The marine isopod genus Limnoria contains algae-eating species. Previous phylogeographic studies have suggested that Limnoria species feeding on buoyant kelp underwent low genetic differentiation on a large spatial scale because rafting on floating host kelps promotes high levels of gene flow. In this paper, we survey the genetic structure of Limnoria nagatai, which bores into the non-buoyant kelps Eisenia bicyclis and E. arborea. We analyze the mitochondrial DNA (cytochrome oxidase subunit I [COI] gene) and morphological traits of L. nagatai, and the host kelps E. bicyclis and E. arborea from 14 populations along the Japanese archipelago of the Pacific Ocean and the Sea of Japan. Four major lineages are recognized within L. nagatai: three lineages in the Pacific Ocean, and one lineage in the Sea of Japan which might be a cryptic species. For L. nagatai, we show high genetic differentiation between geographically separated habitats in the Pacific Ocean, while low differentiation is found among continuous host kelps habitats in the Pacific Ocean as well as the Sea of Japan. L. nagatai in E. bicyclis in the Pacific Ocean has experienced large population expansion after the Last Glacial Maximum (LGM), whereas the lineage in E. bicyclis in the Sea of Japan has not. We suggest that Limnoria feeding on non-buoyant kelps, may attain low genetic differentiation because they might be able to disperse long distance if the habitat of host kelps is continuous. The historical events affecting Limnoria after the LGM may differ between the coasts of the Pacific Ocean and the Sea of Japan.

Plant-pollinator interactions have usually been researched in flowers that have fully opened. However, some pollinators can visit flowers before full opening and contribute to fruit and seed sets. In this paper, we researched the pollination biology of flowers just starting to open in four field experiments. We observed the insect visitors to Lycoris sanguinea var. sanguinea for 3 years at five sites. These observations revealed that only small bees, Lasioglossum japonicum, often entered through tiny spaces between the tepals of 'breaking buds' (i.e. partially opened flowers) and collected pollen. We hypothesized that they can pollinate this species at the breaking-bud stage, when the stigma is located near the anthers. To measure the pollination effect of small bees at the breaking-bud stage, we bagged several breaking buds after small bees had visited them and examined whether these buds were pollinated. In bagging experiments, 30% of the breaking buds set fruit and seeds. Fruit-set ratios of the breaking buds did not differ significantly from those of the fully opened flowers, which had been visited by several insect species. We also counted the pollen grain numbers on the body of L. japonicum and on the anthers of randomly-selected and manipulated flowers. These experiments revealed that all of the captured bees had some pollen of target plants and that L. japonicum collected most of the pollen grains at the breaking-bud stage. Our results showed that the new pollination process, breaking-bud pollination, happened in breaking buds by L. japonicum, although there is no evidence to reveal that this is the most effective pollination method for L. sanguinea var. sanguinea. In principle, this new pollination process can occur in other flowering plants and our results are a major contribution to studies of plant-pollinator interactions.

Plant-pollinator interactions have been studied for approximately one hundred years. During that time, many field methods have been developed to clarify the pollination effectiveness of each pollinator for visited flowers. Pollinator observations have been one of the most common methods to identify pollinators, and bagging and cage experiments have been conducted to show the effectiveness of specific pollinators. In a previous study of Lycoris sanguinea var. sanguinea, its effective pollinators, the visitation frequencies of each floral visitor, and its reproductive strategies were not identified. This study reports the observation that small bees visited flowers that were partially opened (breaking buds). To the best of our knowledge, this phenomenon has not been reported previously. Further, this study investigates the hypothesis that small bees can pollinate at that flowering stage. This study demonstrates the basic methods of field experiments in pollination ecology using L. sanguinea var. sanguinea. Pollinator observations and digital video showed the visitation frequencies of each floral visitor. Bagging and cage experiments revealed that these flowers could be pollinated fully and that breaking-bud pollination could be important for the pollination of this plant species. The advantages and disadvantages of each method are discussed, and recent developments, including laboratory experiments, are described.

No abstract