Kaz Ohashi

@kazohashilab.bsky.social

882 Followers 644 Following

Pollinator foraging behavior & floral evolution | Assistant Professor @UofTsukuba | AE @jpollecol.bsky.social | Author of “Trade-off mitigation” paper (Biol Rev 2021): bit.ly/TradeoffMitigation | ohashilab.com

Book cover of the Japanese edition of Our Fragile Moment: How Lessons from Earth’s Past Can Help Us Survive the Climate Crisis by Michael E. Mann. The cover features the large Japanese title on a white background above an illustration of the Earth with the original English title overlaid. A blue-and-yellow obi (book band) includes the message, “It is not too late—we still have the means to act,” and notes that the book is recommended by Al Gore. Published by NTS.
A mating pair of Japanese Oakblue butterflies (Narathura japonica) on the ground, both with their wings open. The upper butterfly (male) is fully visible, displaying brilliant iridescent blue wings with black borders. The lower butterfly (female) is partly hidden beneath it, with part of its blue wings also visible.
A collage photo of insect pollinators on the while tubular flowers of Ligustrum obtusifolium, including beetles, bumble bees, butterflies, a Noctuid moth, a hover fly, and a blow fly.
A queen of Bombus diversus diversus looking for a place to build a nest.
Functional Ecology’s web page where they show 13 short-listed papers for Haldane Prize 2025.
Four swans swiftly descending toward a pond, with a quote by Gary Snyder reading, “Nature is not a place to visit; it is home.”
A clear-wing hawkmoth (Cephonodes hylas) oviposits on a young shoot of Gardenia jasminoides, its larval host plant. The moth’s body is sharply lit against the glossy green foliage, emphasizing its fuzzy yellow thorax and striped abdomen.
A close-up photograph of a Rosa multiflora blossom being visited by a hoverfly, Episyrphus balteatus. The fly’s translucent wings and vivid orange-and-black striped abdomen contrast beautifully with the delicate white petals and bright yellow stamens of the flower.
A huge ground blue gentian with more than ten flowers.
A male carpenter bee hovers mid-air with its wings outstretched. Its body is robust, with a golden-brown thorax and a shiny, dark abdomen.
A blooming Arisaema urashima. The plant displays its characteristic dark spathe and a long, thread-like appendix curling upward.
Two male Eucera bees resting on bright green lance-shaped leaves.
Two bee-fly-like insects with fuzzy brown and yellow bodies are mating on a rough asphalt surface. Their wings are outstretched, and one insect is perched directly behind the other in alignment.
A custom-made mug with an academic paper design. The left side of the mug features the title, abstract, and author name of a scientific article. The right side showcases an image of a bumble bee foraging on artificial flowers in an experimental setup, along with a graph displaying research results. The mug has an orange handle and rim, sitting on a wooden table with a black chair and green cabinets in the background.
The cover of the latest issue of Functional Ecology. The featured photo shows the indoor experiment using bumble bees (Bombus ignitus) and artificial flowers. The worker gathers sugar solution from an artificial “flower” made of a transparent acrylic jar with a circular-coloured paper attached to the inside.
A close-up photo of a Japanese butterbur (Petasites japonicus) bud emerging from the ground. The green, tightly packed flower buds are surrounded by fresh, bright green leaves, contrasting with the dry brown leaves and grass in the background. Sunlight highlights the textures and details of the plant, signaling the arrival of spring.
A close-up photo of a Japanese butterbur (Petasites japonicus) bud emerging from the ground. The green, tightly packed flower buds are surrounded by fresh, bright green leaves, contrasting with the dry brown leaves and grass in the background. Sunlight highlights the textures and details of the plant, signaling the arrival of spring.
A figure showing how bees adjust their flower constancy based on environmental factors. It consists of four diagrams showing how travel costs (blue), cognitive costs (orange), and total costs (grey) change w/ flower constancy. The red inverted triangles show the optimal level of constancy that minimizes the total costs. When spp. are mixed in space, more constant foragers benefit from reduced cognitive costs but face higher travel costs from skipping more flowers of different spp. When spp have distinct traits, the optimal constancy is intermediate, balancing both costs, but when spp. are similar, cognitive costs are minimal, leading to lower constancy. In patchy environments, constant foragers reduce travel costs and still avoid cognitive costs, resulting in higher flower constancy, regardless of flower trait differences. These patterns match the results in our experiments on spp. mixing and flower color similarities.
(Left) A worker of Bombus ignitus collecting nectar from an array of artificial flowers where blue and yellow flowers are evenly mixed. (Right) Realized flower constancy in our experiments and simulations under three levels of spp. mixing. Gray circles represent results from the experiment w/ blue vs. yellow flowers (dissimilar color pair), and black circles represent those from the experiment w/ yellow vs. golden yellow flowers (similar color pair). Each red bar represents the 5th to 95th percentile range of 500 CI values generated by the simulation, assuming no cognitive costs but travel costs. Realized flower constancy dramatically decreases as flower spp. become more spatially mixed, especially when they look alike, suggesting this behavior is a dynamic optimal strategy rather than a static result of cognitive limitation.
The title of the paper published in Functional Ecology, "Realized flower constancy in bumble bees: Optimal foraging strategy balancing cognitive and travel costs and its possible consequences for floral diversity," authored by Kentaro Takagi and Kazuharu Ohashi.
Two blue-green hummingbirds with white breasts are near three feeders—one in the middle, perching, and the other on the right side, hovering.
The title of the paper, 'A test of the Grant–Stebbins pollinator-shift model of floral evolution,' published in New Phytologist.
Pink morph of Hepatica nobilis var. japonica f. magna.
Three white swans are about to land on the water.
Four white swans in flight, gracefully soaring through a soft blue and cloudy sky.
Purple Erythronium japonicum in bloom, set against a soft forest bokeh.
Cover of Silent Earth by Dave Goulson: a dark background with colorful insects and plants arranged in a circle, highlighting the book’s focus on the insect apocalypse.
A hawkmoth with very green body and clear wings. Actually, the wings are still covered with golden scales bc it was just hatched.
The title of the paper: "Competition for pollen deposition space on pollinators generates last-male advantage".
Tundra swans fly above Sugao-numa Pond in Japan, where they spend the winter.
Yellow flowers of wintersweet (Chimonanthus praecox). A fly (Stomorhina obsoleta) visits the flower, actively probing with its proboscis.
Huge flower heads of Cirsium purpuratum and a bumblebee (Bombus honshuensis). This photo was used as the cover image for the issue featuring my very first publication in the American Journal of Botany 1998.
Title page of the editorial, "How worthwile are pollination networks?" by James D. Thomson.
Counterclockwise from the top left: Elephant hawk moth (Deilephila elpenor), impatiens hawk moth (Theretra oldenlandiae), burnt-spot hummingbird hawk moth (Macroglossum pyrrhosticta), and coffee bee hawk moth (Cephonodes hylas). The first two species are nocturnal, while the latter two are diurnal.
The title slide of my talk, 'Trade-off mitigation: a key concept to understand the evolution and function of convergent floral phenotypes'
Left: Day/night pollinator exclusion experiment. (a) Focal populations used in pollinator exclusion experiment; inset shows species distributions. Species identity of populations is denoted by the first letter of the population abbreviation (S = C. sessiliflora, C = C. citrina, and L = C. lindheimeri). Populations represented by circles have long corolla tubes; those shown by triangles have short tubes. Blue points show populations where the most frequent visitor was nocturnal (Hyles lineata hawkmoths in all cases); orange points are populations where the most frequent visitor was diurnal. (b) Representative photographs of long (i, SBL) and short (ii, SIC) corollas. (c) Hyles lineata hawkmoth probing a long corolla at SBL. The arrow shows the hawkmoth's head contacting the plant's reproductive organs at the mouth of the corolla. (d) Inflorescence bagged to exclude pollinators at SIC. Right: Corolla length, pollinator visitation, and fruit set by population. (a) Corolla length measurements by population, points show mean ± standard deviation. Note SILB12 and SILB13 share the same floral measurements taken from population SILB in 2017 (values plotted twice to align with b, c, and d). (b) Number of floral visits by pollinator functional group observed during exclusion experiments. Diurnal groups are shown in warm colors, nocturnal groups in cool colors. (c) Difference in average fruit set to night-open and day-open plants [normalized by average fruit set to open-treatment plants: that is, (Night FS/Open FS) − (Day FS/Open FS)]. (d) Fruit set to open-treatment plants: individual values (gray points) and mean ± SE (black points and bars). The shape of point shows corolla length category and color shows activity period of the most frequent visitor. For all plots, populations are approximately ordered by species and from south to north.
A male Bombus honshuensis visiting Cirsium makinoii.
Left: A hawkmoth (Macroglossum stellatarum; photo: Andrew Dejardin) approaching a flower and dipping its proboscis into a nectary. Right: Movements of the galeae during fluid uptake. (A) Reversible ‘cracking’ of the dorsal legular band at different locations was observed when the proboscis of M. rustica was lowered into water. The ‘crack’ or opening in the dorsal legular band allowed water to rush into the food canal. The crack was closed again in frame 4. Numbers 1–4 show the sequence of events. (B) Anti-parallel sliding of the galeae in Eumorpha pandorus. Arrows indicate the direction of the movement of each galea. Movie 2 shows the process. (C) Movement of air bubbles (arrows) can be seen in the proboscis of M. sexta (1 and 2) and Manduca quinquemaculata (3 and 4), suggesting that the liquid column in the food canal is partitioned as bubble trains.
Left: Time-calibrated phylogeny of the 699 tree species (database I) used in this study provided by Scenario 3 of S.PhyloMaker. Branch color denotes mycorrhizal type (orange, AM; blue, EcM; green, AM + EcM). Colors of inner and outer labels ringing the tree indicate seed dispersal mode (red, endozoochorous; blue, anemochorous; yellow, unspecialized) and pollination mode (pink, biotic; blue, abiotic). The bar represents 40 million years. Right: (a) Maximum and (b) mean seed dispersal distance of each seed dispersal mode. (c) Maximum and (d) mean pollen dispersal distance of each pollination mode. The central black line in the box plots indicates the median, and the box indicates the upper and lower quartiles. Different letters and asterisks indicate that values differ significantly at P < 0.05 (phylogenetic generalized least squares analysis).
Left: Length and width dimensions of the proximal corolla (a) and the corolla as a whole (b) in Digitalis purpurea. PCL, proximal corolla length; PCW, proximal corolla width; WCL, whole corolla length; WCW, whole corolla width. Right: Phenotypic (upper diagonal elements) and genetic (lower diagonal elements) correlation matrices for the seven morphological and nectar traits analysed in Digitalis purpurea. For phenotypic correlations, significant P-values are represented with asterisks (*: <0.05; ***: <0.001) and numbers show Pearson's correlation r values. For genetic correlations, values show the median of the posterior distribution of the genetic correlations extracted from the multivariate animal model. Dots indicate that the 95% credible intervals do not include zero. CON, nectar sugar concentration; NPR, nectar production rate; PCL, proximal corolla length; PCW, proximal corolla width; SUG, total sugar production; WCL, whole corolla length; WCW, whole corolla width.
Left: a Bombus flavescens collecting pollen; b Bombus flavescens robbing nectar; c Apis cerana scavenging nectar slipping to the entrance of the floral chamber; d Apis cerana collecting pollen; e Cechetra scotti, with legs contacting the anthers and stigma; f Cechetra sp. (scotti or lineosa), with viscin threads indicated by the red arrow; g Acosmeryx naga, with viscin threads indicated by the red arrow; h Notonagemia analis, with legs contacting the anthers. Right: The relative importance of bumblebees and hawkmoths as pol- linators indicated by different proxies for pollinator importance in 2020. The data of visitation rate in 2019 were presented in the text (see Results). Columns with bars indicate mean±SE. The numbers in the column refer to sample sizes, i.e. how many 1-hour observation intervals were made for visitation rates (a) and how many flowers are involved in those treatments (b, c, d). The flowers used to investigate pollen deposition were from 11 individuals. The flowers used to examine fruit set, seed production, and seed production were from 23 individuals. The “Visits per hour” means the number of visits to individual flowers per 1-hour interval within the observed patch con- taining 20 flowers. We use > and < to indicate “significantly more” and “significantly less”. There is no significant difference if neither of them is used. The analysis of visitation rate in A did not include visits by nectar robbers.
Erythronium japonicum
Above: (A) Three species of artificial flowers. From left to right: blue, yellow, and golden yellow. (B) colour dimorphic arrays used in the experiment. From left to right: evenly mixed, moderately patched, and highly patched arrays. Filled and unfilled circles represent yellow and blue (or golden yellow) flowers. Below: Realized flower constancy in two experiments and simulations. The horizontal axis represents the level of species mixing (EM=evenly mixed, MP=moderately patched, HP=highly patched). Grey and black circles represent the first (large color difference: blue and yellow) and second experiments (small color difference: yellow and golden yellow), respectively. Red bars represent the 5th to 95th percentile ranges of CI values generated by computer simulation assuming no cognitive costs.
Scenario 1: Colony nectar and pollen demand (left y-axis), and Scenario 2: Impact of 14-day forage gaps on queen production (right y-axis) for (a) Bombus terrestris and (b) B. pascuorum. Model Scenario 1 (left y-axis): A single food source was present, providing a constant amount of nectar and pollen on each day, and weather conditions allowed 8 h of foraging every day. We show nectar (yellow line) and pollen (orange line) collection (±SD) (relative to the daily available nectar or pollen) of a bumble bee population (all nests present in the model landscape) during year 10 of a 10-year simulation, when we assume that the simulation has reached landscape carrying capacity. On the left y-axis, 1 indicates that the bees are consuming all available resources (model replicates, N = 20). Model Scenario 2 (right y-axis): The total number of hibernating queens produced by the end of the season if a two-week forage gap occurred, starting on a given date during the simulation. Forage gaps start on day 1 (January 1), day 8, day 15, day 22, etc. (model replicates per forage gap date, N = 20).
Relationships between hysteranthy and environmental and biological traits for the 13 species of the American plums. Panel (a) shows the estimated effects of each predictor and their interaction on the likelihood hysteranthy. Points indicate the mean effects and the thick and thin bars represent the 50% and 89% uncertainty intervals, respectively. Panel (b) depicts the conditional effects of each predictor on hysteranthy likelihood. Blue lines indicate the mean estimate and grey fill the 89% uncertainty intervals. Predictor values (x-axis) are in standardized units (z-scored) to allow direct comparisons between predictors. PDSI, Palmer Modified Drought Index.
Figs. 1-5: 1) The optimal allocation of foragers to complex flowers for varying time horizons (colony longevity). The arrows indicate the end of the time horizon for the different values of T. 2) The optimal number of foragers on complex flowers for varying proportion of foragers in the colony (1 − r). 3) The optimal allocation of foraging effort on complex flowers for varying colony sizes (N). 4) The optimal number of foragers on complex flowers at varying learning rates β. 5) The optimal number of foragers on complex flowers at varying probabilities of forager survival (=individual lifespan).
98 full-sib seed families of fast cycling Brassica rapa plants were divided into two replicates (A and B), and spread among treatment groups, with each seed family being represented in each treatment (49 plants were used per treatment per replicate). Differently colored squares represent the two types of soil, tuff- (pale green) and limestone soil (pale blue).
Illustration of the relationship among candidate genes associated with genomic local adaptation to ecological network. Only variables for which a significant enrichment of the selection signature was detected are considered. The left shows the number of candidate genes (set size) identified in genomic local adaptation to the specific variable in GEA analysis. On top, the number of candidate genes associated with a specific variable (single black dot) or shared among variables (multiple black dots linked). The candidate genes are those from the significant zones identified by correcting with the local score method the GEA, and the down and upstream genes.
Cherry trees have started to bloom in Tsukuba, Japan.
A Young (‘green) and old (‘red’) hermaphrodite flowers against the reddish-orange bark of the gynodioecious tree Fuchsia excorticata. B Average reflectance spectra of tree bark, young flowers, transitional flowers, and old flowers.
Types of pollen defences and experimental design for the larval provision manipulation experiment. (a) Osmia lignaria, a generalist species of Osmiini. (b) Artificial nesting structure containing holes lined with paper straws used to collect Osmiini nests. (c) Larva of O. montana on its provision. Larval development and survival were recorded at least every other day. (d) Gas chromatography–flame ionization detector (GC-FID) chromatogram showing the pollen alkaloid profile of a Lupinus argenteus individual (a common host of generalist Osmiini) from our study site showing peaks for α-isolupanine and thermopsine (i.s. = internal caffeine standard), two of the most common quinolizidine alkaloids in L. argenteus pollen. α-Isolupanine and thermopsine were used to manipulate the presence of Lupinus pollen alkaloids in Osmiini provisions. (e) Asteraceae pollen with visible spines, added to provisions for the Asteraceae pollen spines experiment (following treatment shown in (h)). (f) Non-spiny pollen used as a control in the Asteraceae pollen spines experiment. (g) Methodology for manipulating the provisions in the Lupinus alkaloids experiment. (h) Methodology for manipulating the provisions in the Asteraceae pollen spines experiment. Photos and illustrations by Sébastien Rivest.
The number of patch-faithful (lighter shade) and non-faithful (darker shade) observations over time for (A) bumble bees and (B) honey bees.
Photo of vertical array showing 4 Tecoma stans (yellow flowers), 4 T. alata (orange flowers), and 1 test bee in the test arena. Flowers are mounted in customized water tubes that are Velcro mounted to the back wall of the arena. Along the square formed by the flowers, flowers are spaced 7 cm apart on the horizontal and vertical axes.
Mean (+SE) number of pollen grains deposited and germinating among the pollination treatments for nine dominant plant species (a-i). The black diamonds represent the number of pollen grains deposited, and the red circles represent the number of pollen grains germinating on stigmas. Different letters in black or red respectively indicate significant differences using Tukey’s test after GLMM. Mean (+SE) diurnal and nocturnal germination rate of pollen grains deposited on stigmas of 24-h plants (which rely on both nocturnal and diurnal pollinators for reproduction) (j-l). Different letters indicate significant differences.
Graphical abstract of the paper titled, "Pollen diet diversity across bee lineages varies with lifestyle rather than colony size" by Devkota et al. (2024) Journal of Insect Science. The authors found that pollen diet diversity is linked to lifestyle, not colony size.
Results of the experimental assay which tested how the interaction of physical and sensory proximity between deceptive signalling spiders (Gasteracantha fornicata) and flowers shaped capture success. Signalling spiders (n = 24 per treatment group) were paired with inflorescences in a 2 × 2 factorial design to create the four treatments depicted: (1) distant, colour‐matched flowers (2) distant, colour‐mismatched flowers (3) nearby, colour‐matched flowers, and (4) nearby, colour mis‐matched flowers. Letters denote statistically significant differences between treatments identified from post‐hoc comparisons of estimated marginal means. General linear model R 2 = .272.
Inflorescences of Japanese Butterbur (Petasites japonicus). They are one of the earliest flowers in eastern Japan.
Conceptual illustration of the qualitative (A) and the quantitative (B) components of plant–plant pollinator sharing within con- specific and heterospecific contexts. Flower icons represent plant individuals (a–f). The qualitative pollinator sharing between plant indi- viduals (A) is represented with a matrix in which elements depict the presence (1) or absence (0) of shared pollinators between any pair of plant individuals. In this qualitative matrix two plant individuals are linked (1) if they share at least one interaction with the same pollinator species. The quantitative pollinator sharing between plant individuals (B) is represented with a matrix in which elements contain the sum of the lesser values of the number of visits (bold numbers in network links) by each pollinator species shared between two plant individuals, so that sharing is driven by the lowest communality in interactions with pollinator species. Exponential random graph models (ERGMs) fitted with the qualitative pollinator sharing matrix (binary) model the odds of a given plant individual to share at least one pollinator with other plant individuals. Meanwhile, ERGMs fitted with the quantitative pollinator sharing matrix (weighted) model the expected number of shared interactions with pollinators of a given plant individual with all other plant individuals.