1. A distinction is made between the production of tree trunks (economic) and the production of all kinds of primary organic matter (biological) by woodland plants capable of photosynthesis. 2. Yield tables, quality class and site index are considered to be reasonably satisfactory indicators of trunk wood production, but their value as indicators of biological production has not been demonstrated. 3. Some factors, namely geographical, weather, edaphic, genetical, management and biotic, are considered in relation to their influence on economic production. 4. Attention is drawn to the relative paucity of data on biological production. 5. The available information of woodland biomass suggests that the greatest biomass of the world's forests exceeds 500 times 103 kg. of oven‐dry material per hectare. 6. The annual accumulation of plant material in woodland ecosystems can be very rapid, up to about 20 kg./ha. when growth is active. The bulk of the organic matter is in tree trunks but where decomposition is slow large amounts of organic matter are present as litter on the forest floor. 7. Biomass change is considered an unsuitable measure of biological production since it is affected greatly by tree harvesting. 8. Net primary production is regarded as a better measure of biological production. Some of the more serious sources of error in determining net primary production are emphasized. Generally net primary production is underestimated because of failure to take into account the removal of photosynthate by animals and because of infrequent sampling so that the production of plant parts such as inflorescences is missed. 9. Woodlands are shown to be one of the most productive terrestrial communities, giving a range of annual production as 103 kg./ha., from 5 for fir forests of the northern taiga in the U.S.S.R. to 33 in tropical rain forests. The capacity of forests to produce large amounts of organic matter annually is attributed to the large root and leaf mass which permit relatively full use to be made of the soil and incident solar energy. 10. A large part of the organic matter produced by woodland plants is broken down by the secondary producers. Associated with the annual turnover of organic matter in woodlands, nutrient elements circulate through woodland ecosystems. The annual retention of nutrient elements within the biomass is much less than what is returned to the soil by litter fall. Taking a hard beech forest in New Zealand as an example, the woodland plants have an annual uptake as kg. per hectare of potassium, calcium, magnesium, phosphorus and nitrogen of 34, 84, 12, 3‐3 and 40 respectively, but about 30, 74, 11, 2–6 and 37 are returned annually in the litter fall. 11. The annual cycle of mineral elements is regarded as a polycyclic system affected by various input and output factors. The loss of nutrients through the harvesting of tree boles, for instance, is relatively small compared to the addition of nutrients in precipitation. 12. Soil changes resulting from the variable balance in the factors affecting mineral cycling are considered, particularly with respect to general assumptions made about soil deterioration under monocultures of coniferous tree species. From the point of view of the maintenance of soil fertility, general condemnations of conifers and universal recommendations for hardwoods are regarded as unsound, since great variations in their effects on the soil occur in both conifers and broad‐leaved trees as groups. 13. The need for more comprehensive studies, embracing a whole or several forest rotations, is stressed for the evaluation of woodland production, organic turnover, mineral cycling and soil change. Such evaluations are becoming more important as multiple use of woodlands increases. 14. It is suggested that at present the greatest advances in woodland production ecology will be made by studies of primary production in forests known regionally to be the most productive. This may permit meaningful relationships between site factors and production to be recognized. 15. With more complete production and turnover data, it seems likely that existing woodland classification schemes will be improved, since it becomes possible to emphasize both the dynamic nature of woodlands and the long‐term changes taking place in environmental conditions.