Phylogenetic analyses of the Dasyaceae based on sequence analysis of the large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase (rbcL) and 42 morphological characters are presented. Comparative sequence analysis confirms the general view of the Ceramiaceae as a primitive, paraphyletic group giving rise to the Rhodomelaceae, Delesseriaceae and Dasyaceae within the monophyletic Ceramiales. On the basis of both data sets, the Heterosiphonia-like genera (Heterosiphonia, Colacodasya and Dasyella) are the most primitive members of the Dasyaceae, whereas the Dasya-like genera (Dasya, Pogonophorella, Eupogodon and Rhodoptilum) and Thuretia and Dictyurus are of more recent origin. On the basis of morphological data only, Thuretia and Dictyurus form a sister group to Heterosiphonia, and Eupogodon is monophyletic whereas Dasya and Heterosiphonia are not. Primary radial symmetry has arisen once in the Dasya clade but is secondarily obscured in some species by heavy, asymmetrical cortication that gives the appearance of bilateral symmetry. This is illustrated by species of Eupogodon and Rhodoptilum.

We report the cloning and characterization of a 1881 bp Thalassiosira weissflogii (Bacillariophyceae, Centrales) plastid genomic fragment containing the ferredoxin gene (petF) and several other genes. The organization of this fragment is exactly the same as its equivalent region from the plastid genome of another centric diatom, Odontella sinensis, and the sequences share high homology in both coding and non-coding regions. Traces of a probable vestigial protein-coding gene and of a partial trn gene are found just after the trnR gene in both diatoms as well as in the plastid genome of the red alga Porphyra purpurea, illustrating the close relationship between red algal and heterokont algal plastid genomes and arguing in favour of the origin of the latter by secondary endosymbiosis. In accordance with phylogenetic analyses of most plastid genes, phylogenetic trees inferred from 69 ferredoxin sequences show that both the green and the red algal plastid genes arose from cyanobacterial genes; our results furthermore suggest that each of these plastid lineages is related to different extant cyanobacterial genes, implying either that different endosymbiotic events were responsible for the origin of green and red plastids or that each lineage kept only one (and a different one) of several paralogous cyanobacterial genes present in the ancestral cyanobacterial genome.

A new large DNA virus (PlitV-1) infects the marine filamentous brown alga Pilayella littoralis. It was collected in Alaska and infects other P. littoralis isolates of different geographic origin. The virus has an icosahedral capsid of c. 161 nm in diameter, enclosing an electron-dense core. The genome consists of double-stranded DNA and is approximately 280000 base pairs in size. The virus is latent in somatic cells of the host and is propagated only upon induction of the host's reproductive organs. It causes deformed sporangia, resulting in infertility, and is structurally similar to other brown algal viruses. PCR amplification of a genomic sequence coding for part of a structural glycoprotein of the Ectocarpus siliculosus virus EsV-1 produced a fragment of similar size to that obtained with EsV-1.

Small-subunit ribosomal DNA (SSU) sequences for 14 members of the rhodophyte order Acrochaetiales were used to generate phylogenetic trees to determine whether the Acrochaetiales is monophyletic and to assess the relationships of acrochaetioid algae to closely related taxa. Within the Acrochaetiales, two strongly supported groups are recognized: one group that includes, in addition to three other species, the type species of the genera Acrochaetium, Audouinella and Rhodochorton; and a second group that includes an unequivocal representative of Colaconema and nine additional species. The relationships of these groups with the Batrachospermales, Palmariales and Nemaliales were not resolved and the possibility that the Acrochaetiales is polyphyletic was not eliminated. Resolution of relationships for species within these groups was strong in only a few cases, with many of the phylogenetic issues at this level remaining equivocal. Our results do not position Camontagnea in the Acrochaetiales, as has recently been suggested, but strongly ally this genus to the family Rhodothamniellaceae of the Palmariales. The phylogenetic position of Rhododraparnaldia, an alga intermediate between the Acrochaetiales and Batrachospermales, remains unresolved. These molecular results provide a foundation for assessing the taxonomic significance of phenotypic characters (e.g. anatomy, life histories, morphology, phycoerythrin type and plastid details) variously considered taxonomically significant in the Acrochaetiales.

Four populations of the newly described freshwater red algal species Batrachospermum spermatoinvolucrum were sampled from British Columbia on the west coast of North America and Labrador on the east. The rDNA internal transcribed spacer (ITS) 1 and 2 sequences, RUBISCO large subunit (rbcL) gene data for three of the populations and morphometric characteristics were compared with those of B. gelatinosum. In addition, the key feature distinguishing B. spermatoinvolucrum, spermatangia on the involucral filaments of the carpogonial branch, was investigated seasonally in one British Columbia population of this species. The ITS analysis suggested that the B. spermatoinvolucrum populations were more closely related to particular B. gelatinosum populations than to each other and did not form a monophyletic clade. Two populations of B. spermatoinvolucrum from Labrador and British Columbia had identical rbcL sequences that were distinct from that of B. gelatinosum. However, the rbcL sequence from the third population of B. spermatoinvolucrum differed by only one nucleotide substitution from the B. gelatinosum populations. The tree generated from the morphometric data showed two groupings, each containing populations of both taxa. The presence of spermatangia-bearing involucral filaments was a stable character throughout the year. On the basis of these findings, B. spermatoinvolucrum is reduced to a form of B. gelatinosum. This form would encompass variant populations with characteristic carpogonial involucral filaments bearing spermatangia.

The algal vegetation of five diving profiles situated at exposed sites of the outer Öregrund archipelago, southern Bothnian Sea, Sweden, was investigated by SCUBA diving in 1996. The vegetation of the same profiles had been described earlier in 1943–4 and 1984. In the present study, special attention was paid to the occurrence of the structurally important belt-forming brown algae Fucus vesiculosus and Sphacelaria arctica in comparison with the situation 52–53 and 12 years earlier. Prominent differences between the 1940s and 1996 were found. The weighted average depth of the F. vesiculosus belt in 1996 was about 1·7 m shallower than in the 1940s, and the lower distribution limit of this species was about 2·5 m shallower. At its lower end the Fucus belt was replaced by epilithic red algae, mainly Furcellaria lumbricalis, Rhodomela confervoides and Polysiphonia fucoides. These results were similar to those of the 1984 study and show neither an improvement nor a deterioration of the belt during the past 12 years. In four of the five diving profiles, a conspicuous belt of S. arctica occurred in the lower sublittoral, similar to that observed in the 1940s, but in one of the five profiles the S. arctica belt was heavily impoverished. These results contrast with the 1984 study which reported that S. arctica had practically disappeared from all profiles. The epiphytic cover of annual filamentous algae, dominated by Pilayella littoralis and Ceramium gobii, was higher in 1996 than in 1943–4. The results of this comparative study are discussed in relation to the general eutrophication of the Baltic Sea.

In this paper we quantify the interactive effects of wave exposure and intraspecific density on the survivorship and growth of Sargassum muticum in northern Spain. Juveniles settled onto slate plates were experimentally thinned to two densities and transplanted to exposed and sheltered shores. Initially, plants grew faster at the exposed sites but final lengths were similar between exposures after 5 months, largely because of breakages at the top of the main primary lateral of plants at exposed sites. Mean weights of plants did not significantly differ between exposures. The results indicate that S. muticum juveniles growing at high density were longer and thinner but similar in weight to those at low density. Patterns of growth of larger plants transplanted between sheltered sites similarly suggested changes in morphology of S. muticum in response to differences in density. Survivorship of juveniles did not differ between exposures but was significantly poorer at the higher density irrespective of wave exposure. The results of this and previous studies suggest that the abundance of S. muticum in the low intertidal area of exposed shores in northern Spain is limited at several life-history stages. The relative importance of these potential bottlenecks compared with the availability of propagules remains unknown.

The identity of Dinophysis sacculus Stein has proved to be problematic in recent years, during which this species has been associated with diarrhoetic shellfish poisoning (DSP) events along European coasts. To complement the scarce and at times contradictory existing information, we provide details of morphological variation in D. sacculus throughout its geographical range, which includes the Mediterranean Sea and European Atlantic coasts. Our data clarify its relationship with D. pavillardii Schröder. This species differs from D. sacculus only by the presence of a concavity along the dorsal margin, but transitional morphotypes are frequent in almost all the samples examined. We therefore propose merging the two species and classifying morphotypes with a dorsal concavity as D. sacculus f. reniformis (Pavillard) comb. nov. Despite the high variability of Dinophysis sacculus, it was generally possible to separate it from D. acuminata Claparède & Lachmann. The only character distinguishing the two species is the shape of the large hypothecal plates, which in D. sacculus are elongate, almost rectangular or sac-like, whereas in D. acuminata they are shorter, more convex dorsally and often more slender towards the antapex. Thecal areolation and sulcal list ornamentation are more pronounced in D. acuminata, although these characters are somewhat variable. Other thecal details, such as the shape of sulcal platelets and the structure of the apical pore area, were not useful for circumscription of the two taxa. In a few samples from the Galician Rias and from the Spanish Mediterranean coast, we observed morphotypes intermediate between D. sacculus and D. acuminata, which indicates that the two species can have slightly overlapping ranges of variability.