The reduction in feed intake and growth rate that occurs following weaning is of major economic consequence to the pig industry. Currently, a range of antimicrobial products can be used to minimise the impact of weaning on piglet health and subsequent performance. However, the use of these products in pig diets is subject to increasing restriction worldwide because of perceived risks to public health and to the environment. Thus, alternative methods are required to mitigate the growth check that almost invariably occurs after weaning in most production systems. Piglets produced outdoors are claimed to experience less of a growth check at weaning and to be able to thrive in relatively unsophisticated weaner accommodation. However, these claims have not been substantiated under Western Australian conditions, nor a scientific basis for these claims established. Consequently, a series of experiments was designed to test the general hypothesis for this thesis – ‘the gut structure and function, and lifetime performance of the weaned pig are affected by its pre- and post-weaning rearing environments’. Experiment 1 was conducted in two parts to quantify differences in the growth performance, health and gut structure of weaner pigs produced indoors or outdoors and reared in conventional or deep-litter pens. The weaner diet in the first part of the experiment contained 100 ppm of olaquindox and 3,000 ppm of zinc oxide (Exp1a). This experiment was repeated without using dietary antimicrobial products (Exp1b). Experiment 2 was conducted in conventional buildings to examine the effect of exposing piglets in lactation to similar substrates to those available to outdoor piglets used in Exp1a and Exp1b in the absence of other differences in the outdoor production milieu. Pre-weaning environments in Exp1a (indoor production (IP) and outdoor production (OP)) appeared to have little effect on gut structure and overall growth rate but significantly affected carcass composition, whereas post-weaning environments (conventional (C) or deep-litter (DL)) affected both overall growth rate and carcass composition. Although feed disappearance was similar, OP pigs grew faster than IP pigs in the first 47 d after weaning in Exp1a but not in Exp1b. Lifetime growth rate (GR), P2 backfat, feed disappearance and feed conversion ratio (FCR) were not significantly affected by the production environment in Exp1a whereas OP pigs grew slower with higher P2 backfat and FCR in Exp1b. Interestingly, OP pigs had heavier carcass weights and higher dressing percentages than IP pigs in both parts of the experiment. The effects of post-weaning environment were more consistent as DL pigs grew faster, were fatter, and had higher carcass weights and dressing percentages than C pigs. Villus height and crypt depth of IP and OP pigs were not different at 21 (weaning) or 28 d, but villus height decreased and crypt depth increased in the week after weaning. Pigs reared in C pens had greater faecal concentrations of volatile fatty acids than pigs in DL, indicating that the latter ingested sufficient straw to alter fermentation characteristics. In Experiment 2, there were no differences in gut structure or pre-weaning and lifetime GR of pigs offered no creep feed (NC), a commercial creep feed (CF) or an ‘outdoor’ mix (OM) comprising of 1 part straw, 5 parts sow feed and 25 parts of soil taken from paddocks in which OP pigs used in Exp1a and Exp1b were farrowed. However, NC pigs grew slower in the week after weaning than the other two treatments. Backfat and feed disappearance were similar for all treatments but pigs on the OM treatment had higher carcass weights and dressing percentages than pigs on the NC and CF treatments. Villus height and crypt depth were not different between treatments and, although the piglets were weaned at 28 d, villus height decreased and crypt depth increased in the week after weaning to an extent similar to that experienced by piglets weaned at 21 d in Experiment 1. Although all piglets received intramuscular injections of 200 mg iron (Fe) dextran when 1 to 2 days old, piglets offered the OM during lactation had higher serum iron and blood haemoglobin (Hb) levels than those offered NC or CF. Furthermore, half the piglets offered NC or CF had Hb levels indicative of chronic Fe deficiency anaemia. The average parity of sows used in this experiment was 6.3 litters, suggesting that piglets may have been born with low Fe stores, possibly because of low Fe stores in their dams due to sub-optimal mineral nutrition over successive parities. In summary, the findings from these experiments partly supported the general hypothesis for this thesis. Under the conditions of these experiments, access to outdoor substrates in lactation had little effect on gut structure and lifetime growth rate but increased both carcass weight and dressing percentage, whereas rearing in DL pens increased feed intake, FCR, growth rate, P2 backfat, carcass weight and dressing percentage.

Production, egg quality, behaviour, and physical condition were compared from Wk 20 to Wk 50 among three beak- trimmed commercial laying strains, Lohmann White (LW), H & N White (HN), Lohmann Brown (LB), and a non-commercial Cross between Rhode Island Red (male) and Barred Plymouth Rock (female) in conventional cages and in floor pens. All chicks were reared in their respective environments, and 450 and 432 pullets were housed at 18 and 7 weeks of age in cages and floor pens respectively. Hens in cages were provided with 688 cm2/bird and those in pens with over 6,000 cm2/bird, both of which are more than provided by commercial standards. Body weights and eggshell weights were higher for birds in floor pens than those in cages, and although they increased with age, body weight of hens in cages decreased at Wk 50. White-egg layers (LW, HN) used perches and nest boxes more than Brown-egg layers (LB, Cross). During the laying period, mortality was higher for all strains in cages and during the rearing period mortality was higher in floor pens for LB hens but not other strains. No aggressive behaviours were found, but the frequency of gentle feather pecking and pecking at the enclosure was higher in cages than in floor pens. Feather condition deteriorated over time in cages mainly because of contact with the cage wires whereas in floor pens, feather condition of birds at Wk 20 was not different from that at Wk 50. The frequency of keel bone deformities was higher for White-egg layers than for Brown-egg layers in cages and was higher for Cross hens than other strains in floor pens. Claws were longer in cages than in the floor pens. Foot condition was worse in floor pens than in cages. The welfare indicators used in this study showed that cages restricted the hens' behaviour compared to floor pens and resulted in higher laying period mortality, reduced body weight and deteriorated feather condition than floor pens. Both systems had advantages and disadvantages in regard to the hens’ health and welfare. The use of environmental complexities was strain specific in floor pens. The environment by genotype interactions suggests that the strain should be considered when considering alternative housing systems.

Production, egg quality, behaviour, and physical condition were compared from Wk 20 to Wk 50 among three beak- trimmed commercial laying strains, Lohmann White (LW), H & N White (HN), Lohmann Brown (LB), and a non-commercial Cross between Rhode Island Red (male) and Barred Plymouth Rock (female) in conventional cages and in floor pens. All chicks were reared in their respective environments, and 450 and 432 pullets were housed at 18 and 7 weeks of age in cages and floor pens respectively. Hens in cages were provided with 688 cm2/bird and those in pens with over 6,000 cm2/bird, both of which are more than provided by commercial standards. Body weights and eggshell weights were higher for birds in floor pens than those in cages, and although they increased with age, body weight of hens in cages decreased at Wk 50. White-egg layers (LW, HN) used perches and nest boxes more than Brown-egg layers (LB, Cross). During the laying period, mortality was higher for all strains in cages and during the rearing period mortality was higher in floor pens for LB hens but not other strains. No aggressive behaviours were found, but the frequency of gentle feather pecking and pecking at the enclosure was higher in cages than in floor pens. Feather condition deteriorated over time in cages mainly because of contact with the cage wires whereas in floor pens, feather condition of birds at Wk 20 was not different from that at Wk 50. The frequency of keel bone deformities was higher for White-egg layers than for Brown-egg layers in cages and was higher for Cross hens than other strains in floor pens. Claws were longer in cages than in the floor pens. Foot condition was worse in floor pens than in cages. The welfare indicators used in this study showed that cages restricted the hens' behaviour compared to floor pens and resulted in higher laying period mortality, reduced body weight and deteriorated feather condition than floor pens. Both systems had advantages and disadvantages in regard to the hens’ health and welfare. The use of environmental complexities was strain specific in floor pens. The environment by genotype interactions suggests that the strain should be considered when considering alternative housing systems.

In this thesis the selected enterprise indicators of laying were analyzed in terms of the impact of a hen housing system. For commercial egg production in the enterprise the laying hybrid Lohmann brown-LITE hen has been used since 2011. A total of 16 laying turns were included for the observation during the period of 20112014. Due to the sufficient number of data a nine-month laying batch has been chosen. The highest average number of eggs laid by one hen for the observation period was achieved in the Profit-Tier Layer Cage System by Big Dutchman (257.0 pcs.). In the Laying battery Cage System by Kovobel the number of eggs was by 1.3 pcs lower (255.7 pcs). The lowest number of eggs was set down for laying hens kept in a cage free systems (249.2 pcs). It was 7,8 pcs and 6,5 pcs lower respectively than with laying hens housed in the cage systems by Big Dutchman and Kovobel. The highest average intensity of laying for 9-month laying period was found in the cage system by Big Dutchman (93.9%). In the cage system by Kovobel company it was 1.0 % lower (92.9 %). The intensity of laying at a cage free system amounted to 89.2 % pcs. It was 4.7 % and 3.7% respectively lower than when housed in the cage systems by Big Dutchman and Kovobel. For laying hens kept in a cage system the average consumption per hen in a day, for the observed period, was found out to be lower 117.6 g for the cage system by Big Dutchman company and 117.4 g for Kovobel. The highest average feed consumption showed 121.5 g with laying hens kept in a cage free system. The average mortality of laying hens during the observed period was 0.64 %.

Production, egg quality, behaviour, and physical condition were compared from Wk 20 to Wk 50 among three beak- trimmed commercial laying strains, Lohmann White (LW), H & N White (HN), Lohmann Brown (LB), and a non-commercial Cross between Rhode Island Red (male) and Barred Plymouth Rock (female) in conventional cages and in floor pens. All chicks were reared in their respective environments, and 450 and 432 pullets were housed at 18 and 7 weeks of age in cages and floor pens respectively. Hens in cages were provided with 688 cm2/bird and those in pens with over 6,000 cm2/bird, both of which are more than provided by commercial standards. Body weights and eggshell weights were higher for birds in floor pens than those in cages, and although they increased with age, body weight of hens in cages decreased at Wk 50. White-egg layers (LW, HN) used perches and nest boxes more than Brown-egg layers (LB, Cross). During the laying period, mortality was higher for all strains in cages and during the rearing period mortality was higher in floor pens for LB hens but not other strains. No aggressive behaviours were found, but the frequency of gentle feather pecking and pecking at the enclosure was higher in cages than in floor pens. Feather condition deteriorated over time in cages mainly because of contact with the cage wires whereas in floor pens, feather condition of birds at Wk 20 was not different from that at Wk 50. The frequency of keel bone deformities was higher for White-egg layers than for Brown-egg layers in cages and was higher for Cross hens than other strains in floor pens. Claws were longer in cages than in the floor pens. Foot condition was worse in floor pens than in cages. The welfare indicators used in this study showed that cages restricted the hens' behaviour compared to floor pens and resulted in higher laying period mortality, reduced body weight and deteriorated feather condition than floor pens. Both systems had advantages and disadvantages in regard to the hens’ health and welfare. The use of environmental complexities was strain specific in floor pens. The environment by genotype interactions suggests that the strain should be considered when considering alternative housing systems. / Land and Food Systems, Faculty of / Graduate

The world’s population is increasing and every day, more people are coming to the cities. People are constantly relocating their houses for meeting their needs. As societies are growing and mobility in cities is raising, cities would face challenges like accommodating the population inexpensively and providing good quality of life and stable social and environmental infrastructure for them. However, due to the fixed regulated and unadaptable construction market, the architecture doesn’t meet the changing needs of inhabitants, therefore, every day the housing market is more unaffordable because of the high demand for housing developments, lands scarce, etc. Architecture needs to be flexible to adjust itself to the current market. Since most of the relocations and high demands for building new housing projects occur when built spaces fail to meet their occupants’ growing needs. This thesis tries to find an alternative framework for designing a flexible one whereby architecture is shaped by the user’s needs and can adjust itself in a long period. This thesis also works on designing an environment that responds to the social, economic, and environmental needs of residents as one united community. Therefore, it uses a shared living idea as an affordable alternative for making houses cheaper and creating a more stable social environment for inhabitants. The content of this thesis is categorized upon the following narrative. Firstly, this thesis is trying to deepen its understanding of the matter of shared living and its possible advantages and disadvantages for producing stable social and economical infrastructure. For this purpose, it analyses a notion of shared living and shared economy through studying successful case studies. Secondly, by studying and utilizing metabolic design criteria, the thesis will form an understanding of a more sustainable approach toward architecture and built environment to design an affordable housing system that provides high quality of life for inhabitants through responding to their needs. Thirdly, the thesis will frame its design strategy based on shared living perspective and metabolic design criteria. The thesis then will illustrate its steps and methods to achieve flexible architecture through designing its module, structure, and material. Lastly, it will discuss its findings on how the shared living approach and metabolic philosophy can aid architects to design more sustainable housing units that can meet their resident’s needs through a long period and make strong social, economic, and environmental infrastructure.

The aim of this dissertation was to study through meta-analysis the cortisol levels and productivity of sows via a field study the effects of housing on behavior and productivity of sows. The database was composed of 22 articles published between 1996 to 2010, totaling 170 treatments and 1,941 sows. The meta-analysis was performed by sequential analysis: graphical, correlation and variance-covariance. To evaluate the systems of gestation were used 42 sows in a randomized block design with two treatments (T1 - 28 days in gestation crates and 82 days in concrete floor collective pens and T2 - 28 days in gestation crates and 82 days in deep bedding collective pens) three replicates each. Salivary cortisol was positively correlated (P<0.05) with feed intake, of crude protein and lysine and lysine level. Plasmatic cortisol was negatively correlated (P<0.01) with metabolizable energy level and positively correlated (P<0.01) with level and crude protein intake. Individual housing in deep bedding reduced (P<0.01) 47% the salivary cortisol in relation to concrete floor housing. In housing collective on deep bedding this level was 10% higher (P<0.01) in relation housing collective to concrete floor. The piglet birth weight decreased by 0.0002 kg for each increase of one unit nMol.L-1 in plasmatic cortisol (Y=1.6032-0.0002X) or salivary (Y=1.6026-0.0002X). The incidence of stereotyped behaviors was 71% lower (P<0.05) in sows housed in pens of deep bedding, at 90th days gestation. The activities increased on average by 60% (P<0.05) for sows housed in pens of deep bedding during collective pens housing (transfer, after transfer, 90th days of gestation and the average total period). Sows housed in pens concrete floor showed an increase of 66% (P<0.05) in interactions social, the average of the total period. Sows housed in pens of deep bedding had 60% more (P<0.05) skin lesions on the 1st and 2nd days after being transferred to collective pens. The housing system did not affect (P>0.05) backfat thickness and feed conversion the sows during lactation, body weight and weight gain of piglets and the duration total of the farrowing. Sows housed in pens concrete floor during pregnancy reached the end of lactation with 7% more (P<0.05) pigs than sows housed in pens of deep bedding. The pens housing deep bedding for sows in gestation provides better welfare, increases physical activity and does not affect the productive performance of animals. / O objetivo desta dissertação foi estudar através de meta-análise os níveis de cortisol e a produtividade de porcas gestantes e através de um estudo a campo os efeitos do sistema de alojamento sobre o comportamento e a produtividade de porcas gestantes. A meta-análise foi composta por 22 artigos publicados entre 1996 a 2010, totalizando 170 tratamentos e 1.941 porcas. A meta-análise foi realizada por análise sequencial: correlação, gráfica e de variância-covariância. Para avaliar os sistemas de gestação foram utilizadas 42 porcas em um delineamento de blocos ao acaso com dois tratamentos (T1 - 28 dias em celas de gestação e 82 dias em baias coletivas de piso concreto e T2 - 28 dias em celas de gestação e 82 dias em baias coletivas de cama sobreposta) três repetições cada. Para cortisol salivar houve correlação positiva (P<0,05) com consumo de ração, de proteína bruta e de lisina. Para cortisol plasmático a correlação foi negativa (P<0,01) com nível de energia metabolizável e positiva (P<0,01) com nível e consumo de proteína bruta. O alojamento individual em cama sobreposta reduziu (P<0,01) 47% o cortisol salivar em relação ao alojamento individual em piso concreto. No alojamento coletivo em cama sobreposta esse nível foi 10% maior (P<0,01) que o alojamento coletivo em piso concreto. O peso ao nascer dos leitões reduziu em 0,0002 kg a cada aumento de um nMol.L-1 no cortisol plasmático (Y=1,6032-0,0002X) ou salivar (Y=1,6026-0,0002X). A incidência de comportamentos estereotipados foi 71% menor (P<0,05) em porcas alojadas em baias de cama sobreposta, aos 90 dias de gestação. As atividades aumentaram em média 60% (P<0,05) para porcas alojadas em baias de cama sobreposta, durante o alojamento em baias coletivas (transferência, pós-transferência, 90 dias de gestação e na média do período total). Porcas alojadas em baias de piso concreto apresentaram aumento de 66% (P<0,05) nas interações sociais, na média do período total. As porcas alojadas em baias de cama sobreposta apresentaram 60% mais (P<0,05) lesões de pele no 1º e 2º dias após a transferência para baias coletivas. O sistema de alojamento não afetou (P>0,05) a espessura de toucinho e a conversão alimentar das porcas na lactação, o peso vivo e ganho de peso dos leitões e a duração total do parto. Porcas alojadas em baias de piso concreto durante a gestação chegaram ao final da lactação com 7% mais (P<0,05) leitões do que as porcas alojadas em baias de cama sobreposta. O alojamento em baias de cama sobreposta para porcas em gestação proporciona melhor bem-estar, aumenta a atividade física e não altera o desempenho produtivo dos animais.

La présente thèse concerne les mutations sociospatiales introduites par des tendances de financiarisation du système du logement grec au cours des deux dernières décennies. Comment les processus mondiaux de financiarisation se sont-ils accommodés à un contexte local pour développer leur version grecque ? Le but de notre travail est de mettre en lumière la transformation d’une économie autofinancée et basée sur les réseaux familiaux en marché hypothécaire. Notre hypothèse est que ces bouleversements du financement de la fabrique résidentielle, c’est-à-dire l’apparition de la finance comme un « nouvel acteur » sur les circuits de production du logement, vont de pair avec des dynamiques de précarisation du logement et des nouvelles géographies sociospatiales.Dans les années d’après-guerre, Athènes, ville méditerranéenne et périphérique de l’Europe du Sud, est caractérisée par une forte mobilité sociale, sur laquelle les circuits de production du logement ont fonctionné comme un «ascenseur social ». Depuis le début des années 1990, Athènes doit faire face à la fois à un nouveau contexte et à la persistance des structures préexistantes. C’est la première période d’un véritable développement du crédit au logement en Grèce qui fait notablement augmenté les prix des logements. L’analyse quantitative et qualitative, le cas d’étude ainsi que les entretiens semi-directifs mettent en valeur de nouveaux acteurs, représentations et pratiques, mais aussi de tendances de l’exclusion bancaire. Cette étude souligne l’importance de l’aspect géographique ainsi que le rôle signifiant de réseaux familiaux, de l’échelle du quartier et de la vie quotidienne dans l’analyse de la finance ainsi que l’interdépendance entre système financier et l’économie informelle. / The present thesis concerns the sociospatial mutations produced by the financialization trends of the Greek housing system in the past two decades. How are the global processes of financialization accommodated in a local context in order to generate a Greek version of financialization? Aim of the thesis is to shed light on the transformation of a self-financing economy, based on family networks to a mortgage market. The hypothesis is that these changes in the field of financing the residential urban fabric, or the emergence of finance as a "new actor" on housing production circuits, are associated with the precarization of housing and new sociospatial geographies.During the postwar decades, Athens, a Mediterranean city and, at the same time, a south European peripheral urban center, is characterized by strong social mobility, when housing production circuits functioned as a “social elevator”. Since the early 1990s, Athens is facing both new challenges and the persistence of pre-existing structures. It is the first time of a crucial expansion of housing credit in Greece, when housing prices have significantly increased. The quantitative and qualitative analysis, the case study and the semi-structured interviews highlight new actors, their actions, representations and practices as well as tendencies of financial exclusion. This study highlights the importance of the geography as well as the significant role of family networks, the scale of neighborhood and everyday life in analyzing finance and the interconnections of the financial system with the informal economy.

Aufgrund der zu erwartenden Klimaveränderungen, insbesondere durch eine Erhöhung der Temperaturen und einer Zunahme der Extremsituationen (Hitze- und Kälteperioden), sollte vermehrt über Effekte von Hitzestress auf das Wohlbefinden und Leistungsvermögen von Tieren nachgedacht werden. Um den Einfluss von klimatischen Bedingungen (Temperatur und Temperatur-Humiditäts Index) auf die Gesundheit und das Leistungsvermögen von Milchkühen zu untersuchen, wurden drei Studien durchgeführt. Zuerst wurden die Auswirkungen des Temperatur-Humiditäts Index (THI) auf die Behandlungsfrequenz von Milchkühen ermittelt. Hierfür wurden die Daten aus den Bestandsbüchern von 2003 und 2005 von acht Milchviehbetrieben ausgewertet. Die Behandlungen wurden in vier Komplexe unterteilt: Stoffwechsel, Fruchtbarkeit, Euter und Klauen, wobei 37,4% der Behandlungen zum Stoffwechsel-, 32,9% zum Euter-, 21,6% zum Fruchtbarkeits- und 8,1% zum Klauenkomplex gehörten. Es zeigten sich ein tendenzieller Anstieg der Stoffwechselinzidenzen mit steigendem THI und ein nicht signifikanter Anstieg von Euterbehandlungen mit abfallendem THI. Die zu analysierenden Daten bezüglich des Einflusses auf das Leistungsvermögen wurden im Rahmen eines einjährigen Feldversuches vom April 2010 bis März 2011 aufgenommen. Hierfür wurden aus einem Pool von niedersächsischen Milchviehbetrieben 20 Betriebe mit unterschiedlichen Haltungssystemen (gedämmte und ungedämmte Laufställe mit und ohne Weidegang) ausgewählt. Auf diesen Betrieben wurden klimatische Werte (Stalltemperatur, relative Stallluftfeuchtigkeit und THI) als auch Herden- bzw. tierindividuelle Leistungsdaten (4% fettkorrigierte Milchmenge (FCM), Milchinhaltsstoffe und somatischer Zellscore (SCS)) anhand von Tankmilchproben bzw. Einzeltiermilchproben im Rahmen der Milchleistungsprüfung erfasst. Insgesamt wiesen die Untersuchungen unterschiedlich hohe Hitzestresseffekte auf die Milchleistung, Milchinhaltsstoffe und den SCS in den verschiedenen Haltungssystemen auf. Für die Milchmenge ergaben sich Regressionskoeffizienten von -0,12 kg FCM/THI und -0,11 kg FCM/THI bei gedämmten Laufställen ohne und mit Weidegang sowie -0,21 kg FCM/THI für Außenklimaställe ohne und mit Weidegang. Für den SCS ergaben sich bei den Stallhaltungssystemen Regressionskoeffizienten von +0.03 SCS/THI für gedämmte Systeme und +0,07 SCS/THI für Außenklimaställe. Basierend auf den Ergebnissen der Arbeit kann geschlussfolgert werden, dass Milchkühe in Niedersachsen während der Sommermonate Hitzestress ausgesetzt waren, wohingegen sich die Auswirkungen je nach Haltungssystem differenziert darstellten. Des Weiteren konnte gezeigt werden, dass in dem System mit gedämmtem Dach niedrigere Auswirkungen mit steigendem THI zu finden waren.

Die vorliegende Arbeit setzt sich mit verschiedenen Verhaltensparametern bei Schweinen, die in unterschiedlichen Produktionsabschnitten erfasst werden, auseinander. Die Ziele der Arbeit bestanden in der Untersuchung von Merkmalen, die zum einen das agonistische Verhalten und zum anderen mütterliches Verhalten von Schweinen charakterisieren. Des Weiteren wurde eine genetische Analyse der untersuchten Verhaltensparameter durchgeführt und damit die Möglichkeit einer züchterischen Bearbeitung dieser Merkmale evaluiert. Verhaltensparameter beim Schwein erlangen aufgrund der zurzeit stattfindenden Entwicklungen in der Schweineproduktion, wie die Gruppenhaltung von tragenden Sauen und Trends in Richtung alternativer Haltungstechniken von laktierenden Sauen, zunehmend an Bedeutung. Eine Reduzierung von Aggressionen bei Schweinen in der Gruppen¬haltung, sowie geringere Saugferkelverluste durch bessere Muttereigenschaften führen nicht nur zu einem gesteigerten Tierwohl und weniger Tierverlusten, sondern auch zu einer verbesserten Wirtschaftlichkeit der Schweine¬produktion. Im dritten Kapitel wird eine Literaturübersicht über bisher beschriebene Verhaltenstests gegeben, die genutzt wurden, um Muttereigenschaften zu erfassen. Verschiedene Testformen wurden entwickelt um das Maß an Ängstlichkeit und Aggressivität der Tiere gegenüber dem Menschen zu beurteilen. Es gibt unter anderen Tests mit denen die Kontaktfreudigkeit gegenüber Menschen gemessen wurde, sowie Tests zum Meidungsverhalten. Während der Laktation wurde die Aufmerksamkeit der Sau gegenüber ihrem Wurf mit Hilfe verschiedener Formen der Trennung von Sau und Ferkeln bonitiert. Die Empfindsamkeit der Sau gegenüber ihren Ferkeln wurde mittels eines akustischen Signals („Ferkel-Schrei-Test“) oder taktilen Stimulus überprüft. Die in der Literatur beschriebenen Verhaltenstests unterscheiden sich deutlich in ihrem Aufbau, Bewertungsschema und eingesetzten Genetiken. Zudem wurden häufig nur geringe Tierzahlen untersucht. Genetische Parameter der Verhaltenstests sind daher rar. Dennoch erscheinen Verhaltenstests dafür geeignet zu sein, um Muttereigenschaften von Sauen charakterisieren und züchterisch verbessern zu können. Über die Beziehung zwischen Mutter¬eigenschaften und anderen Verhaltensmerkmalen, sowie Produktionsparameter ist bisher wenig bekannt. Im ersten Versuch (viertes Kapitel) wurde untersucht in wie weit sich die genetischen Parameter von agonistischen Verhaltensmerkmalen unterscheiden, wenn eng miteinander verwandte Tiere unter zwei verschiedenen Haltungsumwelten untersucht werden. Es wurden Daten von insgesamt 543 Jungsauen der Linie Pietrain analysiert, 302 Tiere standen auf Betrieb A und 241 auf Betrieb B. Die untersuchten Jungsauen stammten von 96 Ebern ab, von denen 64% Nachkommen auf beiden Betrieben besaßen. Es bestanden deutliche Unterschiede in der Haltung der Jungsauen zwischen den beiden Betrieben. Das Verhalten der 214 ± 12,2 Tage alten Jungsauen wurde über eine Dauer von 30 Minuten beobachtet, nachdem Tiere aus verschiedenen Aufzuchtgruppen zusammengestallt wurden. Die Jungsauen von Betrieb A zeigten weniger unilaterale Aggressionen und bilaterale Aggressionen, wie Jungsauen von Betrieb B. Die Heritabilitäten für die Merkmale unilaterale und bilaterale Aggression lagen für Tiere auf Betrieb A auf einem niedrig Niveau (h² = 0,11 ± 0,07 bzw. h² = 0,04 ± 0,07). Auf Betrieb B konnte für das Merkmal unilaterale Aggression eine Erblichkeit von h² = 0,29 ± 0,13 und für bilaterale Aggression von h² = 0,33 ± 0,12 berechnet werden. Die genetische Korrelation zwischen den gleichen Merkmalen getestet auf Betrieb A und Betrieb B liegt auf einem hohen Niveau. Agonistisches Verhalten scheint daher in dieser Untersuchung nicht nennswert von Genotyp-Umwelt-Interaktionen beeinflusst zu werden. Aus den Ergebnissen lässt sich zudem ableiten, dass eine Selektion auf agonistische Verhaltensmerkmale Erfolg versprechend ist. Ziel des zweiten Versuchs (fünftes Kapitel) war die Analyse von Verhaltenstests während der Laktation, die dazu geeignet sind, das mütterliche Verhalten von Sauen zu beschreiben. Außerdem wurde deren Beziehung mit agonistischen Verhaltensparametern beim Zusammenstallen von einander unbekannten Large White Jungsauen (n = 798) berechnet. Insgesamt wurden drei verschiedene Tests hinsichtlich des Verhaltens durchgeführt. Zum einen wurde zu zwei verschiedenen Zeitpunkten in der Laktation, d.h. innerhalb der ersten zwei bis zwölf Stunden nach der Geburt (SEPD1) und um den zehnten Tag nach der Geburt (SEPD10) die Reaktion von 848 Large White Sauen auf die Trennung von ihren Ferkeln (n = 2,022 Würfe) ermittelt. Zusätzlich wurden die Merkmale unilaterale und bilaterale Aggression, beim Zusammenstallen von Jungsauen, die aus verschiedenen Aufzuchtgruppen stammen, erfasst. Weiter wurde das Geburtsverhalten der Sauen, die Aufzuchtleistung der Sauen, die Gebrauchsfähigkeit der Sauen und das Gesäuge der Sauen während der Laktation bonitiert. Für die Verhaltensmerkmale SEPD1 (h² = 0,03 ± 0,03) und SEPD10 (h² = 0,02 ± 0,03) konnten niedrige Erblichkeiten analysiert werden. Für die zusätzlichen Parameter (Geburtsverhalten, Aufzuchtleistung, Gebrauchsfähigkeit und das Gesäuge), die während der Laktation erfasst wurden, konnten niedrige bis mittlere Erblichkeiten (h² = 0,03 ± 0,02 bis h² = 0,19 ± 0,03) berechnet werden. Die Erblichkeiten für uni- und bilaterale Aggression bei Jungsauen lagen auf einem niedrigen (bilaterale Aggression) bis moderaten Niveau (unilaterale Aggression). Aufgrund von hohen Standardfehlern konnte keine Schlussfolgerung bezüglich der genetischen Beziehung zwischen mütterlichen Verhalten und agonistischen Verhaltensparametern gezogen werden. Abschließend ist zu sagen, dass eine züchterische Bearbeitung von Verhaltensmerkmalen beim Schwein machbar und erstrebenswert erscheint. Die Ansätze dieser Arbeit können genutzt werden um Verhaltensmerkmale in Zuchtprogramme beim Schwein weiter zu integrieren bzw. ihnen eine größere Gewichtung im Zuchtziel zukommen zu lassen.