Horticulture

Accordingly, a synthesis of 600 fields from 41 crop systems showed that only two of the 68 most frequent pollinators globally were specialist species: the weevil Elaeidobius kamerunicus pollinating oil palm and the squash-bee Peponapis pruinosa pollinating pumpkin .Because of differences in species functional traits, greater pollinator richness can lead to foraging complementarity or synergy, improving the quantity and quality of pollination and therefore increasing both the proportion of flowers setting fruits and product quality . Across crop species, insects with contrasting mouth part lengths may be needed for the pollination of flowers not only with easily accessible rewards but also with rewards hidden at the bottom of a tubular corolla . Within a crop species, social and solitary bees visited flowering radish plants at different times of day, suggesting temporal complementarity among these pollinator groups . Flower visiting behavior also differs among pollinators of different body sizes, and visits by a range of differently sized pollinator species increase pumpkin pollination. In addition to functional traits, square pot plastic interspecific differences in response traits to climate and land-use change can increase resilience of pollination services.

The role of diverse assemblages of wild insects in crop pollination is also evident from recent global analyses. Worldwide, incomplete and variable animal pollen delivery decreases the growth and stability of yields for pollinator-dependent crops . This lower yield growth has been compensated for by greater land cultivation to sustain production growth . The consequent reduction in natural areas within agricultural landscapes decreases the richness and abundance of wild pollinators, including bees, syrphid flies, and butterflies , further diminishing crop pollination . A possible solution to this “vicious cycle” is to increase pollinator abundance through single-species management, most commonly European honey bees , which are not greatly affected by isolation from natural areas . However, increasing the abundance of one species may complement but not replace the pollination services provided by diverse assemblages of wild insects, and wild insects pollinate some crops more efficiently than honey bees . Moreover, during the past 50 years, the fraction of animal-pollinator dependent agriculture and the number of managed honey bee hives have increased 300% and 45%, respectively, and honey bees have suffered from major health problems such as colony collapse disorder . All of these factors point to the potential benefit of practices that boost the species richness and abundance of wild pollinators.

Indeed, richness and visitation rate of wild pollinators are strongly correlated across agricultural fields globally . Therefore, practices that enhance habitats to promote species richness are also expected to improve the aggregate abundance of pollinators, and vice versa .Below we describe practices that diversify and improve the abundance of resources for wild insects outside the crop field, without affecting crop management. Practices are ranked from less-to-more required area, with practices covering less area likely to be less costly . Nesting resources – such as reed internodes and muddy spots for cavity nesters, and bare ground for soil nesters – can be enhanced at crop field edges without affecting much of the crop area. Although providing such resources can promote the recruitment of certain bee species , evidence of its effects on crop yield is lacking . Hedgerows and flower strips are woody or herbaceous vegetation, respectively, planted at the edge of a crop field, and generally covering only a small area. If appropriate plant species are chosen and adequately managed through time , hedgerows and flower strips can provide suitable food and nesting resources for, and enhance species richness and abundance of, bees and syrphid flies . These practices also enhance pollinators in adjacent fields – rather than simply concentrating pollinators at dense flower-rich regions – and therefore increase crop yield .

Regional programs that augment the quality and availability of seeds from native flowering plants are important for the success of these practices . Conserving or restoring natural areas within landscapes dominated by crops often provides habitat for wild pollinator populations . In addition, pollinators depend on various types of resources , which are difficult to provide in ways other than by enhancing natural areas. Consequently, these areas also enhance pollination services for nearby crops . Enhancing farmland heterogeneity increases pollinator richness because plant species provide complementary resources over time and space, and insect species use different resource combinations . Also, insects usually require resources for periods longer than crop flowering . In fact, a synthesis of 605 fields from 39 crop systems in different biomes found that diversity of habitats within 4 ha enhanced bee abundance by 76% as compared with bee abundance in monoculture fields . Smaller crop fields increase land-use heterogeneity, and also benefit pollinators because most species forage at distances less than 1 km from their nests . Thus, crops in small fields are more likely to benefit from pollinator enhancements such as nearby field margins and hedgerows . Indeed, pollinator richness, visitation rate, and the proportion of flowers setting fruits decreased by 34%, 27%, and 16%, respectively, at 1 km from natural areas across 29 studies worldwide .In contrast to off-field methods that can be ordered from smaller to larger scale , on-field practices are all applied at a similar spatial scale, ie that of the crop field. Here we discuss practices that reduce the use of insecticides and machinery, enhance the richness of flowering plants, and require greater effort because of changes in the crop species or system . Reducing the use of synthetic insecticides that are toxic to pollinating insects should provide an important benefit . For example, in South Africa, insecticides adversely affected pollinators, impairing rather than enhancing mango yield . Insecticides with low toxicity to pollinators, with non-dust formulations, applied locally through integrated pest management practices, and applied during the non-flowering season are less likely to be detrimental to pollinators than highly toxic, systemic insecticides that are broadly sprayed from airplanes . No-tillage farming may enhance populations of ground nesting bees given that many species place their brood cells <30 cm below the surface . Tillage timing, depth, and method probably have differential impacts on pollinators and pollination, but further studies are required to verify this expectation . Similarly, flood irrigation may be detrimental in comparison to drip irrigation because of the increased likelihood of flooding pollinator nests but, particularly in arid systems, irrigation in general can promote wild-insect abundance through higher productivity of flowering plants or by making the soil easier to excavate . Enhancing flowering plant richness within crop fields can benefit pollinator richness and crop pollination, as demonstrated for mango and sunflower in South Africa. Similar results were found for wild plants within watermelon and muskmelon fields in the US . In Ghana, banana intercropping with cocoa boosted pollinator abundance and cocoa pod set . A diverse set of flower species with different phenologies is likely to increase resource stability for pollinators and thus the resilience of pollination services. Herbicides and mowing can negatively affect pollinators by reducing floral resources provided by weeds , square pots for plants but can be useful for reducing the abundance of invasive grasses that could otherwise displace native flowering plants . Organic farming combines some of the practices described above and can enhance wild pollinator populations in comparison to conventional farming , probably because of the absence of synthetic insecticides and/or greater non-crop floral resources.

Farmland heterogeneity can also be increased by organic management practices, which account for less than 1% of global agriculture . When the extent of organic farming was expanded in a German agrolandscape from 5% to 20%, bee richness rose by 50%, while the density of solitary bees and bumble bees increased by 60% and 150%, respectively . Pollination-related benefits of organic practices were also found for strawberry in Sweden and canola in Canada . Sowing flowering crops, instead of crops that do not offer floral resources for pollinators, may enhance wild pollinators in heterogeneous landscapes . In western France, solitary-bee richness and abundance were higher in margins of canola fields than in fields of other crops . In the UK, bumble bee abundance was higher in areas adjacent to bean fields than to wheat fields but only during crop flowering , suggesting a short-term behavioral response to flower abundance rather than a long-term population enhancement. Similarly, in Germany, canola improved bumble bee early-colony growth but not whole-season sexual reproduction , and greater land cover of mass-flowering crops increased the number of bumble bee workers but not colony numbers . Therefore, although crops can provide abundant resources, the short duration of floral availability, the low diversity of resources, the application of insecticides, and the presence of tillage may limit the capacity of one crop species to support wild pollinator populations on its own . Furthermore, large monocultures of flowering crops can suffer from pollination deficit and trigger indirect negative effects on pollinators . Sowing crops that bloom in different periods may therefore increase wild-insect populations; in Sweden, bumble bee reproduction was improved in landscapes with both late-season flowering red clover and early-season mass-flowering crops . Moreover, managing crop phenology to better match the availability of efficient pollinators should enhance pollination, but we found no studies on this practice .The effectiveness of pollinator-supporting practices is influenced by interactive effects between large and small scale factors. For example, the effects of landscape composition on bee richness are greater on farms with low habitat diversity than on farms with high habitat diversity . Similarly, in Argentina, the importance of wildflower strips as pollinator sources for sunflower increased in the absence of large remnants of natural habitats nearby . In South Africa, the importance of weed richness for enhancing sunflower seed set increased with larger distances from natural areas . Throughout Europe, extensive programs aim to mitigate biodiversity loss on farmland through practices such as organic farming or wildflower strips, thereby offering a unique opportunity to understand interactions among these methods. A meta-analysis showed that these practices enhanced pollinator richness , but their effectiveness varied with the magnitude of increase in flowering plant cover resulting from the practices, farmland type, and landscape context . Because intensively managed croplands are generally devoid of flowering plants, pollinator-supporting practices in these landscapes result in the largest increase in floral resources and thus pollinator richness . On the other hand, conventionally managed grasslands generally contain more flowering plant species than arable fields,making it more difficult to enhance floral resources and pollinators . Finally, local effects were more positive in structurally simple landscapes than in cleared or complex landscapes, presumably because cleared landscapes lack sources of pollinator colonists and complex landscapes have less need of restoration. Recently, researchers have begun to explore the relative effectiveness of different pollinator-supporting practices. In Europe, flower strips were more effective than grass-sown or naturally regenerated strips . Globally, the effect of landscape composition and farm management was more important for improving bee richness than the effect of landscape configuration . Interestingly, conventional farms with high in-field habitat diversity maintained similar pollinator abundance as organic farms with low in-field habitat diversity, across the gradient of heterogeneity in surrounding land use. Thus, different combinations of local and landscape practices can result in similar outcomes in terms of promoting pollinator richness, providing alternative solutions suited to different agricultural settings. The importance of small-scale practices is likely greater for insects with short flight ranges foraging from a fixed nest, such as small- to medium-sized bees, which usually forage within an area of a few hundred meters and comprise the greatest fraction of bee species . Consistent with the idea that small-scale practices alone can have high impact, a study designed to separate the effects of local- versus landscape-scale habitat on pollination services delivered to blueberries found that the local scale had stronger positive effects . Indeed, farmers acting individually are more likely to improve the quality of their own fields and the immediate surroundings than to be able to manage complete landscapes for pollinators. Assuming a foraging range of 200 m from the nest for small bee species , diverse and high quality habitats need to be provided within 13 ha .Understanding the socioeconomic consequences of pollinator-supporting practices is essential to effectively enhancing wild pollinator richness in “real-world” landscapes . Farmers generally face implementation costs, such as those for planting hedgerows, and opportunity costs, such as those for setting aside natural habitats that could otherwise be cultivated .

This indicates that despite the drastic reduction in defense, some mechanisms have remained in domesticated types or have been recently introgressed by geneflow from wild populations. Identifying and amplifying these anti-herbivore defense mechanisms in modern cultivars, as well as introducing new mechanisms from wild relatives and landraces, could greatly improve productivity and profitability, while limiting the use of toxic agrochemicals. Like other crops, Lima bean has lost many of its mechanisms of defense against insect herbivores during the bottlenecks of domestication and modern breeding programs. This is illustrated by the finding that wild P. lunatus seedlings have greater chemical diversity than their domesticated relatives . Lima beans have been used as an experimental model in numerous studies of herbivoreinduced direct and indirect defenses . Previous studies have examined the role of cyanogenesis, volatile organic compounds, blueberry package and extrafloral nectaries as plant defense mechanisms against herbivores . Due to the high metabolic cost to the plant of producing these defensive compounds, P. lunatus makes tradeoffs between direct and indirect defense mechanisms .

The goal of the research presented In this dissertation Is to understand the domestication history, current variability, and breeding potential of the antiherbivore defense traits in Lima bean with a special emphasis on cultivars adapted to the Central Valley of California and cyanogenesis as a mechanism of defense against Lygus hesperus. Lima Beans as the Predominant Grain Legume in California Lima beans are one of five domesticated species in the genus Phaseolus. They were independently domesticated first in Central Mexico and again on the western slope of the Andes Mountains of Ecuador and northern Peru . Dating of starch grains on human dental remains, indicates that beans were domesticated in Northern Peru approximately 8000 years ago . Estimates for the date of domestication of Lima beans in Mexico are less certain but range from 2300-3400 years ago . The domesticated Andean gene pool is characterized by large flat seeds while the Mesoamerican gene pool has smaller round or flat seeds . Lima beans are multi-annual or semi-perennial, with crops typically needing 115- 135 days to reach maturity . They are adapted to a range of climates but are especially suited to warm and humid environments . Lima beans are grown in many regions of the world, including Africa, Asia, and Central and South America . In the United States, California is the primary growing region for mature dry Lima beans, where they are an important crop for the agricultural systems of the Central Valley.

As a nitrogen-fixing, highly vigorous rotation crop for tomatoes and other high-value crops, Lima beans provide an essential service of sustaining soil fertility and breaking pest and weed lifecycles . Additionally, Lima beans are very drought tolerant, making them ideal for the perennial water shortages experienced by California in recent decades . Two market classes – small white and large white – are grown in the state. Baby Limas are grown mostly in the region around Sutter and Colusa Counties. Large Limas, which need cooler nighttime temperatures, are grown mostly in Stanislaus County . Approximately 20,200 acres of Lima beans were grown in 2018, representing nearly half of the dry bean production in the state . Improved cultivars of Lima bean yield approximately 2,500-3,500 pounds per acre . Production of Lima bean in California is limited by its vulnerability to L. hesperus. Regular treatment in the field with pesticides, specifically pyrethroids, is the only known effective method of control . This practice is costly for a low-value crop like Lima beans and is unsuitable for organic production.Herbivory by the polyphagous, native Californian insect Lygus hesperus Knight , negatively affects the yields of several important crops, including alfalfa, strawberries, safflower, peaches, almonds, and dry beans . Current methods of controlling L. hesperus are costly, environmentally toxic, and only moderately effective . In sensitive crops like alfalfa and Lima bean, L. hesperuscan cause up to 70% yield loss as measured in sprayed versus unsprayed plots . There are typically four or five generations of L. hesperus each summer, with variability due to climate . The rate of development is dependent on temperature, but it takes on average 27 days from egg to reproductive adult at 20°C . Each generation develops from egg to adult with five nymphal instars going through incomplete metamorphosis .

All nymphs are flightless, but adults are highly mobile . L. hesperus are omnivorous but feed mostly on plant tissue . Their style of feeding is known as rupture feeding or “lacerate and flush.” Feeding starts with the insect probing the food tissue with its straw-like stylet, causing cells to rupture. Saliva secreted from the stylet has enzymes like polygalacturonase and -amylase, which further break down the tissue, creating a slurry that the insect can ingest through their stylet tube . L. hesperus feeding on Lima beans results in the abscission of flowers or young pods and consequently, yield loss. When feeding occurs on mature pods seed viability may be reduced and scarring can occur on seeds, thereby lowering market value . Adults spend about 20% of their time, and nymphs about 30% of their time, probing plants with their stylets. Actual ingestion represents only 3% of the probing time . The mechanical and chemical damage caused by this frequent probing and feeding behavior contributes to the heavy impact that L. hesperus have on crop yields.Understanding the mechanisms and inheritance patterns of L. hesperus tolerance or resistance in Lima bean will aid efforts to breed new varieties that require fewer pesticides and are suitable for organic production. To start, chapter 1 presents a comprehensive review of the literature on how domestication has affected the genetics of insect defense traits. Chapter 2 details the results of a genome-wide association study of cyanogenesis in Lima bean. The goal of this study was to explain how cyanogenesis has been affected by domestication in the Mesoamerican gene pool of Lima beans, with special consideration of the cyanogenic capacity of California cultivars. Chapter 3 will explore the variation and heritability of the tolerance or resistance to L. hesperus in cultivars of Lima bean adapted to the Central Valley of California. This will include a study of how cyanogenesis in Lima bean is affected by the presence of L. hesperus, specifically to determine the extent to which this trait is constitutive or induced. Analysis of the survival and reproduction of L. hesperus on varieties of Lima bean with variable expression of cyanogenesis will also be presented.The processes of cultivation and domestication have transformed wild species into crop plants that are an invaluable food source for the human population. These same processes have also made crop plants more vulnerable to damage by insect pests than their wild relatives . Globally, blueberry packaging insect herbivory accounts for an estimated 18-20% of yield loss during crop production . It is expected that these losses will increase if current trends in climate change continue . Recovering the defensive abilities of crop wild relatives in domesticated plants would result in reduced need for pesticides – which are harmful to human and environmental health – as well as an increase in global food security. Plant defenses against insect herbivores typically consist of complex suites of traits . These may include resistance traits like chemical deterrents, physical barriers, and reduced palatability, or tolerance traits like increased vigor and delayed phenology . Defensive traits can also involve attraction or resource benefits for beneficial organisms such as parasitoids and predators of insect herbivores .

Defense traits may be constitutively expressed and may also be induced to a higher level of expression by factors such as the presence of the target herbivore or volatile signals from a neighboring plant . Induction of a defense trait may occur immediately or slowly over time . The response can also be localized tothe area immediately surrounding the site of damage or be widespread throughout the plant . Several environmental factors can affect the expression of these traits including light intensity , the interspecific and intraspecific diversity of neighboring plants , photoperiod, temperature, and climate . Control of these traits typically involve the expression of multiple loci, which may be linked to other useful or unfavorable agronomic traits . The process of domestication involves selecting wild plants with desirable agronomic traits. The resulting crop plants typically exhibited some or all a suite of traits known as the domestication syndrome. This can include increased palatability, loss of dormancy, seed retention, increased seed size, and seed number . Selection intensity varied across crops and domestication events and there is serious academic debate about the duration and intentionality of this process . Identifying the genes that control agronomically important traits and understanding the way in which they have been affected by domestication is foundational to conserving, amplifying, and increasing their utility. This review will focus on the ways in which the process of domestication has altered the genes underlying plant defenses against insect herbivores. Prior reviews have focused on the mechanisms and physiology of plant defense against insect pests as well as the ways in which domestication has affected the interactions between crop plants, insect herbivores, and higher trophic levels . To my knowledge this is the first review on the subject of how domestication has affected plant and insect herbivore interactions with a specific focus on crop genetics.Domesticated gene pools have significantly less genetic diversity than wild gene pools . During the transition from wild plant to domesticated crop, diversity is lost during the actual domestication process as well as during cultivation, dispersal outside the center of origin, and later improvement through modern breeding . This loss of diversity within crop species has contributed to a loss of insect herbivore resistance traits . This is illustrated in maize through a comparison of teosinte, maize landraces from within the center of origin, maize landraces from outside the center of origin, and modern maize cultivars. Each stage of transition resulted in changes to the anti-herbivore defense strategies as well as changes in gene expression . A similar pattern can be seen in an analysis of the GsRbohA1 locus in soybean of which the A haplotype confers resistance against common cutworm . This allele was only present in 2.2% of modern cultivars compared with 23.5% of landraces and 95.6% of wild accessions . In the case of some anti-herbivore defense mechanisms, the selection against certain phenotypes had clear benefit to human consumers. For example, selections against genotypes which produce toxic compounds like cyanogenic glycosides, or distasteful physical defenses like trichomes, improved the safety and palatability of food . However, several important defense traits such as herbivore-induced volatile organic compounds cannot be perceived by casual observation and have only recently been revealed by careful scientific study . Despite the impossibility of direct selection for or against these traits prior to their discovery, several studies have identified differences in the expression between wild and domesticated crop plants. For example, in several studies of phytophagous insects and their parasitoids, parasitism was higher for hosts on domesticated rather than cultivated plants . In other studies, reduced predation or parasitism was attributed to the loss of chemical diversity or volatile signals . The loss or amplification of indirect anti-herbivore defense traits may be due to genetic drift, linkage, pleiotropic effects, or selection for alternate resource allocation. However, these were not intentionally selected against as their function has only recently been discovered . Breeding programs with little or no insecticide protection may help maintain insect defense traits compared to programs in which insecticides protect plants and mask susceptibility . Two early steps in the process of domestication were cultivation and storage, the intentional planting of future crop species and the saving of seeds between planting seasons . Both cultivation and storage created novel selection pressures on crop plants and their insect herbivores. Traits changed in frequency from the wild population when under the selection pressures of cultivation and storage . For example, in common bean, Phaseolus vulgaris, resistance to Mexican bean weevil Zabrotes subfasciatus is most likely due to a protein ofthe APA family, arcelin . Arcelin is only found in some accessions of wild Phaseolus vulgaris from Mesoamerica and is controlled by a single mendelian gene for which arcelin production alleles are dominant over alleles for its absence .

The relationship between diversity and the stability of communities and ecosystems is a fundamental question in theoretical and experimental ecology . Beginning in the 1990s, and sparked by clear experimental results showing that species richness decreased the temporal variance of ecosystem function , biodiversity research contributed to a broader discussion about how worldwide declines in biodiversity would affect ecosystem services on which humans rely . A major goal of biodiversity research has involved separating the relative importance of richness , composition , and abundance as drivers of temporal variance in ES. However, both the study designs and the analytical approaches used vary between experimental and observational studies. At smaller scales, the field of biodiversity-ecosystem function research has used controlled experiments and a well-developed body of mathematical theory to explore how species richness and composition affect temporal variance. In contrast, at larger scales, the field of biodiversity-ecosystem services research has been built mostly on correlative studies conducted in real-world systems —where ‘real-world’ means communities that are not directly manipulated—in which it is difficult to rigorously separate the causal roles of richness, composition and abundance .

Because species loss continues to occur at high rates worldwide , nursery grow bag it is critical to gain a better understanding of how species richness affects temporal variance of ES. This requires the development of novel analytical approaches that can separate the effects of richness, composition, and abundance without experimental manipulations, which are difficult if not impossible to conduct at landscape scales.An examination of ES in real-world systems is needed because controlled experiments that typify most biodiversity-ecosystem function research do not fully represent ecological reality ; specifically, these experiments do not mimic realistic species abundance distributions or species loss scenarios. First, a skewed or ‘hollow curve’ species abundance distribution, meaning that communities are composed of few common but many rare species, is ubiquitous in nature yet biodiversity-ecosystem function experiments have not mirrored this pattern and have instead used substitutive designs that equalize initial abundances among species . Although substitutive designs are arguably preferable for isolating the effects of species richness, they do make experimental communities less realistic and decrease the potential for one dominant species to provide the bulk of ecosystem function. Second, species are lost from communities non-randomly, with environmentally-sensitive and rare species being at greater risk of extirpation . In contrast, most biodiversity-ecosystem function experiments have assigned species to plots at random to avoid confounding species richness and species composition.

It is well-known that, because of this design, comparing high- and low-richness experimental plots reveals the effects of random species loss, which will under- or overestimate the effects of non-random species loss, depending on whether species with high or low contributions to function are lost first . However, biodiversity-ecosystem function researchers have countered that understanding the effects of random species loss is an important starting point, given that future patterns of species loss may be unpredictable . Throughout this paper, we use species loss to indicate local extinctions, because ES are delivered by local populations . Despite an awareness of these systematic differences between biodiversity-ecosystem function experiments and real-world ES, there is no consensus on whether species richness will contribute more or less to function when experimental results are scaled up to larger, more complex systems.Scaling up biodiversity-ecosystem function research is further complicated because analyzing observational data creates challenges not present in experiments. First, few if any real-world ecosystems allow researchers to independently assess services provided by each species in the community. This precludes the use of analytical approaches commonly used in bio-diversity ecosystem function research, especially analyses requiring single-species monocultures . Second, communities assemble and disassemble non-randomly with respect to species’ contributions to ES, making it difficult to separate the effect of richness from the effects of species identity. Third, the ubiquity of skewed species abundance distributions in nature makes it hard to separate the effect of abundance from species identity, if the same species are dominant across sites.

Because of these issues, no general analytical method exists for biodiversity-ecosystem services studies and, perhaps for this reason, no consensus exists on the importance of different components of biodiversity for ES . Here, we present a novel version of the Price equation that can analyze temporal variance of any ES, so long as the ES can be expressed as a sum of species-level contributions. Our work builds on the original Price equation from evolutionary biology, and its recent adaptations for biodiversity research . Fox provided the original framework for analyzing temporal variance with the Price equation, and here we extend it so that it can be used with observational data even if species composition is not nested between sites . Our version of the Price equation partitions between-site differences in temporal variance into three additive terms: variance in ES attributable to richness , composition , and context-dependence . Here, we use ‘abundance’ in place of‘context-dependence’ because in our data this term is determined by patterns of abundance fluctuation over time . The richness term establishes an expectation for how species loss and gain would affect ES if species were identical in the ES each provides. The abundance term quantifies the extent to which species present at both sites contribute more to temporal variance of ES at each site.We explore the temporal variance of pollination services using two large, multi-year datasets on pollination provided by bees. Using the new derivation of the Price equation described above, we ask: What is the relative importance of changes in species richness, composition, and abundance to the temporal variance of ecosystem services? Specifically, we compare the relative importance of richness and composition versus abundance.Field surveys. Our study systems consist of the wild bee pollinators of watermelon Matsum. & Nakai) and northern highbush blueberry  plants, both of which rely on insect pollination for successful fruit production. Over five years , we sampled wild bee communities at 10 commercial watermelon fields in central New Jersey and eastern Pennsylvania, USA. We also, over three years , sampled wild bee communities at 16 commercial blueberry fields in southern New Jersey. In a post-hoc analysis, plastic growing bag we confirmed that differences in our results between study systems were not due to length of sampling . Hereafter, we refer to these fields as sites. We ensured that all sites were at least 1 km apart, beyond the typical foraging radius of most bee species in our study . We did not include the honey bee in our data collection, primarily because in our system the honey bee is a managed species that is kept in hives and moved in and out of crop fields by bee-keepers and farmers. Thus, the temporal and spatial variation in honey bee abundance is driven by hive placement rather than ecological factors. In addition, honeybees are the property of bee-keepers and farmers, so we cannot collect them. Finally, honeybees are present at nearly all sites, so including them would likely increase the relative importance of ‘abundance’, making it conservative with respect to our findings to leave the honey bee out of the analysis.

Because watermelon is an annual species, farmers do not necessarily plant it in the same locations each year. To exclude potential effects of spatial variation on wild bee communities, we included watermelon sites in our analyses only if the maximum among-year distance between transects was ≤ 435 meters; this is within the typical foraging radius of all but the smallest bee species in our study . Pollination services. To measure bee richness and the pollination services delivered at each site on each date, we collected two forms of data: the number of individual wild bees visiting flowers, and the number of pollen grains deposited per flower visit. We then multiplied each species’ abundance by the mean pollen deposition of its morphological group to obtain that species’ contribution to pollen deposition. To measure bee abundance, we established a 50 m transect at each watermelon or blueberry site. We collected bees visiting flowers by net throughout the transect and then processed voucher specimens for species-level identification by taxonomists . At each site, data were collected on three days during each plant species’ peak bloom period, with three temporally-stratified 20-minute-long collections during the day; all data collection days were sunny, partly cloudy, or bright overcast with limited wind. Total collection effort was 135 days for watermelon and 144 days for blueberry. We measured the pollination efficiency per flower visit for different pollinator groups in field experiments. We offered a virgin flower to an individual bee foraging in the field, allowed the bee to visit the flower one time, and recorded the pollinator group of the bee . These pollen deposition experiments were conducted in three years for watermelon, and in two years for blueberry. Back in the laboratory, we use a compound microscope to count the number of conspecific pollen grains deposited during the single flower visit. To prepare slides, watermelon stigmas were softened in 10% KOH, and stained with 1% fuchsin. Blueberry stigmas were softened in 1 M NaOH, and then stained for 48h in 0.01% analine blue buffered in 1 M K3PO4. In the pollen deposition experiment, individuals were placed into morphologically similar groups which could be differentiated in the field . Due to the difficulty of collecting single-visit pollen deposition data, we did not collect these at all sites and in all years. However, we do know that that different morphological groups differed significantly in pollen deposition rates, while species within groups did not . Further details of all data collection methods, and site details, are available for both watermelon and blueberry . Because there is substantial variability within each morphological group’s distribution of pollen deposition rates, we conducted a sensitivity analysis to test whether choosing a single pollen deposition value and discarding the remaining variability affected our results. Instead of multiplying each bee species’ abundance by its morphological group’s mean pollination efficiency, for each individual bee, we randomly drew one pollination efficiency value from the correct morphological group’s distribution. We repeated this sensitivity analysis 1000 times, and found that the same results were obtained using either the ‘mean’ or ‘sensitivity’ versions of the analysis .In the watermelon system, we collected 3044 individual wild bees belonging to 59 species , 14 genera, and 5 families. Sites ranged in total temporal variance from 2.3 million to 51.1 million grains of pollen , in richness from 20 to 32 species , and in abundance from 140 to 435 individuals ; richness and abundance values are summed across years. In the blueberry system, we collected 1067 individual wild bees belonging to 36 species , 9 genera, and 4 families. Sites ranged in total temporal variance from 2.6 to 875,154 grains of pollen , in richness from 3 to 19 species and in abundance from 4 to 143 individuals . The range of richness values is similar to the range found in experiments that study various ecosystem services , although only two of our sites had species richness values similar to the lower end of richness manipulations . The rank-abundance distributions for each system are shown in Figure 1.The five Price equation terms are measures of directional effect size, where positive or negative values tell how strongly each term decreases or increases between-site differences in the total temporal variance of ES. For both systems, all richness-loss terms were negative and all richness-gain terms positive, with richness-loss greater in magnitude . This reflects the fact that total temporal variance of ES is generally higher at higher richness sites.All composition-loss terms were positive and all composition-gain terms negative, such that in all cases, composition terms partially cancelled their corresponding richness terms. This means that both the lost and gained species tended to have below-average contributions to total temporal variance of ES. Specifically, the positive composition-loss indicates that observed reductions in species richness resulted in less reduction in total temporal variance than would be the case if species losses had been random with respect to total temporal variance. Similarly, the negative composition-gain indicates that observed increases in species richness resulted in less increase in total temporal variance than would be the case if species gains had been random. In the watermelon system, composition-loss cancelled 86% of richness-loss and composition-gain cancelled 78% of richness-gain; in the blueberry system, these same value were 23% and 76%.

Five common conifer species occur on the northern range of the Archipelago: western hemlock , mountain hemlock , yellow-cedar, Sitka spruce , and shore pine . These coastal forests are simple in composition yet often complex in age and tree structure . Yellow-cedar occurs across a soil-drainage gradient from poorly drained bogs to well-drained soils on steeper slopes that often support more productive stands . This study occurs in the northern portion of the yellow-cedar population distribution and at the current latitudinal limits of forests affected by decline. We centered our investigation on protected lands in four inlets in the Alexander Archipelago on the outer coast of the West Chichag of-Yakobi Wilderness on Chichagof Island in the Tongass National Forest and Glacier Bay National Park and Preserve . Aerial surveys were conducted in 2010 and 2011 to assess the presence of affected forests and to identify the edge of yellow-cedar dieback that occurs south of GLBA on Chichagof Island. Aside from a brief history of small-scale gold mining that occurred in several areas on Chichagof Island between 1906 and 1942, plastic planters there is little evidence of human impact on these lands, making them ideal for studying ecological dynamics.

Drawing upon previous studies that estimated the time-since-death for five classes of standing dead yellow-cedar trees at various stages of deterioration , our plot selection consisted of sequential steps, in the field, to sample forests representative from a range of time-since-death. Not all yellow-cedar trees in a forest affected by mortality die at once; mortality is progressive in forests experiencing dieback . Highly resistant to decay, these trees remain standing for up to a century after their death . As a result, they offer the opportunity to date disturbance, approximately, and to create a long-term chronosequence. First, we stratified the study area coastline into visually distinguishable categories of ‘‘cedar decline status’’ by conducting boat surveys and assessing cedar decline status across 121.1 km of coastline in June 2011 and 2012. We traveled the coastline and made visual observations of live and dead yellow-cedar trees and their snag classes. We assigned cedar decline status to coastal forests at 100 m increments using a GPS Garmin 60 CSx . Next, using the ArcGIS 10.2 Geographic Information System software , we randomly generated plot locations in forests categorized during the coastline survey as follows: live, unaffected by mortality; recent mortality; mid-range mortality; and old mortality. Lastly, we controlled for basal area and key biophysical factors, including elevation and aspect via methods described.

Plots were restricted to elevations less than 150 m, excluding northeast facing plots, to sample from low-elevation plots representative of conditions where yellow-cedar decline commonly occurs at this latitude . Plots were randomly located between 0.1 and 0.5 km of the mean high tide to avoid sampling within the beach fringe area, and on slopes ,72% to limit risk of mass movement . We excluded plots with a total basal area ,35 m2 /ha to avoid sampling below the optimal niche of yellow cedar . This control was performed in the field by point sampling to estimate basal area using a prism with a basal area factor 2.5 . Plots dominated by the presence of a creek bed or other biophysical disturbance were eliminated from plot selection, due to the confounding influence of disturbance on the number of trees standing and species abundance. By restricting our sampling to these controls, our study was designed to examine the process of forest development post-decline in low-elevation coastal forests with plot conditions typical for yellow-cedar mortality excluding bog wetlands, where yellow-cedar may co-occur sparsely with shore pine. After controlling for biophysical factors, 20 plots were sampled in live forests and 10 plots in each of the affected cedar status categories for a total of 50 plots across the study area .Data were collected in fixed, circular nested plots to capture a wide range of tree diameters and in quadrats within each plot to account for spatial variability in understory vegetation.

Forty plots were established and measured during the 2011 field season and 10 plots during the 2012 field season, through the seasonal window of mid-June to mid-August. Nested circular plots were used to sample trees and saplings as follows: a 10.3 m fixed radius plot for trees 25.0 cm diameter at breast height , a 6.0 m fixed radius plot for saplings ,2.5 cm dbh and 1 m height, and trees 2.5–24.9 cm dbh. We counted live saplings of each species to analyze the population dynamics for individuals that survive to this size class. For each tree, we recorded species, dbh to the nearest 0.1 cm, height to the nearest 0.01 m, dead or live, and for dead trees snag classes I–V. Eight quadrats at each plot were utilized to record understory plants and tree seedling densities. To provide an additional longterm view of species changes, we recorded counts for smaller conifer seedlings , identifying western hemlock and mountain hemlock to genus, and other conifers to species. We noted presence/ absence of each conifer species 10–99 cm, but did not sample this size class for individual counts. We recorded maximum height and percentage cover of each plant species observed according to the Daubenmire method on a continuous scale . In unique cases where consistent identification to species was difficult Salisb.; Vaccinium ovalifolium Sm., and V. alaskaense Howell, we combined observations but noted both species presence for total richness across the study area. Blueberries, V. ovalifolium and V. alaskaense, are similar in appearance and often synonymized . Mosses and liverworts were recorded together as bryophytes within the quadrat. Sedges were recorded together but distinguished from true grasses . We used hemispherical photography to assess canopy cover at each plot. Photographing from plot center at dbh camera height, we captured imagery in relatively uniform, overcast skies and consistently avoided any mid-day sun conditions . To prevent diminished sharpness associated with consumer-grade cameras , we used a Sigma 4.5mm fish-eye lens on a professional grade Canon 7D camera . Full-view images were processed using Gap Light Analyzer to yield percentages of canopy openness per plot as a proxy for light in understory analyses .Clustering plots by cedar decline status.—To rigorously account for the timing of mortality relative to the coarse visual cedar decline status categorizations made by boat, we performed kmeans clustering analyses on the yellow-cedar population observed across the chronosequence by partitioning 50 plots into those affected by mortality and live ‘‘controls’’ for subsequent stages of analysis. Using observations of dead and live yellow-cedar trees at each of the 50 plots , we stratified the plots into two groups for unaffected and affected forests. We then performed a k-means clustering analysis with the categorical snag classifications observed at the resulting 30 plots affected by mortality, plastic plant nursery pot assigning the a priori k ¼ 3 for three affected status categories sampled: recent mortality, midrange mortality, and old mortality. We restricted this analysis to yellow-cedar trees .10 cm dbh because the methods of dating time-since-death for yellow-cedar trees rely upon standing, larger trees . We analyzed the cluster stability by computing the Jaccard coefficient to measure similarity between resulting clusters, assessed by the bootstrap distribution of the Jaccard coefficient for each cluster compared with the most similar cluster in the bootstrapped datasets .

Post hoc Fisher’s exact tests further clarified differences in the numbers of observed class I, II, and III snags between recent and mid-range mortality clusters ; observed class II, III, and IV snags between recent and mid-range mortality clusters , and between mid-range and old mortality clusters ; and observed class III, IV, and V snags between mid-range and old mortality clusters . These analyses were performed in R using the GCLUS and FPC packages. This post-field methodology for plot stratification enabled us to refine the visual cedar decline status assigned in the boat surveys by clustering according to the observed populations of live yellow-cedar trees from the plot data. Stand structure and regeneration.—We calculated the importance value for live conifers in the overstory as the sum of relative density, frequency, and basal area per species to characterize the stand structure and conifer composition within each cedar decline status resulting from clustering analyses, and to make comparisons across the chronosequence of cedar decline status. For each species in three size classes , we computed the following variables: density , frequency , and dominance , and with the relative values of these three parameters, the importance value was calculated as IV ¼ DR þ FR þ DoR . Thus, the cumulative value for all tree species per size class in each cedar decline status was 300%. In assessing regeneration, we focused analysis on seedling counts and saplings to consider established plants. We used Krukal–Wallis tests and performed permutation tests on the measure of central tendency to examine differences in mean seedling and sapling abundance across the four cedar decline status categories. Using presence/absence sapling data, we calculated the probabilities of finding each individual conifer species in the sapling life stage in each cedar decline status and generated binomial confidence intervals to estimate uncertainty using the Wilson score interval. We used a two-part modeling approach to determine the probability of species’ occurrence in cedar decline status and to test for significant effects of cedar decline status on each species’ abundance in the sapling stage. This method was selected to account for over dispersion in zeros in the individual abundance data for the conifer species in the sapling life stage . In the first step, the data were considered as zeros versus non-zeros and a binomial model was used to model the probability of observing a zerovalue; in the second step, non-zero observations were modeled with a zero-altered Poisson model . Canopy openness and cedar decline status were included in the models as explanatory variables to predict species presence/absence and sapling abundance. Best models were selected based upon AIC values. These analyses were performed in R using the PSCL, MHURDLE, and BINOM packages. We determined the IV for saplings in each cedar decline status on the basis of relative density and relative frequency , such that the IV of all species would sum 200%. To compare the persistence of saplings to treelets in the early stages of stand development, we calculated the ratio of saplings to live treelets per hectare at each plot and tested for significant differences between live ; recent mortality and live ; mid-range mortality using Wilcoxon rank sum tests. Probabilities calculated for species occurrence in the size class 10–99 cm in each cedar status were used for comparison with seedling and sapling results to assess trends in survival.The changes observed across the chronosequence provide strong evidence that this species dieback associated with climate change can result in a temporally dynamic forest community distinguished by the diminished importance of yellow-cedar, an increase in graminoid abundance in the early stages of stand development, and a significant increase in shrub abundance and volume over time. Tree mortality timing and intensity, as characterized by our stratified sampling of cedar decline status, played an important role in determining the understory community composition and overstory processes of stand re-initiation and development. Our results highlight the ways in which widespread mortality of one species can create opportunities for other species and underscores the importance of considering long-term temporal variation when evaluating the effects of a species dieback associated with climate change. Methods for predicting future changes in species distributions, such as the climate envelope approach, rely upon statistical correlations between existing species distributions and environmental variables to define a species’ tolerance; however, a number of critiques point to many factors other than climate that play an important role in predicting the dynamics of species’ distributions . Given the different ecological traits among species, climate change will probably not cause entire plant communities to shift en masse to favorable habitat . Although rapid climatic change or extreme climatic events can alter community composition , a more likely scenario is that new assemblages will appear . As vulnerable species drop out of existing ecosystems, resident species will become more competitive and new species may arrive through migrations . Individual species traits may also help explain the process of forest development in forests affected by widespread mortality, as the most abundant species may be those with traits that make them well-adapted to changing biotic and abiotic conditions .

California is an agricultural hot-spot of the world, producing large quantities of field crops, fruits, nuts and vegetables. In 2019 the total value of principal crops in California was over $33 billion, the second-most productive state was less than half that amount . There are estimated to be between 1,600 to 2,000 species of bees in California , of which 46 bee species within 17 genera are found commonly in Californian cities . Bees are indispensable drivers of ecosystem pollination, aiding greatly to plant reproductive success and biodiversity , the foundation of an enormous diversity of other organisms. Effective bee pollination across a variety of landscape types helps to ensure self-sustainable cities and human communities . This research focuses on the connection between plants and the attracted foraging bees. This study connects the floral landscape to pollinator foraging events taking place on it. For example, a landscape’s plant community composition is known to be a main predicting factor in the presence or absence of bees .

Many studies focus on bee foraging habitats and have tried to quantify the attractiveness of plants to bees . While plant list short-comings are part of the problem with evaluating bee habitats , french flower bucket understanding how spatial implications play out across a landscape is essential to enhance understanding of patch dynamics and the potential effects of fragmentation. As noted above, in Chapter 1 we determined that actual bee foraging preferences varied greatly from published expected foraging associations. This research builds on Chapter 1’s findings, by examining if spatial and temporal habitat usage was limited by a poor understanding of bee forage plants or not. This research study compares the smaller WHR plant selection with the full plant selection which bees used for forage in Chapter 1. By taking this approach, this research examines spatial habitat utilization and infers potential implications . The study examines whether the differences in plant palette usage made a large difference or not in utilized habitat.Habitat fragmentation, degradation and destruction are cited as the main reasons for declines in California native bee populations .

Drawing from the theories of island biogeography and metapopulation dynamics , fragmentation is defined as suitable habitat patches being too far apart to support sustainable bee populations. On an individual level fragmentation could lead to isolation, or the inability to disperse to another habitat patch for requisite resources, for example, lack of flowers . Spatial isolation of habitat patches has been found to decrease diversity of bees in urban landscapes . Even in naturalistic habitat fragments, there have been shifts of bee abundance and richness due to isolation . Overall, anthropogenic disturbances have significant negative effects on wild bees . However, in areas with less than 50% impervious surface pollinators still seem to provide sufficient pollination services to wild vegetation and crops . Currently bee life history literature does not concur on which scale of habitat fragmentation has the largest impact. For example, some studies concentrate at the entire world scale the North American continent scale , the entire United States country scale , and the smallest studies are said to be at the microsite, consisting of several point locations, several miles across , additionally, Cane et al. compared bee ecological characteristics such as dietary breadth.

Most of the above mentioned research focused more on European honey bees, though studied less often native bees were included in some studies. There are landscape elements which are completely inhospitable to bees and can be defined as habitat destruction, for example: paved surfaces and the footprint area buildings use . Habitat degradation to bees consists of detrimental changes of a site. For example, a dramatic shift in vegetation cover of a formerly naturalistic site will have effects on the bees which utilized the former floral resources. Bees’ responses to such shifts will vary , as their foraging habits, obligate versus generalist, are similar to butterflies in California . In terms of foraging some bees have benefitted from predominantly human shaped and dominated landscapes , while other bees decline . Habitat gaps may also occur temporally, depending on plant phenological patterns, leading to patch isolation or connectivity . Bee foraging preferences, whether polylectic or oligolectic , can also be an origin of habitat fragmentation if a landscape is not suitable. For bees, if sufficient floral resources are not available, that represents a non-foraging area, possibly a population sink. This phenomenon may lead to ecological filtering, reducing bee diversity . Current bee foraging plant lists have limitations in their ability to predict a bee genera’s presence or absence . Nevertheless, measuring bee-to-flower associations is a starting point to understanding the potential that there is habitat fragmentation for bees. Identification of the most basic levels where and when habitat fragmentation occurs is essential to determine before conservation remedies can be prescribed. It is essential to investigate these spatial problems with geographic tools to aid in bee conservation using empirical relationships with which bees experience their world.Bees vary widely in their foraging distance ability from a nest . Bee body size has been shown to be an indicator of forage diameter radii . Based on published body size data , bees may forage up to 2 mi maximum for only European honey bees, 1 mi for large bodied bees such as Bombus and Xylocopa, mid-sized bees are most common and average 0.25 mi , and the smallest bees are estimated at only 0.11 mi from their nests . Bee foraging radii is traditionally measured by the distance between a bee’s nest and its foraging range . However, determining individual nesting locations for solitary native bees is extremely difficult and not attainable or reliable based on current bee research studies , often taking a long time and often not gleaning results quickly enough for this research spatial scale and quantity of data . Given the intractability between locating nests of individual bees observed foraging in gardens, a general assumption is made in this study that foraging instances are relatively closely linked in proximity to nest locations, and therefore the foraging data points are used as a proxy for nesting locations. These foraging radii are for female bees, bucket flower which create and care for nests, rather than males, which utilize the landscape differently . Importantly, maximum foraging ranges represent an extreme limit on travel, as most individual bees will not travel as far as the maximum . In traditional meta population literature, differences between home range movement versus individual dispersal have been studied with more charismatic animals , but these movement dynamics have not been studied thoroughly in relation to bees. Among the limited extant bee dispersal research studies, results align with previously cited research regarding bee body size, degree of host plant specialization and bee sociality .The human landscape’s horticultural gardens can be florally diverse, contributing to an array of potential habitat types for bees . Furthermore, bee foraging preferences are not well understood, but are crucial to mitigate pollinator population declines . Thus, the role that horticultural garden conditions can contribute to increasing bee population numbers needs to be identified. Conversely, horticultural conditions which provide little to no benefit to bees should also be identified and avoided or discouraged. Studies conducted thus far have indicated a variety of responses of bees to managed landscapes, including both increased bee richness in some anthropogenic sites or some have shown decreased bee diversity in anthropogenic sites . However, overall, many land use activities are known to be detrimental to bee communities . Conversely, habitat simplification is known to have a negative effect on bee populations . Furthermore, plant community composition across a spatial gradient has been reported to have an effect on plant-pollinator resilience as a whole .

Therefore, it is important to assess both composition and configuration ofanthropogenic landscapes to determine why pollination ecosystem services occur in greater or lesser frequency across the landscape . Floral abundance, richness and spatial distribution have been noted to affect native bee communities , but the effect of these variables needs to be explored in more detail, and across a larger variety of garden types. Presence of weeds seems to enhance bee presence as well . Overall, bees respond differently to anthropogenic conditions and it is essential to study the variety of responses to maximize conservation efforts . Overall, current habitat remedies involve providing adequate floral resources and increasing floral diversity . Reconciliation ecology, a conservation science technique used in human dominated habitats , has been promising for other organisms and is highly applicable for bees.In recent years, scientists have used the term “Anthropocene” to describe the current geological epoch which by definition states how far humans have influenced Earth’s global biogeochemistry. Similarly, at the landscape scale, humans have dominated and changed Earth’s terrestrial ecosystems dramatically, leading to the term “anthroscape” . Notable in recent times, our infrastructure often excludes wildlife and represents an ecological sink or void. For example, paved roads offer no value to bees. Buildings and their footprints aid in reducing viable bee habitat. Notably, some horticultural planting schemes may represent opportunities to bees. Waterways, ponds, pools represent sinks to nearly all bees, the exception being Apis melifera which has low tolerance for critical water content and need fresh water to survive whereas native California bees have very high tolerance for critical water content and do not need fresh water bodies regardless of being in urban or rural ecosystems . The degree to which bees living in human-dominated areas act as source habitats to adjacent landscapes should be explored further.This research study aims to pinpoint when and where various bee genera experience habitat patch dynamics versus habitat fragmentation . Though bee and flower phenology are discussed at length in the literature, it has not been formally explored from an explicitly spatial perspective at the landscape scale. As such, this is the first study to document bee habitat spatial and temporal dynamics providing important novel insight for future bee conservation efforts or further studies in bee habitat fragmentation. The main research questions covered in this paper include: What constitutes a gap in native bee pollinator habitat? How do bees’ annual spatial patterns of potential versus realized habitat function over a year? How can bee foraging maps be used to conduct conservation gap analysis at a local scale? How do partial and full WHR plant palettes compare in terms of habitat modeling for bees? and, What are the implications of this research for bee habitats?To help solve these complicated topics, the wildlife-habitat relationship models developed for bees in Chapter 1 are used here to spatially estimate and measure gaps in bee genera pollination networks. The spatial analyses presented in this study shed light on the degree of real-world bee habitat fragmentation. Monthly observed bee georeferenced data which were compiled in Chapter 1, are here analyzed via GIS maps. Seasonality of both bees and flowers were integrated in the mapping model analysis in an effort to identify where and when bee potential habitat gaps and potential fragmentation might occur The spatial distribution patterns of suitable, insufficient, or unsuitable habitat for bees are still poorly understood, yet essential to understand for native bee conservation purposes. This study investigates if bee foraging maps can be used to conduct both spatial and temporal habitat gap analysis.The UC Davis Arboretum and Public Garden , located in California’s Central Valley, is a unique environment to study bee patch dynamics and potential habitat fragmentation. The curated plant collection is fully mapped and contains 35 gardens along the linear Arboretum landscape . Each themed garden has a distinct geographically defined border, and plants are identified/labeled to the plant species, subspecies, or cultivar level. Arboretum maps are spatially accurate within two meters .Computer geographic information systems technology, using ArcGIS version 10.7.1 , was employed to investigate and calculate where and when habitat gaps occur for bees in the anthroscape. The theoretical framework was based on wildlife-habitat relationships conservation science modeling . In Chapter 1, monthly WHR foraging models were developed for each bee genus observed in the Arboretum. This chapter applies the predictive foraging models spatially, using the Arboretum’s geodatabase of mapped plants. Using this approach, a GIS predictive bee foraging model was developed and tested on-site with fieldwork to verify and quantify differences .

As such, more work can be done on how disturbances alter seed microbiome assembly processes and outcomes.The genus Vitivirus was created in 1997 for the classification of type member grapevine virus A , a plant virus discovered in grapevine with a filamentous flexuous particle differing from trichoviruses in its genomic arrangement. Vitiviruses have a single-stranded RNA genome encoding five genes: replicase , movement protein, coat protein , nucleic-acid-binding protein and a 20 kDa protein of unknown function. In the 2018 International Committee of Virus Taxonomy Master Species List , nine species of vitivirus infecting grapevine are recognized: Grapevine virus A, Grapevine virus B, Grapevine virus D, Grapevine virus E,Grapevine virus F, Grapevine virus G, Grapevine virus H, Grapevine virus I and Grapevine virus J. Since 2019, two more proposed vitiviruses were discovered in grapevine. Grapevine virus L was initially identified in RNAseq data and later detected in multiple plants in Croatia, New Zealand and the United States. Grapevine virus M was also discovered by high throughput sequencing in an American hybrid grapevine. Three different vitiviruses have been associated with the etiology of rugose wood disease in grapevine, a disease with world-wide distribution.

GVA is associated with stem grooving on the variety Kober 5BB, grapevine virus B was identified as the causal agent of corky bark in the variety LN33, procona florida container and grapevine virus D was implicated in growth reduction in the rootstock Freedom. Additionally, these vitiviruses are frequently detected in coinfection with grapevine leafroll viruses, resulting in synergistic interactions that can lead to lethal effects in several scion and rootstock combinations. The potential pathogenic role of the remaining grapevine vitiviruses, including proposed members, is still unknown. Reliable diagnostic methods are critical in determining the viral infection status of a grapevine. Multiple tests are available for the detection of vitiviruses, including biological indexing, real-time or end-point reverse transcription PCR and HTS. Biological indicators do not show symptoms for all vitiviruses infecting grape, and RT-PCR assays can fail to detect vitivirus variants containing nucleotide differences at critical primer binding locations. HTS is the most effective means of detecting all vitiviruses but its high cost at large scale limits its use as a screening tool. HTS data is helpful to inform and update RT-PCR primer design as new virus strains are continually being characterized. In this study, a universal end-point RT-PCR assay involving degenerate primers with the capacity of detecting all the known grapevine vitiviruses was developed.

To validate the new assay, eleven grapevines each infected with one of the vitiviruses were tested. Moreover, a field survey was conducted of known vitivirus-infected grapevines. Following the first reports of vitiviruses in grapevine, several vitiviruses have been discovered in other hosts; consequently, we investigated if the universal assay can detect these vitiviruses.A universal assay able to detect all known grapevine vitiviruses and potentially other members of the genus Vitivirus was developed here based on sequence data available in GenBank. The presence of highly conserved motifs in the REP protein allowed the design of end-point RT-PCR primers, providing an alternative assay to reduce the work associated with the diagnosis of vitiviruses. The extensive sequence divergence existing among grapevine vitiviruses, observed at the nucleotide and aa levels, makes it difficult to design a test with broad-range detection. RT-PCR with degenerate primers is a simple strategy that is frequently used for the specific and simultaneous detection of multiple viruses. Assays involving degenerate primers targeting grapevine vitiviruses have been described before, however, these studies were conducted in the pre-HTS era, when fewer vitiviruses were known and sequence data was limited.

Although Vitis spp. is recognized as the main host associated with the genus Vitivirus, vitiviruses have been identified in other perennial hosts, the majority of which are woody plants. For example, vitiviruses have been reported in blackberry, mint, agave and recently in blueberry. The universal assay successfully detected MV-2 in mint, however failed to detect BGMaV in blueberry. Additional investigation revealed a variation in motif A of BGMaV , and a similar scenario was observed in AVV. Based on our PCR results we predict that the universal assay will miss AVV during diagnosis, though, the rest of the known vitiviruses do not display any a discrepancy in motifs A or B and they should be detected by the assay. The family Betaflexiviridae comprises twelve different genera , including Vitivirus, Trichovirus and Foveavirus. After in-silico and in-vitro analyses of trichoviruses and foveaviruses, we did not find evidence for cross-reaction by the universal assay. A single test for all known grapevine vitiviruses can be a useful tool for improving efficiency and reducing costs of large-scale surveys. Potentially, this generic assay may detect novel Vitivirus species in grapevine and other hosts given its unbiased nature. Similar assays have been developed for carlaviruses, nepoviruses and different members of the family Betaflexiviridae. Grapevine is clonally propagated, consequently, to prevent the spread of vitiviruses, it is critical to use virus tested material. The assay developed here will be made available to diagnostic labs and will facilitate the production of certified virus-tested propagation material and the effective control of vitiviruses.Over the last few decades, the field of nutrition has grown and evolved. Although we continue to define the critical roles that nutrients play as fuel sources, enzyme cofactors, signaling molecules, and vital infrastructure for our bodies, the cutting edge of nutrition research is pushing beyond simply meeting our bodies’ basic needs. Indeed, as the population is living longer, an emerging focus for nutrition has been on obtaining and maintaining optimal health over the life course. On 10 October, 2022, the Council for Responsible Nutrition held their annual Science in Session conference entitled Optimizing Health through Nutrition – Opportunities and Challenges. The audience consisted of scientists and executives from dietary supplement and functional food companies as well as nutrition graduate student awardees of a CRN and ASN Foundation educational scholarship to attend the symposium. CRN is a trade association representing dietary supplement and functional food companies. The goals for this meeting were to propose a definition for optimal nutrition and identify strategies and tools for evaluating optimal health and nutrition outcomes while highlighting the gaps in this emerging space. Now more than ever in history, our population’s health has emerged as a global priority. Currently, 6 in 10 adults in the United States have a chronic disease, and 4 in 10 have 2 or more. In <10 y, the number of older adults is projected to increase by ~18 million. This means that by 2030, 1 in 5 Americans is projected to be 65 y old. As the major risk factor for many chronic illnesses is age, it is anticipated that the rates of all age-related diseases, especially chronic diseases, will skyrocket, potentially overwhelming the health care system. We need to enable the health care system—and the population—to be more proactive rather than reactive toward health outcomes. There is a critical need to help find solutions to optimize health across the lifespan to support living better longer, i.e., health span. Ensuring optimal nutrition is a significant and easily modifiable variable in the solution for maintaining and improving health span. We need to advance concepts beyond essential health and consider meeting the nutritional needs for optimal health. Although the nutrition science community is moving toward the vision of nutrition to support optimal health, many challenges and gaps still exist, but there are also recent advances and exciting opportunities.

The goal of the CRN “Science in Session” workshop was to discuss these challenges, gaps, and opportunities in order to advance the concept of nutrition for optimal health. This review summaries these findings and discussions.The DRIs for individual nutrients, procona London container including the Estimated Average Requirement and the RDA, are life stage- and sex specific recommendations for Americans and Canadians. These reference intakes were established in the 1990s by the Food and Nutrition Board of the National Academies of Sciences, Engineering, and Medicine to prevent deficiency disease and to reduce the risk of chronic diseases. However, incorporating chronic disease endpoints has been extremely challenging, primarily because data are largely lacking. Thus, the current DRIs, including the RDAs that are aimed to cover the nutrient needs of 98% of the population, do not account for the amount of a nutrient that one needs in order to achieve and maintain ‘optimal’ health.The science of resilience is not a new concept—this scientific concept was documented in the literature as early as the 1800s; the terminology entered the biomedical sciences in the mid- 1900s and emerged in the early 2000s as a concept to be interconnected in multiple health domains. The questions dominating its broad use and applicability tend to focus on how to define resilience. In 2019, the Trans-NIH Resilience Working Group was formed with a goal to develop an NIH-wide definition of resilience and to achieve consistency and harmony on the design and reporting of resilience research studies. In 1993, an introductory manuscript to a special issue published on the science of resilience included a quote stating, “resilience is at risk for being viewed as a popularized trend that has not been verified through research and is in danger of losing credibility within the scientific community”. The authors of the manuscript also warned against definitional diversity with respect to measures of resilience and urged researchers to clearly operationalize the defi- nition of resilience in all research reports. Remarkably, this call to action served as a primary aim of the Trans-NIH Resilience Working Group when it was organized >25 y after the 1993 special issue on resilience. One of the first activities of the Trans-NIH Resilience Working Group was to host a workshop, in March 2020, which led to the development of a definition of resilience and a conceptual infographic. The definition was intended to be applicable and useful across multiple domains, and it states that resilience encompasses “A system’s capacity to resist, recover, grow, or adapt in response to a challenge or stressor” . A system can represent different domains, levels, and/or processes. Over time, a system’s response to a challenge might show varied degrees of reactions that likely fluctuate in response to the severity of the challenge, the length of time exposed to the challenge, and/or innate/intrinsic factors. To show applicability of the definition in resilience research studies, the Resilience Research Design Tool was later developed to help improve consistency in resilience research reports and to facilitate harmony with respect to measures of resilience outcomes. One of the goals of the resilience framework is to reframe the way we ask research questions, particularly about nutritional interventions like dietary supplements, so that we can better understand health outcomes that are not based solely on disease end points. Going forward, as researchers across various scientific domains and sectors come closer to a unified definition of resilience and perhaps agree to the use of a standard checklist for designing and reporting on resilience studies, there is greater opportunity to harmonize the science and develop more empirical evidence of resilience outcomes.Optimizing performance also includes building resilience in order to enhance the ability to perform tasks and ensuring resilience in order to prevent illness, injury, and disease. Within the US Department of Defense, researchers are able to study different models of physical and psychological stress and the application of different nutritional interventions with Service Members throughout their careers. Various models of stress are introduced, including initial military training , advanced military training courses , service academies , and extreme environments , along with examples of various interventions and outcome measures collected to date. The importance of nutrition on readiness and resilience was identified in military populations more than a decade ago and continues to be of interest. Two specific examples are provided to further explore nutrition interventions aimed at optimizing performance in the Department of Defense. The first, a completed double-blind, randomized, placebo-controlled trial, used a calcium and vitamin D fortified food product to optimize bone health during initial military training of Marine Corps recruits. Using a supplement or food intervention for calcium and vitamin D, participants received 2000-mg calcium and 1000-IU vitamin D per day. The primary outcomes of the study showed that bone markers and vitamin D status improve, but the supplementation did not affect skeletal parameters.

In contrast, the floral odors that attract moth pollinators have been more extensively researched. In this study we determined that CMV infection induced changes in olfactory cues emitted by Arabidopsis thaliana and tomato plants in ways that could be perceived by the bumblebee Bombus terrestris, and confirmed in tomato that this was associated with quantitative and qualitative changes in the blend of plant-emitted volatile organic compounds . We also elucidated a role for the host microRNA pathway in regulating the emission of bee-perceivable olfactory cues. Our data indicated that bumblebees possess an innate preference for olfactory signals emitted by CMV-infected tomato plants and we mathematically modeled what the possible wider implications of this might be if a similar preference occurred in wild host plants under natural conditions.In ‘free-choice’ assays, bumblebees encountered flight arenas containing ten tomato plants concealed within towers designed to allow odors to diffuse out but prevent the bees from seeing or touching the plants . Cups that were placed on top of towers hiding plants of both treatment groups offered bumblebees the identical ‘incentive’ of a 30% sucrose solution. Nonetheless, plastic planter pot when presented with mock-inoculated and CMV-infected tomato plants, bumblebees preferred to visit the towers that were hiding infected plants .

Bumblebees showed similar preferences for flowering and non-flowering CMV-infected plants, indicating that leaves were the main source of attractive volatiles . Bumblebees also displayed a preference for CMV-infected tomato plants over plants infected with CMVΔ2b, a viral mutant lacking the gene for the 2b VSR , a factor that also influences CMV-plant-aphid interactions.The results obtained in free-choice assays with tomato plants infected with CMVΔ2b suggested that the 2b protein, which is a VSR, may be exerting effects on the metabolism of plant volatiles by interfering with host small RNA pathways. The model plant Arabidopsis is the best higher plant system to use to investigate the effects of small RNA pathways. However, whilst Arabidopsis plants emit potentially pollinator-influencing volatiles, this species is not bee-pollinated. Consistent with this, bumblebees showed no significant difference in preference for volatiles emitted by CMV-infected versus mock-inoculated Arabidopsis plants in free-choice assays . An alternative approach to investigate the ability of bees to recognise differences in olfactory or other stimuli is to set up a differential conditioning or ‘learning curve’ assay. A differential conditioning assay can reveal whether bees can perceive cues that would not normally induce any behavioural responses and that could not be studied in free-choice assays.

In our differential conditioning assays, cups on towers offered bumblebees either a 30% sucrose solution ‘reward’ for choosing one treatment group or a ‘punishment’ for choosing the other group. Bumblebees cannot distinguish quinine from sucrose except by taste. Thus, increasing frequency of visits to sucrose-offering towers over the course of an experiment indicated that bees have learned to use plant odor as a cue to identify and avoid drinking from cups placed on towers offering quinine solutions. In these assays, a steep learning curve shows that bumblebees can easily distinguish between two treatment groups, and indicates that the volatile blends are likely to be qualitatively and/or quantitatively very distinct, whereas less steep curves indicate that differences between blends are less marked, and that bees find it more difficult to learn to distinguish between them based on odor. An illustration of the power of this approach is shown in Fig 2 . Although bumblebees displayed an innate preference for volatiles emitted by CMV-infected tomato plants in free choice assays , they could be trained by differential conditioning to overcome their innate preference and instead preferentially visit mock-inoculated tomato plants and avoid CMV-infected plants . Although we had observed that bumblebees had no innate preference for, or aversion to, volatiles emitted by Arabidopsis plants , differential conditioning assays revealed that the insects could recognize differences between volatiles emitted by Arabidopsis plants that had been mock-inoculated and by plants that were infected with CMV . Bumblebees could also distinguish between CMV-infected and CMVΔ2b-infected Arabidopsis plants .

Hence, although they exhibit no innate behavioural response to the volatile blends emitted by Arabidopsis plants, differential conditioning assays showed that bumblebees could perceive differences in volatiles emitted by these plants. This meant that differential conditioning assays could permit further dissection of the mechanisms underlying CMV-induced changes in volatile emission using Arabidopsis as a model system. Bumblebees could learn to differentiate transgenic plants constitutively expressing the 2b VSR from non-transgenic plants and from control-transgenic plants that were expressing an untranslatable 2b transcript . However, the insects displayed less ability to learn to distinguish mock-inoculated from CMVΔ2b-infected plants . Comparison of the learning curves in Fig 3A versus Fig 3E by logistic regression indicated that bumblebees were better at distinguishing mock-inoculated plants from CMV-infected plants than from CMVΔ2b-infected plants = 40.17, p < 0.0001. Bees could not be trained to differentiate non-transgenic plants from control-transgenic plants expressing a non-translatable 2b transcript . The results with CMVΔ2b suggested that the 2b VSR plays an important role in altering the emission of bee-perceivable olfactory cues emitted by tomato and Arabidopsis plants . However, CMVΔ2b accumulates to lower levels in plants than wild-type CMV and in previous work it was found that viral titer, as well as the presence of the 2b protein, plays a role in modification of the interactions of Arabidopsis with aphids. Hence, it was conceivable that differences in virus titer might affect the emission of bee-perceivable volatiles by plants infected by CMV or CMVΔ2b and explain why the bees found it difficult to distinguish CMVΔ2b-infected plants from mock-inoculated plants. However, it is known that CMVΔ2b accumulates to levels comparable to those of wild type CMV in Arabidopsis plants carrying mutations in the genes encoding the Dicer-like endoribonucleases DCL2 and DCL4, which are important factors in antiviral silencing. Therefore, we examined the ability of bumblebees to learn to distinguish between volatile blends emitted by CMVΔ2b-infected and mock-inoculated dcl2/4 double mutant plants . The resulting learning curve was not significantly different from that obtained using wild-type plants that had been mock inoculated or infected with CMVΔ2b = 0.66, p = 0.42, indicating that an increase in CMVΔ2b titer did not enhance bee learning. Although we cannot rule out a role for other CMV gene products, the results indicate that the 2b VSR is the most significant viral factor conditioning changes in the emission of bee-perceivable volatiles.One of the host molecules that interact with the 2b VSR is the Argonaute 1 ‘slicer’ protein. AGO1 is required for silencing directed both by short-interfering RNAs and by miRNAs, which are generated by a specific host endoribonuclease from miRNA precursor transcripts encoded by nuclear genes. In differential conditioning assays, bumblebees were able to learn to distinguish between volatiles emitted by wild-type plants versus those emitted by ago1 mutant plants and those emitted by dcl1 mutant Arabidopsis plants . However, bumblebees showed little or no ability to learn to distinguish between volatile blends emitted by ago1 and dcl1 mutant plants, 30 litre plant pots indicating that the volatile blends emitted by plants of these two mutant lines were very similar . Thus, the miRNA-directed silencing pathway regulates the emission of bee-perceivable volatile compounds. Double mutant dcl2/4 plants are unable to generate CMV-derived short-interfering RNAs but are not affected in miRNA biogenesis. In CMV-infected dcl2/4 plants a higher proportion of the 2b protein is available to bind AGO1 and inhibit its miRNA-directed activity, which is likely to enhance virus-induced changes in emission of bee-perceivable volatiles. In line with this, bumblebees were able to learn to distinguish between volatiles emitted by CMV-infected wild-type and dcl2/4 double mutant Arabidopsis plants . As an additional control we showed that bumblebees could not learn to distinguish between volatiles emitted by mock-inoculated plants covered by towers offering sucrose rewards or quinine punishments .

The responses of bumblebees to CMV-infected tomato plants that were hidden from the insects indicated that changes in the emission of volatiles were affecting bee behavior and were responsible for the innate preference of these insects for CMV-infected plants . To confirm that CMV infection caused changes in the emission of VOCs, tomato plant headspace volatiles were collected and analysed by gas chromatography coupled to mass spectrometry . VOCs were collected from non-flowering mock-inoculated plants, plants infected with CMV-Fny and plants infected with the 2b gene deletion mutant of CMV-Fny, CMVΔ2b. The emitted VOCs were distinct from each other when compared by principal component analysis on the relative intensity of ions within the samples . PC1 explained 80.3% of the variation and discriminated between samples from mock-inoculated and CMV-infected plants, whereas PC2 discriminated between samples from mockinoculated and CMVΔ2b-infected plants . Thus, the VOC blend emitted by CMVinfected tomato plants was more distinct from that released by mock-inoculated plants than it was from the volatiles emitted by CMVΔ2b-infected plants. Nevertheless, VOC emission byCMVΔ2b-infected tomato plants was distinct from either mock-inoculated plants or CMV infected plant VOC emission , despite this mutant virus accumulating to markedly lower levels than CMV . Although CMV-infected plants were smaller than either mock-inoculated or CMVΔ2binfected plants, the emission of the combined volatiles on a whole plant basis was similar between mock-inoculated and CMV-infected plants . Indeed, expressing the emission of the combined VOCs by mass of tissue revealed that CMV-infected plants released greater quantities of volatiles compared to mock-inoculated and CMVΔ2b-infected plants . Thus, despite being stunted, CMV-infected plants generated a greater total quantity of VOC than either mock-inoculated or CMVΔ2b-infected tomato plants. Identification by GC-MS of the most abundant VOC by g dry weight of tomato plant tissue showed that terpenoids dominated the profile, with α-pinene, 2-carene, p-cymene, β-phellandrene and the sesquiterpene -caryophyllene being apparent . CMV infection caused quantitative changes in the profile of these VOCs; α-pinene and p-cymene emission increased markedly, whereas 2-carene and β-phellandrene did not, and -caryophyllene almost disappeared from the profile . Isomeric composition was not further determined than that stated here. When VOC emission was compared on a whole plant basis, α- pinene and p-cymene emission rates from CMV-infected plants appeared similar to mockinoculated or CMVΔ2b-infected plants, while 2-carene and β-phellandrene levels appeared to be lower . Bumblebees of a closely related species are known to be repelled by β-phellandrene and 2-carene. Thus, lower emission values of these VOCs from CMV-infected plants may explain why bumblebees displayed an innate preference for CMV-infected tomato plants over mock-inoculated plants in free choice assays . The VOC profiles of mock-inoculated and CMVΔ2b-infected plants were similar, although not identical , and this could explain the bees’ lack of preference in free choice assays . However, optimal self-fertilization requires sonication of the flower to release pollen from the anthers onto the stigma, which is provided either by buzz-pollination from a bee or simulated buzzpollination using mechanical vibration. This is illustrated in Fig 5A, which shows how mechanical buzz-pollination of flowers increased seed production by around a third. Seed production by tomato was very dramatically decreased in plants infected with CMV-Fny to less than 10% of the yield in mock-inoculated plants . Remarkably, artificial buzz-pollination of flowers of CMV-infected plants rescued seed production to a significant degree with seed numbers reaching approximately half the level seen for non-buzzed flowers of mock inoculated plants and about 6- to 7-fold greater than the number of seeds produced in nonbuzzed, CMV-infected plants. The difference in seed yield between mock-inoculated and CMV-infected plants that had been vibrated was less marked than between non-buzzed, mock inoculated and CMV-infected plants . Although CMV-infected plants produced fewer seeds, the mass of individual seeds was unaffected by infection and was not affected whether or not flowers were vibrated . Additionally, the number of flowers produced by CMV-infected plants was similar to the number produced by mock-inoculated plants, and tomato flower morphology was also not markedly altered by infection . Overall plant growth was stunted by CMV infection but, interestingly, virus infection appeared to accelerate the appearance of flowers by a small but statistically significant degree . A recent report indicated that flowers of squash plants infected with the potyvirus zucchini yellow mosaic virus yielded decreased quantities of pollen. However, we found no significant differences in the quantity or viability of pollen released from mock-inoculated and CMV-infected tomato flowers .We investigated the effects of CMV infection on bumblebee-mediated pollination under glasshouse conditions in which the insects could see and interact with flowers .

The bioavailability and tissue distribution of phytochemicals in humans are key factors that need to be clearly established and associated with their biological effects. The fate of phytochemicals in the body, including absorption, metabolism, and distribution, may vary according to the categories of phytochemicals. Ingested polyphenols can be absorbed from the stomach or the small intestine and can undergo conjugation in the intestine and liver to give methyl, glucuronide, and sulfate derivatives . Native polyphenols can also break down, producing smaller phenolic acid derivatives, such as protocatechuic, vanillic, or ferulic acid. These phenolic acids can also undergo phase I and phase II metabolism in the liver . The bioavailability of plant-food bio-active compounds is complex and presents interindividual variation ; however, the extent of such variability and the major determinants involved are currently not established. An example of interindividual variation in the metabolism of plant bio-active compounds is the conversion by the gut microbiota of the soy isoflavone precursors, black plastic nursery pots daidzin and daidzein, to the microbial-derived metabolite equol.

After a soy challenge, 20–30% of Western and 50–60% of Asian populations produce equol. The bacteria involved in the conversion have been identified, but the determinants that govern the daidzein-metabolizing phenotype still have not been fully elucidated. The gut microbiota has also a key role in the metabolism of other plant-food bio-active compounds, such as lignans and ellagitannins . Genetic polymorphisms can also contribute to the interindividual variation in bioavailability. For example, the role of genetic polymorphisms in the interindividual variability in bioavailability of caffeine was demonstrated. Caffeine is mainly metabolized by cytochrome P450 1A2 in the liver, and subjects with the CYP1A2*1F allele variant are considered slow caffeine metabolizers compared with the rapid caffeine metabolizers carrying the wild-type allele . Other factors such as age, sex, and dietary habits may affect the bioavailability of plant-food bio-active compounds. For example, sex differences in the glucuronidation of resveratrol, a polyphenol present in grapes and wine, have recently been observed, which may be explained by sex-specific uridine 59-diphospho–glucuronosyltransferase isoenzyme expression profiles regulated by sex hormones . The existence of an interindividual variability in the bioavailability of plant-food bio-active compounds suggests that there could also exist an interindividual variability in biological response to the consumption of these compounds.

Heterogeneity in the responsiveness to plant bio-active compounds can obscure associations between habitual intakes and health outcomes, resulting in a potential masking of health benefits for specific population groups and thereby limiting our knowledge of the role of the different bio-actives for health. Improving our knowledge of the factors, both genetic and nongenetic [such as age, sex, or genotype], that influence whether plant-food bio-active compounds are more or less effective in individuals will be invaluable to progress in the development of effective and innovative solutions leading to health improvements . However, to date, this interindividual variation in efficacy of plant-food bio-active compounds to modulate physiological outcomes has been little explored. The aim of this review is to provide an overview of the existing studies, both prospective and clinical trials, that has revealed interindividual variability in the responsiveness to the consumption of major plantfood bio-active compounds present in our diet: polyphenols, caffeine, and plant sterols. This review focuses on interindividual variability regarding cardiometabolic outcomes and discloses the potential determinants involved.We identified 6 prospective studies addressing the impact of interindividual variability in biomarkers of cardiometabolic health after habitual intake of a range of different plantbio-active compounds, including coffee and soy .

One area of particular interest relates to the microbially derived soy isoflavone metabolite, equol. In one prospective study, which examined associations between urinary equol excretion, serum lipids, and carotid intima thickness in 572 Chinese participants, 25% were equol excreters on their usual diet. In relation to other characteristics, the number of equol producers was similar between men and women, and there was no significant difference between equol-producer phenotype and age, dietary intakes, blood pressure, or BMI . Equol excreters had significantly lower TG and IMT levels compared to non-equol excreters . Although there was no association between soy isoflavone intake and serum lipids or IMT in the non– equol excreters, equol excreters within the highest quartile of intake had significantly lower IMT and higher HDL cholesterol concentrations than those in the lowest quartile of soy intake. Although this was an Asian population, habitual intakes of isoflavones were low, with a mean intake of 13 mg/d in both the equol- and non–equol-producer groups . The findings are therefore intriguing because data from the extensive literature on soyintervention studies suggest that an isoflavone intake >25 mg/d is required for any biological or clinical effect . The lack of an effect of isoflavone intake on CVD risk in women from the EPIC population was therefore not surprising, given that the median intake of isoflavones was only 0.4 mg/d. This study did not assess equol-producer status, and there was no difference in the association between habitual isoflavone intake and CVD risk when stratified by smoking , BMI, hormone replacement therapy use, age at intake, and hypercholesterolemia . This prospective study also examined associations between habitual lignan intakes and CVD risk in women and observed no association with intake , although the authors suggested a decreased risk of developing CVD in participants who were past smokers and had a higher habitual lignan intake. Therefore, available data on soy and the microbially derived metabolite equol are very limited. The impact of the equol-producer phenotype requires further investigation in population groups in which there is a wide variability in intakes in order to more carefully examine the magnitude of interindividual variability in response to biomarkers of cardiometabolic health and particularly the importance of the microbially derived metabolite equol. Four prospective studies have examined the impact of several factors in explaining the association between coffee intake and CVD risk . Whether polymorphism in the CYP1A2 gene, coding for the main enzyme responsible for the metabolism of caffeine, modulates the association between coffee intake and risk of CVD and related biomarkers was addressed in 3 studies. In one study, the risk of hypertension associated with coffee intake was shown to vary according to CYP1A2 genotype, with carriers of the slow-metabolism *1F allele at increased risk with higher coffee intake but not participants with the fast metabolism *1A/*1A genotype .

In a more recent study from this same hypertensive cohort, the association between coffee intake and impaired fasting glucose was stronger in carriers of the *1F variant, with the highest risk in heavy drinkers [$4 cups/d ] . In relation to myocardial infarction, 30 plant pot in a case-control study coffee intake was only associated with an increased risk of nonfatal myocardial infarction among participants with slow-caffeine metabolism . Only one study examined whether the relation between coffee intake and incident of coronary artery disease is dependent on the metabolism of catecholamines, specifically polymorphisms of the catechol-Omethyltransferase gene. In a cohort of 773 men, the relation between consumption of caffeinated coffee and the incidence of fatal and nonfatal CAD was dependent on COMT genotype. In men who were either homozygous for the high-activity COMT allele or heterozygous, substantial coffee intake did not increase the incidence of acute coronary events. However, for those who were homozygous for the low-activity COMT allele, heavy coffee consumption was associated with a higher incidence of acute coronary events, and the relative CAD incidence was >200% higher among drinkers of >6.5 cups of coffee/d after multivariable adjustment . Taken together, these few prospective studies have shown that there is interindividual variability in response to the consumption of plant-food bio-active compounds and that individuals do not equally benefit from the consumption of these phytochemicals. Different determinants, such as gut microbiota, genetic polymorphism, or smoking, have been suggested to be involved in these between-subject variations. It should also be noted that coffee is a source of not only caffeine, the amount of which can vary depending on brewing , but also of other micronutrients, such as chlorogenic acid, which has been shown to mediate the blood pressure rise caused by coffee intake .Age is the strongest independent cardiovascular risk factor for CVD, as indicated in most methods of risk scoring, such as the Framingham risk score or the European Society of Cardiology SCORE system . Aging is also associated with increased vascular stiffness, endothelial dysfunction, and isolated systolic hypertension . All these age-associated changes in the vascular system are known to have an effect on the bioactivity of some drugs, such as verapamil, albuterol, or benzodiazepines , and potentially could also have an effect on the bioactivity of plant-food bio-actives, which undergo the same conjugation pathways when absorbed.To date, few studies have examined the effects of age on the cardiometabolic effects of food bio-active compounds . Three studies have investigated age-dependent effects of cocoa flavanols on vascular function , with conflicting results. However, only one of them was a controlled study specifically designed to investigate the effects of flavanols in the context of the aged cardiovascular system. A double-blind RCT demonstrated that consumption of a flavanol-rich drink 2 times/d for 2 wk reversed age-related increases in blood pressure together with vascular stiffness in healthy elderly men. CF-intake–associated improvements in the compliance of large arteries were complemented by a decrease in pulse wave velocity and aortic augmentation of systolic blood pressure . Endothelial function in large conduit arteries was also significantly improved in healthy young and elderly individuals. These beneficial effects were associated with an improved dilatory capacity of resistance arteries, lower diastolic blood pressure , and increases in microcirculatory perfusion and RBC deformability. Cardiac output was not affected by CFs. Importantly, despite age-dependent differences in baseline flow-mediated dilatation , PWV, and DBP, the magnitude of the changes in the vascular response to CFs was not significantly different between the young and the elderly. In contrast, flavanol consumption improved only SBP and the augmentation index in the elderly group . This is probably because SBP is slightly higher in the elderly, mainly caused by stiffer arteries. Plasma concentrations of flavanol metabolites were not significantly different between young and elderly individuals, suggesting that differences in bioavailability could not explain the differences observed in biological responses. Of note, endothelial dysfunction is a well-established response to cardiovascular risk factors and precedes the development of atherosclerosis. The measurement of ultrasound-based endothelium dependent FMD in the brachial artery is the more widely used noninvasive measure of endothelial function and constitutes a clinical surrogate marker of vascular health . This technique consists of assessing the change in the diameter of the brachial artery after the increase in shear stress induced by a reactive hyperemia, with the degree of dilatation reflecting arterial endothelial NO release . The aortic AIX is closely related to wave reflections and constitutes a surrogate marker of arterial stiffness . In agreement with previous data, a recent study showed that the absorption, distribution, metabolism, and excretion of CFs was not significantly different between young and elderly healthy subjects after consumption of a similar amount of CFs . However, small but significant differences in metabolism were reported at a higher intake amount of CFs , with higher glucuronidation, lower methylsulfation, and lower urinary excretion of gut microbial g-valerolactone metabolites observed in the elderly. This observation suggests that dose-response studies covering the amounts of bio-active intake that can be achievable through a normal diet are necessary when investigating the interindividual variability in the ADME of plant-food bio-active compounds. A study also investigated whether the consumption of a flavanol-rich cocoa drink could improve blood pressure and endothelial function in healthy young and elderly men . No changes in blood pressure or endothelial function were observed in any group after 4–6 d of daily CF consumption. However, an effect on the last day of the study was seen in both groups after 90–180 min of CF consumption, and when compared with baseline values of day 1, the effect was higher for the elderly volunteers. Pulse wave analysis showed a similar pattern, with higher vascular responses in the elderly after acute consumption. The authors attributed these effects to an increase in NO production because responses to the endothelial NO synthase inhibitor L-nitroarginine-methyl-ester were also greater in the elderly. Nevertheless, the relevance of comparing changes in vascular function after acute consumption on days 4–6 with baseline levels on day 1 remains to be established.

These chemical relationships are represented as a chemical tree that can be visualized in the context of sample metadata and molecular annotations obtained from spectral matching and in silico annotation tools. We show that such a chemical tree representation enables the application of various tree-based tools, originally developed for analyzing DNA sequencing data, for exploring mass-spectrometry data. Here, we introduce Qemistree software that constructs a chemical tree based on predicted molecular fingerprints from MS/MS fragmentation spectra. Molecular fingerprints are vectors where each position encodes a substructural property of the molecule, and recent methods allow us to predict molecular fingerprints from tandem mass spectra. In Qemistree, we use SIRIUS and CSI:FingerID to obtain predicted molecular fingerprints. Users can first perform feature detection to generate a list of observed ions with associated peak areas and MS/MS fragmentation spectra, referred to as chemical features henceforth, raspberry container size to be analyzed by Qemistree . Only chemical features with MS/MS data are included; features with only MS1 are not considered.

SIRIUS then determines the molecular formula of each feature using the isotope and fragmentation patterns and estimates the best fragmentation tree explaining the fragmentation spectrum. Subsequently, CSI:FingerID operates on the fragmentation trees using kernel support vector machines to predict molecular properties . We use these molecular fingerprints to calculate pairwise distances between chemical features and hierarchically cluster the fingerprint vectors to generate a tree representing their chemical structural relationships. Although alternative approaches to hierarchically cluster features based on cosine similarity of fragmentation spectra exist , we use molecular fingerprints predicted by CSI:FingerID for this. Previous work has shown that CSI:FingerID outperforms other tools for automatic in silico structural annotation. Therefore, we leverage it to search molecular structural databases to provide complementary insights into structures when no match is obtained against spectral libraries. Subsequently, we use ClassyFire to assign a 5-level chemical taxonomy to all molecules annotated via spectral library matching and in silico prediction . Phylogenetic tools such as iTOL can be used to visualize Qemistree trees interactively in the context of sample information and feature annotations for easy data exploration. The outputs of Qemistree can also be plugged into other workflows in QIIME 2 or in R, Python, etc. for system-wide metabolomic data analyses.

In this study, we apply Qemistree to perform chemically informed comparisons of samples in the presence of technical variation such as chromatographic shifts that commonly affect mass spectrometry data analysis. Additionally, we exemplify the use of a tree-based representation to visualize and explore chemical diversity using a heterogeneous collection of food products. Qemistree can be used iteratively to incorporate multiple datasets without the need for cumbersome reprocessing , allowing for large-scale dataset comparisons. Qemistree is available to the microbiome community as a QIIME 2 plugin and the metabolomics community as a workflow on GNPS2 . Thechemical tree from the GNPS workflow can be explored interactively using the QemistreeGNPS dashboard. To verify that molecular fingerprint-based trees correctly capture the chemical relationships between molecules, we designed an evaluation dataset using four distinct biological specimens: two human fecal samples, a tomato seedling sample, and a human serum sample. Samples were prepared by combining them in binary, tertiary, and quaternary mixtures in various proportions to generate a set of diverse but related metabolite profiles . Untargeted tandem mass spectrometry was used to analyze the chemical composition of these samples and obtain fragmentation spectra.

The mass spectrometry experiments were performed twice using different chromatographic elution gradients, causing a retention time shift between the two runs . Processing the data of these two experiments with traditional LC-MS-based pipelines leads to the same molecules being detected as different chemical features in downstream analysis. Figure 1 shows the analysis of pure samples to demonstrate this. In Extended Data Figure 4, we highlight how these technical variations make the same samples appear chemically disjointed. Using Qemistree, we mapped each of the spectra in the two chromatographic conditions to a molecular fingerprint, and organized these in a tree structure . Because molecular fingerprints are independent of retention time shifts, spectra are clustered based on their chemical similarity. It is noteworthy that the structural information from chemical features with spectral library matches or other forms of annotation could also be used to compare the chemical composition of samples across different mass spectrometry runs. Qemistree improves upon this by enabling the use of all MS/MS spectra with molecular fingerprints for downstream comparative analyses, by not constraining analysis to the chemical features with spectral matches only. This tree structure can be decorated using sample type descriptions, chromatographic conditions, spectral matches obtained from molecular networking in GNPS , and any other chemical annotations23,28. Figure 1 shows that similar chemical features were detected exclusively in one of the two batches. However, based on the molecular fingerprints, these chemical features were arranged as neighboring tips in the tree regardless of the retention time shifts. This result shows how Qemistree can reconcile and facilitate the comparison of datasets acquired on different chromatographic gradients.Having demonstrated Qemistree’s practical utility on biologically inspired synthetic datasets, we now turned to a conceptual example illustrating the general principle. We demonstrated an application of a chemical hierarchy in performing chemically informed comparisons of metabolomics profiles. In standard metabolomic statistical analyses, eachmolecule is assumed unrelated to the other molecules in the dataset. Some of the pitfalls of this assumption are highlighted in Figure 2a. Consider a scenario where we want to compare samples 1–3. An analysis schema that does not account for the chemical relationships among the molecules in these samples , will assume that the sugars in samples 2 and 3 are as chemically related to the lipids in sample 1 as they are to each other. This would lead to the naive conclusion that samples 1 and 2, and samples 2 and 3 are equally distinct, yet from a chemical perspective they are not. On the other hand, if we account for the fact that sugar molecules are more chemically related to one another than they are to lipids, we can obtain a chemically informed sample-to-sample comparison. The chemical structural compositional similarity metric29 was developed to compute pairwise sample-to-sample comparison by considering cosine similarity of MS/MS spectra from molecular networking. Here, we utilize a tree-based approach to account for chemical relationships, which allows us to adopt phylogeny-based tools for metabolomics analyses . Specifically, we first constructed a tree of chemical similarities by hierarchical clustering molecular fingerprints from CSI:FingerID . This tree is analogous to phylogenetic trees used in ecology, such that the tips of the tree are molecules . In Figure 2a, we show that by using a tree of chemical relationships between molecules in samples 1–3, we can visualize that sample 1 is chemically very distinct from samples 2 and 3. Returning to our evaluation dataset, raspberry plant container we can highlight the importance of comparing samples by accounting for their molecular relatedness. Principal coordinates analysis of the evaluation dataset that ignores the tree structure performs far worse than the Qemistree PCoA that uses the tree .

With the structural context provided by Qemistree, the differences between replicates across batches are comparable to the within-batch differences . The retention time shift in this dataset leads to a strong signal due to chromatography conditions that obscures the biological relationships among the samples . We observed and remediated a similar pattern originating from plate-to-plate variation in a recently published study investigating the metabolome and microbiome of captive cheetahs . In this study, placing the molecules in a tree using Qemistree reduced the observed technical variation , and highlighted the dietary effect that was expected . These results show how systematic and spurious molecular differences can be mitigated in an unsupervised manner using chemically informed distance measures based on a tree structure.As a case study demonstrating the utility of Qemistree on a set of biological specimens, we used the platform to explore chemical diversity in food samples collected in the GlobalFoodOmics initiative . Understanding the chemical relationships between different foods is challenging because most molecules within foods are unannotated. We selected a diverse range of food ingredients to represent animal, plant, and fungal groupings. We first performed feature-based molecular networking using MZmine to obtain spectral library matches for a subset of the chemical features . Using Qemistree, we collated GNPS spectral library matches and in silico predictions from CSI:FingerID to annotate ~91% of the chemical fingerprints with molecular structures. We also retrieved chemical taxonomy assignments for structures that were classified by ClassyFire; the remaining are in the queue to be processed on the ClassyFire server for taxonomy assignment . Labeling annotations allowed us to retrieve subtrees of distinct chemical classes such as flavonoids, alkaloids, phospholipids, acyl-carnitines, and Oglycosyl compounds in food products. We propagated ClassyFire annotations of chemical features to each internal node of the tree and labeled the nodes by pie charts depicting the distribution in chemical superclasses and classes of its tips. The molecular fingerprint-based hierarchy of chemical features agreed well with ClassyFire taxonomy assignment, further demonstrating that molecular fingerprints can meaningfully capture structural relationships among molecules in a hierarchical manner. Furthermore, Qemistree coupled the chemical tree to sample metadata, revealing distinct chemical classes expected for each sample type. Branches representing acyl-carnitines were exclusively found in animal products . In contrast, honey, although categorized as an animal product, shared most of its chemical space with plant products, reflective of the plant nectar and pollen-based diet of honey bees. We observed a clade of flavonoids in both plant products and honey , but no other animal-based foods. While it is expected that a complex food such as blueberry kefir contains molecules from both blueberries and dairy, we can now visualize how individual ingredients and food preparation contribute to the chemical composition of complex foods. We noted that metabolite signatures that stem directly from particular ingredients, such as phosphoethanolamine from eggs, are present in egg scramble , but not in the other two foods highlighted . We can also observe the addition of ingredients in foods that were not listed as present in the initial set of ingredients. We were able to retrieve that there is black pepper in the egg scramble with chorizo and orange chicken, but that this signal is absent from the blueberry kefir .We show that our tree-based approach coherently captures chemical ontologies and relationships among molecules and samples in various publicly available datasets. Qemistree depends on representing chemical features as molecular fingerprints, and does share limitations with the underlying fingerprint prediction tool CSI:FingerID. For example, fingerprint prediction depends on the quality and coverage of MS/MS spectral databases available for training the predictive models, and these will improve as databases are enriched with more compound classes. Nevertheless, the use of CSI:FingerID-predicted molecularfingerprints is highly advantageous. While annotations from spectral matches may be more accurate, their coverage is too low to adequately summarize the chemical content of complex samples. Qemistree is also applicable in negative ionization mode; however, fewer molecular fingerprints can be confidently predicted due to fewer publicly available reference spectra, resulting in less-extensive trees. A key contribution of this work is to introduce the concept of building chemical hierarchies that can be used to leverage phylogeny-based tools , for metabolomics data exploration. Hierarchical relationships have provided a powerful framework to understand the relatedness of organisms. These techniques form a cornerstone for the interpretation of genomics data with phylogenetics and phylogenomics, and even taxonomy. The suite of tools and algorithms that have been developed over the past few decades in these fields, which utilize hierarchical structures, potentially have general relevance to the investigation of mass spectrometry data. Using Qemistree we can begin to explore the applicability of other methods, such as Faith’s Phylogenetic Diversity to understand within-sample complexity, or phylogeneticindependent contrasts with a metabolomics-inspired topology as these representations enter regular use. We showed that a hierarchical representation could be used to infer chemically informed relationships between samples . While we used molecular fingerprints predicted by CSI:FingerID to build chemical hierarchies here, this approach can be extended to incorporate other strategies to compare molecules for building chemical trees. For example, chemical relationships based on assigned chemical classes, spectral motifs, shared biosynthetic origin or other structural comparison methods could also be used as a basis for such a tree. These approaches will result in different tree topologies capturing complementary chemical information for subsequent analyses.

Most pollinators are generalist foragers that can, in some contexts, switch between plant species within a single foraging bout . When pollinators are promiscuous within a single foraging bout, they may transfer heterospecific pollen to floral stigmas, which can have negative effects on both male and female elements of plant reproduction . While heterospecific pollen deposition is highly variable in nature , it can represent a substantial percentage of total pollen on a stigma, often more than 50% of grains . Second, there is the extreme example of an antagonistic interaction from pollinators wherein the visitors do not visit the flower “legitimately” but rather pierce holes in a flower’s corolla to access the nectar rewards without ever touching the reproductive parts of the flower and therefore not acting as a pollinator . Some researchers suggest that most all flowering plants with accommodating floral architecture will experience some degree of nectar robbing . Furthermore, if a pollinator possesses mouth parts capable of robbing, they are likely to act as both legitimate pollinators on some plant species and primary or secondary robbers on others . On the other side of the interaction, blueberry plant pot some plants produce chemicals in pollen or nectar that that can be harmful to the development of the bees that visit their flowers, reducing bee fitness .

It has long been recognized that exploitation of mutualisms is commonplace and can have substantial impacts on the evolutionary persistence of mutualisms . While our understanding of the extent to which antagonistic interactions between plants and their pollinators is not complete, the examples listed above are common enough that the inclusion of such demonstrated negative interactions on network dynamics and how they might impact the robustness of interactions to extinctions warrants exploration. Recent network studies have begun to explore interactions in a continuous, rather than binary positive-or-not framework . While such studies do not take the potential for negative interactions into account, these models allow for some pollinators to be “better” than others in the services they provide or, in the case of Vieira and Neto , to vary the amount of dependence that the mutualistic partners have on one another. Here, we build on binary network simulation modeling approaches to asses show negative interactions impact the effects of pollinator species losses on plant species persistence, i.e. network robustness. In previous simulations where all network interactions are considered positive, the removal of a given pollinator species could result only in the loss of one or more plants. By contrast, after incorporating negative interactions, extinction cascades, also become possible , who produce extinction cascades in an all positive framework by relaxing the assumption that extinctions only take place after all partners are lost). In other words, if the removal of a pollinator species causes plant extinctions, those losses can tip the balance of interactions toward the negative for remaining pollinator species, which can then go extinct, in turn potentially leading to further plant species losses, thus an extinction cascade.

Our study examines the overall robustness to extinctions in two ways R, or, the area under each extinction curve and extinction cascade length –i.e. higher order extinctions that occur beyond the induced pollinator knockout and the resulting plant extinction. To our knowledge there is only one network study that incorporates the possibility for negative interactions between plants and pollinators while examining the robustness of the network when faced with extinctions This model classifies interactions as either mutually beneficial or beneficial for one species and detrimental to the other. This is in contrast to our model that examines robustness of networks with all positive interactions to those that incorporate negative interactions . Importantly the Campbell model focuses on hypothetical networks and does not directly evaluate the role of assigned negative interactions in determining the robustness of the network to extinctions and furthermore does not compare networks with and without negative interactions. Our method of incorporating negative interactions in to empirical networks will allow for higher order extinction cascades, giving us a more realistic impression of what might happen to a network after pollinator extinctions. This is not the case in extinction simulations that simply allow for asymmetric positive interactions , as noted above. We examined the effects of two factors on network robustness: the proportion of negative interactions in the network, including an all-interactions positive control; and the order of extinction, random pollinator losses vs. specialist-to-generalist vs. generalist-to-specialist.

Removing specialists first could be the most probable extinction sequence as specialist pollinators also tend to be the rarest species , but see who show that loss of specialists can accelerate the rate of species loss. By contrast, generalists are thought to be the “backbone” of networks and when highly connected nodes are lost, networks are expected to collapse rather quickly . While losing generalist pollinators first from a network may seem unlikely, we have seen rapid declines and range contractions in several highly generalist bumble bee species which had previously been abundant .First, we hypothesized that increasing the proportion of negative interactions would lead to both a decrease in the robustness of the network and greater number of extinction cascades. Second, we hypothesized that inclusion of negative interactions would not change the effects of extinction order relative to all-interactions-positive networks, in which specialist-to generalist pollinator species removals had the least impact on plant extinctions, generalist-to-specialist removals had the most, and random removals intermediate between the two .Following previous binary network assessments of robustness , we used empirical networks to conduct our robustness assessments. We selected three plant-pollinator networks of varying size and connectance that represent a range of natural plant-pollinator interactions; as in previous assessments, this selection is not meant to be exhaustive . 1) The Clemens and Long network was collected on Pikes Peak, Rocky Mountains, Colorado USA. This is by far the largest with 97 plant species forming 918 unique pairwise interactions with 275 pollinator species. Data were collected in various subalpine habitats at 2500 m elevation over 11 years . 2) The Arroyo et al network data were collected at an elevation between 2200m and 2600m between 1980 and 1981 in the alpine zone of Cordon del Cepo in Central Chile. The network is medium in size with 87 plant species forming 372 unique pairwise interactions with 98 pollinator species. 3) The Dupont network is the smallest as data were collected on the sub-alpine desert above 2000 m on the island of Tenerife, Canary Islands. Data were collect between May 7 and June 7, 2001. The network consists of 11 plant species forming 109 unique pairwise interactions with 38 pollinator species. These networks were retrieved from the NCEAS Interaction Web Database .We simulated extinctions by sequentially removing pollinator species one at a time and recording the number of plant species that were left with a positive sum of pollinator interactions. Plant species left with an interaction sum less than or equal to zero were then considered extinct and removed from the network due to assumed failure to sexually reproduce. Next, we evaluated if the secondary removal of those plant species left a pollinator species with an interaction sum less than or equal to zero. If yes, plastic gardening pots they were then considered extinct and removed from the network . This cycle continued until all plant and pollinator species were left with an interaction sum greater than 0 at which point the simulation moved on to the next pollinator species knockout. We carried out the extinction simulations separately for each of the aforementioned 600 network configurations and each of the three extinction orders.We evaluated network robustness via two response variables: R – this is a quantitative measure of robustness of a network following a species knockout . R is a simple calculation of the sum of the remaining plant species at each time step along the extinction simulation. R is standardized by its maximum value which equals the starting number of plants * the starting number of pollinators . R was calculated for each of the 50 simulations per order and proportion negative for all three networks. 

Extinction cascades – the number of higher order extinction cycles that take place after a single pollinator species knockout.Cascade length is defined as the higher order extinctions that occurred beyond the induced pollinator knockout and the resulting plant extinction. A cascade of length 1 results when such plant extinctions lead directly to subsequent pollinator extinction, while a cascade of length 2 indicates additional subsequent plant extinction. We calculated the total number of cascade events that occurred across the entire knockout sequence for each of the 50 replicate network configurations for each set of starting networks using GLMs with Poisson errors and a log link function. This analysis directly mirrors the modeling approach for robustness discussed above, comprising two sets of models. The first set compared the effect of negative interactions, extinction order, and their interaction within each network , while the second set of models compared the effect of negative interactions, network ID, and their interaction, within each extinction order .Our study examined how both the incorporation of negative interactions into networks as well as order of extinction impacted robustness and total number of extinction cascades in three empirical networks. As expected, we found that incorporating negative interactions leads to lower network robustness via an increased rate of plant species loss in all three of the networks. Furthermore, when compared to random extinction order, simulations with generalist removed first show the lowest overall robustness whereas the removal of specialists first has the least impact on lowering network robustness. This is true for all three networks and followed our expectations based on previous network extinction simulations . While some of the results from our simulations were predictable or even mathematical inevitabilities , not all of our results could be predicted a priori. We did not expect the networks to behave idiosyncratically with respect to how negative interactions and extinction order impacted both R and the total number of extinction cascades. Specifically, we found that in the smallest of the networks , extinction order did not impact the magnitude of the effect of negative interactions on network robustness. The effect of removing generalist-first from the two larger networks seemed to supersede the impact of negative interactions, leading us to conclude that the impact of losing generalists can,in some cases, be so detrimental so as to make inclusion of negative interactions irrelevant. While extinction cascades can only take place in networks that incorporate negative interactions, the impact of negative interactions on total cascade length was unpredictable. Neither increasing negative interactions nor extinction order were a good predictor for how many extinction cascades networks would take place in the simulations. Total number of extinction cascades is likely relate to structural properties unique to each network and warrant further exploration. It difficult to draw conclusions as to why these distinct differences between networks were seen, as this study did not include an exhaustive exploration of other network properties that may influence network robustness and total extinction cascades. Future studies should focus on selecting networks with systematically varying properties such as; network size, nestedness, and connectance. These properties each have the potential to influence the impact that negative interactions have on network robustness. Nestedness is a network property that has been identified in nearly all empirical networks. It is the tendency of specialists to interact with generalists in the network . We would expect that an increase in nestedness could make the networks more robust to extinctions due to the redundancy in the number of pollinators available per plant. Still, Campbell et al. found that high values of nestedness actually can have the opposite effect, and can decrease robustness after the loss of a single species, in some extreme cases lead to the total collapse of the highly nested community. In our simulations, networks with more generalists will be inherently more vulnerable because true specialists were excluded from assignment of negative interactions . While we do not expect this limitation to impact our results dramatically—as true specialists that only interact with one or two plant species are uncommon—to fully explore the role of nestedness one would need to develop a method of assigning negative interactions that distributes the negatives evenly among the specialists and generalists. Connectance is simply the proportion of realized links in a network . In this study we, by chance, selected a range of network size and connectance.