to study the germination capacity of Cryptogramma crispa, spores were cul-'etri dishes with nutritive medium solidified with agar. Germination was and 25°C, and, as in most of homosporous ferns, the germination optunurr bove 20°C. Two light intensities were used, 10 and 40 jjuEm -^ -1 , to reproduce le spores falling on open sites or in rock cracks or hollows. A lower light germination. After sowing, some plates were kept at chilling and other alIn the Iberian Peninsula (Spain and Portugal), Cryptogramma crispa R. Br. grows on ecologically very particular and well defined habitats. It occupies especially siliceous stone fields, such as granite, gneiss, sandstones, quartzites, or slates in high mountain zones, usually above the timber-line. In the Iberian Peninsula, its optimum is about 2000 m. The plants grow preferentially in rock fissures or cracks and in hollows between rock blocks. In these habitats, which are mostly over 2000 m elevation, the growing season may be very short, scarcely two months in extreme conditions, and usually no more than four months (Rivas-Martinez, 1987). The considerations of the distribution of pteridophytes suggest the need for more detailed investigations on the life-cycles of species to determine the importance of specific variations in the life-cycle in limiting the distributions of plants. Variations in the distributions of species might be accounted for by random processes, such as dispersal, or in a more deterministic manner by subtle and specific variations in life cycle characteristics (Woodward, 1987).In ferns, it is important to study the factors that can affect the development of the gametophyte that would lead to the establishment of the sporophytic gen-Probably due to the short period to complete their development, C. crispa sporophytes reach the end of the summer with practically all the spores still retained (Peck et al., 1990; pers. obs.), which are released at about the same time that the leaves shed. Thus, the spores are released at the end of the growing season.It is difficult to reproduce wild conditions in short-term laboratory experiments. But some climate changes, especially temperature ones, that may influence spore germination can be tested in the laboratory, provided that in nature conditions may be operating over a longer period. Thus, a few experiments

The reproductive biology of Cryptogramma crispa, a tetraploid species with a broad circumboreal and alpine distribution, growing mainly in siliceous boulder fields and crevices, was studied in the laboratory by growing gametophytes in plates with both solidified agar media and sterilized soil. In addition, an electrophoretic study of isozymes was carried out on frond samples from five natural populations, as an additional source of evidence concerning the breeding system and the genetic structure of sporophyte populations. Populations throughout the Iberian range of the species were selected for this study, and a Scottish population was included to represent plants from outside our local area and ecology. The morphological development of gametophytes is of the Adiantum type. All multispore cultures developed into a bigametophytic system, consisting in most cases of male and female prothalli. This pattern of sexual expression provides evidence for outcrossing as the main breeding system in this species. Moreover, there is good evidence that the species possesses an antheridiogen system to promote outcrossing. The long time needed by gametophytes to produce gametangia, and afterwards to fertilize and produce sporophytes, might be the primary reason why so few young sporophytes are found in the wild. The values of the percentage of polymorphic loci and the similarity levels obtained from the isozyme analyses indicate a level of genetic variability that would be expected in an outcrossing species. All these characteristics are usually associated with diploid fern species rather than polyploid species.