1. The diversity of species used by preagricultural peoples contracted as a result of domestication, but intraspecific diversity greatly expanded with worldwide migrations of crops, to be drastically contracted as a consequence of modern plant breeding. 2. Wild progenitors and other relatives of crops have much to contribute to plant breeding, but this is inhibited by inadequate collections and quite insufficient information. Although many of these crops are still relatively safe in their natural habitats, they should be extensively collected and studied. 3. The natural gene pools of many of the wild species directly used by humans, and particularly many tropical forest species (including tree fruits as well as forestry species) will be lost through replacement of indigenous forests by farmland and by planting with selected forest species. Extensive nature reserves are essential for gene-pool conservation, which would also help to preserve species that may become useful in the future. 4. In general, wild species are best preserved within the community of which they form a part. Ex situ preservation presents many difficulties, but it may be inevitable in some cases. Seed conservation is a practical alternative and should be more widely explored. 5. Because of the enormous genetic diversity between and within the local land races of crops evolved over long periods in diverse areas with traditional farming systems, they constitute a most valuable source of genetic materials for plant improvement. Such crops are threatened by the rapid advances of much higher-yielding advanced cultivars, and efforts for their preservation are of the highest priority. The situation in Pakistan and neighboring countries is discussed. 6. Advanced cultivars (produced by modern plant-breeding methods) have been the principal genetic resources used in developed countries, but there are good reasons for enlarging gene pools used by plant breeders, one being the danger of “genetic vulnerability” that results from homogeneity. “Primitive” gene pools are of special importance for plant improvement in the countries in which they are situated. 7. Methods for the exploration and collection of genetic materials have been clarified. The urgent need now is to safeguard representative samples of what is left in the field, as well as the material now in collections, much of which is inadequately maintained. 8. Methods and procedures for the long-term conservation of seeds have been worked out, and the organization for international collaboration of seed-storage laboratories is in an advanced stage of preparation. 9. The conservation of genetic resources is solely directed toward utilization by present and future generations; hence, the evaluation and documentation of collected material is of the greatest importance. Evaluation must be directed toward the practical objectives of breeding projects, and information made available to all users. 10. Current developments are briefly recorded.