This review examines the physical function of soil tillage. The two main objectives: 1: best conditions for crop sowing (germination, emergence, early growth, plant survival): 2: best structural conditions for crop root growth and functioning in the tilled layer are discussed. In third part, theoretical considerations are made on the consequences that soil tillage methods and, more generally, cropping systems have on physical components of soil fertility (structure stability, fissurability, compactability, organic matter evolution).

High level of seed dormancy in fresh harvested seeds of S. macrocephala and S. capitata was observed. Several treatments were tested in order to increasing seed germination (loss of dormancy). In the physical methods, the seeds were heated to 60oC for 150 minutes and 4oC for 10 and 14 days. The chemical treatments consisted of soaking the seeds in sulphuric acid (98.08%) for 10, 15 and 20 minutes. The scarification and seed cut in the pod were used as a mechanical treatment. The higher germination was obtained with scarification of the seeds. Scarified seeds were also tested in the presence of KNO3 (0.2%), GA3 (500ppm) and thiourea (0.5 and 1,0%). The most effective treatment for germination and tetrazolium test was thiourea. GA3 containing thiourea increased germination in S. capitata however the seedling growth was affected and the seed pathogens were reduced. Based on this result, the thiourea is suitable for these two species although the concentration needs more studies. | Constatou-se elevado grau de dormência em sementes recém-colhidas de S. macrocephala e S. capitata. Testaram-se alguns tratamentos visando o incremento de germinação para testes de qualidade de sementes. lnicialmente, testaram-se métodos físicos, químicos e mecânicos na germinação de sementes com vagem. Nos tratamentos físicos, as sementes foram submetidas a 60 o C por 150 minutos, e 4 o C durante dez e quatorze dias. Os químicos constaram de imersão de sementes em ácido sulfúrico (98,08%) durante 10, 15 e 20 minutos. Os mecânicos foram: corte de sementes com vagem, e remoção de vagens com posterior lixamento das sementes. Os maiores incrementos na germinação foram obtidos através da escarificação das sementes com lixa. A partir de sementes lixadas, testaram-se os efeitos de tratamentos térmicos e químicos na germinação. Os tratamentos térmicos foram os mesmos já citados, e os químicos foram: nitrato de potássio 0,2%, ácido giberélico 500ppm e tiouréia (0,5 e 1,0%). Para as sementes escarificadas, os únicos tratamentos efetivos na germinação foram os que incluíram tiouréia, obtendo-se resultados semelhantes entre germinação e viabilidade em tetrazólio. O ácido gíberélico adicionado à tiouréia aumentou a germinação em S. capitata. Recomenda-se o uso de tiouréia para testes de germinação em sementes lixadas, das duas espécies. Necessita-se, entretanto, de mais estudos relativos à concentração.

The present state of the Sahelian rangelands and their overexploitation requires control of their use and the development of other management practices. This is the more so, because of the devastating effects of the droughts during the last 15 years. For this purpose description and evaluation of the species composition of the herbage vegetation is needed, which is especially difficult because the predominance of annuals contributes to large differences from year and from place to place.The present thesis concerns research on the interaction between the properties of species, rainfall, substrate and way of exploitation, as reflected in the dynamics of the rangeland vegetation. This research was executed in Niono in Mali in the South of the Sahel at an average rainfall of 600 mm per year. Experiments were done on sand, loam and clay soils. Natural rainfall was varied by shielding against rain and additional sprinkling.The main annual species were studied throughout the growing season from germination to seed production. Auto-ecological experiments in the field and under controlled conditions provided additional information about the species response to environmental differences. The observed changes in the absolute and relative contribution of the species in terms of number and biomass clarified the role of various plant properties in conjunction with rainfall amount and distribution, substrate properties and management.Important properties of the species that govern the dynamics of the vegetation are germination rate, drought resistance of the young seedlings, type of photosynthesis (C 3 or C 4 ), day length, length of the growing cycle, degree in which the plants continue vegetative growth after the onset of flowering and the production of seeds. The main effect of the substrate is its influence on the rain-dependent availability of water throughout the growing season, as affected by physical soil properties and topography. Exploitation by grazing adversely affects this rain-dependent availability of water, but also has a direct effect on height, length of the growing cycle and seed production of the species. Rapid germination, drought resistance of seedlings, C 4 -photosynthesis, a long growing cycle and extension in height are in general advantageous properties to survive, but plants can neither have it all ways. Moreover, slow germination and a short growing cycle are apparently advantageous in long periods of drought, whether these are due to irregular rainfall as such or to substrate properties or overexploitation.Vegetation surveys to support the improvement of rangeland use are of more lasting value, if the dynamics of the herbage cover is interpreted against the background of the results that are presented in this study.

The present state of the Sahelian rangelands and their overexploitation requires control of their use and the development of other management practices. This is the more so, because of the devastating effects of the droughts during the last 15 years. For this purpose description and evaluation of the species composition of the herbage vegetation is needed, which is especially difficult because the predominance of annuals contributes to large differences from year and from place to place.The present thesis concerns research on the interaction between the properties of species, rainfall, substrate and way of exploitation, as reflected in the dynamics of the rangeland vegetation. This research was executed in Niono in Mali in the South of the Sahel at an average rainfall of 600 mm per year. Experiments were done on sand, loam and clay soils. Natural rainfall was varied by shielding against rain and additional sprinkling.The main annual species were studied throughout the growing season from germination to seed production. Auto-ecological experiments in the field and under controlled conditions provided additional information about the species response to environmental differences. The observed changes in the absolute and relative contribution of the species in terms of number and biomass clarified the role of various plant properties in conjunction with rainfall amount and distribution, substrate properties and management.Important properties of the species that govern the dynamics of the vegetation are germination rate, drought resistance of the young seedlings, type of photosynthesis (C 3 or C 4 ), day length, length of the growing cycle, degree in which the plants continue vegetative growth after the onset of flowering and the production of seeds. The main effect of the substrate is its influence on the rain-dependent availability of water throughout the growing season, as affected by physical soil properties and topography. Exploitation by grazing adversely affects this rain-dependent availability of water, but also has a direct effect on height, length of the growing cycle and seed production of the species. Rapid germination, drought resistance of seedlings, C 4 -photosynthesis, a long growing cycle and extension in height are in general advantageous properties to survive, but plants can neither have it all ways. Moreover, slow germination and a short growing cycle are apparently advantageous in long periods of drought, whether these are due to irregular rainfall as such or to substrate properties or overexploitation.Vegetation surveys to support the improvement of rangeland use are of more lasting value, if the dynamics of the herbage cover is interpreted against the background of the results that are presented in this study.

The temporal and spatial dynamics of tree populations in coastal forests of southeastern Long Island, New York, were reconstructed from analyses of (1) historical accounts, maps, and aerial photos, (2) field evidence of forest history (3) age and height class distributions along the moisture gradient, and (4) fossil—pollen content of bordering salt—marsh deposits. Tide—gauge records of sea level provided independent evidence for long—term change in the physical environment (rise in sea level) that caused populations to migrate up the continental shelf. Field evidence for fire and historically documented land—use change demonstrated the important role of disturbance in the dynamics of tree populations. Spatial and temporal patterns in forest, were determined by information on disturbance frequency, dispersal, generation times, and rates of change in the physical environment. Mature cohorts occupied lower elevations than did their progeny because during the period from germination to seed production, suitable soil—moisture conditions shifted upslope in response to sea—level rise. Establishment of new regeneration depended on disturbances that opened the forest canopy and/or prepared seed beds. Frequent fire, land clearance and abandonment, and tree blowdowns insured opportunities for seedling establishment at higher elevations as conditions became unsuitable for regeneration on low—elevation sites. The resultant pattern of range shifts was stepwise, with waves of regeneration following disturbances and subsequent reduced seedling establishment as soil moisture continued to increase. The vegetation dynamics could be explained solely on the basis of life histories of plant species and environmental change; "emergent properties" and "successional stages" in a deterministic sense were not apparent. An understanding of successional change required information on both the spatial and temporal dimensions of vegetation pattern because composition change over time resulted from range shifts in response to environmental transition.