<p dir="ltr">Traditional methods of behavior classification on videos of mice often rely on manually annotated datasets, which can be labor-intensive and resource-demanding to create. This research aims to address the challenges of behavior classification in mouse studies by leveraging an algorithmic framework employing self-supervised learning techniques capable of analyzing unlabeled datasets. This research seeks to develop a novel approach that eliminates the need for extensive manual annotation, making behavioral analysis more accessible and cost-effective for researchers, especially those in laboratories with limited access to annotated datasets.</p>

<p>Home cage aggression in male laboratory mice continues to challenge preclinical researchers. It reduces animal welfare and can alter research parameters, potentially reducing the validity and reliability of study data. While simply reducing aggression would be beneficial, promoting socio-positive, affiliative behaviors would greatly improve mouse welfare as mice are a social species. Mice also use olfaction to communicate, so this sensory modality could be used as a tool to improve social interactions in the home cage. A scoping review of the literature on how mammalian odor signals impact same sex social behavior found that studies are dominated by rodent subjects, treatments from urine, and aggression measures (Chapter 1). As a whole, urine treatments had a variable effect on aggression. This review highlights that treatments from non-urinary sources are not often tested, and affiliative behavior is rarely measured.</p> <p>One murine odor source worth exploring is found in used nesting material. Mice build complex nests for insulation and it has been speculated that the nest holds odor signals that appease home cage aggression, particularly aggression triggered by cage cleaning. It has been suggested that the nest contains secretions from plantar sweat glands, but the chemical content of neither nesting material nor plantar sweat have been examined. The main goals of this dissertation are to identify the odors stored in used nesting material, determine the sources of those odors, and test them for a behavioral role.</p> <p>Samples of used nesting material were collected from cages of group housed male mice. Further, plantar sweat, saliva, and urine were collected from the dominant and subordinate mouse in each cage as plausible odor sources. All samples were analyzed for protein and volatile organic compound content. Home cage aggression and affiliative behavior were also recorded to compare to odor profiles. Protein profiles showed that used nesting material contains a variety of proteins that primarily originate from plantar sweat, saliva, and urine sources (Chapter 2). A large proportion of these proteins contain messages about individual identity and bind volatile compounds that further contribute to identity cues. This suggests that the nest aids in maintaining a familiar odor environment. Analysis of volatile content showed that small compounds in the nest are also traced back to plantar sweat, saliva, and urine sources (Chapter 3). Few of the compounds have a known behavior role. However, one compound detected in nest, sweat, and saliva samples had a negative correlation with home cage aggression and three compounds (two from sweat and one from urine) had a positive correlation with affiliative behaviors, making them potential candidates for controlled studies on social behavior.</p> <p>Before testing the four candidate compounds, a challenge from the correlation study needed to be addressed. Body fluid samples were collected from individual mice based on social status, as this factor impacts production of known murine pheromones. Further, aggression is typically directed from a dominant to a subordinate mouse for territorial reasons. An aggression appeasement signal is likely to be produced by a subordinate to mitigate the dominant mouse’s perceived threat. Data from the correlation study showed no odor profile differences based on social status, and the pheromones that are known to vary with social status did not differ between dominant and subordinate mice. Therefore, Chapter 4 assesses the convergent validity of several dominance measures. Over one week, home cage interactions were observed in group housed male mice. For every aggression occurrence, the aggressor and target mouse was recorded to calculate individual dominance rankings in each cage. Then, individual mice were evaluated for the following measures known to correlate with dominance: levels of urinary darcin (a murine pheromone); scores from three rounds of the tube test; and ratio of preputial gland weight to body length. Postmortem wounding was also compared. Results showed that urinary darcin and preputial gland ratio have strong convergent validity with dominance ranking based on home cage aggression.</p> <p>Finally, the four candidate compounds (identified in Chapter 3) were developed into treatment solutions to assess their effect on home cage social behavior (Chapter 5). Cages of group housed male mice were randomly assigned one of five treatments (four compounds + control) and home cage aggression and affiliative behavior were recorded for one week. Postmortem wounding was recorded as a secondary aggression measure and social stress was measured through fecal corticosterone metabolites from each cage’s dominant and subordinate mouse (rank based on preputial gland ratio). Treatment did not predict changes in most measures. This may be due to limitations in application or from the original correlation study, which are further discussed.</p> <p>Although the final study showed null results, future research is still warranted to fine tune application methods and gain a better understanding of how odor signals impact interactions other than aggression. The relationship between olfaction and affiliative behaviors is largely unexamined and this dissertation is a first step in filling that gap.</p>

Bumble bees (bombus spp.) are significant pollinators of many plants, and are particularly attracted to mass-flowering crops such as Oilseed Rape (Brassica Napus), which they cross-pollinate. B. napus is both wind and insect-pollinated, and whilst it has been found that wind is its most significant pollen vector, the influence of bumble bee pollination could be non-trivial when bee densities are large. Therefore, the assessment of pollinator-mediated cross-pollination events could be important when considering containment strategies of genetically modified (GM) crops, such as GM varieties of B. napus, but requires a landscape-scale understanding of pollinator movements, which is currently unknown for bumble bees. I developed an in silico model, entitled HARVEST, which simulates the foraging and consequential inter-patch movements of bumble bees. The model is based on principles from Reinforcement Learning and Individual Based Modelling, and uses a Linear Operator Learning Rule to guide agent learning. The model incoproates one or more agents, or bees, that learn by ‘trial-and-error’, with a gradual preference shown for patch choice actions that provide increased rewards. To validate the model, I verified its ability to replicate certain iconic patterns of bee-mediated gene flow, and assessed its accuracy in predicting the flower visits and inter-patch movement frequencies of real bees in a small-scale system. The model successfully replicated the iconic patterns, but failed to accurately predict outputs from the real system. It did, however, qualitatively replicate the high levels of inter-patch traffic found in the real small-scale system, and its quantitative discrepancies could likely be explained by inaccurate parameterisations. I also found that HARVEST bees are extremely efficient foragers, which agrees with evidence of powerful learning capabilities and risk-aversion in real bumble bees. When applying the model to the landscape-scale, HARVEST predicts that overall levels of bee-mediated gene flow are extremely low. Nonetheless, I identified an effective containment strategy in which a ‘shield’ comprised of sacrificed crops is placed between GM and conventional crop populations. This strategy could be useful for scenarios in which the tolerance for GM seed set is exceptionally low.

Decision making is a framework we impose on a vast universe of possible behaviors to better understand the advantages and disadvantages of acting in different ways. Decisions under uncertainty are of particular interest because stochasticity is a feature of environments both today and throughout evolutionary history. As a result, we might expect decision makers (DMs) to have evolved mechanisms to handle variability. In this dissertation, I examine common decision processes in several model species: European starlings (Sturnus vulgaris), Banded tetras (Astyanax fasciatus) and humans (Homo sapiens). The broad range of approaches discussed include currencies DMs are expected to maximize (Risk Sensitivity Theory, Expected Utility), the currencies DMs do maximize (e.g. long versus short term rate maximizing models), the representation of outcomes in memory (Scalar Expectancy Theory) as well as explicit choice mechanisms (Sequential Choice Model). The first section of this thesis discusses responses to risk, offering humans and starlings choices between options which deliver certain or variable outcomes. Starlings demonstrate sensitivity to changes in the probability of variable outcomes and strong support for local rate maximization. Humans appear similarly sensitive to their own accuracy when task difficulty is varied. When the DM’s affective state was manipulated, neither humans nor starlings exhibit changes in risk preferences, but the effectiveness of these manipulations used could not be confirmed. Another topic of inquiry is the effect of the DM’s state at the time of valuation learning. State dependent valuation learning is demonstrated for the first time in a fish species, but Within Trial Contrast is not observed in starlings. Lastly, two experiments find strong support for the Sequential Choice Model, a promising new model of the mechanism of choice. Taken together, these experiments offer a glimpse into shared decision processes, but leave open questions about the mechanisms through which value is acquired.

The aim of this thesis was to investigate contrasting pasture management strategies on the foraging behaviour, pasture production, carcass characteristics, herbage intake, faecal gas emissions and methane emission by sheep grazing Italian ryegrass (Lolium multiflorum) pastures. The experiment was carried out in 2014 and 2015. The experimental design was a randomized complete block with four replicates and two grazing strategy treatments (traditional rotational stocking method – RT meaning pre- and post-grazing target heights of 25 and 5 cm, respectively and, “rotatinuous stocking” – RN with pre- and post-grazing target heights of 18 and 11 cm, respectively). The grazing management was based on a 1-day strip-grazing regime. Male castrated sheep were used. The actual average sward heights for the RN treatment were 17.9 and 11.1 cm (pre- and post-grazing, respectively) and 27.1 and 7.8 cm for the RT (pre- and post-grazing, respectively). The stocking period in 2014 was 146 and 140 days (RN and RT, respectively) and in 2015 it was 155 and 146 days (RN and RT, respectively). The diurnal animal activities (grazing, ruminating and resting time) did not differ between treatments, with average of 439.6, 166.9 and 85.0 minutes, respectively. The bite rate, feeding station per min and steps per min were greater at the RN than the RT treatment. Grazing time and bite rate were greater in the afternoon than morning in both treatments. Therefore, the daily herbage intake by sheep grazing Italian ryegrass was greater for the RN than the RT treatment (CHAPTER II). The herbage production in the RN was 28% higher than the RT treatment. Individual average daily gain and live weight gain per hectare were greater in the RN than the RT treatment (CHAPTER III). RN treatment presented greater final live weight, carcass and commercial cut weights from grazing sheep than RT treatment (CHAPTER IV). RN treatment had a faecal chemical quality greater than RT treatment, resulting in a greater daily nitrogen excretion per animal and greater faecal CH4 and N2O emissions (CHAPTER VI). The “rotatinuous stocking” (RN) was the better grazing management strategy for mitigation of CH4 emissions by sheep grazing Italian ryegrass, emitting 64% less CH4 per unit area and 170% less CH4 per unit of animal product than the traditional rotational stocking method (RT) (CHAPTER V).

Este estudo teve como objetivo avaliar a influência de recursos de climatização, ventilação e nebulização sobre a fisiologia e o comportamento de vacas Holandesas alojadas em free-stall, durante o verão do sudeste brasileiro. Foram utilizadas 20 vacas Holandesas lactantes em duas fases experimentais. No experimento I, foram avaliados os efeitos do tratamento ventilação e nebulização e, no experimento II, analisou-se a influência da ventilação isolada. Os parâmetros ambientais registrados foram temperatura do ar, umidade relativa, temperatura de globo negro e pluviosidade, e posteriormente, foi calculado o índice do globo negro e umidade. Os parâmetros fisiológicos avaliados foram temperatura retal, temperatura caudal e freqüência respiratória. Os parâmetros comportamentais observados foram atividade, postura, posição e localização dos animais dentro da instalação. Para obtenção das estatísticas descritivas para as variáveis fisiológicas e comportamentais foi utilizada metodologia de quadrados mínimos. As variáveis fisiológicas para ambas fases experimentais não apresentaram diferenças e indicaram conforto térmico. As respostas comportamentais indicaram que o estudo do comportamento é uma ferramenta importante na avaliação do bem-estar dos animais. O sistema de climatização, com ventilação e nebulização, associadas ou não, podem proporcionar um ambiente de maior conforto, determinando alterações comportamentais. / This studys main goal was to evaluate the influence of resources of climatization, fun and mist, on the physiology and the behaviour of Holstein cows housed in free-stall, during Brazilian southeast summer. Was used twenty Holstein milking cows in two experiments. In Trial one the effects of fun and mist were evaluated and in Trial II the influence of fun was analyzed. The environmental parameters registered were air temperature, relative humidity, black globe temperature and rain precipitation, and subsequently was computed the black globe temperature and humidity index. The physiological parameters evaluated were rectal temperature, tail basis temperature and respiratory rate. The behavioral parameters observed were activity, posture and animal position inside of stall. The obtainment of descriptive statistical for the physiological and behavioral variables was used the minimum square method. There were no differences in the physiological variables for both phases indicating thermal comfort. The behavioral responses indicated that the behaviour study is an important tool to evaluate the animal welfare. The climatization system with fun, with or without mist, can to promote an environment with more comfort, determining behavioral changes.

Chez les pollinisateurs sociaux, comme l'abeille domestique (Apis mellifera L.) et le bourdon terrestre (Bombus terrestris L.), mes deux modèles d'étude, différents individus sont spécialisés dans différentes tâches. Il est admis que différents types de comportement de butinage contribuent à une optimisation des performances de la colonie. Actuellement, les populations de pollinisateurs sont exposées à des stress environnementaux, qui sont connus pour perturber le comportement des individus en visant directement leur cognition. Il est ainsi crucial de mieux comprendre comment les colonies d'abeilles et de bourdons maintiennent une activité de butinage efficace, et quels sont les effets de stress environnementaux sur les butineuses. Dans cette thèse, j'ai donc examiné les différentes stratégies de butinage pour différentes sources de nourriture, pollen et nectar, et les variabilités interindividuelles dans le comportement de butinage. Je me suis aussi intéressé à l'impact de stress tels que les pesticides sur l'efficacité de butinage. J'ai utilisé la technologie RFID pour suivre le comportement des abeilles tout au long de leur vie. J'ai trouvé que les colonies d'abeilles et de bourdons reposent sur un petit groupe d'individus très actifs qui fournissent la majorité de la nourriture pour la colonie. Chez les abeilles, ces individus très actifs sont aussi plus efficaces pour collecter nectar et pollen. J'ai aussi identifié l'existence de différentes stratégies pour la collecte de pollen ou de nectar. Ensuite, j'ai pu montrer que les bourdons ont des différences interindividuelles très marquées dans un test de navigation, une tâche cruciale dans le comportement de butinage. Finalement, j'ai testé l'effet néfaste de pesticides sur l'apprentissage visuel chez l'abeille. Cette thèse a pour but de mieux comprendre les causes de vulnérabilité des pollinisateurs aux stress environnementaux. Mes résultats soulignent le besoin de considérer la diversité comportementale comme une adaptation des espèces de pollinisateurs sociaux, mais aussi comme une potentielle cause de vulnérabilité de la colonie vis-à-vis des stress. / In social insects, such as bees, different individuals specialise in the collection of different resources, and it is assumed that natural behavioural variability among foragers contributes to a self-organised optimisation of colony performance. Currently, bee populations are facing an increasing number of environmental stressors, known to disturb the behaviour of individuals, presumably upon their impact on cognitive capacities. Hence it is important to learn more about how stressors impact on individual foraging behaviour to understand how a colony maintains effective nutrition and development. In this thesis in cognitive ecology, I examined the different foraging strategies for the different macronutrient sources, pollen and nectar, and the inter-individual variation in bee foraging performance. I also looked at how stressors, such as pesticides, can impact on bee foraging efficiency. I compared two social Hymenoptera that vary in their level of social complexity: the European honey bee (Apis mellifera L.) and the buffed-tailed bumblebee (Bombus terrestris L.). I used Radio Frequency Identification (RFID) to automatically track the foraging behaviour of bees throughout their life. I found that honey bee and bumblebee colonies rely on a subset of very active bees to supply the whole colony needs. In honey bees, these foragers are more efficient and collect more pollen. I also identified different strategies for pollen or nectar collection in both species. Using manipulative experiments, I then showed that bees exhibit consistent inter-individual different behaviours in a spatial learning task and that pesticides impair visual learning. My thesis aims at better explaining the causes of vulnerability of pollinators to sublethal pesticides and other environmental stressors. The results highlight the need for considering behavioural diversity as an adaptation for social insects, as well as a potential dimension of colony-level vulnerability to environmental stressors that can impair the whole colony nutritional balance.

Les bancs de poissons sont des entités pouvant regrouper plusieurs milliers d'individus qui se déplacent de façon synchronisée, dans un environnement sujet à de multiples perturbations, qu'elles soient endogènes (e.g. le départ soudain d'un congénère) ou exogènes (e.g. l'attaque d'un prédateur). La coordination de ces bancs de poissons, décentralisée, n'est pas encore totalement comprise. Si les mécanismes sous-jacents aux interactions sociales proposés dans des travaux précédents reproduisent qualitativement les structures collectives observées dans la nature, la quantification de ces interactions et l'accord quantitatif entre ces mesures individuelles et les motifs collectifs sont encore rares dans les recherches récentes et forment l'objet principal de cette thèse. L'approche de ce travail repose sur une étroite combinaison entre les méthodes expérimentales et de modélisation dans l'objectif de découvrir les liens entre les comportements individuels et les structures observées à l'échelle collective. Nous avons caractérisé et quantifié les interactions et mécanismes à l'origine, d'abord, de la coordination des individus dans les bancs de poissons et, ensuite, de la propagation d'information, quand le groupe subit une perturbation endogène ou exogène. Ces travaux, tous réalisés en étudiant la même espèce de poisson d'eau douce, le nez-rouge (Hemigrammus rhodostomus), ont mobilisé une diversité de méthodes expérimentales, d'analyses statistique et de modélisation, à l'interface de l'éthologie, de la physique statistique et des sciences computationnelles. / Fish schools are systems in which thousands of individuals can move in a synchronised manner in a changing environment, with endogenous perturbations (e.g. when a congener leaves the group) or exogenous (e.g. the attack of a predator). The coordination of fish schools, decentralised, is not completely understood yet. If the mechanisms underlying social interactions discussed in previous studies qualitatively match the social patterns observed in nature, the quantification of these interactions and the quantitative match between individual measurements and collective patterns are still sparse in recent works and are the main focus of this thesis. This work combines closely experimental and modelling methods in order to investigate the links between the individual behaviours and the patterns observed at the collective scale. We have characterised and quantified the interactions and mechanisms at the origin of, first, the coordination of individuals in fish schools and, second, the propagation of information, when the group is under endogenous or exogenous perturbations. This thesis focuses on one freshwater fish species, the rummy-nose tetra (Hemigrammus rhodostomus), and is the result of a diversity of experimental methods, statistical analyses and modelling, at the interface of ethology, statistical physics and computational sciences.

Dominance rank often determines the share of reproduction an individual male can secure in group-living animals (i.e. dominance rank-based reproductive skew). However, our knowledge of the interplay between individual and social factors in determining rank trajectories of males is still limited. The overall aim of this thesis was therefore to investigate mechanisms that underlie individual dominance rank trajectories in male crested macaques (Macaca nigra) and to highlight potential individual and social determinants of how males can achieve and maintain the highest rank possible. Data for this thesis were collected on 37 males during a field study on a natural population of crested macaques living in the Tangkoko-Batuangus Nature Reserve in Indonesia. In study 1, I validate Elo-rating as a particularly well suited method to quantify dominance hierarchies in animal species with dynamic dominance relationships. In studies 2 and 3, I suggest a personality structure for crested macaque males consisting of five distinct factors and further demonstrate that two personality factors determine whether males will rise or fall in rank. Finally, in study 4, I present results on how males utilize coalitions to increase their future rank. Together, these results shed light on how individual attributes and social environment both can impact male careers. Ultimately, in order to understand what determines rank-based reproductive skew, we need to consider the complexity and likely diversity of the mechanisms underlying rank trajectories of individual males which are likely to differ across different species.

Práticas de manejo afetam a estrutura do dossel forrageiro, podendo afetar os padrões de deslocamento, procura e ingestão de forragem pelos animais em pastejo. O objetivo deste trabalho foi avaliar o efeito de estratégias de pastejo rotacionado sobre o comportamento animal e padrões de ingestão de forragem de bovinos de corte em pastos de Brachiaria brizantha cv. Marandu. O experimento foi realizado no Departamento de Zootecnia da USP/ESALQ, de novembro de 2005 a abril de 2006. Os tratamentos experimentais compreenderam a combinação entre duas intensidades (altura pós-pastejo de 10 e 15 cm) e dois intervalos entre pastejos (período de tempo necessário para se atingir 95 e 100% de interceptação luminosa pelo dossel durante a rebrotação - IL), e foram alocados às unidades experimentais (piquetes de 1.200 m²) segundo um delineamento inteiramente casualizado e arranjo fatorial 2 x 2, com 3 repetições. As avaliações foram realizadas de forma repetida em duas épocas do ano: (1) primavera (novembro e dezembro de 2005), e (2) verão (janeiro a abril de 2006). A densidade de lotação foi dimensionada para que a duração do período de ocupação fosse de 10 a 12 horas. Foram avaliadas as seguintes variáveis-resposta: distribuição vertical dos componentes morfológicos da massa de forragem, dinâmica do rebaixamento dos pastos, padrões de ingestão (massa do bocado, taxa de bocados e taxa de consumo), comportamento (pastejo, ruminação e outras atividades), composição morfológica da forragem consumida; padrões de deslocamento e busca por alimento (número de estações alimentares por unidade de tempo e número de passos entre estações alimentares), e padrões de desfolhação de perfilhos individuais (profundidade de desfolhação e freqüência e intensidade de desfolhação de folhas em geral e nas categorias folhas em expansão e expandidas). As maiores taxas de rebaixamento foram registradas para os tratamentos 100/10 e 100/15, enquanto as menores para o tratamento 95/15. De modo geral, à medida que os animais permaneceram nos piquetes e ocorria depleção do estrato pastejável, a taxa de rebaixamento diminuiu em virtude da diminuição da altura do dossel e do aumento da presença de colmos e de material morto no horizonte de pastejo. Apesar de a intensidade de pastejo menos severa (altura pós-pastejo de 15 cm) resultar, no início do pastejo, em menor bocado, taxa de consumo e intensidade de desfolhação de folhas em expansão e expandidas, os animais acabaram consumindo uma forragem com maior proporção de folhas. Ao longo do rebaixamento, os tratamentos de 100% IL resultaram em bocados maiores, mas, no entanto, a taxa de consumo e a proporção de folhas na forragem consumida foram menores em relação aos tratamentos de 95% de IL. Os tratamentos 95/10 e 95/15 estiveram associados com maior atividade de pastejo e maior taxa de consumo, sugerindo maior consumo diário. Entretanto, o tratamento 95/15 foi o que resultou na maior proporção de folhas na forragem consumida, e foi o tratamento em que a intensidade e freqüência de desfolhação de folhas foram mais baixas, condicionadas pelas menores densidades de lotação empregadas. Esse padrão de desfolhação resultou em uma maior área foliar remanescente, favorecendo a rebrota seguinte e o rápido retorno dos pastos à utilização. A estratégia de pastejo que resultou em melhor utilização da forragem produzida, foi aquela em que os pastejos foram realizados com 95% de IL até uma altura pós-pastejo de 15 cm. / Management practices affect sward structure and may affect patterns of animal movement, search and forage ingestion during grazing. The objective of this study was to evaluate the effect of rotational grazing practices on animal behaviour and patterns of forage ingestion of beef cattle on Brachiaria brizantha cv. Marandu. The experiment was carried out at Departamento de Zootecnia, USP/ESALQ, from November 2005 to April 2006. Treatments corresponded to combinations between two grazing intensities (post-grazing residues of 10 and 15 cm) and two grazing frequencies (equivalent to the period of time necessary for swards to reach 95 and 100% interception of the incident light during regrowth – LI), and were allocated to experimental units (1200 m2 paddocks) according to a completely randomised design and a 2 x 2 factorial arrangement, with 3 replications. Measurements were performed in two seasons of the year: (1) spring (November and December 2005), and (2) summer (January to April 2006). Stocking density was calculated aiming at an occupation period of paddocks of 10 to 12 hours. The following response variables were analysed: vertical distribution of morphological components of sward herbage mass, dynamics of the grazing down process, patterns of ingestion (bite mass, bite rate, and intake rate), animal behaviour (grazing, rumination and other activities), morphological composition of the consumed herbage, patterns of animal movement and search for forage (number of feeding stations per unit of time and number of steps between feeding stations), and patterns of defoliation of individual tillers (defoliation depth, frequency and intensity of defoliation of leaves – expanding and expanded leaves). The highest rates of decrease in sward height were recorded for treatments 100/10 and 100/15, and the lowest for treatment 95/15. In general, as the occupation period progressed, the grazing stratum was depleted and there was a decrease in the rate swards were lowered, a result of lower sward heights and increased presence of stems and dead material in the stratum used by the animals for grazing. In spite of the lower grazing intensity (post-grazing height of 15 cm) has resulted, at the beginning of grazing, in smaller bites, intake rate and defoliation intensity of expanding and expanded leaves, the ingested herbage had a higher proportion of leaves compared to grazings at 10 cm residue. During the grazing down process, the 100% LI treatments resulted in heavier bites, but lower intake rate and proportion of leaves in the herbage consumed than the 95% LI treatments. Treatments 95/10 and 95/15 were associated with the largest grazing activity and highest intake rate, suggesting higher daily intake. However, 95/15 was the treatment that resulted in the largest proportion of leaves in the herbage consumed, and in the lowest values of frequency and intensity of defoliation of leaves, due to the lowest stocking density used. This pattern of defoliation resulted in a high residual leaf area after grazing, favouring regrowth and quick return of swards to grazing. The grazing strategy that allowed more efficient harvest of herbage was that where grazings were initiated with 95% LI and finished with a 15 cm post-grazing height.