Results are presented of a 20-year investigation into survival of crop and weed seeds in undisturbed soil. Indications are that seeds reveal their survival potential during the first 4 years; only rarely did species which persisted for this initial period lose all viability in the subsequent 16 years.Deterioration occurred more rapidly in acid peat than in loam soil, and the species which persisted in peat also remained viable in loam soil.The Gramineae were short-lived in the soil, with the exception o{ Phleum spp., some seeds of which persisted for 20 years. Most leguminous species persisted for the whole period but not to the same extent as some ofthe weed species. Ofthe weeds, the species with the highest percentage seed survival were Ranunculus repens (53%), Chenopodium album (23%) and Rumex crispus (18%). Longiviid des svmences de plantes cultivies et de mauvaises herbes: survivance apris 20 annies dans le sol. L'auteur pr&ente les r&ultats d'une ^tude poursuivie pendant 20 ann6es sur la survivance de semences de plantes cultiv6es et de mauvaises herbes dans un sol non travailM. II ressort de ce travail que les semences manifestent leur potentiel de survivance durant les 4 premieres annees; rares sont les especes qui ayant persist^ durant cette premiere p^riode ne perdent pas toute viability durant les 16 annees suivantes.La deterioration des semences se produit plus rapidement en sol tourbeux acide qu'en sol argileux, et les especes qui survivent en sol tourbeux restent ^galement viables en sol argileux.Les gramin^es ont ^t^ peu persistantes dans le sol, k l'exception de Phleum spp., dont certaines semences ont surv^cu pendant 20 ans. La plupart des especes I^gumineuses ont persist^ durant toute la p^riode, mais pas dans la meme proportion que certaines autres espfeces de mauvaises herbes. Parmi les mauvaises herbes, les especes avec Ies pourcentages les plus ei6ves de semences survivantes ont €t6: Ranunculus repens (53 %), Chenopodium album (23 %) et Rumex crispus (18 %). Die Langlebigkeit von Kulturpflanzen-und Unkrautsamen: Vberleben im Boden nach 20 JahrenZusammenfassung. Es werden die Ergebnisse einer 20-jahrigen Untersuchung uber das tJberleben von Kulturpflanzen-und Unkrautsamen in nicht-bearbeitetem Boden dargestellt. Das Uberlebenspotential der Samen wurde in den ersten 4 Jahren deutlich. Es kam nur selten vor, dass Samen, die diese Anfangszeit uberstanden, ihre Lebensfahigkeit in den folgenden 16 Jahren verloren.In saurem Torf ging die Lebensfahigkeit schneller zuruck als im Lehmboden und die Arten die im Torf lebensfahig blieben, blieben es auch im Lehm.Die Gramineen waren im Boden kurzlebig, mit Ausnahme von Phleum spp., von denen einige Samen 20 Jahre uberdauerten. Die meisten Leguminosenarten uberdauerten die ganze Versuchszeit, aber nicht im gleichen Masse wie einige der Unkrautarten. Die Unkrautarten mit der hochsten Uberlebensrate waren: Ranunculus repens (53%), Chenopodium album (23%) und Rumex crispus (18%).

The fertile tiller production and seed yields of spaced plants of meadow fescue varieties were found to vary widely with the time of sowing, particularly in the first harvest year. The benefit of early spring sowing is indicated in the total productivity over two harvest years.Varieties differed in the amount of heading in the seeding year; the American type was the most prolific. This had a bearing on the relative yields in the first harvest year, since many of the potentially productive tillers of this variety were vifasted through the plants becoming ripe to flower in the seeding year and, therefore, the rate of tiller production was checked. Competition within the plant appeared to be the cause of tillers remaining vegetative when conditions otherwise seemed suitable for generative development. The smaller-sized plants gave a higher proportional increase of fertile tillers at harvest than would be expected from the vegetative tillers counted in April, while at the same time there was a negative relationship at harvest from plants which had a very high number of vegetative tillers.Infertility as reflected in a poor seed set at the time of harvest is a serious wastage of potential production in meadow fescue and, in selecting for superior seed output, breeders need to take this factor into consideration.The 1000-seed weight varied between varieties, the pasture type being inferior to the two hay types.

A study was made of phenotypic correlations between seed yield and some other traits of determinable value in two bred varieties of meadow fescue. Seed yield and such characters as time of flowering, plant height and dry-matter production of spaced plants in October and at the time of anthesis were found to be closely associated.The method of path coefficient analysis was used to study the interrelationships of seed yield and its components in the causal system: fertile tiller and floret numbers, percentage fertility and thousandseed weight. This analysis corroborates the result of simple correlations in showing the importance of fertile tiller numbers and fertility as major determinants of seed yield.

This investigation was carried out with leafy varieties of perennial ryegrass at four centres near Aberystwyth over a period of three years. Nitrogen applied in spring increased lodging, the number of fertile tillers and also seed yield, except where sowing was made without a cereal cover crop. While reasonable yields were obtained from badly laid crops, it is suggested that lodging prior to anthesis has an adverse effect on seed yield. Clover included as a companion species tended to depress the yield of seed and increase the degree of lodging.Seed rate had less infiuence on lodging and seed yield, but the data show that similar yields were obtained from light seed rates of 4 lb. per acre as from the heavier rates. On the other hand, there was a tendency for the low rate of seeding to produce weedy plots, especially when sown in the absence of a cover crop.

No abstract