The framework is developed for a formal quantitative analysis of the vegetative dynamics of Trifolium repens, based on partitioning the components of its growth. The method is used to describe the vegetative dynamics of T. repens in one pasture during the course of one year. Seasonal and spatial variation were analysed by regression on several environmental variables. The present paper includes only a partial analysis, covering the dynamics of leaves and nodes per shoot axis. The remainder of the analysis will be presented in subsequent papers. The production of modules per shoot axis, and the subsequent mortality of leaves, and burial and mortality of nodes, showed different patterns of spatial and seasonal variability, and different types of response to the environment. There was much seasonal and little spatial variation in rates of birth, burial and death of leaves and nodes. In contrast, there was relatively little seasonal and much spatial variation in the numbers of leaves and nodes, even though these numbers are determined by present and past births and deaths. The rate of production of leaves by individual apices varied from 0.01 per day to 0.19 per day. It appeared to be determined largely by temperature, probably of the apex, but also to some extent by genotype. Leaves survived for 1-21 weeks. The risk of leaf mortality varied with the age of the leaf, the number of sheep in the field, and the season. Leaves born in November tended to live longest, and those born in summer with many sheep shortest. At least 56% of leaves were utilized by sheep, the remainder dying from other causes. Seasonal fluctuations in death rate of leaves tended to lag three weeks behind fluctuations in birth rate, but with additional fluctuations caused by changes in grazing. The lag caused there to be a minimum of 1-5 leaves per axis in January, and a maximum of 3-12 in August, although at all times the number of leaves was increasing on some axes and decreasing on others. Axes were progressively buried by worm casts and leaf litter within 0-39 weeks of birth. The rate of burial appeared to be determined largely by earthworm activity and by treading by sheep. In autumn and early spring, nodes and internodes were buried faster than new ones were born. As a result, as few as three internodes were exposed on average in March, compared with a maximum mean of 20 in August. By late winter, some axes were entirely buried and started to grow vertically upwards towards the soil surface. Nodes Survived for 14-85 weeks, the mean longevity being 51 weeks. On average, each axis bore 25 nodes, with a range of 10-49 nodes. Seasonal fluctuations in death rate lagged some 4-8 weeks behind those in birth rate. Survivial of nodes and internodes was dependent on the establishment and survival of roots. When the oldest rooted node on an axis died, all of the axis between it and the next rooted node also died. The development of pseudo-taproots increased the longevity of nodes by 100 days. Leaves had much shorter lives than their nodes. The 'average' shoot axis bore 3.5 nodes with leaves, 4.2 nodes still visible but without leaves, and 17.6 nodes buried beneath worm casts and litter.