The absence of bumblebees on an oceanic island blurs the species boundary of two closely related orchids.

Oceanic islands offer valuable natural laboratories for studying evolution. The Izu Islands, with their recent geological origin, provide an exceptional opportunity to explore the initial evolution on oceanic islands. Another noteworthy aspect is the absence of bumblebee species on most Izu Islands. We used ecological, morphological, and molecular data to investigate the impact of bumblebee absence on the evolution of two closely related orchid species, Goodyera henryi and Goodyera similis, focusing on Kozu Island, the Izu Islands. Our investigation revealed that while G. henryi exclusively relies on a bumblebee species for pollination on the mainland, G. similis is pollinated by scoliid wasps on both the mainland and the island. Intriguingly, all specimens initially categorized as G. henryi on Kozu Island are hybrids of G. henryi and G. similis, leading to the absence of pure G. henryi distribution on the island. These hybrids are pollinated by the scoliid wasp species that also pollinates G. similis on the island. The absence of bumblebees might result in sporadic and inefficient pollination of G. henryi by scoliid wasps, consequently promoting hybrid proliferation on the island. Our findings suggest that the absence of bumblebees can blur plant species boundaries.

Morphological, ecological, and molecular phylogenetic approaches reveal species boundaries and evolutionary history of Goodyeracrassifolia (Orchidaceae, Orchidoideae) and its closely related taxa.

Species delimitation within the genus Goodyera is challenging among closely related species, because of phenotypic plasticity, ecological variation, and hybridization that confound identification methods based solely on morphology. In this study, we investigated the identity of Goodyeracrassifolia H.-J.Suh, S.-W.Seo, S.-H.Oh & T.Yukawa, morphologically similar to Goodyeraschlechtendaliana Rchb.f. This recently described taxon has long been known in Japan as "Oh-miyama-uzura" or "Gakunan" and considered a natural hybrid of G.schlechtendaliana and G.similis Blume (= G.velutina Maxim. ex Regel). Because the natural hybrid between G.schlechtendaliana and G.similis was described as G.×tamnaensis N.S.Lee, K.S.Lee, S.H.Yeau & C.S.Lee before the description of G.crassifolia, the latter might be a synonym of G.×tamnaensis. Consequently, we investigated species boundaries and evolutionary history of G.crassifolia and its closely related taxa based on multifaceted evidence. Consequently, morphological examination enabled us to distinguish G.crassifolia from other closely related species owing to the following characteristics: coriaceous leaf texture, laxly flowered inflorescence, long pedicellate ovary, large and weakly opened flowers, and column with lateral appendages. Ecological investigation indicates that G.crassifolia (2n = 60) is agamospermous, requiring neither pollinators nor autonomous self-pollination for fruit set, whereas G.schlechtendaliana (2n = 30) is neither autogamous nor agamospermous but is obligately pollinator-dependent. MIG-seq-based phylogenetic analysis provided no evidence of recent hybridization between G.crassifolia and its close congeners. Thus, molecular phylogeny reconstructed from MIG-seq data together with morphological, cytological, and ecological analyses support the separation of G.crassifolia as an independent species.

Polyploid progeny from triploid hybrids of Phegopteris decursivepinnata (Thelypteridaceae).

Phegopteris decursivepinnata includes diploids, tetraploids, and triploid hybrids based on x = 30. We obtained polyploid progeny from triploid hybrids through selfing and crossing experiments. Triploids occasionally formed well-filled spores. The mean occurrence frequencies of well-filled and germinated spores were 2.8% and 0.8%, respectively. Viable spores that succeeded in germinating were regarded as unreduced, triploid spores, because the resulting gametophytes yielded triploid (2n = 86-92) and hexaploid (2n = 170-184) progeny in both isolated and mixed cultures of gametophytes. The triploid and hexaploid progeny likely arose apogamously and sexually, respectively. One of the hexaploid progeny yielded hexaploid sporophytes (2n = 169-180) in the mixed culture of its gametophytes. Artificial crossing between triploid and diploid sporophytes produced tetraploid (2n = 116, 120) and pentaploid (2n = 145-150) progeny that likely arose through the mating of 3x gametes from the triploid with both 1x and 2x gametes from the diploid, respectively. Unreduced spore formation was confirmed in diploid sporophytes. The tetraploid progeny formed viable spores at a frequency of 63-75%. Triploid hybrids of this species are thus expected to produce new triploids, tetraploids, and hexaploids in nature. The wide range of variation in chromosome numbers of hexaploid progeny suggests that viable spores from parental triploid hybrids had unreduced chromosomes, whose numbers, however, deviated considerably from those of the hybrids. This chromosome deviation of viable spores may result from errant movements of chromatids of univalents when unreduced dyads form in meiosis. Downward chromosome deviation from the chromosome number of the parental hybrids may affect the developmental progress of viable spores more tolerantly than upward chromosome deviation.

Evidence of genetic segregation in the apogamous fern species Cyrtomium fortunei (Dryopteridaceae).

In apogamous ferns, all offspring from a parent are expected to be clonal. However, apogamous 'species' frequently demonstrate a large amount of morphological and genetic variations. Cyrtomium fortunei composed of four varieties (C. fortunei var. fortunei, var. clivicola, var. intermedium, and var. atropunctatum), is all reported to be apogamous triploids, but demonstrates large and continuous morphological variation. In previous studies, we showed that considerable genetic diversity was observed in many local populations of the apogamous fern 'species'. We hypothesized that genetic segregation has occurred, because neither sexual type nor intraspecific polyploidy have been observed in C. fortunei in Japan. Of 732 progeny examined (250 gametophytes and 482 sporophytes), obtained from a parental sporophyte whose pgiC genotype was estimated as aab, 11 (4.4%) gametophytes and 8 (1.7%) sporophytes showed a different genotype (aaa) from that of the parent sporophyte. We showed that genetic segregation occurs in apogamous C. fortunei in relatively high frequency. Moreover, we could first show that the segregation frequency in gametophytes is significantly higher than that in sporophytes of the next generation (χ² = 4.90, P = 0.027). It may suggest the existence of deleterious genes, which are expressed during the morphogenesis and growth of sporophytes.

Two types of partial fertility in a diploid population of the fern Thelypteris decursive-pinnata (Thelypteridaceae).

Two types of abnormal sporophytes were observed in a population of diploid Thelypteris decursive-pinnata. Most sporophytes in this population exhibited regular chromosome pairing, resulting in the formation of 30 bivalents in meiosis I; however, they produced abortive spores to various degrees. Some formed large globose spores at low frequencies, most likely to be unreduced diplospores. The other type of abnormal sporophyte underwent synaptic failure to form 60 univalents at meiosis I, but produced fertile spores, mostly large globose ones at low frequencies. The globose spores were considered unreduced diplospores because the gametophytes arising from them produced tetraploid sporophytes by gametophytic selfing. One tetraploid formed only univalents at meiosis I. Allozyme variation was not detected in this population, although neighboring ordinary diploid populations exhibited it to a certain degree. The sympatric occurrence and allozyme uniformity of the two groups suggest that both are offspring of a founder sporophyte, which may have possessed two types of mutated recessive genes responsible for the spore sterility and the synaptic failure in meiosis. Unreduced spores formed by these two types may play an important role in the polyploid speciation of this species.