The improvement of grazing and drought tolerance of temperate (C3) perennial pasture grasses is an important component of plant improvement programmes. The improvement of these characteristics in grasses is best based on an understanding of the physiology of grass growth and development. This paper discusses the components of growth, i.e. buds, leaves, tillers, stems, and plants, how they interact, and how compensation between components occurs. These components are considered in relation to what is known about the grazing and drought tolerance of grasses in Australia where selection for plant survival is as important as for production. It is concluded that tiller and bud densities could be increased in new cultivars. Some existing cultivars have, arguably, too low a tiller density. Plants with more and, or larger tillers have more chances of survival in many circumstances. Reproductive development interacts strongly with grass growth and development. Once apices initiate, growth is stimulated, but the numbers of tillers can then decline and plants become more vulnerable to damage from grazing, especially during stem elongation. Early-flowering cultivars are more winter-active, but they are also likely to be more synchronised in development and more prone to damage from grazing in spring. It is difficult to see that these connections can readily be broken in a positive way. Improvement in drought tolerance is constrained by the limitation that the species currently in use all need to maintain some root contact with water to survive. It is suggested that plants with root systems that can better exploit the available, deeper, soil water should be sought. In addition, plants should have some ability to adjust to water stress to husband water use. Selection of plants that survive regular dry seasons and then grow the fastest when rain falls may be the best initial approach. Unfortunately very little is known about the physiology of grass growth under grazing, with or without water stress, and this should be an active area for research. Selection of plants that can tolerate all sorts of grazing pressures with or without water or nutrient stresses, is however, highly unlikely. In most instances the successful use of improved genotypes will depend upon the development of appropriate management practices. This is more important the less favourable the environment is for plant growth.