2016PA066017
2016PA066017
2016PA066017
Mme Martine Maïbèche Professeur, Université Pierre et Marie Curie, Paris (France) Examinateur
Mme Isabelle Dajoz Professeur, Université Paris Diderot, Paris (France) Directeur de thèse
M. Xavier Raynaud Maître de conférences, Université Pierre et Marie Curie, Paris (France) Directeur de thèse
Université Pierre et Marie Curie
École doctorale 227 « Sciences de la Nature et de l’Homme »
Institut d’Écologie et des Sciences de l’Environnement de Paris (iEES
(iEES-Paris)
Équipes Écologie et Évolution des réseaux d’interactions & Écologie intégrative
Mme Martine Maïbèche Professeur, Université Pierre et Marie Curie, Paris (France) Examinateur
Mme Isabelle Dajoz Professeur, Université Paris Diderot, Paris (France) Directeur de thèse
M. Xavier Raynaud Maître de conférences, Université Pierre et Marie Curie, Paris (France) Directeur de thèse
A mes parents et Benoît
Remerciements
Je souhaite tout d'abord remercier mes co-directeurs de thèse, Isabelle Dajoz et Xavier Raynaud.
Merci à vous de m'avoir proposé ce stage de Master 2 qui m'a ensuite ouvert les portes de la thèse.
Je vous remercie de m'avoir fait confiance pendant ces 3 années et demie. J’ai apprécié d’avoir la
possibilité de prendre des initiatives, de pouvoir mener les expériences par moi-même, tout en
sachant que je pouvais compter sur vos conseils et votre aide en cas de besoin. Merci aussi à vous
deux pour votre écoute et votre humanité tout au long de cette thèse, je l’ai beaucoup appréciée
lorsque j’ai pu vous faire part de mes doutes et appréhensions. Xavier je te remercie aussi pour ton
infinie patience et ta pédagogie face à mes questions (répétées) de statistiques ! Enfin, je vous
remercie pour vos précieux conseils pendant la rédaction, qui, je l’avoue, a été éprouvante ! Le
syndrome de la page blanche est, parait-il, courant mais je ne m’attendais pas à cela ! Merci
d’avoir été si compréhensifs et patients !
Je remercie la région Île-de-France pour le financement de ce travail de thèse via le DIM R2DS.
Je remercie également la Fédération Île-de-France pour de Recherche sur l'Environnement pour le
financement de la première expérience.
Je tiens ensuite à remercier Alan Gange et Richard Michalet pour avoir accepté d’être mes
rapporteurs de thèse et d’avoir ainsi contribué à l’amélioration de ce travail de thèse. Je remercie
également Denis Michez et Martine Maïbeche d'avoir accepté de participer à mon jury de thèse,
respectivement en tant qu’examinateur et présidente du jury, me permettant ainsi d’approfondir et
ouvrir ma réflexion sur ma problématique.
Je souhaite aussi remercier les membres de mon comité de thèse, Mathilde Baude, Laure
Turcati, Sébastien Filoche et Marc-Andre Selosse, qui ont su m’apporter de précieux conseils
tout au long de cette thèse.
Je voudrais ensuite remercier les personnes qui m'ont accompagnée sur le terrain et dans le suivi
quotidien des expérimentations. Je pense tout particulièrement à Amandine Hansart. C'est bien
simple, je n'aurais pas réussi à mener mes expérimentations sans ton aide et ta patience ! Et je
n'aurais pas non plus appris à tirer à l'arc ! Alors merci Amande ! Merci pour ton énergie et ta
motivation sur le terrain. Ce fut un réel plaisir de travailler à tes côtés. Merci aussi pour ta
présence et ton humanité. Avec toutes tes petites attentions, tu as su m'apporter un peu de légèreté
Remerciements | 7
dans les périodes difficiles. Entre les œufs de Pâques, les stylos magiques et les élixirs anti-stress,
j'ai été choyée ! Cela va être difficile de retrouver une telle coéquipière !
Merci également à Eric Motard pour sa bonne humeur et ses connaissances botaniques sans
bornes. Merci à Jacques Mériguet pour ses conseils et son aide (musclée !) sur le terrain.
Je voudrais aussi remercier l'ensemble des stagiaires qui m’ont apporté leur aide. Tout d'abord
Séléné Verstraet pour sa motivation sans faille et son enthousiasme ainsi que Abdou Fofana
pour son implication et sa curiosité débordante. Je remercie aussi Océane Vincent, Alice
Gauthey, Elisa Rousset, Fanny Grossi, Manon Bataille, Ludovico St Amour di Chanaz et
tous les autres stagiaires du CEREEP qui ont pu donner un coup de main pendant les printemps
2013 et 2014. Merci également à toute l’équipe du CEREEP pour son accueil chaleureux et ses
barbecues / tournois de jeux ! Plus particulièrement merci à Beatriz Decencière et Samuel Perret
qui m’ont prêté main forte plusieurs fois.
Je tiens désormais à remercier mes collègues de labo qui m'ont permis de passer de jolis moments.
Je pense d’abord aux collègues de l’ex-BIOEMCO, plus particulièrement à Ambre David, Alix
Sauve, Henri de Parseval, Aleksandar Rankovic, Benoit Gauzens, Stéphane Loisel, David
Carmignac, Catherine Muneghina, Jean-Christophe Lata et Jacques Mériguet. Merci à vous
tous pour l’ambiance si particulière que vous avez su créer au sein du labo. Merci pour tous les
goûtés régionaux, bières et karaokés partagés ! Je pense bien sûr aussi à Kejun Zou, Imen
Louati, Guillaume Chérel et tous les autres. Du côté de Jussieu, je souhaiterais remercier mes
compères de bureau. Je pense notamment à Claudie Haussy, Romain Peronnet, Thomas Garcia
et tonton (!) Timothé Cook qui ont contribué à l’ambiance si chaleureuse du bureau 703. Plus
particulièrement je pense à Marion Chatelain et Thomas Brom. Marion, depuis la licence 2,
nous nous sommes suivies jusqu’à nous retrouver dans le même bureau pour notre thèse ! Cela
nous aura même valu le surnom de « siamoises » ! Que d’après-midi à parler chiffons et cuisine,
que de soirées à aller danser à La Bellevilloise ou au Rosa bonheur. Je suis ravie de tout ce temps
passé en ta compagnie et je compte bien ajouter de nouvelles choses à notre fameuse liste de
sorties ! Merci pour ton amitié et ta présence. Thomas, compère de cantine, c’est toujours un
plaisir de papoter sciences, expos ou …chatons (!) avec toi. Merci pour ton aide sur R (oui
j’installerai un jour NppToR !) et tes conseils pour les enseignements. Merci aussi à toi pour
l’ambiance musicale et grivoise du bureau qui m’a mise de bonne humeur bien des fois ! Il m’aura
fallu un peu de temps pour ne plus m’offusquer de tes blagues, maintenant c’est moi qui en
propose ! Je pense également à Clémentine Renneville. Clémentine, comme tu es souvent par
monts et par vaux, c’est toujours un peu la fête quand on t’aperçoit au labo ! Tu es toujours
Remerciements | 8
partante pour rire, à l’écoute des autres et pleine de bonnes idées. Comme cette expédition à la
Cité des Sciences pour la fameuse braderie de livre ! Merci à toi pour nos séances papotage. Un
grand merci à vous trois mais aussi à Marie Vaugoyeau, Ewen Georgelin et Anaïs Bompard
pour les pique-nique au bord de la Seine, les parties de Molki, et soirées choucroute / tartiflette !
Je pense également à Josefa Bleu, Marta Gallardo Ruiz, Aurore Picot, Andreaz Dupoué,
Pierre Quévreux et aux « petits nouveaux », Gabrielle Ringot, Loïc Prosnier, Eric Tromeur et
Marwan Cheikh Albassatneh. Petite pensée également pour Julien Gasparini et Adrien
Frantz, les deux plaisantins du 7e étage. Je garderais de très bons souvenirs des pauses déjeuner à
la cantine en votre compagnie (malgré quelques incidents « techniques »!).
Enfin merci aussi à l’incroyable équipe administrative qui a su faire face à mes questions et doutes
incessants sur les conventions de stage et autres états de frais. Merci Paola Paradisi, Catherine
Muneghina, Johana Azzi et Carole Bousquet pour votre patience.
Je n’oublie pas non plus les copains et copines de toujours qui a coup de week-end par ci et de
week-end par là m’apportent toujours autant de fous rires et de souvenirs. Merci à vous Romain,
Tiffany, Sophie, Laura, Lucile, Thomas, Ludivine et Sonia.
Je me tourne désormais vers ma famille. Merci à mes parents, Annie et Pierre, pour leur
incroyable soutien et leurs petites attentions qui me portent au quotidien. Merci pour votre
patience, votre générosité. Je remercie également mes frères, Mathieu et Fabien. Merci à vous
pour vos conseils avisés de grands frères ! Merci Mathieu pour ta sensibilité et ta capacité à
m’ouvrir à d’autres mondes que celui des sciences. Merci Fabien pour ton écoute (et ta sagesse
d’ancien thésard!) sans oublier tes invitations du dimanche soir. Merci aussi à leurs charmantes
tribus, Maria, Roxane, Eliott et Pernette, Loreleï, Hypatie. Merci aussi à Tatie Chat, à
Marraine et Patou et à tous celles et ceux que je ne cite pas mais que je n’oublie pas.
Enfin je tiens à te remercie mon câlin. Tout est plus doux avec toi. Merci pour ta présence et ton
soutien au quotidien. Merci pour ta patience et ton écoute. Tu sais me donner confiance,
m’encourager, me bousculer même ! Et surtout tu me fais rire ! Rien de tel après une rude journée
de rédaction ! Mon câlin je te remercie pour ton enthousiasme et toutes tes attentions qui m’ont
portés ces derniers temps. C’est bien grâce à toi que j’ai pu ce relever ce joli défi. Merci pout tout
ce que tu es, pour ton sourire et ta poésie. Que j’ai hâte, désormais, de pouvoir me laisser bercer
avec toi par le doux son de la mer, paisibles…
Remerciements | 9
Sommaire
Remerciements…………………………………………………………………………… 7
I. Introduction générale…………………………………………………………........ 17
II. Chapitre 1
Influence de la compétition induite par la présence d’espèces anémophiles sur des
traits floraux impliqués dans l’attractivité aux pollinisateurs d’espèces
entomophiles
III. Chapitre 2
Compétition entre plantes et traits fins d’attractivité aux pollinisateurs
IV. Chapitre 3
Compétition entre plantes pour les ressources du sol et conséquences sur
l’attraction de pollinisateurs sauvages
Bill Watterson,
Calvin & Hobbes, 1995
Introduction générale
Diversité et complexité sont probablement les termes qui caractérisent le mieux l'ensemble
des interactions entre organismes au sein des écosystèmes. Au cours de cette thèse, j'ai
cherché à mettre ces aspects en évidence en m'intéressant plus particulièrement aux effets des
interactions au sein des communautés végétales et comment ces interactions influencent les
interactions avec d'autres organismes. Je me suis alors concentrée sur l'étude des interactions
de compétition entre plantes et leur influence sur des interactions mutualistes : les relations
plantes-pollinisateurs.
Historiquement, les relations entre espèces ont d'abord été étudiées entre deux espèces. C'est
ainsi qu'une grande diversité d'interactions a pu être décrite (Voir Box 1). L'étude de ces
interactions a permis de mettre en évidence de grands mécanismes et concepts écologiques.
Nous pouvons notamment citer la modélisation des dynamiques proies-prédateurs (Lotka,
1925 ; Volterra, 1926) ou encore les travaux de Gause (1934) sur le principe d'exclusion
compétitive. Par la suite, des systèmes plus complexes ont été étudiés, notamment en
intégrant de nouveaux compartiments ou espèces. De ces travaux, ont émergé les concepts de
chaînes trophiques puis de réseaux, se rapprochant ainsi de la complexité des relations entre
organismes observées au sein des communautés naturelles. Alors que les chaînes trophiques
ne décrivent que les relations « mangeurs-mangés » entre quelques compartiments successifs,
les réseaux sont une représentation plus large des interactions interspécifiques (nombre
d'espèces, types d'interactions concernées, Ings et al., 2009). Via l'étude de ces chaines et
réseaux trophiques, de nouvelles notions ont pu être dégagées. C'est le cas notamment des
effets indirect comme la compétition apparente (Holt, 1977) ou les cascades
trophiques (Paine, 1980). Dans le cas d'une cascade trophique par exemple, un prédateur
(niveau n) en diminuant l'abondance de ses proies (niveau n-1) conduit ainsi à un relâchement
Introduction | 17
de la pression de consommation exercée sur les niveaux trophiques inférieurs (niveau n-2, e.g.
producteurs primaires). Ces effets indirects seraient d’une grande importance dans les
communautés naturelles (Walsh, 2013). L'étude des réseaux d'interaction a, quant à elle,
conduit à s'intéresser à l'architecture des relations entre espèces. Connectance, modularité,
emboîtement, sont autant de caractéristiques pouvant décrire la topologie des réseaux (Massol
et al., 2015). Ces paramètres peuvent être d’une grande importance au sein des communautés.
L'architecture des réseaux pourrait, par exemple, fortement influencer la capacité d'un réseau
à persister dans le temps (persistance) ou à retourner à un état initial après perturbation
(résilience). Thébault & Fontaine (2010), ont ainsi démontré que des réseaux fortement
emboîtés et connectés favorisent la stabilité des réseaux mutualistes alors que les réseaux
trophiques sont plus stables lorsqu'ils sont compartimentés et peu connectés. Pour aller plus
loin dans l'étude des réseaux d'interactions, certains travaux récents se sont focalisés sur
l'étude simultanée de plusieurs types d'interactions étudiés séparément jusqu'à présent
(Fontaine et al., 2011; Kéfi et al., 2012). Ainsi certains auteurs se sont intéressées aux
couplages compétition-prédation (Chase et al. 2002), compétition-herbivorie (van der Wal et
al., 2000), prédation-parasitisme (Lafferty et al., 2008; van Veen et al., 2008) ou mutualisme-
antagonisme (Bronstein et al., 2003; Melián et al., 2009; Georgelin & Loeuille, 2014; Sauve
et al., 2014). Dans la même logique, et afin d'avoir une vision plus globale, certaines études
sont allées jusqu'à relier des interactions entre compartiments d'un écosystème
(aquatique/terrestre : Nakano & Murakami, 2001 ; Knight et al., 2005), terrestre
souterrain/terrestre aérien (Bardgett et al., 1998; Van der Putten et al., 2001, Wardle et al.,
2004; Poveda et al., 2003, 2005; Barber & Soper-Gorden, 2015). L'enjeu actuel semble donc
d'adopter une démarche intégrative, en s'intéressant simultanément à plusieurs types
d'interactions ou en cherchant à relier divers compartiments d'un écosystème.
Outre cette complexité topologique, s'ajoute une variabilité spatiale et temporelle. Une
variabilité spatiale tout d'abord, car selon les conditions (micro)-climatiques, les paramètres
physico-chimiques, la disponibilité des ressources ainsi que la composition des communautés,
qui s'influencent mutuellement, peuvent varier. Par ailleurs, il existe différents types de
distribution spatiale des ressources (homogène, hétérogène) de même que les individus d'une
population peuvent être répartis de manière aléatoire, régulière ou agrégée au sein d'un
écosystème (Ricklefs & Miller, 2005). Ces paramètres peuvent donc influencer les
Introduction | 18
interactions entre organismes. Mais lorsqu’on parle de variabilité spatiale, la notion d’échelle
est également importante. Par exemple, le lien entre le contexte paysager et l’abondance ou la
richesse d’une communauté de pollinisateurs peut fortement varier selon les échelles
considérées (Steffan-Dewenter et al., 2002). Les interactions entre espèces sont également
variables dans le temps. Ces variations peuvent être journalières, saisonnières ou annuelles.
Par exemple, la composition des communautés de plantes et de pollinisateurs peut fortement
varier au sein d'une saison (Olesen et al., 2008) voire d'une année sur l'autre (Alarcon et al.,
2008; Dupont et al., 2009) modifiant ainsi l'architecture des réseaux de pollinisation (Olesen
et al., 2008).
Il apparaît évident que les interactions entre espèces jouent un rôle central dans le
fonctionnement et l'évolution des écosystèmes. Les flux de matière et d'énergie notamment
sont influencés par les relations trophiques (Begon et al., 2006 ; Tirard et al., 2012). De
même, la structure des communautés (richesse et composition spécifique, abondances
relatives), bien que fortement dépendante des conditions abiotiques (conditions climatiques,
disponibilité en lumière, eau, nutriments), est également influencée par les interactions
interspécifiques qui y ont lieu. Enfin, les interactions entre espèces sont un moteur de
diversité. En effet chaque partenaire peut exercer des pressions de sélection sur l’autre,
affectant ainsi la survie et/ou la reproduction. En réponse à ces pressions, une des espèces
partenaires peut subir une modification de traits susceptible d’influencer à son tour des traits
de l’autre espèce (Tirard et al., 2012). Lorsqu’au moins deux espèces influencent ainsi
réciproquement l’évolution de l’autre, on parle de coévolution. Les processus coévolutifs ont
été largement étudiés dans le cas d'interactions antagonistes (hôte-parasite/parasitoïde, plante-
herbivore, proie-prédateur) et mutualistes (plantes-pollinisateurs, plantes-disperseurs de
graine/fruits) (Begon et al., 2006) et seraient à l'origine d'une diversification des espèces chez
certains taxons (Danforth, 2007).
Introduction | 19
Box 1 : Diversité d’interactions interspécifiques au sein des communautés
-/- : compétition
Interactions au cours de laquelle les partenaires
ont une influence négative réciproque. Selon le
principe d’exclusion compétitive de Gause Diminution de la
disponibilité des ressources
(1932), l’utilisation commune d’une ressource
limitée peut mener à l’élimination de l’espèce la
moins compétitive pour cette ressource. Bien
que l'espèce la plus compétitive se maintienne
dans le milieu, ses performances sont Diminution de la
disponibilité des ressources
également diminuées.
A : Compétition
pétition entre plantes pour la lumière
Que ce soit dans la revue de van Dam en 2009 ou le réseau d'interaction publié par Pocock et
al. en 2012, la communauté végétale se trouve au centre d'un ensemble d'interactions diverses,
soulignant ainsi l'importance des communautés de plantes au sein des écosystèmes.
Producteurs primaires, organismes à l'interface entre le milieu hypogé et le milieu épigé, les
plantes font partie de réseaux complexes et interagissent également entre elles et avec leur
milieu. En ce sens, elles illustrent parfaitement la complexité et la diversité des interactions
précédemment évoquées et constituent donc un modèle de choix pour une approche
intégrative.
Bien qu'il existe d'autres types d’interactions entre espèces végétales (Callaway, 1995), la
compétition interspécifique pour les ressources constitue une interaction majeure au sein des
communautés de plantes. En modulant les performances des espèces en interaction, la
compétition influence la distribution, la composition et la diversité des communautés
végétales naturelles (Goldberg & Barton, 1992). A une échelle plus large, les communautés de
plantes sont notamment intégrées dans des interactions mutualistes et antagonistes. Parmi les
partenaires mutualistes, nous pouvons citer les pollinisateurs ou les champignons
mycorhiziens qui vont respectivement avoir un rôle dans la reproduction d'une majorité de
plantes (Ollerton et al., 2011) ou dans leur nutrition (Marschner & Dell, 1994). En ce qui
concerne les interactions antagonistes, les plantes sont principalement confrontées à des
herbivores et parasites. Ces interactions vont moduler les performances de leur plantes hôtes
(Gurevitch et al., 2006) : les herbivores en diminuant la surface des structures d'acquisition
des ressources (feuilles, racines) ; les pathogènes en dérivant une partie de l'énergie produite
pour leur propre bénéfice. Par ailleurs les interactions antagonistes impliquent souvent des
résistances ou mécanismes de défense pouvant rentre en conflit avec des partenaires
mutualistes (Adler & Irwin, 2005). Bien sûr, les interactions entre espèces ne sont pas
toujours aussi clairement définies dans la nature. Dans certains cas il existerait plutôt un
continuum d'interaction. C'est le cas notamment des champignons mycorhiziens dont l'effet
peut aller du mutualisme à des relations proches du parasitisme, notamment selon les
conditions du milieu (Johnson et al., 1977 ; Johnson & Graham, 2013).
Introduction | 21
Box 2 : Les communautés végétales, éléments centraux des écosystèmes.
Introduction | 22
1.4 b) Des interactions entre réseaux par l'intermédiaire d'un compartiment commun
Outre des interactions entre plantes et leurs partenaires, le réseau proposé par Pocock et al.
2012 soulève également la présence d’effets indirects au sein des communautés naturelles. En
effet, les plantes se trouvent bien souvent au sein d'interactions multitrophiques, à plusieurs
niveaux. Bien qu’il existe une ambiguïté dans le cas de la compétition par exploitation
(Strauss 1991), nous définirons les effets indirects comme l'ensemble des relations au sein
desquelles un organisme va avoir en effet sur un autre via une troisième espèce jouant le rôle
d'intermédiaire. Parmi ces interactions, la compétition apparente (Holt, 1977) et les cascades
trophiques (Paine, 1980) sont probablement les plus étudiés (“top down” et “bottom up”
(Hanley & La Pierre, 2015). Dans le cas d'une régulation “top down” c'est le niveau trophique
supérieur qui exerce un contrôle sur les niveaux trophiques inférieurs. Ainsi dans le cas d'une
relation proie-prédateur, c'est le prédateur qui va influencer la structure et la dynamique du
réseau. En interagissant directement avec les proies, celui-ci influence également, mais de
manière indirecte, les niveaux trophiques inférieurs tels que les producteurs primaires. A
l'inverse, dans le cas d'une régulation “bottom-up”, c'est la disponibilité en nutriments et donc
les producteurs primaires qui modulent la structure et la dynamique du réseau jusqu'aux
niveaux supérieurs. Ainsi, l’ensemble de ces observations confirme la complexité et la
diversité des interactions au cœur des communautés naturelles.
Bien que les interactions écologiques aient été largement étudiées, leur diversité et leur
complexité soulèvent encore de nombreuses questions et débats (Agrawal et al., 2007 ; Mayer
et al., 2011, Trinder et al., 2013). A l'heure actuelle, l'enjeu principal reste d'adopter une
démarche intégrative, reliant les interactions et compartiments d'un écosystème. Les
communautés de plantes, au centre d'interactions diverses et complexes constituent un modèle
de choix pour de telles problématiques. Parmi les interactions précédemment citées, les
relations de pollinisation sont d'une importance majeure pour une large part des espèces
végétales actuelles. Il apparaît ainsi essentiel de s'intéresser à leur fonctionnement et réponse
en relation avec d'autres interactions.
Introduction | 23
2. Les plantes au cœur de réseaux d'interaction : cas des interactions
plantes-pollinisateurs
Bien que les premiers événements de pollinisation aient été décrits chez des Gymnospermes il
y a plus de 100 millions d'années (Ren et al., 2009 ; Peñalver et al., 2012), c'est bien au sein
des Angiospermes que ce type d'interaction s'est le plus développé. Les Angiospermes, qui
constituent la majorité des espèces végétales actuelles (Crane et al., 2000), sont caractérisées
par la production de fleurs, comme organes de reproduction. La pollinisation correspond au
transport de pollen des organes mâles jusqu'aux organes femelles d'une même fleur ou d'une
Introduction | 24
fleur différente afin de permettre la fécondation (voir Box 3). Lorsque la pollinisation (et donc
la fécondation) se fait entre fleurs d'un même individu, voire au sein d'une même fleur, on
parle d'autopollinisation (autogamie). Néanmoins, la plupart du temps, le transport de pollen
s'effectue entre fleurs d'individus distincts (allogamie), on parle de pollinisation (et donc
fécondation) croisée. Il existe différents vecteurs de pollen, abiotiques ou biotiques, favorisant
une fécondation croisée. Parmi les différents modes de pollinisation, on rencontre
l’anémogamie où le vecteur est le vent ainsi que l’hydrogamie où le vecteur est l’eau. On
parle alors respectivement d’espèces végétales anémophiles et hydrophiles. En revanche, pour
87,5 % des espèces d'Angiospermes le transport de pollen entre fleurs dépend d'un vecteur
animal (zoogamie ; Ollerton et al., 2011). Sous nos latitudes, et plus généralement dans le
monde, les insectes constituent le principal groupe de pollinisateurs (Danforth, 2007, Rader et
al., 2016) même si certaines espèces végétales peuvent être pollinisées par des oiseaux
(colibris), des mammifères (chauves-souris, rongeurs) ou encore des lézards (Proctor et al.,
1996). Lorsque les insectes assurent le transport de pollen, on parle de pollinisation
entomogame et de plantes entomophiles. Quatre ordres participent plus particulièrement à ce
type de pollinisation: les hyménoptères, les diptères, les lépidoptères et dans une moindre
mesure les coléoptères. Les hyménoptères, parmi lesquels la super famille des Apoidae
(abeilles et bourdons) constituent probablement les pollinisateurs les plus importants pour la
reproduction des Angiospermes. L'abeille domestique (Apis meliferra) et les bourdons
(Bombus spp.) s'avèrent être des pollinisateurs très efficaces pour la majorité des plantes
cultivées. Néanmoins, des études récentes ont également démontré la grande importance
d'espèces d'abeilles solitaires qui part leur abondance et richesse spécifique (et
complémentarité fonctionnelle) assurent la pollinisation de nombreuses espèces végétales
sauvages et cultivées (Greenleaf & Kremen, 2006, Winfree et al., 2007, Garibaldi et al.,
2013). Enfin, parmi les diptères, ce sont les syrphes qui semblent jouer un rôle important pour
certaines cultures comme le colza (Jauker & Wolters, 2008).
Introduction | 25
Box 3 : La reproduction des Angiospermes (« plantes à fleurs »)
La fleur constitue, chez les Angiospermes, l’élément central de la reproduction sexuée. Elles sont
notamment le lieu du processus de pollinisation : le transport des grains de pollen des étamines
(organe mâle) vers un stigmate (organe femelle). Le grain de pollen germe alors du stigmate jusqu’à
l’ovaire, acheminant, entre autres, un des gamètes mâle jusqu’au gamète femelle, l'ovule, pour
fécondation. Des graines se développent à partir des ovules fécondés. Elles contiennent un
embryon végétal et sont protégées au sein d'un fruit issu de la transformation de la paroi de
l'ovaire.
Grain de
pollen
ayant
germé
Stigmate
Style
Etamines
Ovaire
Ovule
Ci-dessus un exemple de pollinisation croisée chez une espèce entomophile (vecteur : abeille)
2.3 Les traits impliqués dans l'attractivité d'une plante aux insectes pollinisateurs
Introduction | 26
et al., 2005 ; Miyake & Sakai, 2005, Makino et al., 2007). Ceci limiterait notamment les coûts
associé au vol ente fleurs et à la revisite de fleur. De même la taille des fleurs a tendance à
influencer le choix des pollinisateurs qui préfèrent souvent de plus grandes corolles (Conner
& Rush, 1996 ; Elle & Carney, 2003). Par ailleurs, l’organisation de l'inflorescence (Pyke,
1978 ; Ishii et al., 2008 ; Iwata et al., 2012) ainsi que la morphologie des fleurs (symétrie
notamment, Möller, 1995 ; Hegland & Totland, 2005) peuvent influencer le comportement de
visite des pollinisateurs. En ce qui concerne la couleur des fleurs, sur la base de différences du
point de vue cognitif, les pollinisateurs ne sont pas tous sensibles aux même longueurs
d'ondes, et seraient donc attirés pas des couleurs différentes (Lunau &Maier, 1995 ; Chittka et
al., 2001). Il en est de même pour les odeurs (Raguso, 2001 ; Dobson, 2006). Alors que les
fleurs influencent l'initiation des visites sur une plante entomophile, les récompenses
favorisent la répétition des visites (Dafni, 1992). En effet, le nectar et le pollen constitue les
principales sources nutritives et énergétiques des pollinisateurs. Ainsi afin de combler leur
besoins énergétiques, certains pollinisateurs privilégient par exemple des visites sur des
espèces entomophiles offrant des récompenses plus concentrées en sucres (Cnaani et al.,
2006). L'ensemble de ces observations soulève donc l'existence, chez les pollinisateurs, de
préférences innées (relatives à leurs capacités cognitives par exemple) mais également
acquises (résultats de leur expérience de butinage). En effet, il a été démontré que certains
groupes d'insectes pollinisateurs sont capables d'associer certains traits floraux à la qualité des
récompenses rencontrées, pouvant ainsi réajuster leurs préférences (Weiss, 1997).
Introduction | 27
spécialiste ne va interagir qu'avec une petite proportion des espèces de plantes (pollinisateurs)
de la communauté voire avec une seule espèce uniquement. Néanmoins certains auteurs ont
émis des réserves à propos de ce concept. Waser et al., (1996) ont notamment suggéré que le
généralisme serait en réalité la règle et non l'exception: plantes et pollinisateurs interagiraient
avec une large part de la communauté. De plus le syndrome de pollinisation ne permettrait pas
de bien décrire la diversité de phénotypes observées chez les espèces entomophile, ni ne
permettrait de prédire les pollinisateurs d’une espèce (Ollerton et al., 2009) faisant que cette
problématique de généralisme/spécialisme est aujourd'hui toujours débattue.
Chez les espèces entomophiles qui en produisent, le nectar est secrété au niveau de structures
florales ou extra-florales, les nectaires. Selon les espèces nectarifères, les nectaires floraux
peuvent se trouver à la base de l’ovaire, des étamines ou des pétales (Heil et al., 2011).
Récompense principale des pollinisateurs (Kearns & Inouye, 1993), le nectar est une solution
principalement composée de sucres (monosaccharides : fructose, glucose et disaccharides :
sucrose). Il contient également des acides-aminées et protéines, des lipides et composés
secondaires tels que des alcaloïdes. Il constitue la source principale d'énergie des
pollinisateurs. Principalement composé de sucres, le nectar peut être relativement coûteux à
produire pour une plante : chez certaines espèces, jusqu'à 37 % du carbone assimilé par jour
via la photosynthèse peuvent être allouée à la production de sucres dans le nectar (Southwick,
1984). En parallèle une augmentation de la production de nectar peut conduire à une
diminution de la production de graines, suggérant ainsi la mise en place d'un trade-off (Pyke,
1991). La composition et la concentration en sucres du nectar peut varier entre espèces mais
aussi entre individus d'une même espèce (Gardener & Gillman, 2001a ; Herrera et al., 2006).
Selon les espèces de plantes, la production de nectar peut varier au cours d'une journée, d'une
saison, l'effet de l'âge de la plante pouvant avoir un effet positif (Pyke, 1978) ou négatif
(Devlin et al., 1987). Les variations observées à l'échelle de la journée présentent souvent un
pic de production (Kearns & Inouye, 1993), pouvant dépendre du pic d'activité du
pollinisateur (Real 1981). La distribution et la production du nectar peut également être
irrégulière au sein d'une communauté de plantes entomophiles, d’une population (Zimmerman
1981) voire au sein même d'une plante (Herrera & Soriguer, 1983 ; Keasar et al., 2008). La
variabilité observée dans la distribution ou encore la qualité du nectar entre fleurs d'une même
plante peut résulter de caractères propres à la plante (dynamique de production du nectar,
Introduction | 28
réabsorption) ou de facteurs extérieurs (comportement non aléatoire des pollinisateurs,
évaporation, levures dans le nectar) (Kearns & Inouye, 1993, Canto et al., 2007). L'ensemble
de ces paramètres peut influencer les communautés de pollinisateurs. Ainsi la production de
nectar par fleur est positivement corrélée au taux de visite des pollinisateurs (Kudo & Harder,
2005). De même la production de nectar à l'échelle d'une plante peut influencer le nombre de
visites totales reçues par une plante entomophile (Leiss & Klinkhamer, 2005). Par ailleurs, la
concentration et composition en sucres ou en acides aminées du nectar peut influencer les
pollinisateurs. Cnaani et al. (2006) ont par exemple démontré une relation positive entre la
concentration en sucre de nectar et le choix d'un bourdon. De même, des papillons peuvent
répondre préférentiellement à la présence de sucrose et de fructose dans le nectar (Romeis &
Wackers, 2000) tandis que d’autres pollinisateurs choisissent plutôt des nectars comprenant
également des acides aminés (Alm et al., 1990). Par ailleurs la composition du nectar en
acides aminés pour jouer un rôle important dans son attractivité au pollinisateur (Gardener &
Gillman, 2002 ; Petanidou et al., 2006). Petanidou et al., 2006 évoquent également l’effet
positif de faibles volumes de nectar (souvent associé à une plus grande concentration en
sucres) sur le choix des pollinisateurs. La diversité des nectars peut également être associée à
la richesse spécifique de certains groupes de pollinisateurs (Potts et al., 2004). Enfin la
distribution du nectar au sein d’une communauté ou d’une inflorescence peut moduler le
comportement de visite et décisions de départ sur une plante entomophile (Waddington &
Heinrich, 1979 ; Pyke, 1982, Keasar et al., 2008).
Afin de mieux comprendre les enjeux liés à l'attractivité d'une plante entomophile, il paraît
pertinent de la mettre en relation avec le succès reproducteur des plantes. De nombreuses
espèces entomophiles présenteraient une limitation par le pollen, c'est à dire une diminution
du succès reproducteur (notamment représenté par une baisse de la production de graine) lors
d'un dépôt insuffisant de pollen (en quantité ou qualité) sur le stigmate (Ashman et al., 2004).
La quantité ou la qualité du dépôt de pollen peut être positivement associé au nombre de
visites reçues (Engel & Irwin, 2003). Ainsi, la taille de la vitrine florale ou la qualité des
récompenses pourraient influencer le succès reproducteur d’une plante entomophile en
modulant la quantité de visites qu’elle reçoit. C'est ce qu'a notamment démontré un ensemble
d'études réalisées dans le contexte d'espèces entomophiles invasives. La présence d’une
Introduction | 29
espèce exotique plus attractive (taille des fleurs, vitrine florale) en focalisant les visites des
pollinisateurs, entraine chez de nombreuses espèces natives une diminution de leur succès
reproducteur (Brown & Mitchell, 2001 ; Muñoz & Cavieres, 2008 ; Kandori et al., 2009). Par
ailleurs, outre le nombre de visite, la diversité des visites peut également avoir un effet positif
sur le succès reproducteur (Klein et al., 2003 ; Albrecht et al., 2012).
Si chaque plante développe une attractivité individuelle, cette attractivité devient relative
lorsque différentes espèces se trouvent dans une même communauté. D'une part, si la quantité
de visites sur une plante dépend de sa vitrine florale, alors deux plantes ayant des vitrines
florales de taille différentes ne recevront pas le même nombre de visites. Par ailleurs, comme
nous l'avons vu dans la partie 2.3, les pollinisateurs peuvent avoir des préférences.
L’attractivité d’un signal visuel ou olfactif émis par une plante repose sur la capacité du
pollinisateur à le discriminer et à l'interpréter. Ainsi certaines plantes seront plus attractives
pour certains groupes de pollinisateurs. L’attractivité d’une plante peut notamment dépendre
de l'attractivité des plantes voisines qu'elles appartiennent à la même espèce (Toräng et al.,
2006) ou a d'autres espèces (Laverty, 1992 ; Moeller, 2004). La présence d'espèces ayant des
morphologies proches (Moeller, 2004) ou distinctes (Lazaro et al., 2009) peut favoriser
l'attraction de pollinisateurs. Ainsi il existerait des relations de facilitation entre plantes via le
partage de pollinisateurs (Moeller, 2004, Michalet et al., 2015). Mais dans certains contextes,
une modification de l'attractivité relative peut avoir des conséquences négatives sur ses
interactions avec des pollinisateurs. Dans le contexte d'introductions biologiques plusieurs
études ont notamment démontré que la présence d'une espèce produisant de nombreuses
fleurs, des fleurs de grande taille et/ou de grandes quantités de nectar attiraient
préférentiellement les pollinisateurs au détriment des espèces natives (Brown & Mitchell,
2002 ; Muñoz & Cavieres, 2008 ; Kandori et al., 2009). Cette baisse d’attractivité peut par la
suite influencer le succès reproducteur de la plante la moins attractive (Brown & Mitchell,
2002 ; Muñoz & Cavieres, 2008 ; Kandori et al., 2009). Il existerait ainsi une compétition
entre plantes pour les pollinisateurs, ressources biotiques essentielles pour la reproduction des
plantes entomophiles. Dans certains cas, en revanche, cette présence peut être positive,
l'espèce plus attractive attirant des pollinisateurs dans des parcelles qui n'auraient reçu que
Introduction | 30
peu de visites (“magnet effect”) (Laverty, 1992 ; Molina-Montenegro et al., 2008). La nature
des interactions entre espèces végétales induites par la présence d’une espèce attractive
(compétition ou facilitation) dépendrait notamment de sa densité au sein d’une communauté
(Seifan et al., 2014). Que l'effet soit positif ou négatif, la composition de la communauté
végétale et donc de l'environnement biotique peut moduler l'attractivité d'une plante aux
pollinisateurs. De même, l'attractivité d'une plante aux pollinisateurs peut être sensible à des
modifications de l'environnement abiotique et notamment des ressources du sol. En effet
plusieurs études ont mis en évidence que l'ajout de nutriments dans le sol via l'utilisation de
fertilisant (Gardener & Gillman, 2001b ; Burkle & Irwin, 2009, 2010 ; Muñoz et al. 2005,
Soper-Gorden & Adler, 2013), de litière (Baude et al., 2011), de compost (Cardoza et al.,
2012) ou encore l’irrigation (Petanidou et al., 1999) peuvent induire une augmentation de la
production de fleurs et/ou de récompenses (voir Tableau 1). Il a même été récemment
démontré que les dépôts azotés issus d'activités anthropiques pouvaient influencer de manière
positive le nombre de fleur, leur diamètre et de manière négative le ratio de certains sucres
dans le nectar (Hoover et al., 2012). Ces modifications d'attractivité peuvent par ailleurs
influencer le nombre de visite des pollinisateurs (Muñoz et al., 2005).
Au cours de cette deuxième partie, nous avons abordé les interactions plantes-pollinisateurs
dans leur diversité et complexité. Essentielles pour la reproduction d'une majorité d'espèces
végétales, ses interactions sont néanmoins sensibles à des variations de leur environnement
biotique et abiotique. La compétition interspécifique 1) jouant également un rôle majeur dans
la structure et le maintien des communautés végétales, et 2) pouvant moduler la disponibilité
des ressources dans le milieu, nous pouvons nous demander si elle est susceptible d'influencer
les interactions plante-pollinisateurs
Introduction | 31
Tableau 1 : Tableau récapitulant un ensemble de travaux ayant étudié l’effet de l’augmentation de la disponibilité des ressources sur
les traits floraux, nectar et pollen de différentes espèces végétales (tiré de Baude 2009)
Introduction | 32
3. La compétition interspécifique dans les communautés végétales et ses effets sur les
interactions plantes-pollinisateurs ?
Dans cette troisième partie nous nous intéresserons à la compétition interspécifique au sein
des communautés de plantes, plus particulièrement pour les ressources du sol. Nous verrons
quelles en sont les principales caractéristiques et les mécanismes, puis nous évoquerons ses
conséquences sur la croissance et la reproduction des plantes, soulevant ainsi la question de
son rôle au sein des interactions plantes-pollinisateurs.
Introduction | 33
l'un des deux partenaires. Dans la suite de ce travail nous nous concentrerons sur les
interactions de compétition entre espèces. Il existe deux grands types de compétition selon
que les espèces entrent directement ou indirectement en compétition pour la ressource. Il
s'agit respectivement de la compétition par interférence et de la compétition par exploitation
(voir Box 4). Parmi les formes de compétition par interférence nous pouvons notamment citer
l'allélopathie qui consiste en la production de molécules inhibitrices par une espèce à
l’encontre d’une autre pour en réduire l'accès aux ressources (de Kroon et al., 2003 ; Schenk,
2006). Dans le cas de la compétition par exploitation, chacune des espèces en compétition
répond à un niveau de ressource diminué par la présence d'une autre espèce. En ce sens on
peut alors distinguer l'effet d'un compétiteur sur la quantité de ressources et la réponse des
autres espèces à ce changement (Goldberg, 1990, voir Box 4). A noter que la notion de
ressources peut faire référence à des facteurs abiotiques, tels que ceux sur lesquels nous nous
focaliserons par la suite (eau, nutriments ou encore lumière) mais des facteurs biotiques
(pollinisateurs, champignons mycorhiziens) peuvent également être au centre d'interactions de
compétition en tant que ressources. Au sein des communautés végétales, la compétition peut
avoir lieu pour les ressources épigées au niveau d’organes aériens (lumière, espace) et/ou
hypogées au niveau d’organes souterrains (nutriments, eau, espace). Si l'intensité de la
compétition varie le long d'un gradient de fertilité, l'importance relative de ses composantes
aériennes et racinaires dépend de la disponibilité en ressources aériennes et souterraines. Une
récente méta-analyse, focalisée sur la compétition dans les systèmes prairiaux, a ainsi montré
que la compétition pour les ressources du sol était généralement plus importante que la
compétition aérienne (Kiaer et al., 2013). Ainsi, nous nous focaliserons sur les interactions de
compétition entre plantes pour les ressources du sol.
3.2 a) Prélèvement des ressources par les plantes et conséquences sur le sol
Dans le sol, les plantes sont en compétition principalement pour l'eau, l'azote et la phosphore,
éléments généralement les plus limitants de la production primaire des communautés
végétales (Aerts & Chapin, 2000). Leurs quantités dans le milieu sont fortement dépendantes
des conditions climatiques, des événements météorologiques (pluies, sécheresses), des
paramètres édaphiques et des interactions biotiques (Gurevitch et al., 2006). Avant d'aborder
Introduction | 34
les mécanismes de la compétition par exploitation, il est important de s'intéresser aux
ressources en elle même et à la manière dont les plantes les prélèvent dans un sol. Du fait de
leur immobilité, les plantes ne peuvent prélever des ressources qu'à la surface de leurs racines.
Ainsi, si une petite part des ressources peut être acquise directement par interception (c'est à
dire que les ressources sont prélevées à mesure que la racine s’agrandit), la grande majorité
des ressources se trouve prélevée par le biais de processus physiques contribuant à déplacer
les ressources vers les surfaces d'absorption (Casper & Jackson, 1997). Ainsi, le prélèvement
de l'eau au niveau des surfaces racinaires conduit à la création d'un flux d'eau du sol vers les
racines. Ce flux contribue à l'alimentation hydrique de la plante mais également à une part de
la nutrition minérale lorsque les nutriments se trouvent dissous dans l'eau du sol (Casper &
Jackson, 1997 ; Aerts & Chapin, 2000). De manière similaire, l'abaissement de la
concentration en nutriments contre les surfaces racinaires crée un gradient de concentration
conduisant à la création d'un flux de nutriments du sol vers les racines (Marschner, 1995).
Bien que tous ces différents processus agissent de manière combinée, l’acquisition des
nutriments se fait principalement par diffusion (Aerts & Chapin, 2000). D’autres mécanismes,
comme la préemption de ressources par l’espèce compétitive, ont également été proposés
(Craine et al., 2005a). La quantité de ressources prélevée par une plante dépend donc de sa
capacité d'absorption par unité de biomasse racinaire (i.e., sa capacité à abaisser la
concentration à la surface des racines), de sa biomasse racinaire (i.e., qui définie la surface
totale d'absorption de la plante) ainsi que de la mobilité des ressources considérées (Aerts &
Chapin, 2000). Par ailleurs, ces changements de concentration en ressources autour des
racines conduisent au développement de la rhizosphère, définie comme la zone du sol sous
l'influence des plantes et partagées avec les microorganismes (Hinsinger et al., 2009).
Sur la base des travaux de Mac Arthur (1972), Tilman (1980) a proposé qu'un mécanisme de
la compétition serait lié à l'abaissement de la concentration en ressource du sol par les plantes.
En effet, en prélevant des nutriments du sol, les plantes abaissent leur concentration dans le
sol. Des plantes ayant des capacités de prélèvement différentes conduiraient ainsi à des
concentrations en nutriments différentes en deçà de laquelle elles ne pourraient persister.
Ainsi, l'espèce qui, dans une communauté pourrait diminuer la concentration moyenne d'une
ressource à son niveau le plus bas tout en persistant dans ce milieu devrait exclure les autres
espèces présentes. La concentration à laquelle une espèce amène le sol, appelée R*, serait
Introduction | 35
alors un indicateur de la compétitivité de cette espèce et l'espèce de la communauté possédant
le R* le plus faible serait plus compétitive (Tilman,1990). Cependant, la capacité d'une espèce
à abaisser la concentration en ressources du sol dépend non seulement de ses propriétés
morphologiques et physiologiques, mais également de la mobilité des ressources considérées.
Ainsi, si les ressources sont peu mobiles, l'effet du prélèvement sur la concentration moyenne
du sol sera fortement limité (Raynaud & Leadley, 2004, Craine et al., 2005a). Cependant, une
plante sera d'autant plus capable de priver ses voisines de ressources qu'elle est capable de
produire de la biomasse. Ainsi, Grime (1979,1988) a proposé une caractérisation des capacité
de compétition des espèces selon différentes stratégies de développement en distinguant des
espèces compétitives (C), des espèces tolérantes au stress (S) et des espèces tolérantes aux
perturbations (R). Ces différentes stratégies seraient associé certains traits. En ce qui concerne
les espèces compétitives, ces traits sont un taux de croissance rapide et des systèmes aériens et
racinaires larges et étendus (Booth et al., 2010). Dans ce cas, la compétitivité d'une espèce
s'exprimerait alors par sa capacité à occuper l'espace et à acquérir des ressources.
3.2 c) Compétitivité d'une espèce pour les ressources du sol : quels traits ?
Introduction | 36
des traits favorisant l’acquisition des ressources, une meilleure compétitivité peut également
s’expliquer par de meilleures capacités de rétention des nutriments (Aerts et al., 1999).
Néanmoins, comme le suggère Grime, la plasticité phénotypique, cette capacité à produire
différents phénotypes selon les conditions du milieu à partir d'un même génotype, joue un rôle
majeur dans la réponse des plantes à la compétition. Bien qu'il existe certains traits
intrinsèque à une plante pouvant lui conférer une avantage sans événement de plasticité (fort
taux de croissance par exemple), une grande partie des traits cités précédemment sont le
résultats d'une plasticité physiologique ou morphologique en réponse à la disponibilité des
ressources dans le milieu : modification dans l’allocation de biomasse (Aerts, 1991 ; Fitter,
1994 ; Müller et al., 2000 ; Hermans et al., 2006), modification de l'architecture et de
l'arrangement spatial des racines (Fitter, 1994 ; Lopez-Bucio et al., 2003), Ainsi, la capacité à
avoir un comportement plastique pourrait être, pour une plante, un trait de compétitivité.
3.3 Quelles conséquences de la compétition entre plantes pour les ressources du sol sur les
interactions plantes-pollinisateurs ?
Introduction | 37
quantité totale de sucre allouée au nectar. Nous avons vu au cours de la partie 2.6 que les traits
d'attractivité aux pollinisateurs peuvent être sensibles à des variations de la quantité de
ressources dans le milieu, ces modifications pouvant à leur tout influencer les visites des
pollinisateurs (Muñoz et al., 2005). Dans le cas de la compétition, la réduction de la
disponibilité des ressources du sol par une espèce compétitrice conduirait à la diminution de
l'attractivité des plantes entomophiles aux pollinisateurs. Ainsi la disponibilité en nutriment,
modulée par la compétition, aurait un effet « bottom up » sur les communautés de
pollinisateurs. Néanmoins, la réponse des pollinisateurs à des modifications de traits
d'attractivité induites par compétition entre plante n'a pas été encore étudiée à notre
connaissance. Dans ce contexte, nous nous sommes alors demandé si la présence d'espèce
anémophiles, très rarement intégrées dans les études sur les réseaux de pollinisation,
pouvait avoir une influence sur l'attractivité aux pollinisateurs de plantes entomophiles,
et ce à travers des interactions de compétition pour les ressources du sol.
Introduction | 38
II. Objectifs de la thèse et méthodes générales
1. Objectifs de la thèse
Ce travail de thèse repose sur trois expérimentations qui constituent les principaux chapitres
du manuscrit. De manière générale, ces expériences visaient à mieux comprendre les
interactions entre communautés végétales et communautés de pollinisateurs. Plus
particulièrement, il s’agissait de déterminer si les espèces anémophiles, qui n'interagissent pas
directement avec la faune pollinisatrice, peuvent néanmoins moduler les interactions plantes-
pollinisateurs par leur présence dans le milieu. La compétition pour l'acquisition des
ressources étant une composante majeure de la structuration des communautés de plantes
(Tilman, 2012), j’ai cherché à étudier l'effet de la compétition induite par la présence
d'espèces anémophiles sur l'attractivité aux pollinisateurs d'espèces entomophiles. Pour
atteindre cet objectif, j’ai adopté une démarche allant de l'échelle de la plante à celle de la
communauté.
Ainsi, la première expérience, qui s'est déroulée sous serre, s’est concentrée sur la mesure
quotidienne de différents paramètres floraux d’intérêt chez 3 espèces entomophiles focales en
compétition avec des espèces anémophiles. Ces dernières ont été choisies afin de soumettre
les espèces focales à un panel d’interactions compétitives. Cette expérience en pot visait à
déterminer 1) si la présence d'une espèce anémophile compétitive pouvait avoir un effet sur la
production de traits floraux d'espèces entomophiles (Fig. 1) et 2) si l'importance de cet effet
dépendait de la compétitivité de l'espèce anémophile. Cette expérience s’intéressait donc à la
réponse individuelle des espèces entomophiles et à leur attractivité potentielle ces plantes
n’étant pas présentées à des pollinisateurs (Voir Chapitre 1).
Méthodes : Pots sous serre Méthodes : Pots sous serre, compétition Méthodes : En champs, compétition souterraine contrôlée
Pollinisateur(s) : aucun souterraine contrôlée Pollinisateur(s) : pollinisateurs sauvages
Variables étudiées : Fleurs, nectar Pollinisateur(s) : B. terrestris Variables étudiées : Fleurs, nectar, visites, fruits/graines
Variables étudiées : Fleurs, nectar, visites
Comme nous l'avons évoqué dans les objectifs, cette thèse se base sur trois expérimentations
similaires, au fur et à mesure desquelles, néanmoins, nous avons apporté des degrés de
complexité supplémentaire. De ce fait, de nombreux paramètres ont été étudiés de la même
manière sur l'ensemble des trois expériences. Afin d'avoir une vision plus globale, l'ensemble
de ces mesures sont ici présentées et résumés dans le tableau 2.
Dans l'ensemble des trois expérimentations, nous avons fait le choix de nous focaliser sur un
ensemble de traits impliqués dans l'attractivité d'une plante aux pollinisateurs. Les fleurs sont
les premiers « indices » perçus par un pollinisateur. Certains de leurs traits (nombre, taille)
peuvent être annonciateurs de la qualité des ressources florales disponible chez d'une plante
(Dafni 1992) favorisant ainsi le nombre de visite reçues par cette plante. Outre cette fonction
de « promotion » auprès des pollinisateurs, les fleurs fournissent des récompenses nutritives.
Certes, le pollen est une importante source de protéines, mais les besoins énergétiques de la
majorité des insectes pollinisateurs reposent sur le nectar, source de glucides (saccharose,
glucose et fructose principalement) et dans une quantité moindre, d'acides aminés (Dafni,
1992). La quantité (volume) et la qualité (composition) du nectar sont donc des paramètres
essentiels de l'attractivité d'une plante aux pollinisateurs (Cnaani et al., 2006)
Sur ces bases, dans les trois expériences réalisées au cours de cette thèse, des mesures
régulières ont été réalisées: la production de fleurs nouvelles, la vitrine florale (ici le nombre
total de fleurs ouvertes au même moment), la taille des fleurs ainsi que la production de nectar
(volume, concentration et quantité en saccharose).
La production de fleurs a été suivie quotidiennement dans les 3 expériences. Sur l'ensemble
des pieds des espèces entomophiles, les boutons floraux ont été marqués la veille. Le
lendemain, les fleurs ayant éclot ont alors été comptées. Ce nombre correspond ainsi au
nombre de fleurs nouvellement produites par plante (production de fleur). Par ailleurs la
Afin d'estimer le volume de nectar produit par fleur par jour, nous avons effectué des
prélèvements à l'aide de microcapillaires calibrés de 0,5µL à 1µL (Minicaps end to end,
Hirschmann laborgeraete).Le prélèvement consiste à introduire le capillaire dans la corolle de la
fleur afin de récolter le nectar produit. Les capillaires étant calibrés en taille et volume, il
suffit de mesure la « longueur » du volume prélevé avec un pied à coulisse pour pouvoir
estimer le volume. Afin d'estimer la concentration en saccharose du nectar nous avons utilisé
des réfractomètres de poche spécialement conçus pour les petits volumes. Ces instruments
permettent de déterminer l'indice de réfraction d'un liquide. L'indice de réfraction du nectar
est utilisé comme mesure d'équivalent saccharose car cette méthode ne prend pas en compte la
diversité de sucres présents dans le nectar (Kearns & Inouye, 1993). Néanmoins, afin d'être
plus concis, nous parlerons par la suite de quantité de saccharose. Les valeurs obtenues sont
exprimées en degrés Brix ( % B) représentant la fraction d'un composé dans un liquide en
gramme de soluté par gramme de solution (ici g saccharose / g nectar). Les mesures de
concentration en saccharose du nectar ont été réalisées au plus tard quelques heures suivant le
prélèvement. Lorsque nécessaire (pour la lecture au réfractomètre), les échantillons ont été
diluée dans de l’eau milliQ (Millipore Corporation). Nous avons ensuite converti ces valeurs
-1
en concentration (g saccharose.L ) via un tableau de correspondance (Kearns & Inouye 1993).
Les instruments sont calibrés au préalable avec une solution en saccharose de concentration
connue à 20°C. Une table de correction est fournie avec les instruments afin de prendre en
compte d'éventuelles variations de températures lors de la mesure.
2.1 c) Estimation de l'investissement global d'une plante dans les traits d'attractivité
Nous nous sommes également intéressés à l'investissement d'une plante dans les traits
d'attractivité sur l'ensemble de la période de floraison. Pour cela nous avons étudié la
production totale de fleurs par plante dans les trois expériences. De même, puisque nous
n'avons pas fait de prélèvement sur l'ensemble des fleurs produites par chaque plante
Indice volume nectar journalier : volume moyen de nectar.fleur. jour-1 x production de fleurs. jour-1
Indice quantité saccharose journalière : quantité moyenne de saccharose allouée au nectar.fleur. jour-1 x
production de fleurs. jour-1
Ces indices ont ensuite été sommés sur l'ensemble de la période de floraison pour chaque
plante entomophile étudiée :
Ces indices ont été calculés pour les expériences 1 et 2. En revanche pour l'expérience 3, les
mesures de taille des fleurs et de nectar ont été faites avant chaque observation, tous les pieds
des plantes entomophiles n'étant alors pas échantillonnés tous les jours. De ce fait ces indices
n'ont pas été réalisés dans cette expérience.
Par ailleurs d'autres traits propre aux objectifs de chaque expérimentation ont été mesurés
(visites, fécondité des plantes focales).
Afin d'estimer les interactions de compétition entre plantes, nous avons récoltés les biomasses
aérienne et racinaire. Ces biomasses ont été récoltées en fin d'expérience
- sur l'ensemble des plantes du dispositif expérimental (espèces entomophiles et espèces
anémophiles) dans les expériences 1 et 2
- sur les plantes entomophiles dans l'expérience 3.
Ces mesures de biomasse nous ont permis de calculer un indice d'intensité de compétition, le
log Response Ratio (d'après Weigelt & Jolliffe, 2003):
Pmix
lnRR=
( )
P control
Traits d'attractivité
Mesures quotidiennes
Présentation du chapitre :
Chapitre 1 | 53
Ces résultats suggèrent que les espèces
anémophiles les plus compétitives ont
réduit la disponibilité des ressources
ress
(espace, nutriments, eau) au détriment
des espèces focales, influençant ainsi
leur allocation aux traits floraux. Dans
le cas des mélanges avec C. album,
album
nous avons observé une augmentation
des paramètres floraux étudiés. Cet
effet résulterait d'unn investissement très
faible dans la biomasse chez C. album
dans cette expérience, les plantes
focales ayant probablement eu accès à
un pool de ressources plus important
qu'en monocultures. En ce qui concerne
la troisième espèce focale, L.
corniculatus, less résultats ne suivent Figure 2 : Cadre d'étude du Chapitre
C 1
pas le même patron de réponse. Les traits
floraux semblent plus particulièrement affectés en conditions de monoculture. Chez cette
espèce, les mesures de biomasse et les indices de compétition ne suggèrent pas de lien entre
l'allocation à la biomasse végétative et l'allocation à la biomasse reproductrice, comme cela
peut être le cas pour les deux espèces annuelles.
ann La réponse de L. corniculatus pourrait donc
être plus « adaptative » que « plastique ». Cependant, L. corniculatus étant une espèce
pérenne, et fixatrice d'azote, il est possible que ces caractéristiques aient joué un rôle dans sa
réponse à la compétition.
étition. Faute d'avoir pu étudier la réponse d'une autre espèce pérenne,
initialement intégrée au protocole expérimental (Mimulus
( ), nous ne pouvons tirer de
guttatus),
conclusions précises pour cette espèce.
Chapitre 1 | 54
www.nature.com/scientificreports
A wide array of plant traits are sensitive to environmental conditions, either abiotic and biotic factors
and their interplay. Modiications of abiotic resources can induce positive to negative efects on plant
vegetative (e.g. plant biomass) and reproductive traits (e.g. lowers, fruits or seeds). For instance, some
studies showed that the addition of soil nutrients (e.g. nitrogen or phosphorous) can lead to an increase
in plant growth rates or biomass1–4, as well as lower3,5–8, fruit6,8,9, pollen9 or seed traits6–8. On the other
hand some studies showed that a high level of nitrogen can lead to a decrease in root biomass4 while
increasing plant growth and lower production7. Likewise, litter and compost addition or irrigation can
induce intricate plant responses. For instance, litter inputs can induce negative to positive efects on
biomass along years10 as well as species-speciic response of loral traits11 just like water addition12. he
efect of resources modiications can also be delayed according to plant life cycles3. hese studies suggest
that plant response to resources availability can be plastic (the change in phenotype being proportional
to changes in environmental conditions) or adaptive with strategies mainly linked to plants’ life cycles.
Within plant communities, interactions among plant individuals, especially competition between roots
systems for water and nutrient acquisition, can lead to changes in the availability of such resources13
and to changes in allocation to vegetative or reproductive plant traits11,14–16. herefore, the composition
1
cnRS, Sorbonne Universités, UPMc Univ Paris 6, inRA iRD, Univ Paris Diderot Paris 7, UPec, institute of ecology
and environmental Sciences - Paris, UMR 7618, 7 Quai St Bernard, f-75005 Paris france. 2Sorbonne Universités,
UPMc Univ Paris 6, cnRS, inRA iRD, Univ Paris Diderot Paris 7, UPec, institute of ecology and environmental
Sciences - Paris, UMR 7618, 7 Quai St Bernard, f-75005 Paris france. 3cnRS, UMS 3194 ceReeP-ecotron ile de
france, f-77140 Saint-Pierre-lès-nemours. 4Univ Paris Diderot Paris 7, Sorbonne Universités, UPMc Univ Paris 6,
cnRS, inRA iRD, UPec, institute of ecology and environmental Sciences - Paris, UMR 7618, 7 Quai St Bernard,
F-75005 Paris France. Correspondence and requests for materials should be addressed to F.F. (email: loriane.
[email protected])
of plant communities can have strong efects on individual plant traits through competitive interactions
for resources.
Variations in plant reproductive traits are especially important in animal-pollinated species because
they condition plant-pollinator interactions17,18. Indeed, in insect-pollinated plants, pollinators are
attracted to lowers and associated rewards. On the one hand, lowers, varying in their number, size,
color or smell, ofer an advertising display that induces visits of various pollinators17,19. On the other
hand, loral rewards (e.g. nectar, pollen) tend to favor the repetition of visits as they are major compo-
nents of pollinators’ diet by supplying proteins, sugars and amino-acids17. he quantity and quality of
such loral traits involved in pollinator attraction can have strong impacts on pollinator behavior. Indeed,
several studies have shown that plant species exhibiting a greater loral display size, (i.e. the total number
of opened lowers at a time) or producing numerous, large lowers and/or greater rewards (in quantity
or quality) are more visited by pollinators than other present plant species20–25. Various experiments
showed that a greater pollinator attractiveness, subsequently to increased loral traits through resource
addition, can enhance pollinator visitation6–8,26 that may lead to better reproductive success6,8. However,
to date, most studies linking changes in soil resources to loral traits and pollinator response considered
the impact of nutrient addition while interactions between plants, especially belowground competition,
could also be of great importance. In an experiment looking at diferent loral traits involved in attrac-
tiveness, Baude et al.11 set up binary mixtures of insect-pollinated plant species and found that loral
traits of one focal species depended on the other species present in the mixtures. More precisely, the
total nectar sugar content of a focal species decreased when growing in presence of a stronger compet-
itor. herefore, interactions between insect-pollinated species could inluence loral traits involved in
pollinator attraction.
Natural plant communities always comprise species with a variety of pollination modes such as
animal-pollinated and abiotically pollinated plants, although the latter are almost never taken into account
in studies on plant-pollinator networks. However, the lower production of a particular insect-pollinated
species can be negatively impacted by competitive interactions induced by a wind-pollinated plant com-
petitor15,27. To our knowledge, the consequences of interactions between insect- and wind-pollinated
species on loral traits involved in attractiveness to pollinators still need investigations, especially with
more focus on loral rewards.
he objectives of our study were to assess how allocation to several loral traits involved in attrac-
tiveness to pollinators (e.g. lower production, lower size and loral rewards) could be afected by the
presence of diferent wind-pollinated species. Especially we wanted to investigate the efect of diferent
intensities of competitive interactions. To do so, we set up a pot experiment in which we grew three
insect-pollinated plant species (Echium plantagineum, Lamium purpureum and Lotus corniculatus) in
binary mixtures with four wind-pollinated species (Agrostis capillaris, Chenopodium album, Holcus lana-
tus and Plantago lanceolata) so that insect-pollinated species were submitted to a panel of belowground
interactions for abiotic resources.
Our hypothesis were that (1) the presence of wind-pollinated competitors should have negative
impacts on loral traits of insect-pollinated species and (2) the magnitude of this efect should difer
according to diferent competition intensities induced by the presence of wind-pollinated species.
Results
Intensity of competitive interactions. Mean log response ratios (ln RR28), estimators of compe-
tition intensity, are summed up in Table 1. As indicated by ln RR values, the three insect-pollinated
focals were submitted to various intensities of competition. E. plantagineum and L. purpureum followed
the same pattern of response. For both species and whatever the biomass measurement (aboveground,
belowground or total biomass), ln RR values (as well as focals’ biomass) were signiicantly higher in
presence of C. album, in opposition to mixtures with H. lanatus for which ln RR had the lowest values
(P ≤ 0.024 for both species, Table 1, Supplementary Fig. S1 for biomass). Mixtures with C. album even
seemed to provide better growth conditions than monocultures of the two focal species with positive
values of ln RR (Table 1). For intermediate levels of competition intensity, ln RR values suggest a global
pattern where intensity of competition is stronger in mixtures with A. capillaris than with C. album but
weaker than with P. lanceolata, the second strongest competitor for L. purpureum and E. plantagineum
(except for ln RR values calculated with belowground biomass of L. purpureum, Table 1). According to
these results, L. purpureum and E. plantagineum experienced the following panel of growing intensity
of competition:
C . album mixtures < monocultures < A . capillaris mixtures < P. lanceolata mixtures < H . lanatus mixtures
he response of L. corniculatus didn’t follow the same pattern. Monocultures showed the highest ln
RR values among all treatments (Table 1). In consequence, all wind-pollinated species induced negative
competitive interactions for this focal species ranging from weak efects in mixture with C. album to
strong efects in mixture with P. lanceolata or A. capillaris (Table 1).
As we are mainly interested in the efect of competition intensity induced by wind-pollinated species
rather than on competitor identity, the log response ratio (ln RR) has been used as an explanatory varia-
ble in the following results. However, because ln RR values were obtained from biomass measurements at
ln RR F value p-value
Monoculture Mixture H. lanatus Mixture P. lanceolata Mixture A. capillaris Mixture C. album
E. plantagineum
Aboveground biomass 0 (+ /− 0.01) − 1.16 (+ /− 0.01) − 0.86 (+ /− 0.01) − 0.60 (+ /− 0.01) 0.28 (+ /− 0.01) F4,20 = 54.31 0.001
Belowground biomass 0 (+ /− 0.04) − 1.96 (+ /− 0.03) − 1.77 (+ /− 0.03) − 1.23 (+ /− 0.04) 0.39 (+ /− 0.04) F4,20 = 20.73 0.001
Total biomass 0 (+ /− 0.02) − 1.43 (+ /− 0.02) − 1.15 (+ /− 0.02) − 0.82 (+ /− 0.02) 0.27 (+ /− 0.02) F4,20 = 46.97 0.001
L. purpureum
Aboveground biomass 0 (+ /− 0.02) − 1.00 (+ /− 0.02) − 0.81 (+ /− 0.02) − 0.22 (+ /− 0.02) 0.44 (+ /− 0.02) F4,20 = 19.08 0.001
Belowground biomass 0 (+ /− 0.05) − 1.32 (+ /− 0.04) − 0.86 (+ /− 0.04) − 1.00 (+ /− 0.04) − 0.42 (+ /− 0.04) F4,20= 3.55 0.024
Total biomass 0 (+ /− 0.03) − 1.15 (+ /− 0.02) − 0.88 (+ /− 0.03) − 0.49 (+ /− 0.03) 0.19 (+ /− 0.02) F4,20 = 12.74 0.001
L. corniculatus
Aboveground biomass 0 (+ − 0.01) 0.03 (+ /− 0.01) − 0.25 (+ − 0.01) 0.06 (+ /− 0.01) 0.03 (+ /− 0.01) F4,20= 6.40 0.002
Belowground biomass 0 (+ /− 0.03) − 0.47 (+ /− 0.03) − 0.74 (+ /− 0.03) − 1.02 (+ /− 0.02) − 0.21 (+ /− 0.03) F4,20 = 6,75 0.001
Total biomass 0 (+ − 0.02) − 0.13 (+ /− 0.02) − 0.41 (+ /− 0.01) − 0.26 (+ /− 0.02) − 0.07 (+ /− 0.01) F4,20 = 6,39 0.002
Table 1. Mean ln RR values (+/−standard error) per treatment for E. plantagineum, L. purpureum and
L. corniculatus in monocultures and mixtures with the wind-pollinated competitors. For each insect-
pollinated focal, p-values indicate signiicant diferences between treatments (ANOVA, E. plantagineum
N = 1125, L. purpureum N = 1125, L. corniculatus N = 1125).
the end of the experiment, we used ln RR (calculated from total biomass, see in Methods) as an explana-
tory variable only for total loral traits (i.e. summed at the end of the experiment, see Flower traits, Nectar
traits and Data analysis sections in Methods). We kept the competitor identity as an explanatory variable
for daily loral traits see Flower traits, Nectar traits and Data analysis sections in Methods). herefore,
in the following results, ‘competition intensity’ will equally refer to 1) ln RR values, 2) the presence of
a particular competitor in the mixture with the focal species. An increase in competition intensity can
thus mean a decrease in ln RR values or the presence of stronger competitors. For the sake of clarity, the
results for daily loral traits will be ordered along the above panel for the three focal species, even if the
pattern of response was diferent for L. corniculatus.
Flower traits. At the end of the experiment, a total of 807, 2053 and 1075 lowers were sampled for
E. plantagineum, L. purpureum, and L. corniculatus, respectively. For E. plantagineum and L. purpureum
there was a signiicant efect of competition leading to a decrease of loral display size, daily lower
production (not shown) and total lower production (total number of lower produced at the end of the
lowering period, see Methods) when competition intensity increased. Indeed, loral display size and
daily lower production tended to be greater in presence of C. album (even greater than in monocul-
tures; although not signiicant) while the presence of stronger competitors, such as H. lanatus, induced
a strong decrease in both traits (Fig. 1 for loral display size, the daily lower production followed the
same pattern). Likewise, for both species, the total lower production decrease as ln RR values decrease,
suggesting lower total lower production in condition of stronger competitive interactions (P < 0.001
Supplementary information Fig. S2). For L. corniculatus, there was also a signiicant efect of the com-
petition treatment. Especially, loral display size and daily lower production were greater in presence of
C. album (Fig. 1 for loral display size, the daily lower production followed the same pattern). However,
both traits tended to be lower in monocultures (Fig. 1 for loral display size, the daily lower production
followed the same pattern). On the other hand, the total lower production decreased according to ln
RR values (P < 0.001, Supplementary information Fig. S2). For all three species, there were signiicant
efects of the date (for the loral display size only, all P < 0.001) and of the interaction terms for both
loral display size and daily lower production (all P < 0.001).
For all three focal species, lower size was afected by the competition treatment (E. plantagineum:
F4,772 = 7.35, P < 0.001, L purpureum: F1,1936 = 8.96, P < 0.001; L corniculatus: F4,1036 = 4.69, P < 0.001,
Supplementary information Fig. S3). Plants of E. plantagineum and L. purpureum produced smaller low-
ers as competition intensiies. For L. corniculatus, monocultures and mixtures with C. album produced
the smallest lowers (see Supplementary information Fig. S3). For lower size, there was also a signiicant
efect of the date for all three species (all P ≤ 0.008).
Nectar traits. here was no signiicant efect of the competition treatment on the daily concentration
of nectar per lower and the daily volume of nectar per lower for both E. plantagineum and L. cornicula-
tus (P > 0.05). L. purpureum, however, was signiicantly afected by the presence of stronger competitors
Figure 1. Mean (+/−standard error) loral display size per plant of E. plantagineum, L. purpureum, and
L. corniculatus in mixture with the competitors. “Mono” refers to monocultures of the focal species. Wind-
pollinated species are ordered according to increasing intensity of competitive interactions (see Results).
Diferent letters indicate signiicant diferences (i.e P ≤ 0.006 for E. plantagineum, P ≤ 0.001 for L. purpureum
and P ≤ 0.040 for L. corniculatus) ater pairwise comparisons (Tukey57) and adjustment of p-values (Holm
method56).
Figure 2. Linear regression between the total sucrose index (mg per plant) and mean ln RR values per
plant calculated from total biomass (P < 0.001 with transformed data, for E. plantagineum (R2 = 0.44)
and L. purpureum (R2 = 0.57) only). he grey line corresponds to the estimated model while dots represent
the data. Ln RR values are associated to mixtures in the legend.
for both variables (F4,1450 = 2.53, P = 0.04 for the daily concentration and F4,1648 = 4.47, P = 0.001 for the
daily nectar volume).
As the daily lower production was afected by the competition treatment, the daily sucrose index of
nectar (taking into account all produced lowers per day and not only sampled lowers, see Methods)
decreased in presence of strongest competitors for both E. plantagineum and L. purpureum (F4,194 = 13.36,
P < 0.001and F4,731 = 5.52, P < 0.001 respectively). herefore, the daily allocation of sucrose to nectar was
lower when competition intensiied. he same pattern was clearly observed for the total allocation of
sucrose to nectar. Indeed, the total sucrose index at the end of the lowering period tended to decrease
with ln RR values (E. plantagineum F1,47 = 57.04, P < 0.001, L. purpureum: F1,48 = 83.81, P < 0.001; Fig. 2,
see Methods). We found the same pattern for daily (E. plantagineum F4,199 = 9.38, P < 0.001; L. purpureum
F4,800 = 8.05, P < 0.001) and total volume index (F 1,46 = 30,71, P < 0.001, F 1,48 = 102,74 P < 0.001 respec-
tively). For L. corniculatus, the daily sucrose index and the daily volume index were signiicantly afected
by the competition treatment (F4,255 = 3.48, P = 0.009 and F4,254 = 3.47, P = 0.009 respectively) with lower
values in monocultures. However both index of nectar were not afected by increasing competition inten-
sity (both P > 0.05, see Fig. 2 for total sucrose index). here were signiicant efects of the date for the
daily concentration of nectar (P ≤ 0.02 for E. plantagineum and L. purpureum) and the daily nectar
volume (L. purpureum only P < 0.001). here was a signiicant efect of the interaction term (competi-
tion:date) for the daily nectar volume (for E. plantagineum and L. corniculatus, all P ≤ 0.01). For both
daily index, there were signiicant efects of the date (for L. purpureum only, P < 0.001) and of the inter-
action term (competition:date) (for E. plantagineum and L. corniculatus, all P ≤ 0.002).
Discussion
Literature data have mostly focused on the relations between plant attractiveness to pollinators and
abiotic conditions and suggest that response of attractiveness traits is complex and species-speciic
as positive, neutral and negative efects have been reported for the efects of water or nutrient addi-
tion on lower productionp3,7,11,12. Because competitive interactions between plants can translate into
changes in resources availability between competitors13,29, we studied their impact on loral traits of three
insect-pollinated species. Moreover, we focused on wind-pollinated species as competitors in order to see
how species that do not interact with pollinators for their reproduction could alter attractiveness traits
of animal-pollinated focal species.
As expected, competition with wind-pollinated species afected plant biomasses and loral traits for
all three investigated insect-pollinated species. Especially, we found that the stronger the competitive
interaction was (i.e., high ln RR values), the stronger the impact on loral traits.
Diferences in biomass allocation patterns, especially belowground, can suggest diferent competitive
abilities in plants29. Indices of competition, such as the log response ratio (ln RR) are frequently used
as they are good tools for summarizing and interpreting competitive interactions between plant spe-
cies28. Here, the ln RR values obtained for each mixture indicate that the wind-pollinated competitors
exposed the insect-pollinated focals to diferent competition intensities. hese ln RR values also sug-
gest that the two annual species of our experiment, L. purpureum and E. plantagineum, faced a similar
panel of growing intensity of competition, with C. album being a weaker competitor, A. capillaris an
average competitor and P. lanceolata and H. lanatus being stronger competitors. Measures of competi-
tors’ biomass grown in mixtures with L. purpureum or E. plantagineum could explain these diferences.
Indeed, in these mixtures, plants of H. lanatus produced the greatest biomass among the four competitor
species. Even though greater biomasses, especially greater root systems, are not always associated with
greater competitive abilities30, larger root systems can be related to greater soil space occupation31 and/
or greater resource uptake29 thus limiting access to resources for neighbouring plants. Likewise larger
individual plants can induce stronger efects on target plants than smaller ones32. Here, H. lanatus is a
strong competitor (so, with high competitive abilities) because its presence (its biomass) probably led
to a strong limitation of the biomass production of the two focal species (lower ln RR values) through
strong space occupation and/or greater nutrient depletion. In contrast, C. album individuals tended to
have a positive efect on the biomass of these two focals species. Facilitative interactions between plants
can be observed through modiications of soil components (e.g. moisture, nutrients)33 or enhancement
of seedling establishment. However here, the positive efect of C. album may probably be due to low
biomass production rather than facilitation. his may have favoured greater space/nutrient exploitation
by the two focal species, and thus higher allocation to biomass than in the other treatments (includ-
ing focals monocultures). So in this study, we consider that the competitive abilities of plant species
are more a consequence of their biomass production (even if biomass production can also result from
higher competitive abilities). In the case of L. corniculatus, response patterns were diferent. Mixture with
C. album appart, H. lanatus behaved as an intermediate competitor in spite of its important biomass
production (especially root biomass). Moreover, even though the range of ln RR values was narrower for
this species (compared to L. purpureum for example), all wind-pollinated species had a small negative
efect on L. corniculatus, compared to monocultures, and the strongest competitors were P. lanceolata and
A. capillaris. However, biomass measurements indicate that only L. corniculatus belowground biomass
was altered by the presence of a competitor and aboveground biomass was unafected (see Supplementary
information Fig. S1). Some characteristics of L. corniculatus could have mediated this diferent response
to competition compared to the two other species. First, L. corniculatus is a legume species and, although
we did not quantify them, L. corniculatus roots showed nodules, indicating that nitrogen ixation did
occur in our experiment. L. corniculatus could thus have accessed to the atmospheric N pool34 so that
it was only slightly afected by competition compared to the two other species. Furthermore, L. cornic-
ulatus has a perennial life cycle that can induce a diferent timing of response to competition as well as
diferent allocation patterns compared to plants having annual life cycles3,35. Initially, the experimental
design contained a second perennial plant, Mimulus guttatus DC. (synonym Erythranthe guttata (Fisch.
ex DC.) G.L. Nesom; see Supplementary Table S1), that might have provided elements to disentangle the
respective roles of perennial life cycle and nitrogen ixation in L. corniculatus. Unfortunately, only 15 out
of 75 Mimulus guttatus plants lowered, and half of the lowers (25/49 total lowers) were produced by
only two plants. As a consequence we decided not to include this species in this work.
Moreover, we should keep in mind that these conclusions rely on inal harvests of biomass when
some studies suggest regular harvest along experiments to better assess the dynamics of competitive
interactions36.
Most of the loral traits measured in this experiment were afected by the competition treatment.
Higher conditions of competition (lower ln RR values induced by the presence of H. lanatus) had the
greatest impact on E. plantagineum and L. purpureum, by reducing lower and reward traits. To date,
studies that have looked for links between attractiveness traits and environmental conditions have mostly
focused on the efects of abiotic conditions and showed a sensitivity of attractiveness traits to nutrients
and water availability3,6,9,12 or litter and compost additions to soil11,26. If modiications of abiotic condi-
tions can alter species attractiveness to pollinators, it is not surprising that biotic interactions such as
competition, that mediate abiotic resources availability, have similar efects15. Flower and nectar pro-
duction can be relatively costly for a single plant37–39 so that allocation to reproductive structures might
be modiied in a context of competition with limited access to nutrients. Here, the lowest ln RR values,
calculated from mixtures in presence of H. lanatus, suggest that this species may have reduced the avail-
ability of soil resources to E. plantagineum and L. purpureum and thus daily as well as total allocation
of plants to loral traits. Overall, this could be responsible for lower resources allocation to reproductive
traits. Conversely, C. album, led to higher lower and nectar production than in monocultures. A greater
resources availability or space, due to the reduced biomass of the competitor might have led to better
growing conditions for the insect-pollinated species, resulting in better resource acquisition (as con-
irmed by higher ln RR values) and increased allocation to reproductive structures.
In the case of L. corniculatus, while ln RR values suggest stronger competition (albeit limited) in pres-
ence of wind-pollinated species compared to monoculture, some loral traits were lowest in monoculture
and in mixtures with P. lanceolata (which is the strongest competitor for L. corniculatus based on ln RR
values) compared to the other mixtures. his suggests that, in contrast to the two other insect-pollinated
species, allocation to reproduction was not related to biomass allocation. In Wurst & Van Beersum40,
monocultures of L. corniculatus can have higher biomass and produce more lowers than in mixture
with H. lanatus, which is not in accordance with the observed pattern here. Considering the cost of N2
ixation suggested in some studies (in term of C allocation to symbiont41), plants in monocultures in our
experiment might have allocated less photosynthetates to loral traits leading to the observed decrease
in monocultures. However, the study of loral traits per unit of biomass (total loral traits divided by the
inal biomass) revealed that L. corniculatus might have a more adaptive response to competition while
the two other focal species might have a ‘purely’ plastic response to resource availability. Indeed, for
L. corniculatus, loral traits per unit of biomass tend to be higher when competition intensity increases,
showing a possible strategy to better attract pollinator in condition of competition. However, as we only
have biomass data at the end of the experiment (and a inal biomass can not only be considered as a
sum of biomass like for total produced lowers, for instance), we believe that further investigations are
needed to conclude on these efects.
Here we focused on lower and nectar production while other attractiveness traits could also be
afected by competitive interactions. For example, plant pigments or volatile compounds involved in
lower colours42 and scents, relative amounts of diferent sugars or amino-acids content in nectar12,43 and
pollen quantity and or quality9, are all sensitive to resource variations, and could be afected by competi-
tive interactions. Even though we observed a negative impact of competition on some loral traits involved
in attractiveness to pollinators, the response of loral traits can be complex and species-speciic3,11,12.
Moreover, we interpret our results in a context of exploitative competition through soil resources deple-
tion while other competitive mechanisms (e.g. interference through allelochemicals)13,29 could conjointly
inluence plant response. As ln RR values did not difer among total or belowground biomass and root
competition is oten stronger than shoot competition (especially with grass competitors44), our results are
mainly interpreted in a context of belowground competition. However further investigations are needed
to better assess the overall impact of plant competition (aboveground as well as belowground) on loral
traits involved in attractiveness to pollinators.
Variations in attractiveness traits are known to strongly impact pollinator visitation patterns and on a
larger scale pollination service. Indeed, greater plant attractiveness can enhance the frequency or number
of lower visits: most pollinators are preferentially attracted to plants producing numerous, large lowers
and/or greater rewards (in quality or quantity)20–25,45. Larger loral display size can also inluence the
abundance of visiting pollinators46. Likewise, the pattern of pollinator visits per plant can be correlated
to the total nectar production per plant47. However, many lowering plants are pollen limited, therefore
an increase in pollination intensity (e.g. a greater pollen deposit on stigmas) can enhance plant fecundity
(i.e. greater fruit and/or seed set)48. As a consequence, our results suggest that a decrease in loral traits
involved in pollinator attractiveness due to plant competition could have negative impacts on pollina-
tor visits, reducing plant reproductive success. However further experiments are needed to test such
hypothesis. Nevertheless, this study emphasises the importance of 1) taking into account species other
than insect-pollinated ones in plant-pollinator network studies, and 2) linking above ground and below
ground interactions to better understand plant-pollinator networks. his is in concordance with some
research initiated on the impact of soil micro-organisms on pollinator visits through variations of loral
traits49,50. Given our results, future research is needed on plant-soil or plant-plant interactions that may
lead to modiications of loral traits involved in attractiveness to pollinators.
Methods
Our objectives were to study how attractiveness traits of insect-pollinated plants are afected by the
presence of neighbouring wind-pollinated plant species. To do so, we set up a greenhouse experiment in
which we grew seven plant species in binary mixtures in pots.
plant species. Seven plant species (3 insect-pollinated plants and 4 wind-pollinated plants) with close
ecological preferences (based on Ellenberg index for British Plants51) were selected: Echium plantagineum
L., Lamium purpureum L., and Lotus corniculatus L., for insect-pollinated focals and Agrostis capillaris
L., Chenopodium album L., Holcus lanatus L. and Plantago lanceolata L. for wind-pollinated competitors
(for plant species description, see Supplementary Table S1).
Experimental set-up. In March 2012, seedlings of all species were planted in plastic pots (14 cm
Ø; 1.5L, Puteaux SA, France) in sandy soil (pH = 6). he soil was taken from a grassland site
(CEREEP-Ecotron Ile-de-France, St Pierre-lès-Nemours, France) and was sieved (< 4 mm) to remove
rocks and plant material. Six plant individuals were placed in each pot to form two-species mixtures
with three individuals of one insect-pollinated species in alternation with three individuals of one
wind-pollinated species. We also set up control monocultures with six individuals of the same spe-
cies (insect-pollinated or wind-pollinated). Each mixture was replicated 5 times, making a total of 95
pots (5 × 3 monocultures of insect-pollinated species, 5 × 4 monocultures of wind-pollinated species,
5 × 4 × 3 binary mixtures). Pots were randomly placed in a greenhouse (CEREEP-Ecotron Ile-de-France,
St Pierre-lès-Nemours, France) and their position was changed each week. Plants were watered daily by
sub-irrigation (lood loors, DIMAC SAS, France). Air temperature in the greenhouse followed outdoor
conditions but was maintained above 18 °C when low temperatures occurred. Photoperiod was initially
set at 12-hours per day through natural light and sodium lamps when necessary (i.e. when solar irra-
diation was under 200 watt/m2/hour; HS2000 Hortilux Schréder, he Netherlands). It was adjusted to
16-hours per day to allow for the blooming of L. corniculatus, a long-day lowering species. Because we
were mainly interested by belowground competition in this study, we took special care to check that plant
foliage did not overlap between individuals all along the experiment. When plant foliage did overlap
(especially in mixtures with L. corniculatus) plant supports were put in to separate plant individuals and
thus limit aboveground competition (i.e. for light).
Nectar traits. For each plant, nectar volume and nectar sugar content were measured on up to three
newly opened lowers, ater lower size measurements (see above). his ensured that nectar traits were
measured on lowers of the same age to limit variations due to lower age52. Nectar was sampled using
microcapillary tubes (0.5 µ L or 1 µ L; Minicaps end to end, Hirschmann laborgeraete, Germany) and
nectar volume was calculated by measuring the length of liquid in the microcapillary tube with a digi-
tal caliper (Digit-Cal MK IV, Brown&Sharpe, USA) (µ L.lower−1.day−1). Daily sugar concentration was
determined with hand-held refractometers (Eclipse 45–81 and Eclipse 45–82, Bellingham+ Stanley Ltd.,
UK) calibrated using sucrose solutions (30% and 50% brix). Because nectar not only contains sucrose
but also other sugars, our concentration measurements correspond to sucrose equivalent. However, for
the sake of brevity, we will only use in the following the term sucrose in reference to “sucrose equiva-
lent”. When nectar volumes were too small to be measured by the refractometer (< 0.5 µ L), samples were
diluted in Milli-Q water before measurement. If concentration measurements could not be done right
ater sampling, microcapillary tubes were stored in a refrigerator at 4°C and measured within the next
two hours. Because only up to three lowers per plant were sampled, we decided to calculate volume
and sucrose indices taking into account the number of lowers produced per plant47. he daily volume
index per plant was assessed by multiplying the average nectar volume per lower per day with the
number of lower produced per plant per day. he daily amount of sucrose allocated to nectar per plant
was estimated by multiplying the average concentration of sucrose in nectar per lower per day by the
average volume of nectar per lower per day. his daily amount was multiplied by the number of lowers
produced per plant per day, giving a daily sucrose index per plant. All daily loral traits (per lower or
at the plant scale) can give information on the plant allocation to reproduction all along the lowering
period. However, in order to have a global assessment of reward production and plant energy allocation,
daily indices were summed along the whole lowering period as total indices11.
Plant traits. At the end of the lowering period of each insect-pollinated plant species (on the 3rd of May
2012 for L. purpureum and on the 2nd of June 2012 for E. plantagineum and L. corniculatus), above- and
belowground biomass of all individuals was harvested. Concerning belowground biomass, we took care
of separating root systems of each species. Plant biomasses were oven-dried (65°C, 48h) and weighted
(g.plant−1).
Competitive interactions. In order to estimate the intensity of competitive interactions between each
focal insect-pollinated plant and its wind-pollinated competitors we calculated the log response-ratio (ln
RR) as an index of competition28. his index is deined as:
P
lnRR = ln mix
Pcontrol
where Pmix is the biomass of a focal plant when grown in mixtures and Pcontrol is the biomass of a focal
plants in monoculture pots. In order to have a good assessment of the ln RR as well as a variance, ln RR
values for each treatment were calculated as means of all possible combinations of each focal plant in a
mixture divided by each focal plant in a monoculture. Because three focal plants were present in mix-
tures, we considered monocultures as ‘mixtures’ of 3 focal plants with 3 ‘competitor’ plants of the same
species. Values of this index are symmetrical around zero with positive values indicating that focals grow
better in mixture (i.e. focals are better competitors) and negative values indicating that focals’ growth
is negatively afected by competitor (i.e. focals are lower competitors). Ln RR values were calculated
from aboveground and belowground biomass but only ln RR calculated from total biomass were used
to study the efect of competition on inal loral traits as it is a better integrator of competition within
both compartments.
Data analysis. All statistical analyses were performed using R 3.153. Linear mixed-efects models were
itted to all measured traits (nlme R package54), with the exception of loral display size and total lower
production that were itted to generalized mixed-efect models with Poisson probability distribution and
log link function (lme4 R package55). As ln RR values are calculated from inal biomass here, this may be
relevant to study the response of total loral traits (values summed all along the lowering period for each
plant to obtain a total value per plant) to competition but not for daily loral traits as competition can
be dynamic along plant lifespan36. As a consequence models were itted with two diferent explanatory
variables: ln RR values calculated from total biomass as a ixed efect for total loral traits (i.e. total lower
production, total sucrose index, total volume index) and wind-pollinated species identity as a ixed efect
for daily loral traits (i.e. loral display size, lower size, daily sucrose concentration in nectar, daily nectar
volume, daily volume index, daily amount of sucrose in nectar, daily sucrose index). he date was also
set as a ixed efect for daily loral traits to take into account the efect of plant age. In all models, pots
and date (for the repeated measures on plants) were set as random efects. For linear mixed models, data
were transformed using log (e.g. loral traits involving nectar volume), square or square root (e.g. lower
size, loral traits involving sucrose concentration) transformations, when necessary. Daily data were then
analysed through analysis of covariance (ANCOVA). For total data, whose values were summed all along
the lowering period for each plant to obtain a total value per plant, analyses of variance (ANOVA) were
performed on these total values. When signiicant diferences were detected, post-hoc comparisons were
performed (Tukey all-pair comparisons, Holm method for p-value adjustment were used56, multcomp
R package57). For the date efect or the interaction term, only signiicant efects are reported. Because
generalized mixed-efect models (glmer, loral display size and total lower production) do not provide
p-values, pairwise comparisons with Holm method for p-value adjustment were used56.
References
1. Xia, J. & Wan, S. Global response patterns of terrestrial plant species to nitrogen addition. New Phytol. 179, 428–439 (2008).
2. Meekins, J. F. & McCarthy, B. C. Responses of the biennial forest herb Alliaria petiolata to variation in population density,
nutrient addition and light availability. J. Ecol. 88, 447–463 (2000).
3. Burkle, L. A. & Irwin, R. E. he efects of nutrient addition on loral characters and pollination in two subalpine plants, Ipomopsis
aggregata and Linum lewisii. Plant Ecol. 203, 83–98 (2009).
4. Müller, I., Schmid, B. & Weiner, J. he efect of nutrient availability on biomass allocation patterns in 27 species of herbaceous
plants. Perspect. Plant Ecol. Evol. Syst. 3, 115–127 (2000).
5. Havens, K., Preston, K. A., Richardson, C. & Delph, L. F. Nutrients afect allocation to male and female function in Abutilon
theophrasti (Malvaceae). Am. J. Bot. 82, 726–733 (1995).
6. Muñoz, A. A., Celedon-Neghme, C., Cavieres, L. A. & Arroyo, M. T. K. Bottom-up efects of nutrient availability on lower
production, pollinator visitation, and seed output in a high-Andean shrub. Oecologia 143, 126–35 (2005).
7. Burkle, L. A. & Irwin, R. E. Beyond biomass: measuring the efects of community-level nitrogen enrichment on loral traits,
pollinator visitation and plant reproduction. J. Ecol. 98, 705–717 (2010).
8. Soper-Gorden, N. L. & Adler, L. S. Abiotic conditions afect loral antagonists and mutualists of Impatiens capensis (Balsaminaceae).
Am. J. Bot. 100, 679–689 (2013).
9. Lau, T.-C. & Stephenson, A. G. Efect of soil nitrogen on pollen production, pollen grain size and pollen performance in
Cucurbita pepo (Cucurbitaceae). Am. J. Bot. 80, 763–768 (1993).
10. Xiong, S. & Nilsson, C. he efects of plant litter on vegetation : a meta-analysis. J. Ecol. 87, 984–994 (1999).
11. Baude, M. et al. Litter inputs and plant interactions afect nectar sugar content. J. Ecol. 99, 828–837 (2011).
12. Petanidou, T., Goethals, V. & Smets, E. he efect of nutrient and water availability on nectar secretion and nectary structure of
the dominant Labiatae species of phrygana. Syst. Geogr. Plants 68, 233–244 (1999).
13. Schenk, H. J. Root competition: beyond resource depletion. J. Ecol. 94, 725–739 (2006).
14. Nötzold, R., Blossey, B. & Newton, E. he inluence of below ground herbivory and plant competition on growth and biomass
allocation of purple loosestrife. Oecologia 113, 82–93 (1997).
15. Partzsch, M. & Bachmann, U. Is Campanula glomerata threatened by competition from expanding grasses? Results from a 5-year
pot-experiment. Plant Ecol. 212, 251–261 (2011).
16. Weiner, J. he inluence of competition on plant reproduction in Plant reproductive ecology: patterns and strategies (eds Lovett-
Doust J. & Lovett-Doust L.) Ch.11, 228–245 (Oxford university press, 1988).
17. Dafni, A. in Pollination ecology: a practical approach (eds Rickwood D. & Hames B. D.) (Oxford university press, 1992).
18. Proctor, M., Yeo, P. & Lack, A. in he natural history of pollination. (eds Timber press) (Timber press, 1996).
19. Lázaro, A., Hegland, S. J. & Totland, O. he relationships between loral traits and speciicity of pollination systems in three
Scandinavian plant communities. Oecologia 157, 249–57 (2008).
20. Mitchell, R. J., Karron, J. D., Holmquist, K. G. & Bell, J. M. he inluence of Mimulus ringens loral display size on pollinator
visitation patterns. Funct. Ecol. 18, 116–124 (2004).
21. Martin, N. H. Flower size preferences of the honeybee (Apis mellifera ) foraging on Mimulus guttatus (Scrophulariaceae). Evol.
Ecol. 6, 777–782 (2004).
22. Conner, J. K. & Rush, S. he efect of lower size and number on pollinator visitation to wild radish, Raphanus raphanistrum.
Oecologia 105, 509–516 (1996).
23. Stang, M., Klinkhamer, P. G. L. & Van der Meijden, E. Size constraints and lower abundance determine the number of
interactions in a plant lower visitor web. Oikos 112, 111–121 (2006).
24. Cartar, R. V. Resource tracking by Bumble Bees : responses to plant-level diferences in quality. Ecology 85, 2764–2771 (2004).
25. Cnaani, J., homson, J. D. & Papajà, D. R. Flower choice and learning in foraging Bumblebees: efects of variation in nectar
volume and concentration. Ethology 112, 278–285 (2006).
26. Cardoza, Y. J., Harris, G. K. & Grozinger, C. M. Efects of soil quality enhancement on pollinator-plant interactions. Psyche a J.
Entomol. 2012, (2012).
27. Partzsch, M., Piesch, C. & Hensen, I. A comparative study of germination biology and plant performance in two dry grassland
species. Folia Geobot. 46, 35–48 (2011).
28. Weigelt, A. & Jollife, P. Indices of plant competition. J. Ecol. 91, 707–720 (2003).
29. Gurevitch, J., Scheiner, S. M. & Fox, G. A. Competition and other interactions among plants in he ecology of plants 2nd edn,
Vol.1 (eds Sinauer, A.D.) Ch.10, 226–256 (Sinauer Associates, 2006).
30. Cahill, J. F. Lack of relationship between below-ground competition and allocation to roots in 10 grassland species. J. Ecol. 91,
532–540 (2003).
31. Raynaud, X. & Leadley, P. W. Soil characteristics play a key role in modeling nutrient competition in plant communities. Ecology
85, 2200–2214 (2004).
32. Rajaniemi, T. K. Evidence for size asymmetry of belowground competition. Basic Appl. Ecol. 4, 239–247 (2003).
33. Callaway, R. M. Positive interactions among plants. Bot. Rev. 61, 306–349 (1995).
34. Gurevitch, J., Scheiner, S. M. & Fox, G. A. Soils, mineral nutrition, and belowground interactions in he ecology of plants 2nd
edn, Vol.1 (eds Sinauer, A.D.) Ch.4, 71–97 (Sinauer Associates, 2006).
35. Pitelka, L. F. Energy allocations in annual and perennial Lupines (lupinus : Leguminosae). Ecology 58, 1055–1065 (1977).
36. Trinder, C., Brooker, R., Davidson, H. & Robinson, D. Dynamic trajectories of growth and nitrogen capture by competing plants.
New Phytol. 193, 948–958 (2012).
37. Pyke, G. H. What does it cost a plant to produce loral nectar? Nature 350, 58–59 (1991).
38. Southwick, E. E. Photosynthate allocation to loral nectar: a neglected energy investment. Ecology 65, 1775–1779 (1984).
39. Snow, A. A. & Whigham, D. F. Costs of lower and fruit production in Tipularia discolor (Orchidaceae). Ecology 70, 1286–1293
(1989).
40. Wurst, S. & van Beersum, S. he impact of soil organism composition and activated carbon on grass-legume competition. Plant
Soil 314, 1–9 (2009).
41. Kaschuk, G., Kuyper, T. W., Lefelaar, P. A., Hungria, M. & Giller, K. E. Are the rates of photosynthesis stimulated by the carbon
sink strength of rhizobial and arbuscular mycorrhizal symbioses? Soil Biol. Biochem. 41, 1233–1244 (2009).
42. De Pascual-Teresa, S. & Sanchez-Ballesta, M. T. Anthocyanins: from plant to health. Phytochem. Rev. 7, 281–299 (2008).
43. Gardener, M. C. & Gillman, M. P. he efects of soil fertilizer on amino acids in the loral nectar of corncockle, Agrostemma
githago (Caryophyllaceae). Oikos 92, 101–106 (2001).
44. Kiaer, L. P., Weisbach, A. N. & Weiner, J. Root and shoot competition: a meta-analysis. J. Ecol. 101, 1298–1312 (2013).
45. Hegland, S. J. & Totland, Ø. Relationships between species’ loral traits and pollinator visitation in a temperate grassland.
Oecologia 145, 586–94 (2005).
46. Goulson, D., Stout, J. C., Hawson, S. A. & Allen, J. A. Floral display size in comfrey, Symphytum oicinale L. (Boraginaceae):
Relationships with visitation by three bumblebee species and subsequent seed set. Oecologia 113, 502–508 (1998).
47. Leiss, K. A. & Klinkhamer, P. G. L. Spatial distribution of nectar production in a natural Echium vulgare population: Implications
for pollinator behaviour. Basic Appl. Ecol. 6, 317–324 (2005).
48. Snow, A. A. Pollination intensity and potential seed set in Passilora vitifolia. Oecologia 55, 231–237 (1982).
49. Gange, A. C. & Smith, A. K. Arbuscular mycorrhizal fungi inluence visitation rates of pollinating insects. Ecol. Entomol. 30,
600–606 (2005).
50. Barber, N. A. & Soper Gorden, N.L. How do belowground organisms inluence plant-pollinator interctions? J. Plant Ecol. 8, 1–11
(2015).
51. Hill, M., Mountford, J., Roy, D. & Bunce, R. Ellenberg’s indicator values for British plants (volume 2a) In Technical Annex to
ECOFACT Research report series, Vol. 2. (Institute of Terrestrial Ecology, 1999).
52. Southwick, A. K. & Southwick, E. E. Aging efect on nectar production in two clones of Asclepias syriaca. Oecologia 56, 121–125
(1983).
53. R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.
(2014). Available at: http://www.R-project.org/. (Accessed: 27th march 2015).
54. Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D. & R Core Team. nlme: Linear and Nonlinear Mixed Efects Models R package version
3.1–117 (2014). Available at: http://CRAN.R-project.org/package= nlme. (Accessed: 27th march 2015).
55. Bates, D., Maechler, M., Bolker, B. & Walker, S. lme4: Linear mixed-efects models using Eigen and S4. R package version 1.1–6
(2014). Available at: http://CRAN.R-project.org/package= lme4. (Accessed: 27th march 2015).
56. Aickin, M. & Gensler, H. Adjusting for multiple testing when reporting research results: he Bonferroni vs Holm methods. Am.
J. Public Health 86, 726–728 (1996).
57. Hothorn, T., Bretz, F. & Westfall, P. Simultaneous Inference in General Parametric Models. Biometrical J. 50, 346–363 (2008).
Acknowledgments
his work was supported by a grant from the Fédération Ile-de-France de Recherche en Environnement
(FIRE) and the Ile-de-France region through the DIM R2DS program. We are grateful to all the members
of the CEREEP Ecotron Ile-de-France. We also thank Jean-Paul Maalouf for statistical advice, Elisa
hébault and Benoît Geslin for helpful comments.
Author Contributions
F.F., X.R. and I.D. conceived and designed the study. All authors (F.F., X.R., A.H., E.M. and I.D.) collected
data. F.F. and X.R. performed statistical analysis. F.F. wrote the irst drat of the manuscript and all
authors (F.F., X.R., A.H., E.M. and I.D.) contributed substantially to revisions.
Additional Information
Supplementary information accompanies this paper at http://www.nature.com/srep
Competing inancial interests: he authors declare no competing inancial interests.
How to cite this article: Flacher, F. et al. Competition with wind-pollinated plant species alters loral
traits of insect-pollinated plant species. Sci. Rep. 5, 13345; doi: 10.1038/srep13345 (2015).
his work is licensed under a Creative Commons Attribution 4.0 International License. he
images or other third party material in this article are included in the article’s Creative Com-
mons license, unless indicated otherwise in the credit line; if the material is not included under the
Creative Commons license, users will need to obtain permission from the license holder to reproduce
the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
Title: Competition with wind-pollinated plant species alters floral traits of insect-pollinated plant
species.Floriane Flacher, Xavier Raynaud, Amandine Hansart, Eric Motard and Isabelle Dajoz
Supplementary Figure S1: Mean (+/-standard error) biomass (g.plant-1) of: E. plantagineum, L.
purpureum and L. corniculatus in mixture with the competitors (three top panels), and competitor
species in mixtures with E. plantagineum, L. purpureum and L. corniculatus (three bottom panels).
Above-ground biomass (light grey, P L. purpureum, and PE. plantagineum <0.01, P L. corniculatus =0.06; Pcompetitors
with L. purpureum, Pcompetitors with E. plantagineum, Pcompetitors with L. corniculatus <0.01), below-ground biomass (dark grey,
P<0.01 for all) (ANOVA, N=75 for each test). “Mono” refers to monocultures of the focal species.
Wind-pollinated species are ordered according to increasing intensity of competitive interactions (see
Results). Different letters are for significant differences (Tukey all pair comparison test and adjustment
of p-values, Holm method56) (bold letters: total biomass; letters at the top: above-ground biomass;
Supplementary Figure S2: Linear regression between the total flower production and mean ln RR
values per plant calculated from final biomass (P<0.001 with transformed data for all three species; E.
plantagineum R²=0.52, L. purpureum R²=0.70 and Lotus corniculatus R²=0.21). The grey line refers to
the model after data transformation while dots represent untransformed data. Ln RR values are
Supplementary Figure S3: Mean (+/- standard error) flower size (mm) of E. plantagineum, L.
purpureum and L. corniculatus in mixture with the competitors. “Mono” refers to monocultures of the
focal species. Wind-pollinated species are ordered according to increasing intensity of competitive
interactions (see Results). E plantagineum: N= 806, F4,772= 7.35, P<0.001; L purpureum: N= 1971,
F1,1936= 8.96, P<0.001; L corniculatus: N= 1070, F4,1036 = 4.69, P<0.001 (ANCOVA). Different letters
are for significant differences (Tukey all pair comparison test and adjustment of p-values, Holm
method56)
Chapitre 1 | 65
Supplementary Figure S1:
Chapitre 1 | 66
Supplementary Figure S3:
Chapitre 1 | 67
Supplementary Table S1: Plant species description 1-3 (Family names are given according to APG III
classification4)
intermediate light
Plantago lanceolata L. native to the grasslands,
perennial intermediate fertility
(Plantaginaceae) region wastelands
intermediate pH
REFERENCES of Supplementary information
Chapitre 1 | 68
IV. Chapitre 2
Présentation du chapitre :
Chapitre 2| 73
cours du Chapitre 1 ont été également mesurés afin de pouvoir les relier aux visites du
pollinisateur étudié : choix de première visite, temps avant la première visite et nombre de
visites (totales et sur des séquences de 10min).
Chapitre 2| 74
quelque soit le traitement de compétition) afin d'étudier la réponse des pollinisateurs à des
variations plus fines des traits
raits d'attractivité (notamment au niveau des récompenses). Par
ailleurs, le nombre de visites ne semble pas non plus être influencé par l’identité du
compétiteur. Cette dernière ayant un effet sur les paramètres floraux, nous nous attendions
cette fois à des différences de visites selon les mélanges. Les différences significatives
observées chez les traits floraux pourraient être trop faibles par rapport à la sensibilité des
pollinisateurs étudiés n’entraînant donc pas de modifications du nombre de visites faites sur
E. plantagineum.. Par ailleurs, ces différences ont été observées lors de prélèvements effectués
sur des fleurs nouvelles. Or la production de récompenses, mais aussi la taille des fleurs
peuvent dépendre de l'âge de la fleur. N'ayant pas pu fournir
fournir uniquement des fleurs nouvelles
aux pollinisateurs, il est possible que l'effet de l'identité du compétiteur ait été dilué.
A l'issue de ce chapitre 2 il semble donc que la compétition induite par la présence d'une
espèce anémophile n’influence pas les interactions plantes-pollinisateurs
pollinisateurs via des traits
d’attractivité fins comme la production de récompenses. Par ailleurs, le contrôle de la vitrine
florale à probablement gommé ici l’effet d’un trait d’attractivité qui semble très important.
Cet aspect mérite
ite donc d’être plus approfondi.
approfondi
Chapitre 2| 75
Does competition with wind-pollinated species alter Echium plantagineum attractiveness
Floriane Flacher1, Amandine Hansart2, Eric Motard3, Benoit Geslin4, Abdou Fofana5, Isabelle
Author affiliations:
1
CNRS, Sorbonne Universités, UPMC Univ Paris 6, Institute of Ecology and Environmental
2
CNRS, UMS 3194 CEREEP-Ecotron Ile de France, F-77140 Saint-Pierre-lès-Nemours
3
Université Paris Diderot Paris 7, Institute of Ecology and Environmental Science - Paris,
4
Université Aix Marseille Université, Institut Méditerranéen de Biodiversité et d’Ecologie
Marine et Continentale 9 (IMBE) CNRS, IRD, Avignon Université Pôle St Jérôme av.
5
Université Memorial de Terre-Neuve 230 Elizabeth Ave, St. John's, NL A1B 3X9, Canada
6
Sorbonne Universités, UPMC Univ Paris 6, Institute of Ecology and Environmental Science
Chapitre 2| 76
ABSTRACT
In order to attract pollinators, insect-pollinated plants display several traits, such as the
number of flowers, their size, colour or odor as well as rewards (i.e. pollen and nectar). These
traits vary among plant species and can lead to differences in attractiveness to pollinators.
These traits also depend on interactions between plants, especially competition. Indeed, at the
species. However, to our knowledge, the influence of floral trait modifications, through
competition induced by wind-pollinated species, on pollinator visits has not been investigated
yet. In this experiment, an insect pollinated species, E. plantagineum, was grown in binary
mixture with three wind-pollinated plants that were selected to exert a panel of competitive
with or without belowground competition. Floral traits as well as pollinator visits were
measured. The “with competition” treatment had no effect on both floral traits and pollinators
visits (except nectar sucrose concentration). However there was an influence of the competitor
identity on flower and reward production with lower rewards and flower production of the
insect-pollinated plant species when grown with strongly competitive wind-pollinated species.
For pollinator visits, we found a tendency for more rapid visits on insect-pollinated plants in
INTRODUCTION
rewarding traits for pollinators, such as flowers, nectar or pollen (Kearns & Inouye, 1993).
Chapitre 2| 77
The number of simultaneously open flowers (floral display size) plays a key role in initiating
visits of a pollinator on a plant (Dafni, 1992) but also influence the number of visitation
events (Conner & Rush, 1996, Mitchell et al., 2004, Grindeland et al., 2005, Miyake & Sakai,
2005, Makino et al., 2007). Nectar and pollen are sources of sugars and proteins influencing
the number and the repetition of pollinator visits (Dafni, 1992). These traits are known to vary
at the plant level (Herrera et al., 2006) or at the population level (Lanza et al., 1995). Most
importantly these traits usually differ between species. Due to pollinator's preferences for
some floral traits (e.g greater floral display size, greater amount of rewards), variations in
these traits induce differences of attractiveness between plant species in a community. Several
studies have shown that the presence of a plant species producing numerous flowers and/or
greater rewards (in quantity and quality) can alter pollinator visits on neighbouring species.
This effect can be positive, the attractive plant being a “magnet” benefiting less attractive
surrounding plant species (Laverty, 1992, Molina-Montenegro et al., 2008) or negative, the
attractive species focusing most pollinator visits at the expense of neighbouring species
(Brown & Mitchell, 2001; Muñoz & Cavieres, 2008 ; Kandori et al., 2009). Such competition
for pollination between plant species can lead to modification in species reproductive success
(Bell et al., 2005; Mitchell et al., 2009). While these differences in individual attractiveness
are dependent of the neighbouring context, some studies have shown that the nutritional
context of a plant can also lead to differences in floral traits involved in attractiveness to
pollinator. Indeed, the addition of nutrients (e.g. nitrogen or phosphorus: Lau & Stephenson,
1993,1994; Burkle & Irwin, 2009, 2010; Muñoz et al., 2005; Soper-Gorden & Adler, 2013) as
well as litter (Baude et al., 2011), compost (Cardoza et al., 2012) or water (Petanidou et al.,
1999) can have a positive impact on flower production as well as pollen and nectar quantity or
quality. As the composition of plant communities can alter the availability of resources to
plant species, it has been suggested that competitive interactions between plants for abiotic
Chapitre 2| 78
resources may also alter plant attractiveness traits to pollinators. Some studies demonstrated
that competition could alter allocation to reproduction (Weiner 1988). Working with mixtures
of insect-pollinated plants, Baude et al., 2011 found that the presence of a strong competitor
led to a decrease of sugar allocated to nectar in a focal species. Likewise, the presence of a
Bachmann, 2011) and Flacher et al. 2015 found that competition induced by wind-pollinated
plants led to a decrease in floral display size, flower size but also total flower production and
total sugar allocated to nectar in two annual insect-pollinated species. In this last experiment,
the stronger the effect on attractiveness traits. Overall, the composition of the plant
community could play a key role in the maintenance of plant-pollinator interactions both
through “direct” (competition for pollination) and “indirect” (competition for resources)
interactions. Moreover, because competition for resources can occur between insect-
pollinated and wind-pollinated plants species, Partzsch & Bachmann (2011) and Flacher et al.
(2015) strongly suggest that the whole plant community, and not only insect-pollinated
species, can influence plant-pollinator interactions. However little attention has been paid to
the role of competition between plants on interactions with pollinators. In order to understand
album, Agrostis capillaris, Holcus lanatus). These last three species were selected to create a
competition as it is often cited to be the most important part of competition in natural plant
communities (Kiaer et al., 2013). Floral traits (flower and nectar production) as well as visits
by a common pollinator Bombus terrestris were followed on the focal species. The objective
Chapitre 2| 79
species could influence pollinator visits through modifications of floral traits involved in plant
Plant species
A total of 4 plant species were selected, based on similar ecological preferences: one focal
Chenopodium album, Agrostis capillaris, and Holcus lanatus. These last three species were
plant, from positive (Chenopodium album) to strongly negative (Holcus lanatus, see Flacher
et al. 2015) and are named “competitors” in the following. Plant species descriptions are
Experimental set-up
The experiment took place in a greenhouse at the CEREEP Ecotron Ile-de-France (Foljuif, St
Pierre-lès-Nemours, France). In spring 2013, seedlings of all plant species were planted in
plastic pots (14 cm Ø; 1.5L, Puteaux SA, France) containing sandy soil (pH=6). Soil was
taken from a nearby grassland site and was sieved (<4 mm) to remove rocks and plant
material. Each pot was divided by plastic dividers that were glued inside the pots into two
interactions between root systems of plants from the two compartments (“with competition”
treatment) whereas in the other half were not, dividers were not perforated preventing
Chapitre 2| 80
competition (“without competition” treatment). In pots containing mixtures, one compartment
was planted with 3 focal plants while the other was planted with 3 plants of one competitor
species. Monocultures contained 6 plants of the focal species, 3 plants in each compartment.
All in all, the experiment consisted in 4 plant treatments (3 mixtures and 1 monoculture)
mixture and competition was replicated 15 times, making a total of 120 pots (4x2x15 pots).
Pots were randomly placed in the greenhouse and their position was changed each week. Air
temperature in the greenhouse followed outdoor conditions but was maintained above 18 °C
when low temperatures occurred. Photoperiod was initially set at 12-hours per day using
natural light and sodium lamps when necessary (i.e. when solar irradiation was lower than
200 watt/m2/hour; HS2000 Hortilux Schréder, The Netherlands). Among those 120 pots, 40
were used to measure floral traits (“floral trait pots” see “Floral traits” section) and 80 were
used to observe pollinator visits (“visited pots”, see “Pollinator visits” section). Indeed,
pollinator visits as well as sampling deplete floral resources (e.g. nectar) or damage flowers
(bumblebee marks, flower handling) so that using the same pots to follow both floral traits
measurements and pollinator visits could have led to a misestimation of floral traits
Floral traits
Flower production was measured each day on all focal plants growing in the pot set dedicated
to floral traits measurements. Buds ready to open were marked the day before each sampling.
Flowers actually opened the day after were counted. The total number of opened flowers (new
and old) was also counted to estimate the daily floral display size per plant. For each plant, 3
Chapitre 2| 81
newly open flowers were selected at most for flower size measurements and nectar sampling
(see “Nectar traits”) Flower size was measured from the bottom of the corolla to the tip of the
Nectar traits
Nectar production was followed every day on all focal plants of the same pot set. On a
maximum of 3 newly open flowers (see “Flower production and flower size”), nectar was
sampled using calibrated microcapillary tubes of 0.5µL and 1µL (Minicaps end to end,
Hirschmann laborgeraete, Germany). Nectar volume per flower was estimated by measuring
the length of sampled nectar in the microcapillary tube with a digital calliper (Digit-Cal MK
IV, Brown&Sharpe, USA). Right after sampling, nectar sugar concentration per flower was
measured with hand-held refractometers for small volumes (Eclipse 45-81 and Eclipse 45-82,
Bellingham+ Stanley Ltd., UK) and expressed in g.L-1 thanks to a conversion table (Kearns &
Inouye 1993). When nectar volumes were too small, samples were diluted with MilliQ water
sucrose solution at 20°C. Because nectar contains other sugars than sucrose (Kearns & Inouye
1993), our concentration measurements correspond to sucrose equivalent. However, for the
sake of brevity, in the following, “sucrose” will refer to “sucrose equivalent”. Measurements
were corrected according to temperature when necessary. The amount of sucrose allocated to
nectar per flower was calculated by multiplying nectar volume per flower and nectar
concentration per flower. To estimate the amount of sucrose allocated to nectar per plant, the
amount of sucrose allocated to nectar per flower was multiplied by the number of produced
flowers per plant and to produce a “daily sucrose index”. This index was also summed all
along the flowering period (“total sucrose index”) to estimate the whole plant allocation to
nectar. The same index was calculated for nectar volume. As flowers submitted to pollinators
Chapitre 2| 82
could not be all newly open flowers, the production of nectar per flower was also followed
from 1 to 2 days after the opening day to assess if nectar production varied with flower age in
Bumblebee colonies
We studied visits of a bumblebee species, Bombus terrestris subspecies terrestris. This species
is a common pollinator in France, providing a good pollination service and is easy to breed.
Colonies of B. terrestris are widely used in agriculture for crop pollination in greenhouses
(Velthuis & van Doorn, 2006). Colonies were supplied by KOPPERT Biological Systems
(Netherlands). They were stored in a room with controlled temperature (20°C). Bumblebees
were fed daily (2h, Baude et al., 2008; after all observations) with a sucrose solution supplied
supplement every 3 to 5 days. A total of 9 colonies were used all along the experiment.
Pollinator visits
From May to July 2013, we followed pollinator visits on plants from the “visited pots” set.
Observations took place in a flying arena, made of a nylon mesh cage (150x80x130cm; mesh
size 0.93x1.6mm, TIP 1700, Texinov). We worked with naïve bumblebees in order to exclude
acquired preferences from a foraging experience that could have interfered with our
treatments. Before observations, the colony was starved for 12h at least. As the starvation
period grew along hours, we took care to randomize different starvation status among the
treatments. For each observation, 2 pots of the same plant treatment, one of each competition
treatment (with and without competition), were placed in the flying arena. As much as
possible, we chose pots with newly opened flowers to match measurement of nectar made on
“floral traits pots” with nectar parameters of “visited pots”. As floral display size (total
Chapitre 2| 83
number of open flowers) can be involved in attractiveness, therefore we chose pots with
similar number of flowers to test the effect of modifications in fine attractiveness traits such
as rewards. As a result, for each observation, pollinators would be submitted to one plant
treatment, with similar number of flower per plant, but potentially different rewards induced
by belowground competition. As bumblebees can lay scented marks (Stout & Goulson 2001)
or damages on flower that may influence the visit of a congener pots were changed for every
bout. In the same way, a delay of 48h was settled before reusing a pot for observation. For
each observation, the colony was opened to let an individual go out and be transported to the
flying arena a few meters away. An observation began as soon as the pollinator was released
in the flying arena. Visits were recorded using the Jwatcher software (version 0.9) for 41
minutes. For each mixture type, we recorded the time to first visit, the number of visits and
Plant biomass
At the end of the experiment, we carried out several biomass measurements to confirm the
gradient of competitive interactions previously found (see Flacher et al., 2015). In “floral
traits pots”, we harvested aboveground and belowground biomass of all plants species (focal
hypothesized that pollinators might be sensitive to aboveground biomass when foraging (i.e.
focals with greater biomass may attract more pollinators in our experimental design). For
belowground biomass we took special care to separate roots of each plant individual.
Competitive interactions
In order to estimate the intensity of competitive interactions between each focal insect-
Chapitre 2| 84
pollinated plant and its wind-pollinated
pollinated competitors we calculated the log response-ratio
response (ln
RR) as an index of competition (Weigelt & Jolliffe, 2003). This index is defined as:
where Pmix is the biomass of a focal plant when grown in mixtures and Pcontrol
cont is the biomass
of a focal plants in monoculture pots. In order to have a good assessment of the ln RR as well
as a variance, ln RR values for each treatment were calculated as means of all possible
combinations of each focal plant in a mixture divided by each focal plant in a monoculture.
‘mixtures’ of 3 focal plants with 3 ‘competitor’ plants of the same species. Values of this
mixture (i.e. focals are better competitors) and negative values indicating that focals’ growth
is negatively affected by competitor (i.e. focals are lower competitors). Ln RR values were
biomass were used to study the effect of competition on final floral traits as it is a better
Statistical analysis
All statistical analysis were performed using R 3.2.2 (R Core Team 2015).
Plant traits
Biomass, flower size and nectar traits were analyzed through linear mixed models (nlme
(
package, Pinheiro et al., 2015). As both linear and generalized mixed models did not fit well
daily flower production and floral display size we used non parametric test (Kruksal Wallis)
for these two variables. However, generalized mixed models were used for the total flower
Chapitre 2| 85
production (lme4 package, Bates et al., 2015). ANCOVA were performed with competitor and
competition treatment set as fixed effects while pots and plants were set as random effects.
The date was set as a covariate (fixed effect) to take into account the influence of plant age on
flower and reward production. It was also set as a random effect to take into account repeated
measures on same plants. Because generalized mixed-effect models (total flower production)
do not provide p-values, pairwise comparisons with Holm method for p-value adjustment
were used. Data were transformed when not normal (log and square root transformations for
Pollinator visits
As only 31% to 46% of observations lead to visits depending on mixture type, (NCH=12,
NEC=11, NAG=13, NHO=10) we performed non parametric tests to study the effect of our two
treatments (competition and competitor identity) on pollinator visits. For the first choice of
visit on a pot, Pearson's Chi-squared was performed for each mixture. The null hypothesis
was that the number of first visits would be equally distributed among pots submitted or not to
competition (p=0.5). Simulations (N=2000, Monte Carlo simulations) were run as the number
of visits for each category was not always superior or equal to 5. Time to first visit was
analysed thanks to survival analysis (Coxph function of the R Survival package, Therneau &
Grambsch, 2000). Kurskal-Wallis tests were performed for the total number of visits and the
number of visit per 10 minutes sequences. We tested the competition effect on one hand, the
competitor identity effect on the other hand and the interaction between both was tested
through differences of visits between competition levels among each mixture. P-values were
Chapitre 2| 86
RESULTS
Plant traits
and total biomass in “floral traits pots” and “visited pots” (Table 1 and see Supplementary
information, Fig S1). However, there was a significant effect of the competitor identity and of
the competition: competitor identity interaction (Table 1) meaning that the impact of
mixture with C. album (even compared to monocultures for aboveground and total biomass)
in opposition to mixtures with H. lanatus (Fig. S1). Moreover there were no significant
calculated on aboveground biomass there was no effect of competition but a significant effect
of both the competitor identity and the competition:competitor identity interaction (Table 1).
C. album was the weakest competitor (even compared to monocultures) while H. lanatus
acted as the strongest competitor of this competitive gradient (see Table 2). A. capillaris
C. album mixtures < Monocultures < A. capillaris mixtures < H. lanatus mixtures. For lnRR values
bases on belowground and total biomasses, once again we found an effect of the competitor
identity and of the competition:competitor identity interaction (Table 1) but also a significant
effect of competition alone (Table 1). While the same panel of competitive interaction was
found in presence of belowground competition (Table 2), C. album acted as the strongest
Chapitre 2| 87
Flower traits
Daily flower production, total flower production, floral display size and flower size were not
affected by the separate effect of competition (all P>0.05). However these traits were
significantly affected by the competitor identity with a decrease along the competitive
gradient. Indeed, daily flower production, total flower production, floral display size and
flower size were significantly higher when E. plantagineum was in presence of C. album
compared to mixtures with H. lanatus (χ²=77.09, df=3, P<0.01, χ²=8.37 df=3, P=0.04 and
χ²=226.11, df=3, P<0.01 and F3,116=3.48, P=0.018 respectively, see Fig. 1 to 4). Moreover
interactions (C+) with C. album compared to plants without competition (C-) while the
opposite pattern was found in competition with H. lanatus (K-W χ²=20.58, df=1, P<0.01 and
identity. Likewise, this interaction was significant for total flower production (χ²=19.59, df=3,
P<0.01). The date also had an impact on daily flower production, floral display size and
flower size (all P<0.01) especially with a decrease of flower size across time.
Nectar traits
The competition treatment had no influence on any of the nectar traits except for the daily
sucrose concentration (F3,36=2.95, P=0.05). Competitor identity had an influence on daily and
total indices only (i.e when flower production was taken into account, see Material and
Methods, section “Nectar traits). Both daily volume and sucrose indices tended to decrease
along the competitive gradient (however marginal for daily volume index F3,36=2.95, P=0.05;
F3,36=3.74, P=0.02 for daily sucrose index). Especially, these two variables increased in
Chapitre 2| 88
induced a decrease of those traits. The same pattern was found during the whole flowering
period when considering total volume and sucrose indices (F3,35=7.79, P<0.01 and F3,32=9.90,
P<0.01 respectively, Fig. 5 and Fig. 6). We found no effect of the competition:competitor
interaction. For daily measurements, there was a significant effect of the date (all P<0.01)
with an increase of daily nectar volume, daily sucrose amount, daily volume index and daily
Pollinator visits
First visit
Within each mixture, the competition treatment did not influence the first visit of a bumblebee
on a pot (χ², all P>0.05). However, in all mixtures there was a tendency for more first visits on
pots in which focal plants were not submitted to belowground competition (except for
mixtures with C. album). Likewise first visits tended to be made more rapidly on
mixture with H. lanatus, however not significantly (Likelihood ratio test=6.83, df=3, P=0.08,
Fig. 7).
Depending on mixture type, only 31% to 46% of observations lead to visits (NCH=12, NEC=11,
NAG=13, NHO=10). There was no overall impact of the competitor identity on the total number
of visits received by E. plantagineum (K-W χ²=6.21, df=3, P=0.10, Fig. 8). This variable was
neither influenced by the competition treatment (whatever the competitor, K-W χ²=0.04, df=1,
P=0.84, Fig. 8). Moreover the apparent contrasted effects of competition within mixtures was
Chapitre 2| 89
Visit sequences
Competition treatment and competitor identity had no influence on the distribution of visits
across time (all P>0.05). However the number of visits received by a plant of E.
plantagineum was affected by time (K-W χ²= 71.66, df=3, P<0.01). Bumblebee made a
greater number of visits on the focal plant between 30 and 40 minutes compared to the 10 first
minutes.
DISCUSSION
plant through modifications of floral traits involved in fine attractiveness. Our results
demonstrated that competitor identity can influence floral traits involved in attractiveness
while the effect of belowground competitive status depended on competitor identity for some
Competitive interactions
context of the focal plant, E.plantagineum. Globally, it produced greater biomass in mixtures
with C. album (even greater than in monoculture for aboveground biomass) in opposition to
competitive (with H. lanatus). These results are in accordance with Flacher et al. (2015)
which found a similar pattern, although they did not controlled belowground competition. In
this study the significant competition:competitor interaction for all biomass and lnRR
Chapitre 2| 90
variables suggest a different effect of belowground competition among mixtures. Indeed, the
presence of belowground interactions had a positive effect on all biomass and lnRR values in
monocultures. However we cannot talk about facilitation per se as this process implies a
modification of the soil environment by the facilitative species. Instead, the positive effect of
C. album might be due to the relative small amounts of above- and belowground biomass it
produced: when the dividers were open (C+) the focal species probably had more space than
resources, enabling more allocation to biomass production. The opposite pattern was observed
in presence of H. lanatus when dividers were open. This wind-pollinated can produce great
album, but see also Flacher et al., 2015). Even though biomass allocations and competitive
abilities are not always associated, it has been reported that larger plants can be greater
competitors through resource depletion or space occupancy (Raynaud & Leadley, 2004,
Gurevitch et al., 2006). Therefore, the presence H. lanatus could have limited the access of E.
plantagineum to soil resources leading to a decrease in biomass production for the latter.
Besides, even if we took care to limit plant overlapping, we cannot deny that E. plantagineum
with grass competitors (Kiaer et al., 2013). Moreover, pairwise comparisons revealed that
belowground competition. This suggests that, if indeed there was aboveground competition, it
was of the same intensity whatever the competitor. The only exception concerns lnRR values
Chapitre 2| 91
calculated on aboveground biomass with differences among the two extremes of the panel:
well (e.g. through shading) while C. album did not. It would be interesting in future
Finally, the observed negative effect of C. album on belowground and total biomass in
absence of competition was not expected. C. album can have an allelopathic effect on other
plants (Batish et al., 2006) especially through shoot residues (Qasem & Hill, 1989). We can
on E. Plantagineum by limiting its growth through leaves residues on soil for instance.
However in presence of belowground competition with C.album, this effect might have been
Competitor identity had a negative influence on flower production, floral display size, flower
size and nectar indices but not on daily nectar production. Several studies showed that floral
availability: the addition of nutrients in soil (nitrogen, phosphorous, or through litter, Muñoz
et al., 2005; Burkle & Irwin, 2009,2010 ;Baude et al., 2011, Soper-Gorden & Adler, 2013) as
well as water irrigation (Petanidou et al., 1999), can lead to an increase in flower and/or
nectar production. Indeed such resources can play a key role in plant metabolic pathways
(Gurevitch et al.,2006), especially those involved in the production of flowers and their
associated rewards. Moreover nitrogen is a key component of pollen and nectar (through
proteins and amino-acids, Dafni, 1992). It is therefore consistent that competition, which
causes resource depletion (Casper & Jackson, 1997), has a negative influence on floral and
Chapitre 2| 92
nectar traits. Several studies showed that allocation to reproductive structures can be
negatively affected by competition (Weiner 1988, Nötzold et al. 1997). Likewise, floral
display size, flower size, total flower production and total amount of sugar allocated to nectar
pollinated (Baude et al., 2011, Partzsch & Bachmann, 2011, Flacher et al., 2015). Especially
Flacher et al. (2015) found that the stronger the wind-pollinated competitor, the stronger the
decrease. In our study, E. plantagineum probably had greater access to resources in mixture
with C.album leading to an increase in allocation to both flower and nectar traits. On the
contrary, mixtures with H. lanatus may have led to resource depletion in pots inducing a
expected to find no overall impact of the competition treatment on floral traits (the only
belowground competition would rise through an interactive effect with the competitor
identity. All traits followed the same pattern as biomass with greater to lower traits in
presence of belowground competition with C. album and H. lanatus respectively, however not
significantly (except for floral display size and total flower production). The absence of such
interaction for most traits was unexpected. This suggests that, within each mixture, E.
plantagineum allocate the same amount of energy to the production of attractiveness traits,
whatever its belowground competitive status with neighbours. One explanation would be that
there was a trade off in energy allocation between vegetative and reproductive structures
(Obeso, 2002): because E. plantagineum complete its life cycle within a year, it may favour
energy allocation to reproductive parts (such as flowers and associated rewards) to ensure its
reproductive success, leading to a trade off with vegetative structures (such as biomass).
Chapitre 2| 93
Therefore energy allocation to vegetative parts might suffer more from competition. However,
floral display size and total flower production did suffer from competition. As rewards ensure
the repetition of visits on a plant (Dafni, 1992), one strategy for E. plantagineum in conditions
of competition could be to favour smaller floral display size but with constant reward
production at the flower level. The fact that we observed a significant effect of competitor
identity on nectar production at the plant level and on the whole flowering period (daily and
total nectar indices) was mostly due to modifications of flower production. Flacher et al.
(2015) found the same patter for two annual species, including E. plantagineum, with no
impact of competition on daily nectar production but a negative influence at the plant level
and on the whole flowering period. These results suggest a strong importance of flower
production for plant attractiveness. The study of L. purpuruem, another annual insect-
pollinated plant that was initially part of the experimental design but that never bloomed,
could have confirmed such conclusions. Moreover, other traits involved in attractiveness to
pollinators such as flower colour and pollen quantity or quality might be worth studying as
they can be sensitive to resource availability (Lau & Stephenson, 1993, 1994; de Pascual &
Sanchez-Ballesta, 2008).
Pollinator visits
We found no effect of the belowground competitive status on the first visit choice. The
number of open flower (or floral display size) is one of the cues pollinators rely on to assess
plant attractiveness. As we decided to choose pots with a similar number of flowers between
both competition statuses (with and without competition) for every observation in order to
focus on fine attractiveness, this result was then expected. However we hypothesized that
other floral traits might be used by pollinators. For instance flower size or flower colour can
be involved in pollinators' choice (Conner & Rush, 1996; Elle & Carney, 2003; Chittka &
Chapitre 2| 94
Thompson, 2001). However there were no significant differences in flower size of E.
investigation might be needed to conclude properly, we can assume that the absence of clear
visual cues, such as differences in floral display size, did not allow bumblebees to
discriminate among plants with different belowground competitive status. As there were
differences in flower production and floral display size among mixtures it could be interesting
competition. The time to first visit did not differ according to competitors' identity. Again we
bigger flowers and the competitor smaller amount of biomass, we assumed that bumblebees
would make their choice more rapidly as flowers would be easier to discriminate. Either these
differences were too small (only 2 mm) to act as cues for pollinators, either they use other
cues for their first visits (such as floral display size that was controlled here).
For the total number of visits, the lack of a competition effect was not surprising according to
our results for nectar traits. Indeed, the repetition of visits on a plant relies on reward quantity
and quality (Dafni ,1992). In our study there was neither an overall nor an interactive effect of
competition on most nectar traits (except for daily sucrose concentration). As there was no
strong differences between pots with or without competition (or maybe too small for daily
visited plants of E.plantagineum equally within each mixtures. The lack of effect of the
competitor identity was more unexpected as there were differences of floral and nectar traits
between plant treatments. This underlines that significant differences in floral traits does not
necessarily lead to differences in pollinator visits. The observed modifications of floral and
Chapitre 2| 95
nectar traits might not have been strong enough to induce contrasted pollinator foraging
minutes to study the distribution of visits across time did not bring more information about
Overall, these conclusions are made on a relative small number of observations. Indeed, only
31% to 42% of observations lead to visits. In the rest of the observations, the focal pollinator
was flying or stayed immobile on the nylon mesh of the cage. Some studies indicate that E.
Morever, as foraging experience can modify innate pollinator preferences (Weiss 1997,
Chittka & Thompson , 2001), we took care of working with naïve bumblebees so it would not
visits on our focal plant. Several hypotheses were tested to understand why we were
encountering such difficulties to observe visits: 1) the flying arena was exported outdoor as
we suspected the glass of the greenhouse to filter UV light, a part of the light spectrum used
by hymenopterans for visual detection (Chittka et al., 1994) 2) colonies were put in the flying
arena for habituation to avoid stress and 3) starvation duration was modulated. In spite of all
these trials, we still had difficulties to record observations with active visits. One hypothesis
would be that, in spite of all the care we took to the colonies, they were not in good conditions
to visits the focal plant. The alternative hypothesis would be that the density of flowers in the
flying arena was relatively low. As contrast with the green background can be important for
flower detection (Chittka & Raine, 2006), a small number of flowers might have reduced the
ability of bumblebees to discriminate them from the green background we set up to mimic
outdoor conditions. Finally, reward production can be negatively (Devlin et al., 1987) or
positively (Pyke, 1978) influenced by flower age. Additional measurements on 2 days and 3
days aged flowered revealed that nectar volume and nectar concentration tend to be lower
Chapitre 2| 96
compared to new flowers, especially in presence of belowground interactions. In
consequences, as we could not offer only newly open flowers to bumblebees, contrast of
floral and nectar traits among mixtures might have been diluted by flower age leading, to the
To resume, we showed that the presence of a wind-pollinated plant can influence floral traits
though its identity. The effect of belowground competition however is contrasted and
probably depend in competitor identity, especially for flower production and floral display
size. The modification of reward production at the plant level and on the whole flowering
period but not at the flower level suggest that competition for resources might not influence
might be more influenced through modification of visual cues such as flower production and
floral display size. However, in this study we focused on only one pollinator species. In
nature, most insect-pollinated species interact with a whole community of pollinators that may
have different preferences (Lunau, 1995; Chittka & Thompson, 2001) and so different
responses to floral and nectar traits modifications. Future research should therefore focus on
field experiment to study the effect of competition induced by wind-pollinated plants on wild
Aknowledgment
This work was supported by a grant from the Ile-de-France region through the DIM R2DS
program. We are grateful to all the members of the CEREEP Ecotron Ile-de-France. We are
thankful to Océane Vincent for her help on the field. A special thanks to Martin Perrigault
Chapitre 2| 97
REFERENCES
Bates, D., Maechler, M., Bolker, B. & Walker S. (2015). Fitting linear mixed-effects models
using lme4. Journal of Statistical Software, 67(1), 1-48. doi:10.18637/jss.v067.i01
Batish, D. R., Singh, H. P., Rana, N., & Kohli, R. K. (2006). Assessment of allelopathic
interference of Chenopodium album through its leachates, debris extracts, rhizosphere
and amended soil. Archives of Agronomy and Soil Science, 52(6), 705-715.
Baude, M., Dajoz, I., & Danchin, É. (2008). Inadvertent social information in foraging
bumblebees: effects of flower distribution and implications for pollination. Animal
Behaviour, 76(6), 1863-1873.
Baude, M., Leloup, J., Suchail, S., Allard, B., Benest, D., Mériguet, J., Nunan, N., Dajoz, I. &
Raynaud, X. (2011). Litter inputs and plant interactions affect nectar sugar content.
Journal of Ecology, 99, 828–837.
Bell, J. M., Karron, J. D., & Mitchell, R. J. (2005). Interspecific competition for pollination
lowers seed production and outcrossing in Mimulus ringens. Ecology, 86(3), 762-771.
Brown, B. J., & Mitchell, R. J. (2001). Competition for pollination: effects of pollen of an
invasive plant on seed set of a native congener. Oecologia, 129(1), 43-49.
Burkle, L. A., & Irwin, R. E. (2009). The effects of nutrient addition on floral characters and
pollination in two subalpine plants, Ipomopsis aggregata and Linum lewisii. Plant
Ecology, 203(1), 83-98.
Burkle, L.A. & Irwin, R.E. (2010). Beyond biomass: measuring the effects of community-
level nitrogen enrichment on floral traits, pollinator visitation and plant reproduction.
Journal of Ecology, 98, 705–717.
Cardoza, Y.J., Harris, G.K. & Grozinger, C.M. (2012). Effects of soil quality enhancement on
pollinator-plant interactions. Psyche: a journal of entomology, 2012, 1–8.
Casper, B.B. & Jackson, R.B. (1997). Plant competition underground. Annual review of
Ecology and Systematics, 28, 545–570.
Chittka, L., Shmida, A., Troje, N., & Menzel, R. (1994). Ultraviolet as a component of flower
reflections, and the colour perception of Hymenoptera. Vision research, 34(11), 1489-
1508.
Chittka, L., & Thomson, J. D. (2001). Cognitive ecology of pollination. Animal behaviour
and floral evolution. Cambridge University Press, Cambridge.
Chittka, L., & Raine, N. E. (2006). Recognition of flowers by pollinators. Current opinion in
plant biology, 9(4), 428-435.
Conner, J.K. & Rush, S. (1996). The effect of flower size and number on pollinator visitation
to wild radish, Raphanus raphanistrum. Oecologia, 105, 509–516.
Chapitre 2| 98
Dafni, A. (1992). Pollination Ecology: A Practical Approach. IRL press at Oxford university
press, New York.
Devlin, A.B., Horton, J.B., Stephenson, A.G., Devlin, B., Horton, J.B. & Stephenson, A.G.
(1987). Patterns of nectar production of Lobelia cardinalis. The American Midland
Naturalist, 117, 289–295.
Dupont, Y. L., & Skov, C. (2004). Influence of geographical distribution and floral traits on
species richness of bees (Hymenoptera: Apoidea) visiting Echium species
(Boraginaceae) of the Canary Islands. International Journal of Plant Sciences, 165(3),
377-386.
Elle, E. & Carney, R. (2003). Reproductive assurance varies with flower Size in Collinsa
parviflora (Scrophulariaceae ). American journal of botany, 90, 888–896.
Flacher, F., Raynaud, X., Hansart, A., Motard, E., & Dajoz, I. (2015). Competition with wind-
pollinated plant species alters floral traits of insect-pollinated plant species. Scientific
reports, 5.
Harder, L. D., & Cruzan, M. B. (1990). An evaluation of the physiological and evolutionary
influences of inflorescence size and flower depth on nectar production. Functional
Ecology, 559-572.
Herrera, C.M., Pérez, R. & Alonso, C. (2006). Extreme intraplant variation in netcar sugar
composition in an insect-pollinated perennial herb. American journal of botany, 93, 575–
581.
Kandori, I., Hirao, T., Matsunaga, S. & Kurosaki, T. (2009). An invasive dandelion
unilaterally reduces the reproduction of a native congener through competition for
pollination. Oecologia, 159, 559–69.
Kearns, C.A. & Inouye, D.W. (1993).Techniques for Pollination Biologists (ed University
Press of Colorado). Niwot Colorado.
Kiaer, L.P., Weisbach, A.N. & Weiner, J. (2013). Root and shoot competition: a meta-analysis.
Journal of Ecology, 101, 1298–1312.
Grindeland, J.M., Sletvold, N. & Ims, R.A. (2005). Effects of floral display size and plant
density on pollinator visitation rate in a natural population of Digitalis purpurea.
Functional Ecology, 19, 383–390.
Gurevitch, J., Scheiner, S.M. & Fox, G.A. (2006). The Ecology of Plants, Second edition (ed
Sinauer Associates Inc.Publishers). Sunderland Massachusetts.
Lanza, J., Smith, G. C., Sack, S., & Cash, A. (1995). Variation in nectar volume and
composition of Impatiens capensis at the individual, plant, and population levels.
Oecologia, 102(1), 113-119.
Laverty, T.M. (1992). Plant interactions for pollinator visits: a test of the magnet species
effect. Oecologia, 89, 502–508.
Chapitre 2| 99
Lau, T. C., & Stephenson, A. G. (1993). Effects of soil nitrogen on pollen production, pollen
grain size, and pollen performance in Cucurbita pepo (Cucurbitaceae). American Journal
of Botany, 763-768.
Lau, T. C., & Stephenson, A. G. (1994). Effects of soil phosphorus on pollen production,
pollen size, pollen phosphorus content, and the ability to sire seeds in Cucurbita pepo
(Cucurbitaceae). Sexual Plant Reproduction, 7(4), 215-220.
Lunau, K., & Maier, E. J. (1995). Innate colour preferences of flower visitors. Journal of
Comparative Physiology A, 177(1), 1-19.
Makino, T.T., Ohashi, K. & Sakai, S. (2007). How do floral display size and the density of
surrounding flowers influence the likelihood of bumble bee revisitation to a plant?
Functional Ecology, 21, 87–95.
Mitchell, R.J., Karron, J.D., Holmquist, K.G. & Bell, J.M. (2004). The influence of Mimulus
ringens floral display size on pollinator visitation patterns. Functional Ecology, 18, 116–
124.
Mitchell, R. J., Flanagan, R. J., Brown, B. J., Waser, N. M., & Karron, J. D. (2009). New
frontiers in competition for pollination. Annals of Botany, 103(9), 1403-1413.
Miyake, Y.C. & Sakai, S. (2005). Effects of number of flowers per raceme and number of
racemes per plant on bumblebee visits and female reproductive success in Salvia
nipponica (Labiatae). Ecological Research, 20, 395–403.
Molina-Montenegro, M. a., Badano, E.I. & Cavieres, L. a. (2008). Positive interactions among
plant species for pollinator service: assessing the “magnet species” concept with invasive
species. Oikos, 117, 1833–1839.
Muñoz, A.A., Celedon-Neghme, C., Cavieres, L.A. & Arroyo, M.T.K. (2005). Bottom-up
effects of nutrient availability on flower production, pollinator visitation, and seed output
in a high-Andean shrub. Oecologia, 143, 126–35.
Muñoz, A.A. & Cavieres, L.A. (2008). The presence of a showy invasive plant disrupts
pollinator service and reproductive output in native alpine species only at high densities.
Journal of Ecology, 96, 459–467.
Nötzold, R., Blossey, B., & Newton, E. (1997). The influence of below ground herbivory and
plant competition on growth and biomass allocation of purple loosestrife. Oecologia,
113(1), 82-93.
Obeso, J. R. (2002). The costs of reproduction in plants. New Phytologist, 155(3), 321-348.
Chapitre 2| 100
Phytochemistry reviews, 7(2), 281-299.
Petanidou, T., Goethals, V. & Smets, E. (1999). The effect of nutrient and water availability on
nectar secretion and nectary structure of the dominant Labiatae species of phrygana.
Systematics and Geography of Plants, 68, 233–244.
Pinheiro, J., Bates, D ., DebRoy, S., Sarkarn, D. & R Core Team (2015). nlme:
Linear and Nonlinear Mixed Effects Models_. R package version 3.1-122,
<URL: http://CRAN.R-project.org/package=nlme>.
Pyke, G. H. (1978). Optimal foraging in bumblebees and coevolution with their plants.
Oecologia, 36(3), 281-293.
Qasem, J. R., & Hill, T. A. (1989). Possible role of allelopathy in the competition between
tomato, Senecio vulgaris L. and Chenopodium album L. Weed Research, 29(5), 349-356.
Raynaud, X. & Leadley, P.W. (2004). Soil characteristics play a key role in modeling nutrient
competition in plant communities. Ecology, 85, 2200–2214
Soper-Gorden, N.L. & Adler, L.S. (2013). Abiotic conditions affect floral antagonists and
mutualists of Impatiens capensis (Balsaminaceae). American Journal of Botany, 100,
679–689.
Stout, J. C., & Goulson, D. (2001). The use of conspecific and interspecific scent marks by
foraging bumblebees and honeybees. Animal behaviour, 62(1), 183-189.
Therneau, T. M. & Grambsch, P.M. (2000). Modeling Survival Data: Extending the Cox
Model_. Springer, New York. ISBN 0-387-98784-3.
Velthuis, H. H., & Van Doorn, A. (2006). A century of advances in bumblebee domestication
and the economic and environmental aspects of its commercialization for pollination.
Apidologie, 37(4), 421.
Weigelt, A. & Jolliffe, P. (2003) Indices of plant competition. Journal of Ecology, 91, 707–
720.
Weiner, J. A. C. O. B. (1988). The influence of competition on plant reproduction. Plant
reproductive ecology: patterns and strategies, 228-245.
Weiss, M.R. (1997). Innate colour preferences and flexible colour learning in the pipevine
swallowtail. Animal Behaviour, 53, 1043–1052.
Chapitre 2| 101
Figure 1: Mean floral display size of E. plantagineum (+/- s.e.) per plant according to
competition (full grey, without competition C-; hatched grey, with competition C+) and
competitor identity. “Mono” refers to monocultures of E. plantagineum and wind-pollinated
species are ordered according to increasing intensity of competitive interactions (see Results,
CH: mixtures with C. album, AG: mixtures with A. capillaris, HO: mixtures with H. lanatus).
Different letters indicate significant differences between plant treatments (normal for the
“without” competition treatment, bold for the “with competition” treatment). The stars
represent the significant effect of competition within plant treatments.
Chapitre 2| 102
Figure 2: Mean daily flower production of E. plantagineum (+/- s.e.) per plant according to
competition (full grey, without competition C-; hatched grey, with competition C+) and
competitor identity. “Mono” refers to monocultures of E. plantagineum and wind-pollinated
species are ordered according to increasing intensity of competitive interactions (see Results,
CH: mixtures with C. album, AG: mixtures with A. capillaris, HO: mixtures with H. lanatus).
Different letters indicate significant differences between plant treatments (normal for the
“without” competition treatment, bold for the “with competition” treatment). The stars
represent the significant effect of competition within plant treatments.
Chapitre 2| 103
Figure 3: Mean total flower production of E. plantagineum (+/- s.e.) per plant according to
competition (full grey, without competition C-; hatched grey, with competition C+) and
competitor identity. “Mono” refers to monocultures of E. plantagineum and wind-pollinated
species are ordered according to increasing intensity of competitive interactions (see Results,
CH: mixtures with C. album, AG: mixtures with A. capillaris, HO: mixtures with H. lanatus).
Different letters indicate significant differences between plant treatments (normal for the
“without” competition treatment, bold for the “with competition” treatment). The stars
represent the significant effect of competition within plant treatments.
Chapitre 2| 104
Figure 4: Mean flower size (mm) of of E. plantagineum (+/- s.e.) per plant according to
competition (full grey, without competition C-; hatched grey, with competition C+) and
competitor identity. “Mono” refers to monocultures of E. plantagineum and wind-pollinated
species are ordered according to increasing intensity of competitive interactions (see Results,
CH: mixtures with C. album, AG: mixtures with A. capillaris, HO: mixtures with H. lanatus).
Different letters indicate significant differences between plant treatments (normal for the
“without” competition treatment, bold for the “with competition” treatment).
Chapitre 2| 105
Figure 5: Mean total volume index of E. plantagineum (+/- s.e.) per plant according to
competition (full grey, without competition C-; hatched grey, with competition C+) and
competitor identity. “Mono” refers to monocultures of E. plantagineum and wind-pollinated
species are ordered according to increasing intensity of competitive interactions (see Results,
CH: mixtures with C. album, AG: mixtures with A. capillaris, HO: mixtures with H. lanatus).
Different letters indicate significant differences between plant treatments (normal for the
“without” competition treatment, bold for the “with competition” treatment).
Chapitre 2| 106
Figure 6: Mean total sucrose index of E. plantagineum (+/- s.e.) per plant according to
competition (full grey, without competition C-; hatched grey, with competition C+) and
competitor identity. “Mono” refers to monocultures of E. plantagineum and wind-pollinated
species are ordered according to increasing intensity of competitive interactions (see Results,
CH: mixtures with C. album, AG: mixtures with A. capillaris, HO: mixtures with H. lanatus).
Different letters indicate significant differences between plant treatments (normal for the
“without” competition treatment, bold for the “with competition” treatment).
Chapitre 2| 107
Figure 7: Time to first visit on E. plantagineum (+/- s.e.) per plant according to competitor
identity. “Mono” refers to monocultures of E. plantagineum.
Chapitre 2| 108
Figure 8: Mean number of visits on E. plantagineum (+/- s.e.) per plant according to
competition (full grey, without competition C-; hatched grey, with competition C+) and
competitor identity. “Mono” refers to monocultures of E. plantagineum and wind-pollinated
species are ordered according to increasing intensity of competitive interactions (see Results,
CH: mixtures with C. album, AG: mixtures with A. capillaris, HO: mixtures with H.
lanatus).Different letters indicate significant differences.
Chapitre 2| 109
Table 1: Anova table for biomass measurements and lnRR values
Chapitre 2| 110
Table 2: Mean ln RR values (+/−standard error) per treatment for E. plantagineum in
monocultures and mixtures with the wind-pollinated competitor (C+ “with competition”, C-
“without competition”).
lnRR values
Competition Mono Mixture Mixture Mixture
C. album A. capillaris H. lanatus
Floral traits pots
Aboveground c+ 0 0.453 -0.150 -0.710
biomass (+/- 0.003) (+/- 0.002) (+/- 0.002) (+/- 0.002)
Belowground c+ 0 0.036 -0.459 -1.292
biomass (+/-0.004) (+/-0.003) (+/-0.003) (+/-0.004)
Total biomass c+ 0 0.289 -0.273 -0.899
(+/-0.002) (+/-0.002) (+/-0.001) (+/-0.002)
Aboveground c- 0 0.043 -0.161 -0.215
biomass (+/- 0.0008) (+/-0.001) (+/-0.001) (+/-0.002)
Belowground c- 0 -0.159 -0.063 0.0212
biomass (+/-0.003) (+/-0.002) (+/-0.003) (+/-0.003)
Total biomass c- 0 -0.192 -0.145 -0.151
(+/-0.001) (+/-0.001) (+/-0.001) (+/-0.001)
Visited pots
Aboveground c+ 0 0.387 0.009 -0.442
biomass (+/-0.0008) (+/-0.0008) (+/-0.0004) (+/-0.0004)
Aboveground c- 0 0.203 -0.115 -0.176
biomass (+/-0.0004) (+/-0.0004) (+/-0.0003) (+/-0.0004)
Chapitre 2| 111
SUPPLEMENTARY INFORMATION
Supplementary Table S1: Plant species description 1-3 (Family names are given according to
APG III classification4) From Flacher et al. 2015
** focal insect-pollinated species initially integrated in the experimental design but that never
bloomed
Habitat preference
Life
Plant species Status Habitat based on Ellenberg Other characteristics
cycle
values5
Insect-pollinated
species
native to the violet tubular
western flowers mainly
Echium plantagineum annual along full light
Mediterranean pollinated by
L. or coasts, intermediate fertility
basin honeybees,
(Boraginaceae) biennial sandy places intermediate pH
subspontanous bumblebees,
in the region butterflies
pink tubular flowers
Lamium purpureum grasslands, intermediate light
native to the mainly pollinated
L. annual forest high fertility
region by honeybees,
(Lamiaceae)** hedges intermediate pH
bumblebees
Wind-pollinated
species
intermediate light
Agrostis capillaris L. native to the perenni grasslands, intermediate fertility
(Poaceae) region al wastelands low to intermediate
pH
intermediate to full
Chenopodium album
native to the grasslands, light
L. annual
region wastelands high fertility
(Amaranthaceae)
intermediate pH
intermediate to full
Holcus lanatus L. native to the perenni grasslands, light
(Poaceae) region al wastelands intermediate fertility
intermediate pH
Chapitre 2| 112
Figure S1: Mean biomass (g) of E. plantagineum (+/- s.e.) per plant according to competition
and competitor identity: a. aboveground and belowground biomass in “traits pots”, b.
aboveground biomass only in “visited pots”. “Mono” refers to monocultures and wind-
pollinated species are ordered according to increasing intensity of competitive interactions
(see Results, CH: mixtures with C. album, AG: mixtures with A. capillaris, HO: mixtures
with H. lanatus). When indicated different letters indicate significant differences between
plant treatments. Stars represent significant differences within plant treatment. For the panel a.
light grey is for aboveground biomass, dark grey is for belowground biomass and black is for
total biomass.
Chapitre 2| 113
V. Chapitre 3
Présentation du chapitre :
Chapitre 3| 117
contrôler la pollinisation. Pour chaque observation, la toile était soulevée afin de laisser la
faune pollinisatrice sauvage visiter les plantes S. alba. Le temps avant la première visite, le
nombre de visites (totales et sur des séquences de 5min) ainsi que la diversité des visites ont
été étudiés. En parallèle et afin de pouvoir relier les visites aux traits floraux, la production de
fleurs et l’écran floral ont été étudiés tout au long de l’expérience. La taille des fleurs et la
production de nectar (volume et concentration) ont de nouveau été mesurés, notamment car
nous changions de conditions initiales (serre vs champ) et que nous n’avions aucune donnée
pour S. alba. Enfin les fruits et graines ont été récoltées à la fin de l'expérience afin d’estimer
la fécondité de S. alba.
Ce chapitre 3 confirme l'effet négatif que H. lanatus peut avoir sur des espèces
entomophile focales. En effet, la compétition hypogée induite par la présence de cette espèce
anémophile a affecté à la fois des structures végétatives et reproductrices de S. alba. Ainsi la
biomasse, la hauteur de tige mais aussi la production de fleur, la vitrine florale et la taille des
fleurs sont autant de traits qui ont diminué en présence de compétition hypogée. Alors que la
production de nectar n'a pas été affectée. En réponse à ces modifications de traits
d'attractivité, nous avons pu observer une diminution du nombre de visite de pollinisateur
reçue par S. alba à l'échelle individuelle et à l'échelle une parcelle, en condition de
compétition pour les ressources du sol. La diversité des visites quant à elle n'a pas été
influencée. Ainsi, une modification des traits floraux en présence d compétition pour les
ressources du sol peut également influencer le comportement de visites de pollinisateurs
sauvages. Les conséquences sur la fécondité sont quant à elles plus faibles : seul le nombre
d'ovules non fécondés était plus important en conditions de compétition, modifiant ainsi le
rapport de fécondité de S. alba
Les résultats du Chapitre 3 suggèrent que la présence d'une espèce anémophile compétitrice
puisse influencer les interactions plantes-pollinisateurs, via des interactions de compétition
pour les ressources du sol. L'effet semble principalement se produire via une modification de
la production de fleur et de la vitrine florale plutôt que via les récompenses produites qui
restent relativement constantes quelque soit le traitement.
Chapitre 3| 118
Figure 4 : Cadre d'étude du Chapitre 3
Chapitre 3| 119
Belowground competition induced by wind-pollinated species influence attractiveness of an
Floriane Flacher1, Amandine Hansart2, Benoit Geslin3 & Isabelle Dajoz4 / Xavier Raynaud5
Author affiliations:
1
CNRS, Sorbonne Universités, UPMC Univ Paris 6, Institute of Ecology and Environmental Sciences
2
CNRS, UMS 3194 CEREEP-Ecotron Ile de France, F-77140 Saint-Pierre-lès-Nemours
3
Université Aix Marseille Université, Institut Méditerranéen de Biodiversité et d’Ecologie Marine et
Continentale 9 (IMBE) CNRS, IRD, Avignon Université Pôle St Jérôme av. Escadrille N. Niemen
4
Université Paris Diderot Paris 7, Institute of Ecology and Environmental Science - Paris, UMR 7618,
5
Sorbonne Universités, UPMC Univ Paris 6, Institute of Ecology and Environmental Science - Paris,
Chapitre 3| 120
ABSTRACT
Competitive interactions between plants are known to affect both vegetative (e.g. biomass)
and reproductive structures (e.g. flowers). However as many plants also interact with
pollinators, modifications of floral traits through plant competition could have a strong
floral and nectar traits of insect-pollinated plant in presence of a competitor plant species, the
impact on wild pollinators has never been investigated to our knowledge. To do so, we set up
involved in attractiveness to pollinators (i.e. flowers and nectar) were measured as well as
pollinator visits. Fruits and seeds were also studied to assess the fecundity of S. alba.
Belowground competition induced by H. lanatus had a significant negative effect on daily and
total flower production, daily floral display size and flower size. However daily nectar traits
were not affected. The total number of pollinator visits received at the plot level and at the
floral traits. Therefore the observed decrease of pollinators visits on S. alba is associated to
the modifications of floral traits induced by competition. For the fecundity of S. alba,
competition, suggesting pollen limitation. This study demonstrate that pollination networks
might be modulated by competitive interactions between plants, even with non insect-
pollinated species.
Chapitre 3| 121
INTRODUCTION
In the current context of global change, plant and insect pollinator communities are submitted
agricultural practices or biological invasions (Potts et al., 2010), leading to a parallel decline
al., 2007; Aizen et al., 2008). Interactions between insect-pollinated plant and their
pollinators depend on the production of floral traits such as visual or olfactory cues attracting
pollinators seeking for trophic resources (Dafni, 1992; Proctor et al., 1996). These cues are
generally related to flower characteristics such as their number, size, color or scent and their
associated rewards (i.e. pollen and nectar). Therefore, the attractiveness of a plant depends on
pollinators' preferences. Indeed most pollinators have innate preferences for flower colours
(Lunau & Maier, 1995), scents (Dobson, 2006) or morphologies (Rodriguez et al., 2004).
Likewise, most insect pollinators tend to visit more plant species displaying large flowers
(Conner & Rush 1996), large floral display (Mitchell et al., 2004) or greater rewards (Cnaani
et al., 2006). Moreover, such preferences can be modulated by their foraging experience. For
example, most bee species are able to learn associations between floral traits and the reward
status of a plant (Weiss,1997; Chittka & Thompson, 2001). As a result, plant species that are
identified as more attractive are more visited. This is of special importance at the plant
community level: the individual plant attractiveness depends on the attractiveness of other
plant species of the community that display floral traits of varying attractiveness to
pollinators. For example, many studies on invasive plant species showed that the presence of
a showy and/or highly rewarding non native plant species in a plant community can lead to a
decrease in pollinator visit on less attractive species (Muñoz & Cavieres, 2008). Subsequent
Chapitre 3| 122
pollinators (Brown et al., 2002; Kandori et al., 2009). In addition to modifying pollinator
visits, the composition of plant communities can also influence the identity and diversity of
pollinators (Lazaro et al., 2009) than can in turn affect plant fruit and seed set (Albrecht et al.,
2012). Therefore, the neighbouring context of a plant species can have great consequences on
If the neighbouring context of a plant can alter its reproductive success through competition
for pollinators, it can also influence individual plant attractiveness through competition for
soil resources. Indeed, some studies showed that attractiveness traits (such as flowers size and
numbers as well as reward quantity and quality) are sensitive to soil resource availability with
positive effects of nutrient addition (e.g. nitrogen, phosphorous; Lau & Stephenson, 1994 ;
Muñoz et al., 2005; Burkle & Irwin, 2009; Burkle & Irwin, 2010, Soper-Gorden & Adler,
2013), soil amendments (i.e. litter, vermicompost, Baude et al., 2011; Cardoza et al., 2012) or
irrigation (Petanidou et al., 1999) and negative effects of competitive interactions with plant
neighbours (Partzsch & Baschmann, 2011; Baude et al., 2011). In particular, Flacher et al.
(2015) demonstrated that the presence of wind-pollinated competitors could alter several
floral parameters such as floral display size, total flower production and total amount of
sucrose allocated to nectar of insect-pollinated focals. Especially, they showed that the impact
on these floral traits and associated rewards was stronger as competition intensity increased.
Because floral traits are closely linked to the foraging behaviour of pollinators we can assume
that plant competition for soil resources could have a bottom-up effect on plant-pollinator
interactions which could, in return, alter the reproductive success of insect-pollinated species.
However, to our knowledge, the impact of competition for soil resources between plants on
pollinator behaviour and the subsequent plant fecundity, through modifications of floral traits,
has not been investigated yet. In this study we set up a field experiment in which we
submitted or not an insect-pollinated plant (Sinapis alba) to competition for soil resources
Chapitre 3| 123
with a wind-pollinated plant (Holcus lanatus). We followed plant attractiveness along the
flowering period (flower production, floral display size, flower size, nectar production) as
well as pollinator visits. At the end of the experiment, fruits and seeds were collected to
estimate plant fecundity. Through this experiment our objectives were to answer the following
question: how does belowground competition affect 1) floral traits involved in plant
attractiveness to pollinators, 2) pollinators visits and 3) subsequent plant fecundity, all at the
plot and the plant level? We expected belowground competition to have a negative effect on
pollinator visits and subsequent plant fecundity through modifications of floral traits involved
in attractiveness to pollinators.
Plant species
Insect-pollinated species
Sinapis alba is an annual forb that grows along roads, in wastelands or near crops (CR
strategist, Kühn et al., 2004). It is considered naturalized in the Ile-de-France region, were the
experiment took place (CBNBP, 2016). It is an obligate outcrossing species (Olsson, 1960 ;
Cheng et al., 2012). From May to July, it produces yellow open flowers that can be visited by
honey bees and Bombus species (Corbet 1978) as well as shorter-tongued pollinators like
solitary bees (Dukas & Schmida, 1989) or hoverflies (Kunin, 1993). Fruits are siliques
Competition mixture
Sinaps alba was grown in mixture with two other species: Holcus lanatus and Echium
Chapitre 3| 124
vulgare. Holcus lanatus is a wind-pollinated perennial grass that grows in grasslands,
wastelands, or along road crops (C strategist, Kühn et al., 2004). It produces great biomass
and can negatively affect floral traits involved in attractiveness to pollinators of insect-
pollinated species through competitive interactions (Flacher et al., 2015). Echium vulgare is a
biannual species that grows along roads (CR strategist, Kühn et al., 2004). It is insect
pollinated when flowering. However, due to its biannual life cycle, none of the plant
individuals bloomed during the experiment. Therefore, we focused on Sinapis alba for the
Experimental set-up
The experiment took place in a grassland site (CEREEP Ecotron Ile-de-France, Saint Pierre
lès Nemours, FRANCE) with sandy soil (pH =6). Before setting up the experimental plots,
the top 10 cm of soil was ploughed and steam-sterilized to remove the seed bank and existing
vegetal material (e.g. plant stolons), that could both interfere with experimental treatments.
In March 2014, 20g of Holcus lanatus seeds per experimental plots (0.9x0.9m) were sown. In
April 2014, 12 seedlings of Sinapis alba and Echium vulgare were planted in each
experimental plot along parallel diagonals. Belowground competition was controlled thanks to
PVC tubes (30cm deep) buried in plots to isolate S. alba roots from other root systems. Two
treatments were set up: 1. the “without competition” treatment (C-) plots in which roots of
Sinapis alba are isolated from competition and 2. the “with competition” treatment (C+) plots
in which roots of Sinapis alba are not isolated from competition (see supplementary
information, Fig. S1). These treatments were organized in blocks and replicated 5 times, for a
total of 10 experimental plots. All experimental plots were enclosed in a nylon mesh cage
experimental surveys. When necessary, all plots were watered. Aboveground competition
Chapitre 3| 125
induced by the competition mixture was limited thanks to regular cuts of aboveground
biomass of Holcus lanatus (10-15cm high, all experimental plots at the same time).
For each plant, the time to first flowering was recorded. This date was also used to calculate
the duration of flowering until the last produced flower. Flower production of each plant
individual was assessed every day (except for weekends) along the flowering period: newly
opened flowers were marked and counted. At the same time, the total number of opened
flowers per day, defined here as the floral display size, was estimated. Additional traits
(flower size and nectar traits) were measured, on a maximum of 6 flowers, randomly selected
on 3 plants (3 plants, 2 flowers per plant). These measurements were made 1) on newly
opened flowers to limit variations of flower size and reward production induced by flower age
(Southwick & Southwick 1983) and 2) before each behavioural observation to avoid a
shortage due to pollinator foraging. No measurements were made if the number of newly open
flowers was insufficient (i.e. less than 3 newly open flowers) so pollinators could have
resources to seek. Flower size was estimated on the selected flowers (see above) by measuring
the width of their corolla between the tip of two opposed petals (mm) with a digital calliper
(Digit-Cal MK IV, Brown&Sharpe). The floral display size was also checked before each
observation in case of newly open flowers between the measurement of flower production and
the observation.
Concerning nectar traits, we measured the volume of nectar and its sugar concentration on the
selected flowers (6 at most, see above). Nectar was sampled using calibrated microcapillary
tubes (0.5µL, 32mm- Minicaps end to end, Hirschmann Laborgeraete). Right after sampling
nectar volume was estimated by measuring the length of liquid in the microcapillary with a
digital calliper (Digit-Cal MK IV, Brown&Sharpe). Nectar sugar concentration was measured
Chapitre 3| 126
using hand-held refractometers (Eclipse 45-81 and Eclipse 45-82, Bellingham+ Stanley Ltd.).
When nectar volumes were too small, samples were diluted with MilliQ water (Millipore
solution of pure sucrose, we will refer to sucrose equivalent (rather than sugar content) in the
following.
In all plots, at the end of the flowering period, fruits were collected. Because of fruit
herbivory and late harvest, we were not able to estimate the total number of fruit produced per
plant. However, for each plant we measured the length (mm) and weight (mg) of six fruits at
most (randomly chosen among intact fruits) when possible. After being dried (see below,
Plant height and biomass), fruits were measured from the bottom to the top of the silique with
(Pioneer OHAUS). For each sampled fruit, seeds were counted, weighted (after being dried)
on a microbalance (mg) and their diameter was measured (mm) with a digital calliper (Digit-
Cal MK IV, Brown&Sharpe) by taking the hilum (former attachment of the ovule to the
ovary) as a reference point. For each fruit, the number of unfertilized ovules as well as
aborted seeds was also counted. Based on these numbers, we calculated a “fecundity ratio” by
dividing the number of viable seeds by the total number of ovules (i.e. sum of viable, aborted
seeds and unfertilized ovules). Therefore this ratio varies from 0 to 1 (from no ovules to all
In all plots, we measured the height of four plants (randomly chosen, cm) along the
experiment (at 8, 35, 63 days). Moreover, at the end of the experiment, aboveground as well
Chapitre 3| 127
as belowground biomass of each possible insect-pollinated individual was weighed after being
Behavioural observations
Before each behavioural observation, and after floral and nectar traits were measured, the
nylon mesh cage was lifted up, allowing wild pollinators to visit the plot. Observations were
made to the naked eye, therefore pollinators were divided into easily recognizable groups:
honeybees (Apis mellifera), solitary bees, bumblebees (Bombus spp), butterflies, syrphids,
other diptera and beetles. A session of behavioural observation began with the first visit of a
pollinator on a plant. Then, during 20 minutes, every floral visit was noted to the plant
individual level. Visits made by the same individual pollinator or by different individuals of
the same group were not discriminated. At the end of the session, the nylon mesh cage was
put back and any remaining pollinator was removed from the cage. The time between the
lifting of the nylon cage and the first visit was also recorded to determine a time to first visit
on plots. We waited 30 minutes maximum for a first visit before stopping an observation
session. To have a greater overview, we focused on the total number of visits (on the
experiment) received at plot and plant levels. From the plant's point of view, the total number
of visit is more likely to influence its fecundity, and therefore is of more interest than per
observation visits. In order to estimate the diversity of visit an individual of S. alba could
receive, we adapted Shannon diversity's and Piélou' evenness indices to our data following the
formulas:
Diversity index (D')= - Σ (Ni /N) log (Ni/N) (adapted from Shannon's)
where Ni corresponds to the number of visits made by one group of pollinators, N the total
number of visits by all pollinators and P the number of pollinator groups. Meteorological
Chapitre 3| 128
parameters (air temperature, air humidity, cloud cover and wind speed) that may influence
Data analysis
All statistical analysis were performed with R 3.2.2 (R core Team 2015). Plant stem height,
plant biomass, flower size, nectar traits (i.e. volume, concentration and amount of sucrose),
fruit size, fruit weight, seed size and seed weight were analysed using linear mixed models
(lme, nlme package, Pinheiro et al., 2015). Competition treatment was set as a fixed effect. As
all meteorological parameters (air temperature, air humidity, cloud cover and wind speed)
were correlated with each other and with the date, we used the date in the following as an
integrative covariate in the case of daily measurements. This also enable taking into account
for plant age and it was set as a fixed effect also for daily parameters. Plots and date were set
as random effects as there were repeated measures. Data were transformed when necessary
(square for flower size and square root transformations for nectar traits, fruit weight). All
count data (floral display size, daily flower production, total flower production, visits, and
seed number) as well as the fecundity ratio and indices of diversity and evenness were
analysed using non parametric tests (Kruskal-Wallis) as generalized linear models did not fit
well the data. As we could not test the effect for all explaining variables (i.e. competition
treatment, pollinator type and date for daily variables) at the same time or for interactions, we
performed a test for the overall effect of competition treatment, the effect of pollinator type
among competition levels and the overall effect of the date (for daily measures). Moreover we
did some pairwise test especially to study differences among pollinator types. Therefore, p-
values were adjusted with the Holm method as we used subsets of the same dataset. For the
time to first flowering and the time to first visit we used survival models, especially non-
Chapitre 3| 129
parametric Cox proportional hazard regression models to test the effect of the competition
treatment on the “risk” of first flowering and first landing across time (coxph, Survival
package, Therneau & Grambsch, 2000). Finally we tested the association between floral traits
and visits using Spearman rank correlation tests. Because of a strong edge effect, 1 plot of
each treatment had to be removed from the experiment, leading to 8/10 studied plots in total.
A total of 4450 flowers were counted from June, 12th to September, 11th 2014 with Nno
competition = 3827 and Ncompetition= 623. Among these flowers, 131 were sampled for flower size
and nectar measurements (Nno competition = 97, Ncompetition= 34). For behavioural observations, a
total of 66 sessions were run from the 18th of june 2014 to the 31st of july 2014 with Nno
RESULTS
There was no significant effect of competition on stem height at the beginning of the
experiment (F3,30=1.03, P=0.39). However there was a significant effect of competition later
in the experiment. (F3,30=17.27, P<0.01, see supplementary information, Fig. S2). Indeed,
after 63 days of growth, plants submitted to root competition had shorter stems than plant
isolated from competition, meaning that the competitor mixture had a negative effect on the
growth of S. alba. The same pattern was observed for aboveground and belowground biomass
(F1.6=13.21, P=0.01 and F1,6=34.59, P<0.01 respectively). Moreover, stem height of S. alba
was not influenced by the presence of E. vulgare (all P>0.05, see supplementary informations
Fig. S2).
Chapitre 3| 130
Impact of belowground competition on attractiveness to pollinators at the plot level
Flower traits
At the plot level, floral display size was significantly reduced by the competition treatment
(K-W, χ²=112.76 df=1 P<0.01, Fig. 1). Plots were less attractive when plants of Sinapis alba
were competing with H. lanatus for belowground resources, with fewer open flower per plots.
Daily and total flower production of S. alba were also significantly reduced by competition
(K-W, χ²=117.29 df=1 P<0.01 and K-W, χ²=5.33 df=1 P=0.02 respectively). There was also a
significant effect of the date for daily variables (P<0.05) with a decrease 40 days after the
Belowground competition had a negative impact on the time to first visit on a plot (Z=3.5,
P<0.01, Fig. 2). The first visit on a plot occurred in less than 10 minutes after the removal of
the nylon mesh cage for all C- plots, whereas only 50% of C+ plot had their first visits during
this time. Overall, if all C- plots were visited during behavioural observation rounds, 10% of
C+ plots had no visit during the first 30min after nylon mesh removal.
Number of visits
First, as 97.7% of visits were made by solitary bees, hoverflies and bumblebees we focused
on those 3 pollinator types among the 7 previously defined. Concerning, the total number of
visits per plot, there was a tendency for fewer visits when S. alba was submitted to
competition (K-W, χ²=3, df=1 P=0.08). There was an effect of the pollinator type on the total
number of visits received by S. alba in absence of competition only (K-W, χ²=7.38, df=1
P=0.02). Especially, bumblebees did less visits compared to solitary bees and hoverflies. In
Chapitre 3| 131
presence of competition however there was no difference in the distribution of visits among
pollinator types.
Diversity of visits
There was no influence of the competition treatment on the diversity or evenness of visits at
the plot level (P>0.05 for both indices). In both plots with or without competition, plants of S.
alba received relatively the same amount of visit from the different pollinators groups
Flower traits
The time to first flowering as well as flowering duration were influenced by the competition
competition, about 40% of plants had not bloomed yet 20 days after the beginning of the
experiment and still 20% had not bloomed at all at the end of the experiment. In comparison,
all plants of S. alba without belowground competition had bloomed by the 35th day of
experiment (Z=2.08, P=0.04, Fig. 3). Moreover, plants of S. alba tended to have shorter
flowering duration in the “with competition” treatment (K-W, χ²=12.92, df=1 P<0.01). Plants
of S. alba produced fewer flowers and had a smaller flower display size in presence of
belowground competition with H. lanatus (K-W, χ²= 523.01, df =1, P<0.01 and K-W,
χ²=563.66, df=1, P<0.01 respectively, Fig. 1 for floral display size). The total flower
production was also negatively impacted by the competition treatment (K-W, χ²=42.82 df=1
P<0.01). In addition, plants of S. alba produced significantly smaller flowers in the “with
competition” treatment (F1.6=11.15, P=0.02). There was also a significant effect of the date for
Chapitre 3| 132
the daily flower production, the floral display size and flower size (all P<0.05).
Nectar traits
The daily nectar volume, the daily nectar concentration and the daily amount of sucrose
allocated to nectar were not affected by the competition treatment (all P>0.05). However there
was a contrasted but significant effect of the date for all three variables (all P<0.01). Along
the experiment, daily nectar volume increased while daily amount of sucrose decreased
Number of visits
At the level of plant individual, the total number of visits received per plant of S. alba all
along the experiment was significantly lower in the “with competition” treatment than in the
“without competition” treatment (K-W, χ²=36.44, df=1 P<0.01, Fig. 4). Moreover, among
each treatment of competition, there was an influence of the pollinator type (without
competition: K-W, χ²=42.49, df=1 P<0.01; with competition: K-W, χ²=8.21 df=1 P=0.02).
Similarly to the plot level, the effect of competition seems to be mainly due to bumblebees
visiting less S. alba plants than the other two groups of pollinators.
Fecundity of S. alba
At the end of the experiment, a total of 200 fruits were collected (N with competition=47 on 23
plants; N witthout competition = 153 on 32 plants). There was no influence of the competition
treatment on fruit size, fruit weight, seed number per fruit, seed size or seed weight (all
P>0.05) and a marginally non significant effect for fruit size F1.6=4.80, P=0.07 and seed
number K-W, χ²= 3.19, df=1, P=0.07). On the other hand, belowground competition with H.
lanatus had a negative influence on the fecundity ratio of S. alba (K-W, χ²= 12.73, df=1,
P<0.01, Fig. 5). This difference was mainly due to a greater number of unfertilized ovules for
Chapitre 3| 133
plants in the “with competition” treatment (K-W, χ²= 4.67 df=1, P=0.03).
There was a significant negative correlation between the floral display size and the time to
first visit (S=45.16x103, P<0.01, ρ=-0.46). When the floral display size increases, pollinators
come more rapidly on the plot. There was also a positive correlation between the total flower
production and the total number of visits at the plot level (however marginally significant,
S=22, P=0.05, ρ=0.74). For each observation, the same pattern was observed between floral
display size and the number of visits received by S. alba, at the plot level (S=19.36x103,
P<0.01, ρ=0.60).
DISCUSSION
In this study we wanted to test if the presence of a wind-pollinated species could influence the
production of floral traits in S. alba and 2) these modifications in turn affected pollinator
visits received by this focal plant. However, the effect on the fecundity of S. alba was less
obvious.
In this study, we focused on the effect of the presence of H. lanatus, a wind- pollinated plant,
while an insect-pollinated plant E. vulgare was also present in the competition mixture. This
is justified as the presence of E. vulgare had no influence on the focal plant (here S. alba)
whether on its vegetative (stem height) or reproductive structures (flower production, data not
Chapitre 3| 134
shown). The competition induced by H. lanatus however was clear with the reduction of both
biomass and floral traits of the insect-pollinated plant, S. alba. Indeed, floral traits such as
flower production, floral display size and flower size tended to decrease in presence of
belowground competition while daily nectar traits (volume, sucrose concentration and sucrose
amount allocated to nectar) were not affected A similar pattern was found for E.
traits decreased in presence of a strong competitor (Flacher et al., 2015, Flacher et al.,
unpublished data). However, in the first study, belowgound competition was not controlled
and in the second study, the effect of belowground competition depended on the competitor
competition (be it aboveground, belowground or both) decreased floral traits involved in plant
attractiveness to pollinators. Flacher et al. 2015 suggest that, for some focal plants,
stem height here) and/or reproductive structures (flowers). The results obtained here by
controlling belowground competition are in agreement with such assertion. Even though
greater root biomass is not always associated to greater competitive abilities (Cahill 2003),
larger plants such as H. lanatus can be advantaged especially through soil space occupation or
nutrient acquisition (Gurevitch et al., 2006, Raynaud & Leadley, 2004). As floral traits of
insect-pollinated plant can be sensitive to soil resources amount (Muñoz et al., 2005; Burkle
& Irwin, 2009), a decrease of their production is expected in presence of a competitor plant.
Concerning rewards, daily nectar traits remained constant whatever the competitive context.
This pattern is consistent with some previous results (Flacher et al., 2015, Flacher et al.,
unpublished data). For annual species (such as S. alba) that have to reproduce within a season,
Chapitre 3| 135
maintain floral rewards constant. Indeed rewards quantity and quality have a strong influence
on the repetition of pollinator visits on a plant (Dafni, 1992). The number of visits a plant
receive can in turn influence pollen receipt (Engel & Irwin, 2003), ovule fertilization and
hence seed production (Young & Stanton, 1990). Therefore, annual insect-pollinated plants
might favour floral rewards rather than flower number, leading to a trade-off in the production
on rewards at the plant level and on the whole flowering period. Indeed, for some insect-
pollinated plants, the total amount of sucrose allocated to nectar (the product of mean sucrose
amount per flower and total flower production per plant) as well as the total nectar volume
(the product of mean nectar volume per flower and total flower production per plant) are
reduced in presence of a competitor plant (Baude et al., 2011; Flacher et al., 2015) while they
were not affected at the flower level (Flacher et al., 2015). Therefore rewards might be
affected by competition only if flower production is taken into account simultaneously. Such
pollinator visits (Leiss & Klinkhamer, 2005). In our experiment, we were not able to calculate
it as we did not follow nectar traits every day but only before an observation.
The positive correlations between floral traits (i.e. flower production and floral display size)
and the number of visits confirm that, whether on the whole experiment or at each observation
modifications of floral traits. This is in accordance with many studies showing that pollinators
(e.g. bees or hoverflies) tend to visit more plants with greater floral display size (Conner &
Rush, 1996 ; Mitchell et al., 2004 ; Grindeland et al., 2005 ; Miyake & Sakai, 2005, Makino
Moreover, it can be a reliable cue for pollinators to assess if a plant is rewarding as floral
Chapitre 3| 136
display size can be sometimes correlated to reward production (Harder & Cruzan, 1990). At
the plant level, a greater floral display size can also be associated to lower movement costs
between flowers and higher rewards as re-visitation might be low (Mitchell et al., 2004)
For the fecundity of S. alba, the main result concerns the fecundity ratio, declining in
presence of competition. This decrease was mainly due to more unfertilized ovules per fruits
for plants in presence of belowground competition. As S. alba produced the same amount of
ovules also means a decrease in fertilized ovules leading to seeds. This was represented by the
tendency for seed number to be lower in presence of belowground competition. While the
availability of soil resources can influence the number of seed produced per fruit (Campbell &
Halama, 1993; Burkle & Irwin, 2009), our results suggest rather that the decrease of the
fecundity ratio observed in presence of belowground competition could have been induced by
modifications of pollinator visits (i.e. pollen limitation, Campbell & Halama, 1993; Ashman
et al., 2004 ). In this experiment, we did not discriminate the number of visits at the flower
level. Therefore we could not test for correlation between the number of received visits per
flowers and this number of unfertilized ovules per plants. However, some studies
demonstrated that the number of visits a plant receives can influence seed set (Karron et al.,
2006). Likewise experiments of hand pollination showed that several plant species are pollen
limited, producing fewer fruit and/or seed with inadequate pollen receipt (Ashman et al.,
competition suggests that there might be some pollen limitation. Concerning fruit and seed
morphometrical traits (i.e length and weight) we expected that competition would have a
“direct” effect as soil resource availability (Campbell & Halama, 1993) and competition can
influence seed traits (Weiner 1988, Aguiar et al., 2001). However, we observed constancy in
Chapitre 3| 137
those traits whatever the competition level. Seed size constancy has sometimes been
canalization, Silvertown, 1989). Some authors have suggested that there is high inter- and
intraplant variation among plant species, leading to apparent constancy in mean seed size
(Fenner, 2012). Even though this was observed for very large samples, we did observe intra-
and interplant variation for both competition levels in our study that could explain the absence
of overall competition effect. The effect of competition through modification of visits number
could have been greater on fruit number. Indeed several studies showed that fruit set depends
on the number of received visits and pollen receipt (pollen limitation, Ashman et al., 2004).
However, both plant mortality along the experiment and late harvest lead to the impossibility
to assess fruit number correctly. In future experiments this trait should be followed. Moreover,
plant fecundity and more globally plant reproductive success is not only represented by fruit
and seed sets. It would have been interesting to study the recruitment rate (number of juvenile
plants / total number of seeds) of Sinapis alba according to the competition treatment.
In most of our conclusions above, we estimated that the competition induced was mainly
belowground, for soil resources. Of course, we cannot deny that there was aboveground
than aboveground competition (Casper & Jackson,1997) especially when the competitor is a
grass (Kiaer et al., 2013). Moreover, we tried to limit aboveground competition through
behaviour, several studies revealed a decrease in above- or belowground biomass, soil nutrient
assimilation and even an increase in nitrogen availability in soil (Shahzad et al., 2012;
Youngner, 2012). Therefore, if the behaviour of H. lanatus was modified because of regular
cuts it is likely that clipping reduced its natural competitive abilities leading to an
Chapitre 3| 138
underestimation of its impact on S. alba here.
Conclusion:
Overall, our results demonstrate that the presence of a wind-pollinated plant species in a
community, through competitive interactions among species, can impact the communitie’s
the attractiveness and reproductive success of a focal, insect pollinated plant species. In the
current context of a biodiversity crisis, these results have potential strong implications for
management practices aimed at preserving the diversity of both plant and pollinator
communities.
Acknowledgments: This work was supported by a grant from the Ile-de-France region
through the DIM R2DS program. We are grateful to all the members of the CEREEP Ecotron
Ile-de-France. We also thank Séléné Verstraet, Elisa Rousset, Alice Gauthey, Manon Bataille
and Ludovico St Amour di Chanaz for their help on the field and in the lab.
Author contributions statement: FF, XR and ID conceived and designed the study. FF, AH,
ID and XR collected data. FF, XR and BG performed statistical analysis. FF wrote the first
draft of the manuscript and all authors (FF, ID, AH, BG and XR) contributed substantially to
revisions.
Additional information:
Chapitre 3| 139
REFERENCES
Aguiar, M.R., Lauenroth, W.K. & Peters, D.P. (2001) Intensity of intra- and interspecific
competition in coexisting shortgrass species. Journal of Ecology, 89, 40–47.
Aizen, M.A., Morales, C.L. & Morales, J.M. (2008) Invasive mutualists erode native
pollination webs. PLoS biology, 6, 396–403.
Albrecht, M., Schmid, B., Hautier, Y. & Müller, C.B. (2012) Diverse pollinator communities
enhance plant reproductive success. Proceedings of the Royal Society B: Biological
Sciences, 279, 4845–52.
Albrecht, M., Schmid, B., Hautier, Y. & Müller, C.B. (2012) Diverse pollinator communities
enhance plant reproductive success. Proceedings of the Royal Society B: Biological
Sciences, 279, 4845–52.
Ashman, T.-L., Knight, T.M., Steets, J.A., Amarasekare, P., Burd, M., Campbell, D.R.,
Dudash, M.R., Johnston, M.O., Mazer, S.J., Mitchell, R.J., Morgan, M.T. & Wilson,
W.G. (2004) Pollen limitation of plant reproduction: ecological and evolutionary causes
and consequences. Ecology, 85, 2408–2421.
Baude, M., Leloup, J., Suchail, S., Allard, B., Benest, D., Mériguet, J., Nunan, N., Dajoz, I. &
Raynaud, X. (2011) Litter inputs and plant interactions affect nectar sugar content.
Journal of Ecology, 99, 828–837.
Biesmeijer, J.C., Roberts, S.P.M., Reemer, M., Ohlemüller, R., Edwards, M., Peeters, T.,
Schaffers, a P., Potts, S.G., Kleukers, R., Thomas, C.D., Settele, J. & Kunin, W.E. (2006)
Parallel declines in pollinators and insect-pollinated plants in Britain and the
Netherlands. Science, 313, 351–354.
Brown, B.J., Mitchell, R.J. & Graham, S.A. (2002) Competition for pollination between an
invasive species (Purple Loosestrife) and a native congener. Ecology, 83, 2328–2336.
Burkle, L.A. & Irwin, R.E. (2009) The effects of nutrient addition on floral characters and
pollination in two subalpine plants, Ipomopsis aggregata and Linum lewisii. Plant
Ecology, 203, 83–98.
Burkle, L.A. & Irwin, R.E. (2010) Beyond biomass: measuring the effects of community-
level nitrogen enrichment on floral traits, pollinator visitation and plant reproduction.
Journal of Ecology, 98, 705–717.
Cahill, J. F. (2003). Lack of relationship between below‐ground competition and allocation to
roots in 10 grassland species. Journal of ecology, 91(4), 532-540.
Campbell, D. R., & Halama, K. J. (1993). Resource and pollen limitations to lifetime seed
production in a natural plant population. Ecology, 1043-1051.
Cardoza, Y.J., Harris, G.K. & Grozinger, C.M. (2012) Effects of soil quality enhancement on
pollinator-plant interactions. Psyche: a journal of entomology, 2012.
Chapitre 3| 140
Casper, B.B. & Jackson, R.B. (1997) Plant competition underground. Annual review of
Ecology and Systematics, 28, 545–570.
Chittka, L. & Thomson, J.D. (2001) Cognitive Ecology of Pollination (eds L Chittka and JD
Thomson).
Cnaani, J., Thomson, J.D. & Papajà, D.R. (2006) Flower Choice and Learning in Foraging
Bumblebees : Effects of Variation in Nectar Volume and Concentration. Ethology, 112,
278–285.
Conner, J.K. & Rush, S. (1996) The effect of flower size and number on pollinator visitation
to wild radish, Raphanus raphanistrum. Oecologia, 105, 509–516.
Conservatoire Botanique National du Bassin Parisien. (2016) CBNBP. Available at:
<http://cbnbp.mnhn.fr/cbnbp/ >
Corbet, S.A. (1978) Bee visits and the nectar of Echium vulgare L. and Sinapis alba L.
Ecological Entomology, 3, 25–37.
Dafni, A. (1992) Pollination Ecology: A Practical Approach. IRL press at Oxford university
press, New York.
Dobson, H. E. M. (2006). Floral fragrance composition and type of pollinator in Biology of
floral scent. CRC Press.
Dukas, R. & Schmida, A. (1989) Correlation between the color, size and shape of Israeli
crucifer flowers and relationships to pollinators. Oikos, 54, 281–286.
Engel, E.C. & Irwin, R.E. (2003) Linking pollinator visitation rate and pollen receipt.
American journal of botany, 90, 1612–1618.
Flacher, F., Raynaud, X., Hansart, A., Motard, E. & Dajoz, I. (2015) Competition with wind-
pollinated plant species alters floral traits of insect-pollinated plant species. Scientific
reports, 5, 1–10.
Flacher F., Hansart A., Motard E., Geslin B., Fofana A., Dajoz I. & Raynaud X. (unpublished).
Does competition with wind-pollinated species alter Echium plantagineum attractiveness
to a common pollinator, Bombus terrestris?
Chapitre 3| 141
Ecology, 4, 559–572.
Jauzein, P. (2011) Flore Des Champs Cultivés, 2e ed (ed Quae). versailles.
Kandori, I., Hirao, T., Matsunaga, S. & Kurosaki, T. (2009) An invasive dandelion unilaterally
reduces the reproduction of a native congener through competition for pollination.
Oecologia, 159, 559–69.
Karron, J. D., Mitchell, R. J., & Bell, J. M. (2006). Multiple pollinator visits to Mimulus
ringens (Phrymaceae) flowers increase mate number and seed set within fruits. American
Journal of Botany, 93(9), 1306-1312.
Kiaer, L.P., Weisbach, A.N. & Weiner, J. (2013) Root and shoot competition: a meta-analysis
(ed D Gibson). Journal of Ecology, 101, 1298–1312.
Kühn, I., Durka, W., & Klotz, S. (2004). BiolFlor: a new plant-trait database as a tool for plant
invasion ecology. Diversity and Distributions, 10(5/6), 363-365.
Kunin, W.E. (1993) Sex and the single mustard: population density and pollinator behavior
effects on seed-set. Ecology, 74, 2145–2160.
Lau, T. C., & Stephenson, A. G. (1994). Effects of soil phosphorus on pollen production,
pollen size, pollen phosphorus content, and the ability to sire seeds in Cucurbita pepo
(Cucurbitaceae). Sexual Plant Reproduction, 7(4), 215-220.
Lázaro, A., Lundgren, R., & Totland, Ø. (2009). Co-flowering neighbors influence the
diversity and identity of pollinator groups visiting plant species. Oikos, 118(5), 691-
702.
Leiss, K.A. & Klinkhamer, P.G.L. (2005) Spatial distribution of nectar production in a natural
Echium vulgare population: Implications for pollinator behaviour. Basic and Applied
Ecology, 6, 317–324.
Lunau, K., & Maier, E. J. (1995). Innate colour preferences of flower visitors. Journal of
Comparative Physiology A, 177(1), 1-19.
Makino, T.T., Ohashi, K. & Sakai, S. (2007). How do floral display size and the density of
surrounding flowers influence the likelihood of bumble bee revisitation to a plant?
Functional Ecology, 21, 87–95.
Memmott, J., Craze, P.G., Waser, N.M. & Price, M. V. (2007) Global warming and the
disruption of plant-pollinator interactions. Ecology letters, 10, 710–7.
Mitchell, R.J., Karron, J.D., Holmquist, K.G. & Bell, J.M. (2004) The influence of Mimulus
ringens floral display size on pollinator visitation patterns. Functional Ecology, 18, 116–
124.
Miyake, Y.C. & Sakai, S. (2005). Effects of number of flowers per raceme and number of
racemes per plant on bumblebee visits and female reproductive success in Salvia
nipponica (Labiatae). Ecological Research, 20, 395–403.
Chapitre 3| 142
Morales, C.L. & Traveset, A. (2009) A meta-analysis of impacts of alien vs. native plants on
pollinator visitation and reproductive success of co-flowering native plants. Ecology
letters, 12, 716–28.
Muñoz, A.A., Celedon-Neghme, C., Cavieres, L.A. & Arroyo, M.T.K. (2005) Bottom-up
effects of nutrient availability on flower production, pollinator visitation, and seed output
in a high-Andean shrub. Oecologia, 143, 126–35.
Muñoz, A.A. & Cavieres, L.A. (2008). The presence of a showy invasive plant disrupts
pollinator service and reproductive output in native alpine species only at high densities.
Journal of Ecology, 96, 459–467.
Olsson, G. (1960) Self-incompatibility and outcrossing in rape and white mustard. Hereditas,
46, 241–252.
Partzsch, M. & Bachmann, U. (2011) Is Campanula glomerata threatened by competition
from expanding grasses? Results from a 5-year pot-experiment. Plant Ecology, 212,
251–261.
Petanidou, T., Goethals, V. & Smets, E. (1999) The effect of nutrient and water availability on
nectar secretion and nectary structure of the dominant Labiatae species of phrygana.
Systematics and Geography of Plants, 68, 233–244.
Pinheiro J, Bates D, DebRoy S, Sarkar D and R Core Team (2015). nlme: Linear and
Nonlinear Mixed Effects Models_. R package version 3.1-122, Available online at:
<http://CRAN.Rproject.org/package=nlme>.
Potts, S.G., Biesmeijer, J.C., Kremen, C., Neumann, P., Schweiger, O. & Kunin, W.E. (2010)
Global pollinator declines: trends, impacts and drivers. Trends in ecology & evolution,
25, 345–53.
Proctor, M., Yeo, P. & Lack, A. (1996) The Natural History of Pollination (ed Timber Press
Inc.). Portland, Oregon.
R Core Team (2015). R: A language and environment for statistical computing. R Foundation
for Statistical Computing, Vienna, Austria. Available online at: https://www.R-
project.org/
Raynaud, X. & Leadley, P.W. (2004) Soil characteristics play a key role in modeling nutrient
competition in plant communities. Ecology, 85, 2200–2214.
Rodríguez, I., Gumbert, A., de Ibarra, N. H., Kunze, J., & Giurfa, M. (2004). Symmetry is in
the eye of the ‘beeholder’: innate preference for bilateral symmetry in flower-naïve
bumblebees. Naturwissenschaften, 91(8), 374-377.
Shahzad, T., Chenu, C., Repinçay, C., Mougin, C., Ollier, J.L. & Fontaine, S. (2012) Plant
clipping decelerates the mineralization of recalcitrant soil organic matter under multiple
grassland species. Soil biology and biochemistery, 51, 73–80.
Silvertown, J. (1989) The paradox of seed size and adaptation. Trends in ecology & evolution,
Chapitre 3| 143
4, 24–26.
Soper-Gorden, N.L. & Adler, L.S. (2013) Abiotic conditions affect floral antagonists and
mutualists of Impatiens capensis (Balsaminaceae). American Journal of Botany, 100,
679–689.
Southwick, A.K. & Southwick, E.E. (1983) Aging effect on nectar production in two clones of
Asclepias syriaca. Oecologia, 56, 121–125.
Therneau, T. M. & Grambsch, P.M. (2000). Modeling Survival Data: Extending the Cox
Model_. Springer, New York. ISBN 0-387-98784-3.
Young, H.J. & Stanton, M.L. (1990) Influences of floral variation on pollen removal and seed
production in wild radish. Ecology, 71, 536–547.
Youngner, V.B. & McKell, C.M. (2012) Physiology of defoliation and regrowth. The biology
and utilization of grasses, p. 446.
Weiner, J. (1988). The influence of competition on plant reproduction. Plant reproductive
ecology: patterns and strategies, pp. 228–245.
Weiss, M.R. (1997) Innate colour preferences and flexible colour learning in the pipevine
swallowtail. Animal Behaviour, 53, 1043–1052.
Chapitre 3| 144
Figure 1: Mean floral display size and mean total flower production of S. alba (+/- s.e.)
without (light grey, C-) and with competition (dark grey, C+) at a) the plot level and b) the
plant level. Different letters are for significant differences.
Chapitre 3| 145
Figure 2: Probability of first visit on a plant of S. alba across time, without (C-) and with
competition (C+).
Chapitre 3| 146
Figure 3: Probability of first flowering for a plant of S. alba across time, without (C-) and
with competition (C+).
Chapitre 3| 147
Figure 4: Mean total number of visits received by a plant of S. alba, without (C-) and with
competition (C+). The star represents a significant difference between competition levels (i.e.
C- or C+) while different letters are for significant differences observed per pollinator type
among competition levels (i.e C- or C+).
Chapitre 3| 148
Figure 5: Mean fecundity ratio (+/- se) of S. alba without and with belowground competition
with H. lanatus. Different letters are for significant differences.
Chapitre 3| 149
Supplementary information
Figure S1: Illustration of the initial experimental design. This represents one block containing
the four different treatments of competition with Holcus lanatus: a). Sinapis alba (E1)
isolated from belowground competition, b). Echium vulgare (E2) isolated from belowground
competition, c). both Sinapis alba and Echium vulgare isolated from belowground
competition, and d). both Sinapis alba and Echium vulgare not isolated from belowground
competition (see Material and Methods) (not to scale). We focused on treatment a. (the
“without competition” treatment C-). and d. (he “with competition” treatment C+).
Chapitre 3| 150
Figure S2: Mean stem height (+/- s.e.) of S. alba (cm) after 63 days of interaction with the
competition mixture. Compared to the Figure S1, “all competition” refers to both S. alba and
E. vulgare not isolated from belowground competition (treatment d.), “S.alba isolated” refers
to S. alba isolated from belowground competition (treatment a.), “E. vulgare isolated” refers
to E. vulgare isolated from belowground competition (treatment b.) and “all isolated” refers to
both S. alba and E. vulgare isolated from belowground competition (treatment c.). In the
study, we focused on “all in competition” (treatment d. in Figure S1, C+) and “S. alba
isolated” (treatment a. in Figure S1, C-). Different letters are for significant differences.
Chapitre 3| 151
VI. Conclusion générale et perspectives
Au cours de cette thèse nous nous sommes intéressés à l'effet des interactions de compétition
entre plantes, notamment via la présence d'espèces anémophiles, sur l'attractivité aux
pollinisateurs de plantes entomophiles. D'un système simple, à l'échelle d'une plante, jusqu’à
l'étude des interactions entre une communauté de plantes et de pollinisateurs, nous avons
démontré que la compétition induite par la présence d'espèces anémophiles, notamment pour
les ressources du sol, pouvait diminuer l’attractivité de plantes entomophiles annuelles. Cela
semble notamment influencer leur production de fleurs et leur vitrine florale. L'effet de la
compétition sur les récompenses associées, des traits plus fins d'attractivité comme la
production de nectar, ne semble se révéler que lorsque la production de fleurs est prise en
compte, qui plus est sur l'ensemble de la période de floraison. A l'échelle de la fleur, les
récompenses en nectar semblent rester constantes quelque soit le traitement de compétition.
Ces résultats nous ont ainsi permis de suggérer un possible trade-off dans la production de
traits d'attractivité, notamment chez les espèces entomophiles annuelles. Concernant
l'attractivité aux pollinisateurs en elle même, elle semble être également influencée par la
présence d'espèces anémophiles compétitives dans le milieu, la diminution du nombre de
visites en conditions de compétition étant associés à la diminution de la production de fleurs
précédemment énoncée. Les conséquences sur la fécondité de la plante visitée concernent le
nombre d'ovule non fécondés, qui diminue en conditions de compétition, suggérant ainsi une
probable limitation dans la quantité reçue de pollen due à la réduction des visites.
Globalement, l'intensité de ces effets dépend fortement de la compétitivité de l'espèce
anémophile. Ainsi, certaines espèces peu agressives favorisent les traits d'attractivité chez des
espèces entomophiles lorsque ces dernières sont de meilleurs compétiteurs. Puisque la
compétition entre plantes peut moduler la disponibilité des ressources (Casper & Jackson,
1997; Gurevitch et al., 2006), elle même source de variation des patrons d'allocation au sein
des plantes (Aerts, 1991 ; Fitter, 1994 ; Müller et al., 2000 ; Hermans et al., 2006), l'effet
négatif de la présence d'espèces anémophiles compétitives observé dans cette étude est
probablement le résultat d'une diminution des ressources du sol, limitant ainsi l'allocation à la
fois aux structures végétatives (biomasse) et reproductrices (fleurs, notamment). De manière
globale, cette étude soulève donc l'importance de la compétition, notamment pour les
ressources du sol, dans les interactions plantes-pollinisateurs, son effet pouvant se transmettre
Bien sûr, ce travail de thèse présente certaines limites, ouvrant ainsi de nouvelles pistes de
recherche. Tout d'abord, nous nous sommes basés sur un design substitutif afin d’étudier les
interactions de compétition interspécifique. Ce protocole repose sur l'étude des interactions de
compétition où la densité totale est maintenue constante: les mélanges sont le résultat du
remplacement de certains individus par des individus d'autres espèces. L’ensemble des plantes
a ainsi accès au même volume de sol (et donc à la même quantité de ressources) entre les
monocultures et les mélanges. Néanmoins, ce design ne permet pas, dans le cas d’une
meilleure performance de la plante focale en mélange, d’estimer si cela est le résultat d’une
meilleure compétitivité de l’espèce focale ou si l’autre espèce en mélange est facilitatrice.
C’est le cas notamment pour les espèces entomophiles étudiées ici dont les performances
peuvent être meilleures en mélange avec C. album (cf Chapitre 1 notamment). La production
d’une faible biomasse racinaire chez C. album suggère fortement que les espèces
entomophiles étudiées sont de meilleures compétitrices, néanmoins, nous ne pouvons
objectivement pas écarter l’hypothèse d’une relation de facilitation. Cette distinction pourrait
être faite à l'aide d'un design additif, qui consiste à ajouter, à une densité constante de plants
de l’espèce focale, une certaine densité d’une autre espèce. Ainsi, afin de mieux séparer les
effets de la facilitation de celle de la compétition où une espèce est très supérieure à l'autre, il
pourrait être judicieux de combiner ces deux types d’expérimentation. Par ailleurs, les
interactions de compétition ont été estimées à partir des mesures de biomasses finales des
différentes plantes en interaction. Il s'agit d'une méthode couramment utilisée (Casper &
Jaskson, 1997 ; Gurevitch et al., 2006) sur laquelle se basent un grand nombre d'indices
permettant de quantifier le sens, l'intensité, ou le résultat de la compétition induite par une
espèce végétale sur une autre (Weigelt & Jolliffe, 2003). De plus, selon certains auteurs la
biomasse ou la taille d'une plante peut être associée à la compétitivité d'une espèce (Gurevitch
et al., 2006, Rajianemi 2007), la compétition pour les ressources du sol étant principalement
Malgré ces limites, les résultats obtenus suggèrent qu'une espèce végétale compétitive pour
les ressources, bien qu'elle n’interagisse pas directement avec la faune pollinisatrice, peut
affecter l'attractivité et la fécondité d'une plante entomophile. A terme, les conséquences pour
ces espèces, et plus généralement pour les réseaux plantes-pollinisateurs, pourraient être
importantes. D'une part la composition des communautés de plantes pourrait être modifiée. En
effet, si la présence d'une espèce anémophile compétitive pour les ressources du sol, peut
diminuer le succès reproducteur d'une plante entomophile via une diminution du nombre de
visites reçues, cela pourrait diminuer l'abondance de l'espèce entomophile et à terme (selon la
taille de la population, la diversité génétique) menacer son maintien dans la communauté
(Thomann et al., 2013). En parallèle, cela pourrait influencer la communauté des
pollinisateurs, notamment selon leur degré de spécialisme vis à vis de la plante entomophile
menacée Ainsi au sein des communautés naturelles, il est probable que la composition des
communautés végétales et la compétition entre plantes pour les ressources du sol, influencent
(et ait influencé) la structuration des communautés végétales de manière indirecte, via des
interactions plantes-pollinisateurs. Cela pourrait être particulièrement le cas dans le contexte
actuel de modification des communautés végétales (Lavergne et al., 2006) susceptible de
modifier les interactions de compétition entre plantes. En diminuant l'attractivité d'une plante
entomophile, la compétition entre plantes pour les ressources du sol peut ainsi réduire la
quantité de ressources florales offerte aux pollinisateurs, souvent cité comme une des causes
de leur déclin actuel (Potts et al., 2010). Les conséquences pour la plante entomophile en
compétition pourraient être d'autant plus importantes si l'espèce compétitrice est elle même
entomophile et invasive. En effet en plus d'entrer en compétition pour les ressources du sol,
les espèces entomophiles de la communauté entreraient en compétition pour les pollinisateurs
(Chittka & Schürkens, 2001). Par ailleurs, la topologie des réseaux de pollinisation pourrait
également être modifiée selon le contexte de compétition au sein d'une communauté de
plante. Dans le contexte d'invasions biologiques (impliquant souvent des espèces
entomophiles plus attractives) certaines études ont déjà démontré une modification de la
structure des réseaux de pollinisation (Aizen et al., 2008, Bartomeus et al., 2008). Ainsi, dans
un contexte de compétition entre plantes pour les ressources du sol réduisant l'attractivité de
certaines espèces entomophiles, les réseaux de pollinisation pourraient également subir des
changements. Néanmoins, il ne s'agit là que d'un ensemble d'hypothèses qui mériteraient
d'être étudiées plus en détail.
Outre ces perspectives de recherche, ce travail de thèse a des implications potentielles pour
des problématiques plus appliquées. La fragmentation des habitats (dont l'urbanisation et le
remembrement agricole) est souvent citée comme une des causes majeures des modifications
observées au sein des communautés naturelles, et notamment du déclin des pollinisateurs
(Potts et al. ,2010 ; Geslin et al., 2016). Ainsi de nouvelles initiatives ont émergé ces dernières
années afin de restaurer ressources et habitats. Dans notre cas, nous pouvons notamment
penser à l'élaboration de mélanges de graines pour la mise en place de prairies ou jachères
fleuries, voire de « bandes enherbées », sensées servir de réservoir de nourriture mais aussi
Adler, L. S., & Irwin, R. E. (2005). Ecological costs and benefits of defenses in nectar.
Ecology, 86(11), 2968-2978.
Aerts, R. (1999) Interspecific competition in natural plant communities : mechanisms , trade-
offs and plant – soil feedbacks. Journal of Experimental Botany, 50, 29–37.
Aerts, R., & Chapin III, F. S. (2000). The mineral nutrition of wild plants revisited. Adv Ecol
Res, 30, 1-67.
Agrawal, A.A., Ackerly, D.D., Adler, F., Arnold, A.E., Caceres, C., Doak, D.F., Post, E.,
Hudson, P.J., Maron, J., Mooney, K.A., Power, M., Schemske, D., Stachowicz, J.,
Strauss, S., Turner, M.G. & Werner, E. (2007) Filling key gaps in population and
community ecology. Frontiers in Ecology and the Environment, 5, 145–152.
Aguiar, M.R., Lauenroth, W.K. & Peters, D.P. (2001) Intensity of intra- and interspecific
competition in coexisting shortgrass species. Journal of Ecology, 89, 40–47.
Aizen, M. A., Morales, C. L., & Morales, J. M. (2008). Invasive mutualists erode native
pollination webs. PLoS Biol, 6(2), e31.
Alarcon, R., Waser, N.M. & Ollerton, J. (2008) Year-to-year variation in the topology of a
plant pollinator interaction network. Oikos, 1–12.
Albrecht, M., Schmid, B., Hautier, Y. & Müller, C.B. (2012) Diverse pollinator communities
enhance plant reproductive success. Proceedings of the Royal Society B: Biological
Sciences, 279, 4845–52.
Allen, E. B., & Allen, M. F. (1984). Competition between plants of different successional
stages: mycorrhizae as regulators. Canadian Journal of Botany, 62(12), 2625-2629.
Alm, J., Ohnmeiss, T. E., Lanza, J., & Vriesenga, L. (1990). Preference of cabbage white
butterflies and honey bees for nectar that contains amino acids. Oecologia, 84(1), 53-57.
Ashman, T.-L., Knight, T.M., Steets, J.A., Amarasekare, P., Burd, M., Campbell, D.R.,
Dudash, M.R., Johnston, M.O., Mazer, S.J., Mitchell, R.J., Morgan, M.T. & Wilson,
W.G. (2004) Pollen limitation of plant reproduction: ecological and evolutionary causes
and consequences. Ecology, 85, 2408–2421.
Barber, N.A. & Soper-Gorden, N.L. (2015) How do belowground organisms influence plant –
pollinator interactions ? Journal of Plant Ecology, 8, 1–11.
Bardgett, R.D., Wardle, D.A. & Yeates, G.W. (1998) Linking above-ground and below-ground
interactions: how plant response to foliar herbivory influence soil organisms. Soil
biology and biochemistery, 30, 1867–1878.
Bibliographie | 164
Bartomeus, I., Vilà, M., & Santamaría, L. (2008). Contrasting effects of invasive plants in
plant–pollinator networks. Oecologia, 155(4), 761-770.
Baude, M. (2009) Information et compétition dans les systèmes plantes-pollinisateurs : rôle de
la structure spatiale et de la diversité des communautés végétales
Baude, M., Leloup, J., Suchail, S., Allard, B., Benest, D., Mériguet, J., Nunan, N., Dajoz, I. &
Raynaud, X. (2011) Litter inputs and plant interactions affect nectar sugar content.
Journal of Ecology, 99, 828–837.
Begon, M., Townsend, C.R. & Harper, J.L. (2006) Ecology - From Individuals to Ecosystems,
fourth edition (ed Blackwell Publishing). Carlton.
Biesmeijer, J.C., Roberts, S.P.M., Reemer, M., Ohlemüller, R., Edwards, M., Peeters, T.,
Schaffers, a P., Potts, S.G., Kleukers, R., Thomas, C.D., Settele, J. & Kunin, W.E. (2006)
Parallel declines in pollinators and insect-pollinated plants in Britain and the
Netherlands. Science, 313, 351–354.
Blackmore, L. M., & Goulson, D. (2014). Evaluating the effectiveness of wildflower seed
mixes for boosting floral diversity and bumblebee and hoverfly abundance in urban
areas. Insect Conservation and Diversity, 7(5), 480-484.
Blaauw, B. R., & Isaacs, R. (2014). Flower plantings increase wild bee abundance and the
pollination services provided to a pollination‐dependent crop. Journal of Applied
Ecology, 51(4), 890-898.
Booth, B. D., Murphy, S. D., & Swanton, C. J. (2010). Structure and dynamic of population in
Invasive plant ecology in natural and agricultural systems 2nd edition, CABI
Broeckling, C.D., Broz, A.K., Bergelson, J., Manter, D.K. & Vivanco, J.M. (2008) Root
exudates regulate soil fungal community composition and diversity. Applied and
Environmental Microbiology, 74, 738–744.
Bronstein, J.L., Wilson, W.G. & Morris, W.F. (2003) Ecological dynamics of mutualist /
antagonist communities. The American naturalist, 162, S24–S39.
Brown, B. & Mitchell, R. (2001) Competition for pollination: effects of pollen of an invasive
plant on seed set of a native congener. Oecologia, 129, 43–49.
Brown, B. J., Mitchell, R. J., & Graham, S. A. (2002). Competition for pollination between an
invasive species (purple loosestrife) and a native congener. Ecology, 83(8), 2328-2336.
Burkle, L. A., & Irwin, R. E. (2009). The effects of nutrient addition on floral characters and
pollination in two subalpine plants, Ipomopsis aggregata and Linum lewisii. Plant
Ecology, 203(1), 83-98.
Burkle, L.A. & Irwin, R.E. (2010) Beyond biomass: measuring the effects of community-
level nitrogen enrichment on floral traits, pollinator visitation and plant reproduction.
Journal of Ecology, 98, 705–717.
Bibliographie | 165
Burkle, L. A., & Alarcón, R. (2011). The future of plant–pollinator diversity: understanding
interaction networks across time, space, and global change. American Journal of Botany,
98(3),
Cahill Jr, J. F., & Casper, B. B. (2000). Investigating the relationship between neighbor root
biomass and belowground competition: field evidence for symmetric competition
belowground. Oikos, 90(2), 311-320.
Cahill Jr, J. F., Elle, E., Smith, G. R., & Shore, B. H. (2008). Disruption of a belowground
mutualism alters interactions between plants and their floral visitors. Ecology, 89(7),
1791-1801.
Callaway, R.M. (1995) Positive interactions among plants. Botanical Review, 61, 306–349.
Callaway, R., Newingham, B., Zabinski, C. A., & Mahall, B. E. (2001). Compensatory growth
and competitive ability of an invasive weed are enhanced by soil fungi and native
neighbours. Ecology Letters, 4(5), 429-433.
Callaway, R.M. (2007) Positive Interactions and Interdependence in Plant Communities (ed
Springer). Dordrecht.
Canto, A., Pérez, R., Medrano, M., Castellanos, M.C. & Herrera, C.M. (2007) Intra-plant
variation in nectar sugar composition in two Aquilegia species (Ranunculaceae):
Contrasting patterns under field and glasshouse conditions. Annals of Botany, 99, 653–
660.
Cardoza, Y.J., Harris, G.K. & Grozinger, C.M. (2012) Effects of soil quality enhancement on
pollinator-plant interactions. Psyche: a journal of entomology, 2012, 1–8.
Casper, B.B. & Jackson, R.B. (1997) Plant competition undergound. Ecology, 28, 545–570.
Chase, J.M., Abrams, P.A., Grover, J.P., Diehl, S., Chesson, P., Holt, R.D., Richards, S.A.,
Nisebt, R.M. & Case, T.J. (2002) The interaction between predation and competition : a
review and synthesis. Ecology letters, 5, 302–315.
Chittka, L., & Schürkens, S. (2001). Successful invasion of a floral market. Nature,
411(6838), 653-653.
Chittka, L., Spaethe, J., Schmidt, A. & Hickelsberger, A. (2001) Adaptation, constraint, and
chance in the evolution of flower color and pollinator color vision in Cognitive ecology
of pollination. Cambridge : Cambridge University Press
Cnaani, J., Thomson, J.D. & Papajà, D.R. (2006) Flower choice and learning in foraging
bumblebees : effects of variation in nectar volume and concentration. Ethology, 112,
278–285.
Conner, J.K. & Rush, S. (1996) The effect of flower size and number on pollinator visitation
to wild radish, Raphanus raphanistrum. Oecologia, 105, 509–516.
Bibliographie | 166
Craine, J.M., Fargione, J. & Sugita, S. (2005a) Supply pre-emption, not concentration
reduction, is the mechanism of competition for nutrients. The New phytologist, 166, 933–
40.
Craine, J. M. (2005b). Reconciling plant strategy theories of Grime and Tilman. Journal of
ecology, 93(6), 1041-1052.
Craine, J. M., & Dybzinski, R. (2013). Mechanisms of plant competition for nutrients, water
and light. Functional Ecology, 27(4), 833-840.
Crane, P. R., Friis, E. M., & Pedersen, K. R. (2000). The origin and early diversification of
angiosperms. Shaking the Tree: Readings from Nature in the History of Life, 233-250.
Dafni, A. (1992) Pollination Ecology: A Practical Approach. IRL press at Oxford university
press, New York.
van Dam, N.M. (2009) How plants cope with biotic interactions. Plant Biology, 11, 1–5.
Danforth, B. (2007) Bees. Current biology, 17, 156–161.
Darwin, C. 1862 on the various contrivances by which British and foreign orchids are
fertilised by insects, and on the good effects of intercrossing
Devlin, A.B., Horton, J.B., Stephenson, A.G., Devlin, B., Horton, J.B. & Stephenson, A.G.
(1987) Patterns of nectar production of Lobelia cardinalis. The American Midland
Naturalist, 117, 289–295.
Dobson, H. E. (2006). Relationship between floral fragrance composition and type of
pollinator. Biology of floral scent, 147-198.
Dupont, Y.L., Padrón, B., Olesen, J.M. & Petanidou, T. (2009) Spatio-temporal variation in
the structure of pollination networks. Oikos, 118, 1261–1269.
Elle, E. & Carney, R. (2003) Reproductive assurance varies with flower Size in Collinsa
parviflora (Scrophulariaceae ). American journal of botany, 90, 888–896.
Engel, E.C. & Irwin, R.E. (2003) Linking pollinator visitation rate and pollen receipt.
American journal of botany, 90, 1612–1618.
Feltham, H., Park, K., Minderman, J., & Goulson, D. (2015). Experimental evidence of the
benefit of wild flower strips to crop pollination. Ecology and Evolution.
Fenner, M. W. (2012). Seed ecology. Springer Science & Business Media.
Fenster, C. B., Armbruster, W. S., Wilson, P., Dudash, M. R., & Thomson, J. D. (2004).
Pollination syndromes and floral specialization. Annual Review of Ecology, Evolution,
and Systematics, 375-403.
Bibliographie | 167
Flacher, F., Raynaud, X., Hansart, A., Motard, E., & Dajoz, I. (2015). Competition with wind-
pollinated plant species alters floral traits of insect-pollinated plant species. Scientific
reports, 5.
Fontaine, C., Dajoz, I., Meriguet, J., & Loreau, M. (2005). Functional diversity of plant–
pollinator interaction webs enhances the persistence of plant communities. PLoS Biol,
4(1), e1.
Fontaine, C., Guimarães, P.R., Kéfi, S., Loeuille, N., Memmott, J., van der Putten, W.H., van
Veen, F.J.F. & Thébault, E. (2011) The ecological and evolutionary implications of
merging different types of networks. Ecology Letters, 14, 1170–1181.
Gange, A. C., & Smith, A. K. (2005). Arbuscular mycorrhizal fungi influence visitation rates
of pollinating insects. Ecological Entomology, 30(5), 600-606.
Gardener, M.C. & Gillman, M.P. (2001a) Analyzing variability in nectar amino acids:
composition is less variable than concentration. Journal of chemical ecology, 27, 2545–
58.
Gardener, M.C. & Gillman, M.P. (2001b) The effects of soil fertilizer on amino acids in the
floral nectar of corncockle, Agrostemma githago (Caryophyllaceae). Oikos, 92, 101–106.
is less variable than concentration. Journal of chemical ecology, 27, 2545–58.
Gardener, M.C. & Gillman, M.P. (2002) The Taste of Nectar : A Neglected Area of Pollination
Ecology. Oikos, 98, 552–557.
Garibaldi, L.A., Steffan-Dewenter, I., Winfree, R., Aizen, M.A., Bommarco, R., Cunningham,
S.A., Kremen, C., Carvalheiro, L.G., Harder, L.D., Afik, O., Bartomeus, I., Benjamin, F.,
Bireux, V., Cariveau, D., Chacoff, N.P., Dudenhöffer, J.H., Freitas, B.M., Ghazoul, J.,
Greenleaf, S., Hipolito, J., Holzschuc, A., Howlett, B., Isaacs, R., Javorek, S.K.,
Kennedy, C.M., Krewenka, K., Krishnan, S., Mandelik, Y., Mayifield, M.M., Motzke, I.,
Munyuli, T., Nault, B.A., Otieno, M., Petersen, J., Pisanty, G., Potts, S.G., Rader, R.,
Ricketts, T.H., Rundölf, Ma., Seymour, C.L., Schüepp, C., Szentgyörgyi, H., Taki, H.,
Tscharntke, T., Vergara, C.H., Viana, B.F., Wanger, T.C., Westphal, C., Williams, N. &
Klein, A.-M. (2013) Wind pollinators enhance fruit set of crops regardless of honey bee
abundance. Science, 339, 1608–1611.
Gause, G. (1934). The struggle for existence. 1st edn. Williams and Wilkins, Baltimore.
Georgelin, E. & Loeuille, N. (2014) Dynamics of coupled mutualistic and antagonistic
interactions, and their implications for ecosystem management. Journal of Theoretical
Biology, 346, 67–74.
Geslin B., Le Féon V., Kulhmann M., Vaissière B., & I. Dajoz. (2016). The bee fauna of large
parks in downtown Paris, France. Annales de la Société Entomologique de France. In
press.
Goldberg, D. E. (1990). Components of resource competition in plant communities. in
Bibliographie | 168
Perspectives on plant competition, 27-49.
Goldberg, D. E., & Barton, A. M. (1992). Patterns and consequences of interspecific
competition in natural communities: a review of field experiments with plants. American
naturalist, 771-801.
Goldberg, D. E., Rajaniemi, T., Gurevitch, J., & Stewart-Oaten, A.. (1999). Empirical
Approaches to Quantifying Interaction Intensity: Competition and Facilitation along
Productivity Gradients. Ecology, 80(4), 1118–1131
Grace, J. B. (1990). On the relationship between plant traits and competitive ability. in
Perspectives on plant competition, 51-65.
Greenleaf, S.S. & Kremen, C. (2006) Wild bees enhance honey beess ’ pollination of hybrid
sunflower. Proceedings of the National Academy of Sciences, 103, 13890–13895.
Grime, J. P. (1979). Primary strategies in plants. In Transactions of the Botanical Society of
Edinburgh (Vol. 43, No. 2, pp. 151-160). Taylor & Francis Group.
Grime, J. P. (1988). The CSR model of primary plant strategies—origins, implications and
tests. In Plant evolutionary biology (pp. 371-393). Springer Netherlands.
Grindeland, J.M., Sletvold, N. & Ims, R.A. (2005) Effects of floral display size and plant
density on pollinator visitation rate in a natural population of Digitalis purpurea.
Functional Ecology, 19, 383–390.
Gurevitch, J., Scheiner, S.M. & Fox, G.A. (2006) The Ecology of Plants, Second edi (ed SA
Inc.Publishers). Sunderland Massachusetts.
Hanley, T. C., La Pierre, K. J. 2015 Trophic ecology : bottom-up and top-down interactions in
aquatic and terrestrial systems cambridge : Cambrige University Press 410p
Hegland, S.J. & Totland, Ø. (2005) Relationships between species’ floral traits and pollinator
visitation in a temperate grassland. Oecologia, 145, 586–94.
Heil, M. (2011) Nectar: generation, regulation and ecological functions. Trends in plant
science, 16, 191–200.
Hermans, C., Hammond, J. P., White, P. J., & Verbruggen, N. (2006). How do plants respond
to nutrient shortage by biomass allocation?. Trends in plant science, 11(12), 610-617.
Bibliographie | 169
biogeochemistry and ecological relevance. Plant and Soil, 321, 117–152.
Hodge, A. (2004). The plastic plant: root responses to heterogeneous supplies of nutrients.
New phytologist, 162(1), 9-24.
Holt, R.H. (1977) Predation, apparent competition and structure of prey communities.
Theoretical population biology, 12, 197–229.
Hoover, S.E.R., Ladley, J.J., Shchepetkina, A.A., Tisch, M., Gieseg, S.P. & Tylianakis, J.M.
(2012) Warming, CO2, and nitrogen deposition interactively affect a plant-pollinator
mutualism. Ecology Letters, 15, 227–234.
Ings, T.C., Montoya, J.M., Bascompte, J., Blüthgen, N., Brown, L., Dormann, C.F., Edwards,
F., Figueroa, D., Jacob, U., Jones, J.I., Lauridsen, R.B., Ledger, M.E., Lewis, H.M.,
Olesen, J.M., van Veen, F.J.F., Warren, P.H. & Woodward, G. (2009) Review: Ecological
networks - beyond food webs. Journal of Animal Ecology, 78, 253–269.
Ishii, H.S., Hirabayashi, Y. & Kudo, G. (2008) Combined effects of inflorescence architecture,
display size, plant density and empty flowers on bumble bee behaviour: experimental
study with artificial inflorescences. Oecologia, 156, 341–350.
Iwata, T., Nagasaki, O., Ishii, H.S. & Ushimaru, A. (2012) Inflorescence architecture affects
pollinator behaviour and mating success in Spiranthes sinensis (Orchidaceae). New
Phytologist, 193, 196–203.
Jauker, F. & Wolters, V. (2008) Hover flies are efficient pollinators of oilseed rape. Oecologia,
156, 819–823.
Johnson, N.C., Graham, J.H. & Smith, F.A. (1997) Functioning of mycorrhizal associations
along the mutualism – parasitism continuum. New Phytologist, 135, 575–585.
Johnson, N.C. & Graham, J.H. (2013) The continuum concept remains a useful framework for
studying mycorrhizal functioning. Plant and Soil, 363, 411–419.
Kandori, I., Hirao, T., Matsunaga, S. & Kurosaki, T. (2009) An invasive dandelion unilaterally
reduces the reproduction of a native congener through competition for pollination.
Oecologia, 159, 559–69.
Kearns, C.A. & Inouye, D.W. (1993) Techniques for Pollination Biologists (ed University
Press of Colorado). Niwot Colorado.
Keasar, T., Sadeh, A., & Shmida, A. (2008). Variability in nectar production and standing
crop, and their relation to pollinator visits in a Mediterranean shrub. Arthropod-Plant
Interactions, 2(2), 117-123.
Kéfi, S., Berlow, E.L., Wieters, E.A., Navarrete, S.A., Petchey, O.L., Wood, S.A., Boit, A.,
Joppa, L.N., Lafferty, K.D., Williams, R.J., Martinez, N.D., Menge, B.A., Blanchette,
C.A., Iles, A.C. & Brose, U. (2012) More than a meal… integrating non-feeding
interactions into food webs. Ecology Letters, 15, 291–300.
Bibliographie | 170
Kiaer, L.P., Weisbach, A.N. & Weiner, J. (2013) Root and shoot competition: a meta-analysis
(ed D Gibson). Journal of Ecology, 101, 1298–1312.
Klein, A.-M., Steffan-Dewenter, I. & Tscharntke, T. (2003) Fruit set of highland coffee
increases with the diversity of pollinating bees. Proceedings. Biological sciences / The
Royal Society, 270, 955–61.
Knight, T.M., McCoy, M.W., Chase, J.M., McCoy, K.A. & Holt, R.D. (2005) Trophic
cascades across ecosystems. Nature, 437, 880–883.
de Kroon, H., Mommer, L. & Nishiwaki, A. (2003) Root Competition : Towards a mechanistic
understanding. Ecological studies, 168.
Krupnick, G. A., Weis, A. E., & Campbell, D. R. (1999). The consequences of floral herbivory
for pollinator service to Isomeris arborea. Ecology, 80(1), 125-134.
Kudo, G. & Harder, L.D. (2005) Floral and inflorescence effects on variation in pollen
removal and seed production among six legume species. Functional Ecology, 19, 245–
254.
Lafferty, K.D., Allesima, S., Arim, M., Briggs, C.J., De Leo, G., Dobson, A.P., Dunne, J.A.,
Johnson, P.T.J., Kuris, A.M., Marcogliese, D.J., Martinez, N.D., Memmott, J., Marquet,
Pablo, A., McLaughlin, J.P., Mordecai, E.A., Pascual, M., Poulin, R. & Thieltges, D.W.
(2008) Parasites in food webs : the ultimate missing links. Ecology letters, 11, 533–546.
Laird, R. A., & Addicott, J. F. (2007). Arbuscular mycorrhizal fungi reduce the construction of
extrafloral nectaries in Vicia faba. Oecologia, 152(3), 541-551.
Lavergne, S., Molina, J. & Debussche, M. (2006) Fingerprints of environmental change on
the rare mediterranean flora : a 115-year study. Global change biology, 12, 1466–1478.
Laverty, T.M. (1992) Plant interactions for pollinator visits: a test of the magnet species effect.
Oecologia, 89, 502–508.
Lau, T. C., & Stephenson, A. G. (1993). Effects of soil nitrogen on pollen production, pollen
grain size, and pollen performance in Cucurbita pepo (Cucurbitaceae). American
Journal of Botany, 763-768.
Lau, T. C., & Stephenson, A. G. (1994). Effects of soil phosphorus on pollen production,
pollen size, pollen phosphorus content, and the ability to sire seeds in Cucurbita pepo
(Cucurbitaceae). Sexual Plant Reproduction, 7(4), 215-220.
Lázaro, A., Lundgren, R., & Totland, Ø. (2009). Co‐flowering neighbors influence the
diversity and identity of pollinator groups visiting plant species. Oikos, 118(5), 691-702.
Leiss, K.A. & Klinkhamer, P.G.L. (2005) Spatial distribution of nectar production in a natural
Echium vulgare population: Implications for pollinator behaviour. Basic and Applied
Ecology, 6, 317–324.
Bibliographie | 171
Lever, J.J., Nes, E.H., Scheffer, M. & Bascompte, J. (2014) The sudden collapse of pollinator
communities. Ecology letters, 17, 350–359.
López-Bucio, J., Cruz-Ramı́rez, A., & Herrera-Estrella, L. (2003). The role of nutrient
availability in regulating root architecture. Current opinion in plant biology, 6(3), 280-
287.
Luo, E. Y., Ogilvie, J. E., & Thomson, J. D. (2014). Stimulation of flower nectar
replenishment by removal : a survey of eleven animal-pollinated plant species. Journal
of Pollination Ecology, 12(7), 52-62.
MacArthur, R. H. (1972). Geographical ecology: patterns in the distribution of species.
Princeton University Press.
Makino, T.T., Ohashi, K. & Sakai, S. (2007) How do floral display size and the density of
surrounding flowers influence the likelihood of bumble bee revisitation to a plant?
Functional Ecology, 21, 87–95.
Marschner, H. & Dell, B. (1994) Nutrient uptake in mycorrhizal symbiosis. Plant and Soil,
159, 89–102.
Marschner, H. (1995). Functions of mineral nutrients: macronutrients. Mineral nutrition of
higher plants 2nd Edition. Academic Press, NY, 299-312.
Massol, F., Jabot, F., Manel, S. & Munoz, F. (2015) Analyse des réseaux sociaux appliquée à
l’éthologie et à l'écologie. , 50.
Mayer, C., Adler, L., Armbruster, W., Dafni, a, Eardley, C., Huang, S.-Q., Kevan, P., Ollerton,
J., Packer, L., Ssymank, a, Stout, J. & Potts, S. (2011) Pollination ecology in the 21st
century: key questions for future research. Journal of Pollination Ecology, 3, 8–23.
Melián, C.J., Bascompte, J., Jordano, P. & Křivan, V. (2009) Diversity in a complex
ecological network with two interaction types. Oikos, 118, 122–130.
Michalet, R., Schöb, C., Xiao, S., Zhao, L., Chen, T., An, L. Z., & Callaway, R. M. (2015).
Beneficiary feedback effects on alpine cushion benefactors become more negative with
increasing cover of graminoids and in dry conditions. Functional Ecology.
Mitchell, R.J., Karron, J.D., Holmquist, K.G. & Bell, J.M. (2004) The influence of Mimulus
ringens floral display size on pollinator visitation patterns. Functional Ecology, 18, 116–
124.
Miyake, Y.C. & Sakai, S. (2005) Effects of number of flowers per raceme and number of
racemes per plant on bumblebee visits and female reproductive success in Salvia
nipponica (Labiatae). Ecological Research, 20, 395–403.
Bibliographie | 172
Molina-Montenegro, M. a., Badano, E.I. & Cavieres, L. a. (2008) Positive interactions among
plant species for pollinator service: assessing the “magnet species” concept with invasive
species. Oikos, 117, 1833–1839.
Moeller, D. A. (2004). Facilitative interactions among plants via shared pollinators. Ecology,
85(12), 3289-3301.
Møller, A.P. (1995) Bumblebee preference for symmetrical flowers. Proceedings of the
National Academy of Sciences of the United States of America, 92, 2288–2292.
Moora, M., & Zobel, M. (1996). Effect of arbuscular mycorrhiza on inter-and intraspecific
competition of two grassland species. Oecologia, 108(1), 79-84.
Mothershead, K., & Marquis, R. J. (2000). Fitness impacts of herbivory through indirect
effects on plant-pollinator interactions in Oenothera macrocarpa. Ecology, 81(1), 30-40.
Müller, I., Schmid, B., & Weiner, J. (2000). The effect of nutrient availability on biomass
allocation patterns in 27 species of herbaceous plants. Perspectives in plant ecology,
evolution and systematics, 3(2), 115-127.
Muñoz, A.A., Celedon-Neghme, C., Cavieres, L.A. & Arroyo, M.T.K. (2005) Bottom-up
effects of nutrient availability on flower production, pollinator visitation, and seed output
in a high-Andean shrub. Oecologia, 143, 126–35.
Muñoz, A.A. & Cavieres, L.A. (2008) The presence of a showy invasive plant disrupts
pollinator service and reproductive output in native alpine species only at high densities.
Journal of Ecology, 96, 459–467.
Nakano, S. & Murakami, M. (2001) Reciprocal subsidies : Dynamic interdependence between
terrestrial and aquatic food webs. Proceedings of the National Academy of Sciences, 98,
166–170.
Nötzold, R., Blossey, B. & Newton, E. (1997) The influence of below ground herbivory and
plant competition on growth and biomass allocation of purple loosestrife. Oecologia,
113, 82–93.
Obeso, J. R. (2002). The costs of reproduction in plants. New Phytologist, 155(3), 321-348.
Bibliographie | 173
Ollerton, J., Winfree, R. & Tarrant, S. (2011) How many flowering plants are pollinated by
animals? Oikos, 120, 321–326.
Ordano, M., & Ornelas, J. F. (2004). Generous-like flowers: nectar production in two
epiphytic bromeliads and a meta-analysis of removal effects. Oecologia, 140(3), 495-
505.
Paine, R.T. (1980) Food webs: linkage, interaction strength and community structure. The
Journal of animal ecology, 49, 666–685.
Partzsch, M. & Bachmann, U. (2011) Is Campanula glomerata threatened by competition
from expanding grasses? Results from a 5-year pot-experiment. Plant Ecology, 212,
251–261.
de Pascual-Teresa, S., & Sanchez-Ballesta, M. T. (2008). Anthocyanins: from plant to health.
Phytochemistry reviews, 7(2), 281-299.
Peñalver, E., Labandeira, C.C., Barron, E., Delclos, X., Nel, P., Nel, a., Tafforeau, P. &
Soriano, C. (2012) Thrips pollination of Mesozoic gymnosperms. Proceedings of the
National Academy of Sciences, 109, 8623–8628.
Petanidou, T., Goethals, V. & Smets, E. (1999) The effect of nutrient and water availability on
nectar secretion and nectary structure of the dominant Labiatae species of phrygana.
Systematics and Geography of Plants, 68, 233–244.
Petanidou, T., Van Laere, A., N Ellis, W., & Smets, E. (2006). What shapes amino acid and
sugar composition in Mediterranean floral nectars?. Oikos, 115(1), 155-169.
Pitelka, L. F. (1977). Energy allocations in annual and perennial Lupines (lupinus:
Leguminosae). Ecology, 1055-1065.
Pocock, M.J., Evans, D.M. & Memmott, J. (2012) The robustness and restoration of a
network of ecological networks. Science, 335, 973–977.
Potts, S.G., Vulliamy, B., Roberts, S., O’Toole, C., Dafni, A., Ne’eman, G. & Willmer, P.G.
(2004) Nectar resource diversity organises flower-visitor community structure.
Entomologia Experimentalis et Applicata, 113, 103–107.
Potts, S.G., Biesmeijer, J.C., Kremen, C., Neumann, P., Schweiger, O. & Kunin, W.E. (2010)
Global pollinator declines: trends, impacts and drivers. Trends in ecology & evolution,
25, 345–53.
Poveda, K., Steffan-Dewenter, I., Scheu, S. & Tscharntke, T. (2003) Effects of below- and
above-ground herbivores on plant growth, flower visitation and seed set. Oecologia, 135,
601–605.
Poveda, K., Steffan-Dewenter, I., Scheu, S. & Tscharntke, T. (2005) Effect of decomposers
and herbivores on plant performance and aboveground plant-insect interactions. Oikos,
108, 503–510.
Bibliographie | 174
Proctor, M., Yeo, P. & Lack, A. (1996) The Natural History of Pollination (ed Timber Press
Inc.). Portland, Oregon.
van der Putten, W.H., Vet, L.E.M., Harvey, J.A. & Wäckers, felix L. (2001) Linking above-
and below ground multitrophic interactions of plants, hervibores, pathogens and their
antagonists. Trends in ecology & evolution, 5347, 547–554.
van der Putten, W. H., Bardgett, R. D., De Ruiter, P. C., Hol, W. H. G., Meyer, K. M.,
Bezemer, T. M., ... & Hemerik, L. (2009). Empirical and theoretical challenges in
aboveground–belowground ecology. Oecologia, 161(1), 1-14.
Pyke, G. H. (1978). Optimal foraging in bumblebees and coevolution with their plants.
Oecologia, 36(3), 281-293.
Raynaud, X. & Leadley, P.W. (2004) Soil characteristics play a key role in modeling nutrient
competition in plant communities. Ecology, 85, 2200–2214.
Real, L. A. (1981). Nectar availability and bee-foraging on Ipomoea (Convolvulaceae).
Biotropica, 64-69.
Ren, D., Labandeira, C.C., Santiago-blay, J.A., Rasnitsyn, A., Shin, C., Bashkuev, A., Logan,
M.A. V, Hotton, C.L. & Dilcher, D. (2009) A probable pollination mode before
Angiosperms: Eurasian, long-proboscid Scorpionflies. Science, 326, 840–847.
Rickelfs, R.E. & Miller, G.L. (2005) Ecologie, 4e edition (ed D Boeck). Bruxelles.
Bibliographie | 175
Romeis, J. & Wäckers, F.L. (2000) Feeding responses by female Pieris brassicae butterflies
to carbohydrates and amino acids. Physiological Entomology, 25, 247–253.
Sauve, A.M.C., Fontaine, C. & Thébault, E. (2014) Structure-stability relationships in
networks combining mutualistic and antagonistic interactions. Oikos, 123, 378–384.
Schenk, H.J. (2006) Root competition: beyond resource depletion. Journal of Ecology, 94,
725–739.
Shumway, S. W. (2000). Facilitative effects of a sand dune shrub on species growing beneath
the shrub canopy. Oecologia, 124(1), 138-148.
Soper-Gorden, N.L. & Adler, L.S. (2013) Abiotic conditions affect floral antagonists and
mutualists of Impatiens capensis (Balsaminaceae). American Journal of Botany, 100,
679–689.
Southwick, E.E. (1984) Photosynthatse allocation to floral nectar : a neglected energy
investment. Ecology, 65, 1775–1779.
Steffan-Dewenter, I., Münzenberg, U., Bürger, C., Thies, C. & Tscharntke, T. (2002) Scale-
dependent effects of landscape context on three pollinator guilds. Ecology, 83, 1421–
1432.
Steffan-dewenter, I., Potts, S.G. & Packer, L. (2005) Pollinator diversity and crop pollination
services are at risk. Trends in ecology & evolution, 20, 651–652.
Strauss, S.Y. (1991) Indirect effects in community ecology: Their definition, study and
importance. Trends in Ecology and Evolution, 6, 206–210.
Tansley, A. G. (1917). On competition between Galium saxatile L.(G. hercynicum Weig.) and
Galium sylvestre Poll.(G. asperum Schreb.) on different types of soil. The Journal of
Ecology, 173-179.
Thébault, E. & Fontaine, C. (2010) Stability of ecological communities and the architecture of
mutualistic and trophic networks. Science, 329, 853–856.
Tilman, D. (1980). Resources: a graphical-mechanistic approach to competition and
predation. American Naturalist, 362-393.
Tilman, D., & Grace, J. B. (1990). Mechanisms of plant competition for nutrients: the
elements of a predictive theory of competition (pp. 117-141). Academic Press, Inc.
Tilman, D., & versus Complexity, V. A. (2012). Competition for Nutrients: The Elements of
Theory of Competition. Perspectives on Plant Competition, 117.
Tirard, C., Barbault, R., Abbadie, L. & Loeuille, N. (2012) Mini Manuel D’écologie (ed
Dunod). Paris.
Thomann, M., Imbert, E., Devaux, C., & Cheptou, P. O. (2013). Flowering plants under
global pollinator decline. Trends in plant science, 18(7), 353-359.
Bibliographie | 176
Toräng, P., Ehrlén, J., & Ågren, J. (2006). Facilitation in an insect-pollinated herb with a floral
display dimorphism. Ecology, 87(8), 2113-2117.
Trinder, C., Brooker, R., Davidson, H., & Robinson, D. (2012). Dynamic trajectories of
growth and nitrogen capture by competing plants. New Phytologist, 193(4), 948-958
Trinder, C.J., Brooker, R.W. & Robinson, D. (2013) Plant ecology’s guilty little secret:
Understanding the dynamics of plant competition. Functional Ecology, 27, 918–929.
Tylianakis, J.M. (2013) The Global Plight of Pollinators. Science, 339, 1532.
UICN, CBNBP & MNHN. (2012) Flore vasculaire de France métropolitaine : Premiers
résultats pour 1000 espèces, sous - espèces et variétés. [En ligne].
http://www.uicn.fr/liste-rouge- flore.html (consulté en janvier 2016).
Vanbergen, A.J. & the Insect Pollinator initiative. (2013) Threats to an ecosystem service:
pressures on pollinators. Frontiers in Ecology and the Environment.
van Veen, F.J.F., Müller, C.B., Pell, J.K. & Godfray, H.C.J. (2008) Food web structure of three
guilds of natural enemies : predators , parasitoids and pathogens of aphids. Journal of
Animal Ecology, 77, 191–200.
Violle, C., Garnier, E., Lecoeur, J., Roumet, C., Podeur, C., Blanchard, A., & Navas, M. L.
(2009). Competition, traits and resource depletion in plant communities. Oecologia,
160(4), 747-755.
Volterra, V. (1926). Fluctuations in the abundance of a species considered mathematically.
Nature, 118, 558560.
van der Wal, R., Egas, M., van der Veen, A. & Bakker, J. (2000) Effects of resource
competition and herbivory on plant performance along a natural productivity gradient.
Journal of ecology, 88, 317–330.
Waddington, K. D., & Heinrich, B. (1979). The foraging movements of bumblebees on
vertical “inflorescences”: an experimental analysis. Journal of Comparative Physiology,
134(2), 113-117.
Walsh, M.R. (2013) The evolutionary consequences of indirect effects. Trends in Ecology and
Evolution, 28, 23–29.
Wardle, D. a, Bardgett, R.D., Klironomos, J.N., Setälä, H., van der Putten, W.H. & Wall, D.H.
(2004) Ecological linkages between aboveground and belowground biota. Science, 304,
1629–33.
Waser, N.M., Chittka, L., Price, M. V, Williams, N.M. & Ollerton, J. (1996) Generalization in
pollination systems, and why it matters. Ecology, 77, 1043–1060.
Weigelt, A. & Jolliffe, P. (2003) Indices of plant competition. Journal of Ecology, 91, 707–
720.
Bibliographie | 177
Weiner, J. (1988) The influence of competition on plant reproduction. Plant reproductive
ecology: patterns and strategies, pp. 228–245.
Weiss, M.R. (1997) Innate colour preferences and flexible colour learning in the pipevine
swallowtail. Animal Behaviour, 53, 1043–1052.
Wilson, S.D. & Tilman, D. (1993) Plant competition and resource availability in response to
disturbance and fertilization. , 599–611.
Winfree, R., Williams, N.M., Dushoff, J. & Kremen, C. (2007) Native bees provide insurance
against ongoing honey bee losses. Ecology Letters, 10, 1105–1113.
Xi, X., Griffin, J.N. & Sun, S. (2013) Grasshoppers amensalistically suppress caterpillar
performance and enhance plant biomass in an alpine meadow. Oikos, 122, 1049–1057.
Zimmerman, M. (1981) Patchiness in the dispersion of nectar resources: probable causes.
Oecologia, 49, 154–157.
REFERENCES Tableau 1
Boose, D. L. (1997). Sources of variation in floral nectar production rate in Epilobium canum
(Onagraceae): implications for natural selection. Oecologia, 110(4), 493-500.
Burkle, L. A., & Irwin, R. E. (2009). The effects of nutrient addition on floral characters and
pollination in two subalpine plants, Ipomopsis aggregata and Linum lewisii. Plant
Ecology, 203(1), 83-98.
Campbell, D. R., & Halama, K. J. (1993). Resource and pollen limitations to lifetime seed
production in a natural plant population. Ecology, 1043-1051.
Crone, E. E., & Lesica, P. (2006). Pollen and water limitation in Astragalus scaphoides, a
plant that flowers in alternate years. Oecologia, 150(1), 40-49.
Juenger, T., & Bergelson, J. (1997). Pollen and resource limitation of compensation to
herbivory in scarlet gilia, Ipomopsis aggregata. Ecology, 78(6), 1684-1695.
Lau, T. C., & Stephenson, A. G. (1993). Effects of soil nitrogen on pollen production, pollen
grain size, and pollen performance in Cucurbita pepo (Cucurbitaceae). American Journal
of Botany, 763-768.
Leiss, K. A., & Klinkhamer, P. G. L. (2005). Genotype by environment interactions in the
nectar production of Echium vulgare. Functional Ecology, 19(3), 454-459.
Muñoz, A. A., Celedon-Neghme, C., Cavieres, L. A., & Arroyo, M. T. (2005). Bottom-up
effects of nutrient availability on flower production, pollinator visitation, and seed output
in a high-Andean shrub. Oecologia, 143(1), 126-135.
Bibliographie | 178
Wyatt, R., Broyles, S. B., & Derda, G. S. (1992). Environmental influences on nectar
production in milkweeds (Asclepias syriaca and A. exaltata). American Journal of
Botany, 636-642.
Zimmerman, M. (1983). Plant reproduction and optimal foraging: experimental nectar
manipulations in Delphinium nelsonii. Oikos, 57-63.
Zimmerman, M., & Pyke, G. H. (1988). Reproduction in Polemonium: assessing the factors
limiting seed set. American Naturalist, 723-738.
PHOTOGRAPHIES
• Page 20
Bibliographie | 179
%29.jpg
http://creativecommons.org/licenses/by-sa/3.0/
p20 Trichonympha campanula : Tai V, James ER, Perlman SJ, Keeling PJ - Tai V, James ER,
Perlman SJ, Keeling PJ (2013). "Single-cell DNA barcoding using sequences from the small
subunit rRNA and internal transcribed spacer region identifies new species of Trichonympha
and Trichomitopsis from the hindgut of the termite Zootermopsis angusticollis". PLoS ONE 8
(3): e58728. (avec modifications)
http://creativecommons.org/licenses/by/2.5/
p43, 54, 75 et 119 Agrostis sp. (tiré de: Icones et descriptiones Graminum austriacorum de
Nikolai T. Host (1801); avec modifications)
Public domain : https://commons.wikimedia.org/wiki/File:Apera_spica-venti.jpg
p20, 43, 54, 75 et 119 Echium vulgare. (tiré de : Dictionnaire des plantes suisses (1853);
avec modifications)
Public domain : https https://commons.wikimedia.org/wiki/File:Echium_vulgare_L_ag1.jpg
Bibliographie | 180
https://creativecommons.org/licenses/by-nc-sa/2.0/
P20, 43 et 119 Sinapis alba. (tiré de: Köhler's Medizinal-Pflanzen de Franz Eugen Köhler
(1897) ; avec modifications)
Public domain : https://commons.wikimedia.org/wiki/File:Sinapis_alba_-
_K%C3%B6hler%E2%80%93s_Medizinal-Pflanzen-265.jpg
Bibliographie | 181
- Résumé -
Influence des interactions entre espèces végétales sur les relations plantes-pollinisateurs: cas de la
compétition induite par la présence d'espèces anémophiles sur l'attractivité aux pollinisateurs d'espèces
entomophiles
Afin d'attirer les pollinisateurs, les espèces végétales entomophiles produisent des fleurs et des récompenses
associées comme le nectar et le pollen. Plusieurs études ont démontré que la production de ces traits d'attractivité
est sensible aux variations de ressources dans le milieu (azote, phosphore). Puisque la compétition entre plantes
peut modifier la disponibilité et la quantité de ressources dans le milieu, nous nous sommes intéressés à son effet
sur les traits d'attractivité aux pollinisateurs.Dans cette thèse nous nous sommes concentrés sur la compétition
entre des espèces entomophiles et anémophiles, ces dernières étant rarement intégrées aux études des réseaux
plantes-pollinisateurs. Après une première expérience en serre, nous avons démontré que la présence d'une espèce
anémophile compétitive réduisait la production de fleurs et de nectar chez une espèce entomophile. Cette
diminution est d'autant plus forte que l'espèce anémophile est compétitive. Afin de savoir si ces modifications
d’attractivité potentielle pouvaient influencer l'attractivité réelle d'une plante entomophile, nous avons réalisé deux
expériences en conditions contrôlées, en serre et au champ. Au cours de ces expériences, nous avons étudié le
comportement de visite de pollinisateurs sur des espèces entomophiles en compétition avec une espèce
anémophile. Ces deux études suggèrent un rôle majeur de la compétition sur la production de fleur et de la vitrine
florale dans l'attraction des pollinisateurs. Plus précisément, nous avons observé dans l'expérience en champs une
diminution du nombre total de visites des pollinisateurs sauvages, lorsque les plantes étaient en compétition, en
relation avec une diminution de la production de fleurs et de la vitrine florale. Concernant la production de nectar,
nous n’avons observé aucun effet au niveau de la fleur, suggérant que les espèces entomophiles pouvaient
favoriser la production de récompenses au détriment de la production de fleurs. Néanmoins, le nombre de fleurs
étant réduit par la compétition, cette dernière semble avoir un effet sur la production de nectar à l’échelle de la
plante. Enfin, la modification du nombre de visites de pollinisateurs, via cette compétition induite par la présence
d'une espèce anémophiles, pourrait affecter le succès reproducteur des espèces visitées. En effet, chez une des
espèces entomophiles étudiées, nous avons pu noter une diminution du taux de fécondité en conditions de
compétition. Par conséquent, bien que n'interagissant pas directement avec les pollinisateurs, les espèces
anémophiles peuvent moduler les interactions plantes-pollinisateurs via des relations de compétition. Ainsi, cette
thèse soulève de nouvelles perspectives pour l'étude des réseaux plantes-pollinisateurs, qui se focalisent
généralement sur les plantes entomophiles, en les ouvrants à l'ensemble de la communauté végétale.
- Abstract -
Influence of plant interactions on insect-pollinated species and their attractiveness to pollinators: the case of
competition induced by wind-pollinated plant species
In order to attract pollinators, insect-pollinated plants display several traits such as flowers, and associated
rewards (i.e. pollen and nectar). Several studies have demonstrated that these traits can be sensitive to variation of
resources (e.g. nitrogen, phosphorous). As competition between plants can modify resource availability, it also has
the potential to influence floral traits involved in attractiveness to pollinators.In this thesis, we focused on
competitive interactions between insect-pollinated and wind-pollinated plants as the latter are almost never taken
into account in plant-pollinator networks. Therefore, in a first pot experiment in a greenhouse, we showed that the
presence of a competitive wind-pollinated species could indeed alter flower and nectar production. Especially, the
stronger the competitor, the stronger is the effect on floral traits. To test if such modifications of potential
attractiveness could have an influence on actual attractiveness we set up two more experiments, one in a
greenhouse experiment and one outdoor, in which we studied pollinator visits on insect-pollinated plants
competing with wind-pollinated plants for soil resources. For pollinator visits, these two experiments suggest an
important role of competition on pollinator attraction through modification of flower production. Especially in the
outdoor experiment, there was a decrease in the total number of pollinator visits in presence of belowground
competition, in relation to a decrease in flower production and floral display size. Concerning nectar production,
we found no influence of competition at the flower level, meaning that insect-pollinated plants might favour
reward production to promote pollinator revisitation at the expense of flower production. However, the number of
flowers being reduced by competition, the latter seems to have an effect on the production of nectar at the plant
level. Finally the modification of pollinator visits through competition with a wind-pollinated plant might have an
influence on insect-pollinated plant reproductive success as we observed a decrease of the fecundity ratio of the
studied insect pollinated plant species. Therefore, even though they do not interact directly with pollinators, wind-
pollinated plant species can modulate plant-pollinator interactions, through competitive interactions. This thesis
thus raises new perspectives for the study of plant-pollinator networks, which are generally focused on insect-
pollinated plants, by opening them up to the whole plant community.