Germinability and the rate of germination were related to seed mass in Lithospermum arvense L. and Anchusa arvensis{L.) Bieb. Prior to germination experiments, seeds from two populations of each species had been sorted into nine similarly sized groups according to their seed mass. Germinability varied greatly among the different seed-mass fractions in L arvense, while differences were smaller in A. arvensis. Germination rate varied with seed mass in some cases. However, in many cases where clear patterns in the relationship between germination characteristics and seed mass were found, these patterns differed both between species and between populations. Furthermore, the character of the relationship varied depending on the experimental conditions used. It is hard to identify the cause of these relations since within-and between-plant variability can affect both seed mass and germination characteristics independently. It is concluded that the results of a germination test can differ depending on the seed mass fractions used, but that the magnitude and direction of such differences are difficult to foresee. Therefore, seed batches should be prepared with care, and experiments designed with awareness of the importance of the germination conditions chosen.

No abstract

No abstract

Five different barley cultivars, including covered and naked samples containing low, normal, or high‐amylose starches were fractionated by weighing, pneumatic classification, sieving, or sorting on a specific gravity table, and analyzed for content of starch, protein, ash, and β‐glucan. For ash content, almost no variation could be found between different fractions. Protein content was minimum in the intermediate fractions for all cultivars when sorted by weighing. For the other fractionation methods, the differences in protein content were small. A tendency for decreasing content of starch with increasing grain mass and size could be seen when fractionating grains by weighing and sieving, respectively. The clearest trend was seen in differences in β‐glucan content for all cultivars and all methods used. The main interpretation of our results is, however, that the chemical composition within the cultivars studied is very similar for all fractions, and that the differences between the unfractionated barley samples are larger.