein according to the same dynamics during the day. In our trial,

performances are not strictly similar depending on different times of energy supply

and protein throughout the day. The following ranking on laying performances (egg

weight, egg mass and FCR) can be proposed: best: E+P- / E-P+; E0P0 / E0P0; E+P0

162
/ E-P0; middle: E+P+ / E-P-; E0P+ / E0P-; low: E0P0 / E-P-. However, this hierarchy

of diets is not true for BW or BWG. For the “best” group, there is no P- in the

afternoon and no P + in the morning. Therefore it might be relevant to supply the

quantity of protein required for hens, continuously throughout the day (P0 in the

morning and afternoon), or concentrated in the fraction of the afternoon according to

Keshavarz, (1998). He concluded that hens need an adequate level of protein for

optimum performances regardless of the time of distribution while Penz and Jensen,

(1991) concluded not to concentrate protein in the morning distribution. Regarding

energy, conclusions are not obvious. However, supplying energy in the morning

seems to give good results in SF in our work, as previously observed (Umar Faruk et

al., 2010, 2011; Traineau et al., 2013).

Acknowledgements

We thank Philippe Didier and Michel Couty (INRA) for their technical assistance. We

are grateful to the experimental unit (UE PEAT) for its help in the set up of the

experiment. Scientific and financial support of INZO° (France) was highly

appreciated. Barbara Konsak was thanked for the English corrections.

References

Benyi K, Akinokun O. and Lebbie S H B 1981. Effects of feed restriction during the

laying period on the performance of two strains of mature caged. The Journal of

Agricultural Science 96, 195-201

Berg EP and Marchello MJ 1994. Bioelectrical-impedance analysis for the prediction

of fat-free mass in lambs and lamb carcasses. Journal of Animal Science 72,

322-329.

163
Chah CC and Moran ET 1985. Egg characteristics of high-performance hens at the

end of lay when given cafeteria access to energy, protein, and calcium. Poultry

Science 64, 1696-1712.

Chauve C 1998. The poultry red mite Dermanyssus gallinae: current situation and

future prospects for control. Veterinary Parasitology 79, 239-245.

Grobas S, Mendez J, De Blas C and Mateos GG 1999. Laying hen productivity as

affected by energy, supplemental fat, and linoleic acid concentration of the diet.

Poultry Science 78, 1542-1551.

Gunawardana P, Roland DA and Bryant MM 2009. Effect of dietary energy, protein,

and a versatile enzyme on hen performance, egg solids, egg composition, and

egg quality of Hy-Line W-36 hens during second cycle, phase two. Journal of

Applied Poultry Research 18, 43-53.

Heywang BW 1940. The effect of restricted feed intake on egg weight, egg
production, and body weight. Poultry Science 19, 29-34.
Jackson N 1970. The effect of restricting the individual daily energy intake of caged
layers on the efficiency of egg production. British Poultry Science 11, 93-102.
Keshavarz K 1998. Investigation on the possibility of reducing protein, phosphorus,

and calcium requirements of laying hens by manipulation of time of access to

these nutrients. Poultry Science 77, 1320-1332.

Lesna I, Wolfs P, Faraji F, Roy L, Komdeur J and Sabelis MW 2009. Candidate

predators for biological control of the poultry red mite Dermanyssus gallinae.

Experimental and Applied Acarology 48, 63-80.

Mongin P and Sauveur B 1974. Voluntary food and calcium intake by the laying hen.

British Poultry Science 15, 349-359.

Nahashon SN, Adefope N, Amenyenu A and Wright D 2005. Effect of dietary

metabolisable energy and crude protein concentrations on growth performance

164
and carcass characteristics of French guinea fowl broilers. Poultry Science

Journal 84, 337-344.

Nir I, Melcion JP and Picard M 1990. Effect of particle-size of sorghum grains on

feed-intake and performance of young broilers. Poultry Science 69, 2177-2184.

Penz AM and Jensen LS 1991. Influence of protein-concentration, amino-acid

supplementation, and daily time of access to high-protein or low-protein diets on

egg weight and components in laying hens. Poultry Science 70, 2460-2466.

Perez-Bonilla A, Novoa S, Garcia J, Mohiti-Asli M, Frikha M and Mateos GG 2012.

Effects of energy concentration of the diet on productive performance and egg

quality of brown egg-laying hens differing in initial body weight. Poultry Science

91, 3156-3166.

Plavnik I, Wax E, Sklan D, Bartov I and Hurwitz S 1997. The response of broiler

chickens and turkey poults to dietary energy supplied either by fat or

carbohydrates. British Poultry Science 76, 1000-1005.

Rutter K, Hennoste L, Ward LC, Cornish BH and Thomas BJ 1998. Bioelectrical

impedance analysis for the estimation of body composition in rats. Laboratory

Animals 32, 65-71.

Safaa HM, Jimenez-Moreno E, Valencia DG, Frikha M, Serrano MP and Mateos GG

2009. Effect of main cereal of the diet and particle size of the cereal on

productive performance and egg quality of brown egg-laying hens in early phase

of production. Poultry Science 88, 608-614.

Sauvant D, Perez J.-M, and Tran G 2004. Tables of composition and nutritional value

of feed materials. INRA Editions and AFZ. Wageningen Academic Publishers,

Paris, France.

165
Shim MY, Song E, Billard L, Aggrey SE, Pesti GM and Sodsee P 2013. Effects of

balanced dietary protein levels on egg production and egg quality parameters of

individual commercial layers. Poultry Science 92, 2687-96.

Short FJ, Gorton P, Wiseman J and Boorman KN 1996. Determination of titanium

dioxide added as an inert marker in chicken digestibility studies. Animal Feed

Science and Technology 59, 215-221.

Summers JD and Lesson S 1985. Commercial poultry nutrition, 3rd edition.

Nottingham University Press Editions, University of Guelph, Ontarion, Canada.

Traineau M, Bouvarel I, Mulsant C, Roffidal L, Launay C and Lescoat P 2013. Effects

on performance of ground wheat with or without insoluble fiber or whole wheat in

sequential feeding for laying hens. Poultry Science 92, 2475-2486.

Umar Faruk M, Bouvarel I, Meme N, Rideau N, Roffidal L, Tukur HM, Bastianelli D,

Nys Y and Lescoat P 2010. Sequential feeding using whole wheat and a

separate protein-mineral concentrate improved feed efficiency in laying hens.

Poultry Science 89, 785-796.

Umar Faruk M, Bouvarel I, Mallet S, Ali MN, Tukur HM, Nys Y and Lescoat P 2011. Is

sequential feeding of whole wheat more efficient than ground wheat in laying

hens? Animal 5, 230-238.

Valkonen E, Venalainen E, Rossow L and Valaja J 2008. Effects of dietary energy

content on the performance of laying hens in furnished and conventional cages.

Poultry Science 87, 844-852.

166
Table 1 Composition of experimental diets given to laying hens between 19 and 39 weeks of age with a combination of 62.5 g of morning
fraction between 8:30am and 3:30 pm and 62.5 g of afternoon fraction between 3:30 pm and 8:30 am (no light between 8:00 pm and 04:00 am).
167
E0P0 / E0P0 E0P0 / E-P- E+P+ / E-P- E+P- / E-P+ E0P+ / E0P- E+P0 / E-P0
E0P0 E0P0 Ca+ E0P0 E-P- Ca+ E+P+ E-P- Ca+ E+P- E-P+ Ca+ E0P+ E0P- Ca+ E+P0 E-P0 Ca+
morning afternoon morning afternoon morning afternoon morning afternoon morning afternoon morning afternoon
Ingredient (%)
Corn 31.1 30 31.1 28.0 33.0 28.0 39.7 23.2 27.0 33.1 38.0 25.0
Wheat 30.0 29.0 30.0 28.7 34.6 28.7 39.0 22.0 27.9 33.0 36.3 26.0
Soybean meal 9.5 17.0 9.5 4.0 17.0 4.0 6.0 19.0 16.5 13.5 13.0 11.0
Rapeseed meal 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0
Sunflower meal 10.0 10.0 10.0 3.7 10.0 0.3 9.1 10.0 4.8 10.0
Wheat bran 10.0 10.0 4.5 4.5 7.0 10.0 2.0
Calcium carbonate 1.57 15.36 1.57 15.35 15.35 15.24 15.38 15.3
Corn gluten 4.8 1.4 3.7 1.4 3.4 0.8 1.3 0.2 2.7
Dicalcium phosphate 1.01 1.34 1.01 1.19 1.15 1.19 1.1 1.23 0.97 1.35 1.17 1.21
Soybean oil 0.5 1.2 0.5 0.5 0.5 0.5 0.5 0.5 0.5 1.1 0.5 0.5
1
Premix 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0
DL-Methionine 0.17 0.17 0.17 0.12 0.20 0.12 0.15 0.21 0.22 0.15 0.18 0.16
L-Lysine 0.19 0.15 0.19 0.22 0.16 0.22 0.20 0.11 0.14 0.12 0.14 0.18
L-Tryptophane 0.01
L-Threonine 0.005 0.005 0.02
Nutritional values2
ME calculated (Kcal/Kg) 2601 2600 2601 2303 2901 2303 2905 2302 2599 2597 2903 2302
CP calculated (g/kg) 16.5 16.6 16.5 13.5 19.5 13.5 13.5 19.4 19.6 13.6 16.4 16.5
Lysine dig. (%) 0.72 0.72 0.72 0.59 0.85 0.59 0.59 0.85 0.86 0.60 0.71 0.72
Methionine dig. (%) 0.42 0.42 0.42 0.34 0.49 0.34 0.34 0.51 0.50 0.34 0.42 0.42
Threonine dig. (%) 0.51 0.53 0.51 0.42 0.62 0.418 0.42 0.64 0.62 0.42 0.52 0.53
Tryptophan dig. (%) 0.17 0.16 0.17 0.13 0.19 0.13 0.13 0.19 0.20 0.13 0.16 0.16
Calcium (%) 1.2 6.6 1.2 6.6 0.6 6.6 0.5 6.6 0.6 6.6 0.6 6.6
Total P (%) 0.65 0.52 0.65 0.59 0.59 0.59 0.56 0.62 0.67 0.50 0.59 0.60
Available P (%) 0.32 0.32 0.32 0.32 0.32 0.32 0.32 0.32 0.32 0.32 0.32 0.32
Coarse particles >2.5mm 19.3 26.3 19.3 16.5 31 16.5 29 21.6 22.2 22.5 28.3 22
Fine particles < 1.18 mm 48.7 28.9 48.7 52.6 28.5 52.6 31.2 44 37.9 32.2 26.7 35.2
168
1
Premix : Supplied per kg of diet at Vitamin A 750000 UI, Vitamin D 3 300000 UI, Vitamin E (tocopherol acetate 3a700) 1000 UI, Vitamin K3 100 mg, Zinc
10000 mg, Manganese 10000 mg, Copper 1000 mg, Iron 4000 mg, Iode 150.0 mg, Selenium 30.0 mg ; Canthaxanthin 1950.0 mg, ß-apocarotenoïde acid
2000.0 mg, lutein 3560.0 mg, zeaxanthin 200.0 mg, Cryptoxanthin 80.0 mg, Citric acid 400.00 mg, orthophosphoric acid 660.00 mg, Ethoxyquin 400 mg,
Propyl gallate (E310) 400 mg
2
Nutritional values were calculated out of Sauvant et al., (2002).
169
Table 2 Effects on egg production, feed conversion ratio (FCR), body weight and body composition of laying hens from 20 to 39 weeks old
offered a combination of 62.5 g of morning fraction between 8:30am and 3:30 pm and 62.5 g of afternoon fraction between 3:30 pm and 8:30am
(no light between 8:00 pm and 04:00 am). Fractions were a combination of Energy (E) and Protein (P) levels, with low (-), medium (0) or high
(+) values.
170
1
E0P0 / E0P0 / E+P+ / E+P- / E0P+ / E+P0 / 1 1 Diet x
Room 1 Room 2 SEM Diet Room
E0P0 E-P- E-P- E-P+ E0P- E-P0 Room
Week 20 to 26
Egg production (%) 82.4 79.4 86.0 80.7 79.3 82.8 85.0 79.3 1.1 ns * ns
Egg weight (g) 53.4 53.3 50.9 51.0 53.7 53.8 52.2 53.1 0.7 ns ns ns
Egg mass (g/hen per d) 43.1 40.8 43.8 41.1 41.9 43.8 44.1 41.2 0.5 ns ** ns
FCR 2.42 2.56 2.36 2.54 2.68 2.42 2.46 2.52 0.04 ns ns ns
Initial Body weight wk 19 (g) 1421 1399 1419 1394 1427 1412 1425 1403 7.0 ns ns ns
Body weight gain (g) 272 237 216 a 283 b 260 273 269 247 6.2 * 0.08 *
Body composition : % fat wk 19 11.4 11.3 11.3 11.0 11.6 11.2 11.3 11.3 1.71 ns ns ns
Body composition : % protein wk 19 18.3 18.3 18.3 18.3 18.2 18.3 18.3 18.3 0.32 ns ns ns
Week 27 to 39
Egg production (%) 98.1 A 94.7 B 96.8 98.1 A 95.9 95.4 98.2 95.2 0.4 * ** ns
Egg weight (g) 58.0 55.9 b 57.0 57.7 58.6 58.9 a 58.7 56.9 0.3 * ** ns
Egg mass (g/hen per d) 56.9 53 b 55.2 56.7 a 56.2 56.2 57.7 54.2 0.4 * ** ns
FCR 2.04 a 2.2 b 2.11 ab 2.07 a 2.07 a 2.06 a 2.06 2.12 ** ** ** ns
Body weight wk 27 (g) 1692 1636 1635 1677 1687 1685 1694 1650 8.6 ns * ns
Final Body weight wk 39 (g) 1699 1668 1699 1761 1733 1695 1771 1679 14.8 0.07 ** ns
Body weight gain (g) 68 33 64 81 a 45 9 b 78 29 6.8 * ** ns
Body composition : % fat wk 27 16.1 15.3 15.2 16.0 16.0 16.1 16.1 15.5 0.19 ns * ns
Body composition : % protein wk 27 17.3 17.4 17.4 17.3 17.3 17.3 17.3 17.3 0.38 ns ns ns
Body composition : % fat wk 35 16.8 15.2 16.5 16.6 16.9 16.4 17.3 15.7 0.24 ns ** ns
Body composition : % protein wk 35 17.0 17.3 17.0 17.1 17 17.1 16.9 17.1 0.46 ns ** ns
171
FCR = feed conversion ratio (feed intake/egg mass).
1
**P<0.01 ; * P<0.05; ns (non-significant) P>0.05 ; values within a row with common letters (a, b, c) differ significantly using Bonferroni–Dunnet test at 5%
significance level; values within a row with capital letters (A, B, C) differ significantly using Bonferroni–Dunnet test at 10% significance level.
172
Table 3 Effects on feed, ME and CP intakes of laying hens from 20 to 39 weeks old offered a combination of 62.5 g of morning fraction
between 8:30am and 3:30 pm and 62.5 g of afternoon fraction between 3:30 pm and 8:30am (no light between 8:00pm and 04:00 am).
Fractions were a combination of Energy (E) and Protein (P) levels, with low (-), medium (0) or high (+) values.
173
1
E0P0 / E0P0 / E+P+ / E+P- / E0P+ / E+P0 / Diet x
E0P0 E-P- E-P- E-P+ E0P- E-P0 Room 1 Room 2 SEM 1Diet 1
Room Room
Week 20 to 26
Total feed intake (g/b/d) 102.8 102.5 102.3 103.0 103.1 104.9 106.1 100.9 0.7 ns ** ns
Morning 48.3 b 48.5 ab 49.5 ab 52 a 47.7 b 50.7 ab 50.7 48.5 0.4 ** ** ns
Afternoon 54.5 54.0 52.8 51.0 55.4 54.2 55.4 52.4 0.4 0.06 ** ns
Total ME intake (kcal/b/d) 267 251 b 265 268 268 272 a 273 260 1.9 * ** ns
Morning 126 b 126 b 143 a 151 a 124 b 147 a 140 134 1.3 ** ** 0.07
Afternoon 141 b 125 a 122 a 117 a 144 b 125 a 133 126 1.3 ** ** ns
Total CP intake (g/b/d) 17 a 15.3 b 16.7 a 16.9 a 16.9 a 17.2 a 17.1 16.3 0.1 ** ** ns
Morning 7.9 c 8 c 9.6 a 7b 9.3 a 8.3 c 8.6 8.2 0.1 ** ** ns
Afternoon 9.1 c 7.3 a 7.1 a 9.9 b 7.6 a 8.9 c 8.5 8.1 0.1 ** ** ns
Week 27 to 39
Total feed intake (g/b/d) 115.7 115.9 116.1 116.8 115.8 115.4 118.3 114.2 0.6 ns ** ns
Morning 53.4 53.5 53.8 54.5 51.8 53.6 54.2 52.9 0.3 ns * *
Afternoon 62.3 62.4 62.3 62.3 64.0 61.8 64.1 61.3 0.4 ns ** ns
Total ME intake (kcal/b/d) 301 a 283 b 300 a 302 a 301 a 298 a 303 293 1.6 ** ** ns
Morning 139 b 139 b 156 a 156 a 135 b 156 a 149 145 1.2 ** * *
Afternoon 162 b 144 a 144 a 146 a 166 b 142 a 154 148 1.2 ** ** ns
Total CP intake (g/b/d) 19.1 a 17.2 b 18.9 a 19.4 a 18.8 a 19 a 19.1 18.5 0.1 ** ** ns
Morning 10.8 a 8.8 c 10.5 a 7.3 b 10.1 a 8.8 c 9.2 9 0.1 ** 0.06 0.06
Afternoon 8.3 a 8.4 a 8.4 a 12.1 b 8.7 a 10.2 c 9.9 9.5 0.1 ** * ns
174
175
1
**P<0.01 ; * P<0.05; ns (non-significant) P>0.05 ; values within a row with common letters (a, b, c) differ significantly using Bonferroni–Dunnet test at 5%
significance level; values within a row with capital letters (A, B, C) differ significantly using Bonferroni–Dunnet test at 10% significance level.Figure
captions
Figure 1 Effects of different diets on feed intake (g/h) during a whole day at 27 weeks old for laying hens fed between 19
and 39 weeks of age with a combination of 62.5 g of morning fraction between 8:30 am and 3:30 pm and 62.5 g of
afternoon fraction between 3:30 pm and 8:30 am (no light between 8:00 pm and 04:00 am).
1
Morning distribution
2
Afternoon distribution
176
Chapitre 6
Discussion Générale

177
178
En conditions expérimentales, le système d’alimentation séquentielle (AS) chez les
poules pondeuses a permis de maintenir un niveau de production équivalent à une
alimentation continue tout en réduisant l’ingestion. Les travaux d’Umar Faruk et al, (2010 ;
2011) sur l’AS mais également les premiers travaux de la thèse (Traineau et al., 2013) ont mis
en évidence que l’amélioration de l’indice de consommation était supérieure lorsque se
combinaient dans la fraction du matin du blé entier et un complément riche en protéines et
calcium l’après midi. Cependant il parait difficile de dépendre uniquement d’une seule
céréale, enlevant toute souplesse au système dans le cadre d’une forte volatilité du prix des
céréales, en particulier de forte hausse du prix du blé. La possibilité d’incorporer d’autres
matières premières en AS est donc nécessaire, le but restant d’obtenir des effets positifs
identiques à ceux du blé apporté sous forme de graine entière dans l’aliment du matin. En
effet l’apport de cette graine sous forme entière a un effet positif sur le développement du
gésier (Nir et al., 1990) en augmentant le temps de rétention du bolus alimentaire dans le
gésier (Vergara et al., 1989). Deux hypothèses ont été proposées. D’une part, l’inclusion de
fibres insolubles dans la fraction du matin pourrait avoir le même effet sur le temps de
rétention et sur le développement du gésier que du blé entier (Longe, 1983 ; Hetland et al.,
2005). D’autre part, l’incorporation de blé ou de maïs broyés grossièrement avec une petite
proportion de blé entier dans la fraction du matin en AS, permettrait également un bon
développement du gésier (Nir et al., 1990). Les résultats obtenus dans nos travaux ont montré
que l’utilisation de blé entier le matin donnait encore une fois les meilleurs résultats par
rapport aux autres régimes distribués de façon séquentielle qui donnent eux-mêmes des
meilleurs résultats par un IC inférieur que les poules nourries avec un aliment complet.
L’utilisation de céréales broyées grossièrement ainsi que de fibres insolubles a aussi permis
un développement important du gésier, du même ordre que les poules recevant du blé entier et
supérieur aux poules nourries avec un aliment continu. Ces résultats suggèrent donc qu’il est
possible d’utiliser d’autres matières premières et sous d’autres formes en AS tout en
améliorant l’IC par rapport à une alimentation continue. Cependant, ces premiers travaux
(Traineau et al., 2013) ont mis en évidence quelques faiblesses potentielles du système. En
effet, les poules placées en AS ont moins consommé d’aliment et avaient un poids vif
inférieur aux poules alimentées classiquement, cette baisse de consommation expliquant
largement l’amélioration de l’IC. Les poules en AS étant moins lourdes qu’en alimentation
continue, elles pourraient être plus sensibles à des épisodes environnementaux difficiles
comme des fortes chaleurs. Ainsi au cours de ces travaux, les poules en AS ont subi une plus
179
forte perte de poids que les poules alimentées de façon classique lors de températures en
augmentation dans le bâtiment. Une autre question soulevée est la difficulté des poules à
ingérer la fraction du matin (céréales), riche en énergie et pauvre en protéines. Une des
explications pour cette plus faible consommation le matin que l’après midi est peut être en
lien avec la régulation de l’ingestion des poules sur le niveau énergétique de l’aliment
(Nahashon et al., 2005). Ces deux interrogations quant à la fragilité des poules placées en AS
lors d’épisode de forte chaleur mais également la difficulté à comprendre et maîtriser
l’ingestion de la fraction du matin riche en énergie sont celles auxquelles nous avons voulu
répondre au cours des seconds travaux de la thèse.
Le travail de méta-analyse a confirmer l’ajustement du niveau d’ingestion en fonction
de la nature des régimes avec le rôle premier du niveau énergétique de l’aliment (comme
suggéré par Summers et Leeson 1993; Van Krimpen et al., 2007). Les résultats par méta
analyse ont également mis en évidence un ajustement de l’ingestion sur le niveau énergétique
de l’aliment. Le niveau de protéines, comme ceux de methionine et lysine, ne semblent pas
avoir d’impact sur l’ingestion (en accord avec Summers et Leeson, 1993 ; Mohiti-Asli et al.,
2012). La régulation de l’ingestion sur le niveau énergétique permet donc d’émettre
l’hypothèse que la fraction du matin étant riche en énergie, les poules en limitent l’ingestion,
en lien avec leur besoin énergétique « instantané ». Concernant la fraction de l’après midi,
riche en protéines et pauvre en énergie, les poules qui doivent chercher à combler un besoin
en énergie consomment donc toute cette fraction à basse teneur en énergie, l’absence de
régulation signifiante sur le niveau de protéines ne limitant pas l’ingestion.
Cette capacité de modulation de l’ingestion sur le niveau d’énergie du régime, va
permettre de « conduire » l’ingestion des poules. Pour cela, deux régimes déséquilibrés en
énergie et en protéines étaient donnés aux poules pondeuses en croissance placées en AS, ce
qui devait permettre d’obtenir des poules avec des poids et des compostions corporelles
différentes. Les poules ont bien consommé les aliments déséquilibrés en protéines et en
énergie dans les proportions proposées tout en conservant un niveau de production équivalent
aux poules placées en alimentation continue. L’ingestion des poules placées en AS étant
encore inférieure à celle des poules en alimentation standard, l’IC était aussi amélioré dans ce
cas. En comparant uniquement les régimes distribués en séquentiel, l’ingestion de protéines
est en cohérence avec la mise à disposition des nutriments : faible, moyenne et forte, induisant
trois niveaux différents de protéines brutes ingérées. Ce résultat suggère que les poules
régulent peu leur ingestion sur le niveau de protéines dans le régime (Mohiti-Asli et al.,

180
2012). Cependant, pour le niveau d’énergie, les résultats ne sont pas similaires. En effet,
lorsque les trois régimes étaient distribués, aucune différence d’énergie totale ingérée sur la
journée n’a pu être observée. Deux hypothèses peuvent être avancées. La première est que les
poules régulent et ajustent leur consommation en fonction du niveau d’énergie de l’aliment
offert (Bouvarel et al., 2010). La seconde provient de la différence plus importante entre les
niveaux de déséquilibre de protéines et d’énergie, rendant difficile un rééquilibrage sur les
protéines. En effet, la différence entre le régime à bas niveau d’énergie et haut niveau de
protéines et le régime opposé (haut niveau de protéine et bas niveau d’énergie) est de +59%
pour les protéines contre -37.3% pour l’énergie. Cependant, il ne semble pas y avoir d’effet
irréversible conséquence de cet apport déséquilibré, puisque les poules sont capables de
rapidement s’adapter à des changements d’aliment: lorsque les poules qui recevaient des
aliments déséquilibrés préalablement se voient offrir un régime équilibré, elles retrouvent
quasi immédiatement un niveau d’ingestion d’énergie et de protéines similaire à celles qui
ont reçu un aliment équilibré. Ces travaux ont mis en évidence la capacité à piloter
partiellement l’ingestion des poules en AS, cependant les déséquilibres alimentaires
provoqués n’ont pas permis de faire varier de façon contrôlée le poids des poules. Une des
hypothèses qui peut expliquer cette absence de différence de poids vif entre les poules en AS
peut être un maintien prioritaire du poids vif au détriment de la production (Heywan, 1940).
Néanmoins, comme lors de tous les précédents travaux sur l’AS, les poules en alimentation
continue ont un poids vif supérieur aux poules placées en AS. Simultanément, la possibilité de
performances inférieures de poules en AS par rapport à des poules alimentées en continue
dans des conditions de température élevées et continue (26°C) a été étudiée aux cours de ces
travaux. La hiérarchie observée pour les performances et les poids restent les mêmes qu’en
température standard avec une diminution en valeur absolue pour tous les régimes (absence
d’interractions). Il semble donc qu’il soit possible d’utiliser la technique d’AS, même dans
des conditions de températures élevées. Une constante à travers les expérimentations est,
malgré la variation dans les apports en énergie et en protéines, que le niveau d’ingestion du
matin est systématiquement inférieur à celui de l’après midi. Ces résultats suggèrent peut être
que la fraction du matin, pour ne plus être sous consommée ne devrait pas forcément être très
riche en énergie et pauvre en protéines. Les travaux sur la cinétique des apports en énergie et
en protéines au cours de la journée à permis de répondre à ces interrogations sur la cinétique
des apports journaliers.

181
Précédemment, la fraction offerte le matin était riche en énergie et pauvre en protéines
et inversement l’après-midi. La troisième expérimentation a étudié des combinaisons
variables de niveaux d’apports d’énergie et de protéines au cours d’une journée, pour
déterminer si des cinétiques intradiennes de besoins pouvaient être identifiées sous
l’hypothèse de besoins vraiment ciblés au sein d’une partie de la journée. Pour l’ensemble des
régimes, le calcium était apporté dans la fraction de l’après-midi et donc même le régime où
les poules recevaient deux aliments avec les mêmes niveaux d’énergie et de protéines matin et
après-midi (E0P0 / E0P0 + Ca) étaient deux aliments différents dans leur composition en
matière première et dans leur granulométrie. Cette différence de granulométrie peut apporter
une explication quant à la plus faible consommation de cette fraction du matin E0P0 par
rapport à la fraction E0P0 +Ca de l’après midi. En effet cette fraction de l’après-midi ayant
plus de grosse particules, a été plus consommée car les poules consomment plus d’aliment si
celui-ci contient plus de larges particules (Safaa et al., 2009). Cette ingestion de grosses
particules a pu également causer des déséquilibres dans l’ingestion (le phosphore, les
pigments, les acides aminés de synthèse étant apporté dans le prémix qui est sous forme de
farine). Pour l’ensemble de 6 régimes, le niveau d’ingestion et la nature des ingérés ne
présentaient pas de différences significatives au cours de la journée excepté pour le régime
E0P0 / E-P- qui étaient moins consommé et pour lequel les poules avaient un niveau
d’énergie et de protéines ingérées inférieur à l’ensemble des autres régimes. Pour les 5
régimes où aucune différence d’ingestion n’était observée, les performances de ponte, et plus
précisément l’indice de consommation, variaient : trois régimes présentaient de meilleures
performances (le régime E0P0 / E-P- présentaient les moins bonnes performances de ponte et
l’indice de consommation le plus élevé). Pour deux de ces régimes le niveau de protéines était
continu au cours de la journée tandis que pour le troisième régime les poules recevaient le
matin un aliment pauvre en protéines et l’après midi une fraction riche en protéines.
Concernant le niveau d’énergie de ces trois régimes performants, deux proposaient le matin
un aliment riche en énergie et l’après midi un aliment pauvre en énergie, le troisième régime
ayant un niveau constant d’énergie au cours de la journée. Aucun de ces trois régimes ne
proposaient de niveaux haut en protéines dans la fraction du matin ni de niveau bas en
énergie. Il parait donc possible de conclure qu’en AS la fraction du matin doit être riche en
énergie et la fraction de l’après-midi symétriquement pauvre. Pour les protéines il ne semble
pas qu’il y ait de moment optimal pour l’apporter mais plutôt qu’il faut que les poules en ait
une quantité donnée au cours de la journée en accord avec Keshavarz, (1998) qui conclut que

182
les poules ont besoin de recevoir un niveau adéquat de protéines au cours de la journée peu
importe le moment de distribution. Penz and Jensen, (1991) proposent néanmoins comme
préférable d’éviter de concentrer les apports protéiques le matin. Il semble donc que pour
optimiser les apports nutritionnels chez les poules placées dans un système d’AS, il faille
apporter le matin une fraction riche en énergie avec soit un niveau moyen de protéines soit un
niveau faible et dans la fraction de l’après midi un aliment avec le carbonate de calcium,
pauvre en énergie et avec un niveau moyen ou haut en protéines. Cette troisième
expérimentation a également mis en avant que lorsque les poules sont dans des contextes
environnementaux difficiles (ici infections subies de poux rouges) les performances
(ingestion, ponte, croissance) sont altérées mais les poules alimentées en séquentiel ne
semblent pas plus impactées que les poules qui ont des apports continus en énergie et en
protéines au cours de la journée. Il est cependant difficile de généraliser cette observation à
l’ensemble des situations environnementales qui pourraient être rencontrées lors de la
conduite d’un élevage de poules pondeuses.

Valorisation et Perspectives
L’alimentation séquentielle est un système d’alimentation innovant et prometteur au
vu des résultats observés. Il permet de conserver des performances de ponte, avec une
ingestion moindre ce qui améliore l’IC. Cependant certaines questions se posent encore afin
de pouvoir utiliser cette distribution à plus grande échelle et donc pourvoir faire des
économies réelles de céréales pour les producteurs d’œuf.

Afin d’expliquer la non divergence des poids des poules (expérience 2), pourtant
nourries avec des régimes très différents en AS, il faudrait pouvoir évaluer l’interaction entre
le poids des poules avec l’exportation d’œuf. En effet une « non-différence » de poids vif peut
s’expliquer par une exportation d’œuf différente avec des nutriments ingérés qui auraient été
utilisés préférentiellement dans l’entretien de la poule plutôt que dans la fabrication d’un œuf.
Les futurs travaux devraient avoir pour but de définir et de quantifier les différents flux dans
la poule (flux énergie, protéines, acides aminés, minéraux) en fonction de différents contextes
environnementaux. Ces connaissances sur les flux permettraient d’adapter encore un peu plus
les apports de nutriments aux poules en fonction de différents contextes. L’aboutissement de
ces travaux futurs pourrait être un modèle de prédiction des performances de ponte en

183
fonction des flux entrants (aliment, eau, …) de l’état de la poule (souche, âge, poids vif,
composition corporelle,…) et des conditions environnementales (température, hygrométrie,...)

Une autre question qui n’a jamais été évoquée en AS est la quantité d’eau consommée
par les poules. En effet la question mérite d’être évoquée car le fractionnement des apports à
probablement un effet sur la consommation de celle-ci. De plus l’utilisation de fibres
insolubles ou de graines entières doit également augmenter les besoins en eau. Des mesures
simples de consommation d’eau permettraient de répondre facilement à cette interrogation et
d’évaluer l’impact d’une distribution séquentielle sur la consommation d’eau de boisson.

L’alimentation séquentielle pratiquée au cours des expérimentations, était une


distribution non automatisée et nécessitait une distribution manuelle de la ration. Afin
d’utiliser ce système en élevage, il est obligatoire de réussir à automatiser les distributions.
Cependant certaines contraintes techniques devront être respectées. En effet, le bâtiment
d’élevage devra être équipé d’au moins deux silos comprenant chacun une des deux fractions.
Afin de profiter de l’effet positif du fractionnement de la ration journalière, il faudra un
pilotage fin des distributions. Les quantités qui sortent du silo devant être maitrisées, la
présence de d’une benne peseuse sera donc nécessaire. Les quantités distribuées doivent être
maitrisées afin d’éviter tout dérive de consommation de l’une des deux fractions, comme cela
a pu être observé dans des travaux antérieurs sur l’AS (Umar Faruk et al, 2010).

Les deux fractions ne devront également pas se mélanger sinon le gain dû au


fractionnement des apports nutritionnels sera surement perdu. En effet un mélange de ces
deux fractions ne reviendrait qu’à reconstituer le régime complet et donc les gains obtenus par
le fractionnement des apports paraissent perdus. De plus si les deux fractions se mélangent,
dont une fraction riche en énergie et donc principalement composée de céréales broyées avec
un complémentaire sous forme de farine grossière les poules consommeraient donc
principalement les grosses particules et moins la farine (qui contient les acides aminés de
synthèse, les pigments, et beaucoup de protéines). Ce tri pourrait alors être à l’origine de
possibles déficits en certains acides aminés et donc avoir des conséquences négatives sur la
ponte. Il sera également important de ne pas inverser les rations, le calcium étant distribué
uniquement dans la fraction de l’après midi, correspondant au moment de la formation de la
coquille, une inversion provoquerai probablement des incidences sur l’ingestion et surtout sur
la qualité des coquilles.

184
Afin de pouvoir utiliser cette distribution novatrice de l’aliment dans des bâtiments de
production, il est nécessaire que le gain réalisé sur le coût de l’aliment soit supérieur face à la
logistique particulière mise en place : une formulation spécifique de deux aliments différents,
des contraintes technologiques relativement importantes et un pilotage quotidien et précis
demandé à l’éleveur. Cependant dans le contexte de production actuel, cette technique
mériterait d’être utilisée. Afin de mieux se rendre compte des gains qui pourraient être
effectués grâce à des régimes moins coûteux, il faudrait essayer dans des travaux mêlant
mesures en élevages et calculs bio-économiques de mettre en évidence les gains effectués ou
les gains possibles en fonction de différents contextes de prix des matières premières. Cette
question doit être évaluée car en fonction des différents contextes de prix, l’argument
financier et le coût des régimes auront plus de poids pour prendre la décision de formuler des
régimes « économiques » avec soit un bas niveau d’énergie soit un bas niveau de protéines.
L’alimentation séquentielle, avec ce fractionnement et les niveaux variables en énergie et
protéines, pourrait permettre l’utilisation de matières premières alternatives et notamment des
coproduits issus de la fabrication d’agro carburants. L’utilisation de graines entières, sans
transformation et sans coût de transport permettrait à l’éleveur de faire des économies et
améliorerait l’impact sur l’environnement. L’apport de graines entières peut en effet être une
solution pour répondre et contrecarrer cette fluctuation du prix des matières premières. En
effet, si l’éleveur peut injecter dans sa ration, une proportion de blé entier, produit chez lui et
sans transformation, le gain économique est certain pourrait être intéressant. Un arbitrage
devra cependant être fait entre le gain obtenu si l’éleveur vend ces céréales si il utilises ces
céréales pou nourrir ses animaux. De plus, comme vu dans la première expérience, la
possibilité de pouvoir changer la matière première principale dans le régime est très important
dans des contextes ou le blé ou le maïs aurait des écarts de prix très importants. L’intérêt étant
de pouvoir s’adapter à ces variations et de passer d’une formulation sur une céréale à une
autre. Le but étant d’avoir un système et une méthode non figée et donc adaptable à
l’évolution du marché.

185
Conclusion générale

Ces différents travaux ont permis d’obtenir et de confirmer de nombreuses


connaissances quant à l’utilisation de la distribution séquentielle de l’aliment. En effet
l’utilisation de blé entier dans la fraction du matin, donne des poules avec des performances
de ponte équivalente à l’alimentation continue mais avec un niveau d’ingestion inférieur ce
qui permet une amélioration significative de l’IC (Umar Faruk et al 2010, expérience 1). Lors
d’une distribution de l’alimentation sous forme séquencée, la fraction du matin est
systématiquement moins consommée que la fraction de l’après midi (expériences 1, 2, 3).
Cependant cette ingestion inférieure totale par rapport à une alimentation continue
(expériences 1 et 2) avec un niveau de production équivalent permet d’améliorer l’IC des
poules en séquentiel (expériences 1 et 2). Les poules en séquentiel ayant une ingestion
inférieure, elles ont un poids vif inférieur aux standards de la souche ou aux poules en
alimentation classique (expériences 1, 2 et 3). Les poules en séquentielle n’ont également pas
la même composition corporelle que les poules en alimentation classique. En effet, le
séquentiel permet d’obtenir des poules moins grasses (expériences 1, 2 et 3). Ce poids
inférieur des poules peut néanmoins rendre ces poules moins robustes à un stress
environnemental soudain (forte températures, expérience 1) mais lorsque celles-ci sont
placées dans des conditions difficiles prolongées (chaleur expérience 2 ; infection de poux
rouges expérience 3) elles ne semblent pas présenter plus de difficulté que les poules en
alimentation continue.
Cependant, une des interrogations portait sur la substitution du blé entier dans la
fraction du matin par l’utilisation d’autres céréales. Les résultats ont montré que l’utilisation
de blé broyé ou de maïs broyé (expérience 1), l’utilisation de fibres insolubles ou d’une faible
proportion de blé entier dans cette fraction (expérience 1) pouvait permettre de ne plus utiliser
uniquement du blé entier le matin. Un mélange de blé et de maïs dans des proportions
équivalentes (expériences 2 et 3) permettait de faire varier les apports en nutriments dans la
journée (expérience 3) mais également une amélioration de l’IC (expériences 1 et 2) par
rapport à une alimentation continue. Il parait donc raisonnable de penser que l’utilisation de
blé entier dans la fraction du matin n’est plus systématique.
Une seconde question portait sur la justification et l’intérêt d’apporter la fraction la
plus énergétique le matin et la fraction la plus protéique l’après midi. Les poules semblent
capables de réguler leur ingestion sur le niveau d’énergie dans le régime mais pas sur le
186
niveau de protéines (méta-analyse, expérience 2). Lorsque les apports en protéines sont
concentré l’après midi ou bien continue au cours de la journée, il ne semble pas qu’il y ait
d’incidence sur l’ingestion, les performances et la croissance (expérience 3). Au contraire,
l’énergie semble devoir être apportée principalement dans la fraction du matin en non de
façon continue au cours de la journée (expérience 3). Les connaissances sur l’alimentation
séquentielle ont donc progressé mais il reste néanmoins d’autres connaissances à acquérir afin
d’utiliser cette technique à échelle commerciale.

187
188
Références

Arima, Y., Mather, F. B. and Ahmad, M. M. (1976) Response of egg-production and shell
quality to increases in environmental-temperature in 2 age-groups of hens. Poult. Sci.,
55: 818-820.
Bell DD and William Daniel Weaver Book, Commercial Chicken Meat and Egg
Production 2002, 5th edition ed Springer
Bizeray D, Leterrier C, Constantin P, Picard M and Faure JM 2002. Sequential feeding can
increase activity and improve gait score in meat-type chickens. Poultry Science 81,
1798-1806.
Blair R, Dewar WA and Downie JN 1973. Egg production responses of hens given a
complete mash or unground grain together with concentrate pellets. British Poultry
Science 14, 373-377.
Blesbois, E 2010. Quelles recherches pour l’aviculture en France ? Inra Production Animale
5 427-432
Bokkers EAM and Koene P 2003. Eating behaviour, and preprandial and postprandial
correlations in male broiler and layer chickens. British Poultry Science 44, 538-544.
Bouvarel I, Nys Y, Panheleux M and Lescoat P 2010 How diet influences the quality of eggs.
INRA Productions Animales 23, 167-181.
Bouvarel I, Barrier-Guillot B, Larroude P, Boutten B, Leterrier C, Merlet F, Vilarino M,
Roffidal L, Tesseraud S, Castaing J and Picard M 2004. Sequential feeding programs
for broiler chickens: Twenty-four- and forty-eight-hour cycles. Poultry Science 83, 49-
60.
Braine A and Magdelaine P The poultry sector in 2011. TeMA - Techniques et Marches
Avicoles, 18-25.
Chah CC and Moran ET 1985. Egg characteristics of high-performance hens at the end of lay
when given cafeteria access to energy, protein, and calcium. Poultry Science 64, 1696-
1712.
Chauve C 1998. The poultry red mite Dermanyssus gallinae (De Geer, 1778): current
situation and future prospects for control. Veterinary Parasitology 79, 239-245.

189
Clifton PG 1979. Temporal patterns of feeding in the domestic chick - ad-libitum. Animal
Behaviour 27, 811-820.
Collins LM and Sumpter DJT 2007. The feeding dynamics of broiler chickens. Journal of the
Royal Society Interface 4, 65-72.
De Araujo JA, Vilar da Silva JH, Perazzo Costa FG, Batista de Sousa JM, Naves Givisiez PE
and Sakomura NK 2011 Effect of the levels of calcium and particle size of limestone
on laying hens. Revista Brasileira De Zootecnia-Brazilian Journal of Animal Science
40, 997-1005.
Etches R 2008. Breeding technology and public opinion. World Poultry 24, 25-27.
Gous RM and Dupreez JJ 1975. Sequential feeding of growing chickens. British Journal of
Nutrition 34, 113-118.
Hetland H, Svihus B and Choct M 2005. Role of insoluble fiber on gizzard activity in layers.
Journal of Applied Poultry Research 14, 38-46.
Heywan BW 1940. The effect of restricted feed intake on egg weight, egg production, and
body weight. Poultry Science 19, 29-34.
Keshavarz K 1998. Further investigations on the effect of dietary manipulation of protein,
phosphorus, and calcium for reducing their daily requirement for laying hens. Poultry
Science 77, 1333-1346.
Krimpen MMv, Kwakkel RP, Peet-Schwering CMCvd, Hartog LAd and Verstegen MWA
2011 Effects of dietary energy concentration, nonstarch polysaccharide concentration,
and particle sizes of nonstarch polysaccharides on digesta mean retention time and gut
development in laying hens. British Poultry Science 52, 730-741.
Krimpen MMv, Kwakkel RP, Andre G, Peet-Schwering CMCvd, Hartog LAd, Verstegen
MWA, der Peet-Schwering CMCv, van Krimpen MM, van der Peet-Schwering CMC
and den Hartog LA 2007. Effect of nutrient dilution on feed intake, eating time and
performance of hens in early lay. British Poultry Science 48, 389-398.
Lesna I, Wolfs P, Faraji F, Roy L, Komdeur J and Sabelis MW 2009. Candidate predators for
biological control of the poultry red mite Dermanyssus gallinae. Experimental and
Applied Acarology 48, 63-80.
Lewis PD, Perry GC and Morris TR 1996. Effects of changes in photoperiod and feeding
opportunity on the performance of two breeds of laying hen. British Poultry Science
37, 279-293.

190
Longe OG 1983. Varietal differences in chemical characteristics related to cooking quality of
cowpea. Journal of Food Processing and Preservation 7, 143-150.
Magdelaine P and Braine A International and European overview of egg production and
consumption. INRA Productions Animales 23, 111-122.
Mohiti-Asli M, Shivazad M, Zaghari M, Rezaian M, Aminzadeh S and Mateos GG 2012
Effects of feeding regimen, fiber inclusion, and crude protein content of the diet on
performance and egg quality and hatchability of eggs of broiler breeder hens. Poultry
Science 91, 3097-3106.
Mongin P and Sauveur B 1974. Voluntary food and calcium intake by the laying hen. British
Poultry Science 15, 349-359.
Morris BA and Taylor TG 1967. The daily food consumption of laying hens in relation to egg
formation. British Poultry Science 8, 251-257.
Morris TR and Butler EA 1995. New intermittent lighting program (the reading system) for
laying pullets. British Poultry Science 36, 531-535.
Nahashon SN, Adefope N, Amenyenu A and Wright D 2005. Effects of dietary metabolizable
energy and crude protein concentrations on growth performance and carcass
characteristics of French guinea broilers. Poultry Science 84, 337-344.
Nir I, Melcion JP and Picard M 1990. Effect of particle-size of sorghum grains on feed-intake
and performance of young broilers. Poultry Science 69, 2177-2184.
Noirot V, Bouvarel I, Barrier-Guillot B, Castaing J, Zwick JL and Picard M 1998. Whole
cereal grains in broiler nutrition: the come back? Productions Animales 11, 349-357.
Nys Y, Bain M and Van Immerseel 2011. Improving the safety and quality of eggs and egg
products, Woodhead Publishing, 2011
Olver MD and Malan DD 2000. The effect of choice-feeding from 7 weeks of age on the
production characteristics of laying hens. South African Journal of Animal Science-
Suid-Afrikaanse Tydskrif Vir Veekunde 30, 110-114.
Ouhida I, Perez J, Piedrafita J and Gasa J 2000. The effects of sepiolite in broiler chicken
diets of high, medium and low viscosity. Productive performance and nutritive value.
Animal Feed Science and Technology 85, 183-194.
Penz AM and Jensen LS 1991. Influence of protein-concentration, amino-acid
supplementation, and daily time of access to high-protein or low-protein diets on egg
weight and components in laying hens. Poultry Science 70, 2460-2466.

191
Picard ML, Uzu G, Dunnington EA and Siegel PB 1993. Food-intake adjustments of chicks -
short-term reactions to deficiencies in lysine, methionine and tryptophan. British
Poultry Science 34.
Picard M, Le Fur C, Melcion JP and Bouchot C 2000. Feed particles: what chickens see and
touch! Productions Animales 13, 117-130.
Reichmann KG and Connor JK 1979. Effects of meal feeding of calcium, protein and energy
on production and calcium status of laying hens. British Poultry Science 20, 445-452.
Richardson AJ 1970. The role of the crop in the feeding behaviour of the domestic chicken.
Animal Behaviour 18, 633-639.
Robinson D 1985. Performance of laying hens as affected by split time and split composition
dietary regimens using ground and unground cereals. British Poultry Science 26, 299-
309.
Rose SP, Fielden M, Foote WR and Gardin P 1995. Sequential feeding of whole wheat to
growing broiler-chickens. British Poultry Science 36, 97-111.
Rosebrough RW, McMurtry JP and Steele NC 1989. Protein and energy relations in the
broiler chicken - chronic or acute effects of alternating protein or intermittent feeding
regimens on broiler lipid-metabolism. British Journal of Nutrition 61, 223-233.
Safaa HM, Jimenez-Moreno E, Valencia DG, Frikha M, Serrano MP and Mateos GG 2009.
Effect of main cereal of the diet and particle size of the cereal on productive
performance and egg quality of brown egg-laying hens in early phase of production.
Poultry Science 88.
Sauveur B, Reviers Md and De Reviers M 1988. Reproduction and egg production in poultry.
Reproduction des volailles et production d'oeufs., 449pp.-449pp.
Savory CJ 1980. Diurnal feeding patterns in domestic-fowls - review. Applied Animal
Ethology 6, 71-82.
Shanawany MM 1982. The effect of ahemeral light and dark cycles on the performance of
laying hens - (a review). Worlds Poultry Science Journal 38, 120-126.
Shelton JL, Dean DW, Southern LL and Bidner TD 2005. Effect of protein and energy
sources and bulk density of diets on growth performance of chicks. Poultry Science
84, 1547-1554.
Shevchenko, V. G., and G. G. Sherapanov (1986). Evaluation of the egg protein synthesis
dynamics in the oviduct of laying hens. Sel’ Skokhozyaistvennaya Biologiva. 7, 110-114.

192
Summers JD and Leeson S 1993. Influence of diets varying in nutrient density on the
development and reproductive-performance of white leghorn pullets. Poultry Science
72, 1500-1509.
Syafwan, S., Wermink, G. J. D., Kwakkel, R. P. and Verstegen, M. W. A. (2012) Dietary self-
selection by broilers at normal and high temperature changes feed intake behavior,
nutrient intake, and performance. Poult. Sci, 91: 537-549.
Taylor TG 1970. How an eggshell is made. Scientific American 222, 88-95.
Traineau M, Bouvarel I, Mulsant C, Roffidal L, Launay C and Lescoat P 2013 Effects on
performance of ground wheat with or without insoluble fiber or whole wheat in
sequential feeding for laying hens. Poultry Science 92, 2475-2486.
Umar Faruk M, Bouvarel I, Mallet S, Ali MN, Tukur HM, Nys Y and Lescoat P 2011 Is
sequential feeding of whole wheat more efficient than ground wheat in laying hens?
Animal : an international journal of animal bioscience 5, 230-238.
Umar Faruk M, Bouvarel I, Meme N, Rideau N, Roffidal L, Tukur HM, Bastianelli D, Nys Y
and Lescoat P 2010 Sequential feeding using whole wheat and a separate protein-
mineral concentrate improved feed efficiency in laying hens. Poultry Science 89, 785-
796.
Umar Faruk MU, Bouvarel I, Meme N, Roffidal L, Tukur HM, Nys Y and Lescoat P 2010
Adaptation of wheat and protein-mineral concentrate intakes by individual hens fed ad
libitum in sequential or in loose-mix systems. British Poultry Science 51, 811-820.
Vergara P, Ferrando C, Jimenez M, Fernandez E and Gonalons E 1989. Factors determining
gastrointestinal transit-time of several markers in the domestic-fowl. Quarterly Journal
of Experimental Physiology and Cognate Medical Sciences 74, 867-874.
Wauters A.M., G. Guibert, A. Bourdillon, M.A. Richard, J.P. Melcion, M. Picard, 1997.Choix
de particules alimentaires chez le poussin : effet de la taille et de la composition 2èmes
JRA, Tours, 201-204
Wood-Gush, D.G.M., Horne, A.R., 1970. The effect of egg formation and laying on the food
and water intake of brown leghorn hens. Br. Poult. Sci. 11, 459-466.
Woods SC, Seeley RJ, Porte D and Schwartz MW 1998. Signals that regulate food intake and
energy homeostasis. Science 280, 1378-1383.
Yo T, Siegel PB, Guerin H and Picard M 1997. Self-selection of dietary protein and energy
by broilers grown under a tropical climate: Effect of feed particle size on the feed
choice. Poultry Science 76, 1467-1473.

193
Zhou WT and Yamamoto S 1997. Effects of environmental temperature and heat production
due to food intake on abdominal temperature, shank skin temperature and respiration
rate of broilers. British Poultry Science 38, 107-114

194
195
Maxime TRAINEAU
Détermination du champ d’application
sur système d’alimentation séquentielle
chez les poules pondeuses

L’alimentation séquentielle (AS) consiste à distribuer de manière cyclique deux aliments différents dont la
combinaison permet des apports nutritionnels adaptés. L’objectif de la thèse était d’évaluer la robustesse et les
limites de ce système. Trois expérimentations ont étéréalisées. Les questions portaient sur la variation de la
composition en matières premières des régimes, la capacité d’adaptation des poules à des situations de
déséquilibre alimentaire et de stress thermique et enfin la modulation au cours de la journée des apports en
énergie et en protéines. Un travail de méta-analyse apportait des réponses sur la régulation de l’ingestion chez
les poules pondeuses ainsi que sur l’impact de la nature des ingérés sur les performances de ponte et de
croissance.

Les expérimentations (Ex.1, 2, 3) ont permis de mettre en évidence la possibilité de distribuer d’autres céréales
(blé et/ou maïs broyé, fibres insolubles) à la place du blé entier distribué dans la fraction du matin. L’ingestion
totale inférieure par rapport à une alimentation continue (Ex. 1 et 2) et un niveau de production équivalent
permettent d’améliorer l’IC des poules en AS par rapport à des poules alimentée de façon classique. Les poules
semblent capables de réguler leur ingestion sur le niveau d’énergie de l’aliment mais pas sur son niveau de
protéines dans les conditions étudiées (méta-analyse, Ex. 2). Cependant, les apports en énergie doivent se
concentrer dans la fraction du matin afin d’obtenir les meilleures performances en AS tandis que les apports
protéiques doivent soit être concentrés dans la fraction de l’après midi soit continus au cours de la journée (Ex.
3). Ces connaissances sur l’AS doivent encore être complétées afin de pouvoir utiliser ce système
d’alimentation en élevage néanmoins l’AS est prometteur en terme d’économies de ressources.

Mots clés : Alimentation séquentielle, poules pondeuses, énergie, protéines, céréales.

Sequential feeding (SF) consists in splitting the nutrients into two different diets. The goal of this thesis was to
explore the limits of this feeding system. Three experiments were performed. The major questions go on
utilization of other feedstuffs, capacity of hens to adapt their feed intake on unbalanced diet and hot
temperatures and responses to modulation of energy and protein supplies. A meta-analysis gave responses on
the regulation of feed intake and the impact on nutrient intake on laying and growth performances in laying
hens.

Experiments (Ex.1, 2, 3) shown off the possibility to substitute whole wheat with other cereals or feedstuffs
(ground wheat and/or corn, insoluble fiber) in the morning fraction. Lower total feed intake in SF compared to
continuous feeding (Ex. 1 et 2) and same laying performances improved the FCR. Laying hens were able to fit
their feed intake on energy in the diet but not on protein content (meta-analysis, Ex. 2). Nevertheless, energy
supplies had to be given in the morning fraction while protein could be given concentrate in the afternoon or
throughout the day (Ex. 3). Knowledge on SF has to be completed in order to use this feeding system in
commercial conditions. SF can help sparing resources and improve the on farm economical efficiency.

Key words: Sequential feeding, laying hens, energy, protein, cereals

196

S-ar putea să vă placă și