Relative Pollen Research Articles

Pollen quality for pollinators tracks pollen quality for plants in Mimulus guttatus

Pollen viability ranges substantially among and within plant species, and can be reduced by various factors, including pollution and inbreeding. While it is evident that a reduction in gamete quality reduces the male component of plant fitness, the implications of changes in pollen viability on the nutritional value of pollen for pollinators has not been studied. To test this, we created a greenhouse population of Mimulus guttatus plants that ranged in pollen viability from 17% to 98.5% and related both pollen quantity and pollen protein content to pollen viability. We found that crude protein concentration ranged from 15% to 45% and was positively associated with pollen viability. We also found that pollen mass per flower ranged 5‐fold, also positively related to pollen viability. Across experimental plants, there was an 11‐fold difference in protein mass per flower between plants of the lowest and highest pollen viabilities. We conclude that conditions that lead to reduced pollen viability, especially early in pollen development, may greatly reduce the awards available to plant pollinators.

An unexpected record of the PETM in terrestrial and organic sediments of Avesnois, between the Paris and Belgian Basins, NW Europe

An unexpected record of the PETM in terrestrial and organic sediments of Avesnois, between the Paris and Belgian Basins, NW Europe

Present-day vegetation helps quantifying past land cover in selected regions of the Czech Republic.

The REVEALS model is a tool for recalculating pollen data into vegetation abundances on a regional scale. We explored the general effect of selected parameters by performing simulations and ascertained the best model setting for the Czech Republic using the shallowest samples from 120 fossil sites and data on actual regional vegetation (60 km radius). Vegetation proportions of 17 taxa were obtained by combining the CORINE Land Cover map with forest inventories, agricultural statistics and habitat mapping data. Our simulation shows that changing the site radius for all taxa substantially affects REVEALS estimates of taxa with heavy or light pollen grains. Decreasing the site radius has a similar effect as increasing the wind speed parameter. However, adjusting the site radius to 1 m for local taxa only (even taxa with light pollen) yields lower, more correct estimates despite their high pollen signal. Increasing the background radius does not affect the estimates significantly. Our comparison of estimates with actual vegetation in seven regions shows that the most accurate relative pollen productivity estimates (PPEs) come from Central Europe and Southern Sweden. The initial simulation and pollen data yielded unrealistic estimates for Abies under the default setting of the wind speed parameter (3 m/s). We therefore propose the setting of 4 m/s, which corresponds to the spring average in most regions of the Czech Republic studied. Ad hoc adjustment of PPEs with this setting improves the match 3–4-fold. We consider these values (apart from four exceptions) to be appropriate, because they are within the ranges of standard errors, so they are related to original PPEs. Setting a 1 m radius for local taxa (Alnus, Salix, Poaceae) significantly improves the match between estimates and actual vegetation. However, further adjustments to PPEs exceed the ranges of original values, so their relevance is uncertain.

The Safety of Specific Immunotherapy for Patients Allergic to House-Dust Mites and Pollen in Relation to the Development of Neoplasia and Autoimmune Disease: A Long-Term, Observational Case-Control Study

Background: Specific immunotherapy (SIT) safety has been well documented. However, the prolonged late side effects in patients who terminated SIT several years previously have been reported on in only a limited number of studies. The aim of this study was to perform a 20-year post-SIT observational evaluation for the assessment of any manifestation of serious immunological disease. Materials and Methods: In total, 1,144 patients (521 women and 623 men), with a mean age of 22.8 ± 16.9 years (at the moment of SIT completion) and who had atopic bronchial asthma and/or allergic rhinitis, were observed 20 years after immunotherapy. New neoplastic and autoimmune disease cases were monitored. The SIT group was compared to a control group consisting of 1,154 allergic patients who had never received SIT and had only had symptomatic treatment. Results: There was an inverse association between SIT treatment and the prevalence of new chronic myeloid leukemia and chronic lymphoblastic leukemia cases (OR 0.32, 95% CI 0.18-0.81 and OR 0.58, CI 0.44-0.78, respectively). In other neoplastic diseases, however, prevalences similar to those observed in the control group were confirmed. There were also no significant differences in the autoimmune disease prevalence between the analyzed groups. Conclusion: The results of this long-term observational study indicate a lack of a significant prevalence for new instances of neoplastic and autoimmune diseases, which suggests that SIT in the long term is indeed safe.

A process-based approach to predicting the effect of climate change on the distribution of an invasive allergenic plant in Europe.

Ambrosia artemisiifolia is an invasive weed in Europe with highly allergenic pollen. Populations are currently well established and cause significant health problems in the French Rhône valley, Austria, Hungary and Croatia but transient or casual introduced populations are also found in more Northern and Eastern European countries. A process-based model of weed growth, competition and population dynamics was used to predict the future potential for range expansion of A.artemisiifolia under climate change scenarios. The model predicted a northward shift in the available climatic niche for populations to establish and persist, creating a risk of increased health problems in countries including the UK and Denmark. This was accompanied by an increase in relative pollen production at the northern edge of its range. The southern European limit for A.artemisiifolia was not expected to change; populations continued to be limited by drought stress in Spain and Southern Italy. The process-based approach to modelling the impact of climate change on plant populations has the advantage over correlative species distribution models of being able to capture interactions of climate, land use and plant competition at the local scale. However, for this potential to be fully realised, additional empirical data are required on competitive dynamics of A.artemisiifolia in different crops and ruderal plant communities and its capacity to adapt to local conditions.

Cross-Sensitization To Rosaceae Fruits and Their Molecular Components In Japanese School Children

Cross-reactivity between plant allergens is a central pathogenesis of fruit allergy. Cross-sensitization status to both fruits and related pollen in general population is to be clarified. Objectives: To describe the prevalence of allergic sensitization to Rosaceae fruits in general risk children in Japan and to evaluate the cross-sensitization characteristics between the fruit molecular components and its related birch pollen proteins.

Impacts of Canopy Position, Plants’ Age and Pollinators on PrePollination Pollen Viability of Tectona grandis Linn

Tropical trees may change in pre-pollination pollen viability according to the types of pollinator, flower exposure, canopy position, plants' age and pollen predacious ants and thrips. A canopy study system was used and in vitro pollen viability measures, to evaluate relative impact of such factors in one forest tree species, monsoon forest teak, Tectona grandis Linn. of the family Verbenaceae. Relative pre-pollination pollen viability was measured with Alexander's stain, TTC, MTT stain and in vitro pollen germination. The impact of ants and thrips was gauged from bagged and open flowers, and direct counts on flowers in Tectona grandis; canopy position impact (affecting light, exposure, temperature) was gauged using collections made with pole clippers, and by reciprocal transfer of inflorescences on branches. A few native insect species were scored for viable pollen as they first left their nests in early morning. Viability decreased according to ants and thrips presence in Tectona grandis. Pre-pollination pollen viability was higher on the higher flowers. Pollen viability in transplanted inflorescences changed. Relative pollen viability was in agreement using different in vitro pollen viability tests. The frequency of pollen viability increased steadily from low to high canopy height. The percentage of viable pollen was different for different assays showing similar trend to increase. The middle and high canopy height did not show much variation in pollen viability compared to low canopies. So the tree canopy height may attain a vital role in determining pollen viability. The proportion of viable pollen, varied greatly, showing a steady increase with plant age, and the number of pollen grains per flower also increased with age. These results support the qualitative prediction that pre-pollination pollen viability gradually decreases with ageing, reflecting an increase in genetic load with age.

Relative pollen productivities of typical steppe species in northern China and their potential in past vegetation reconstruction

The Relative Pollen Productivities (RPPs) of common steppe species are estimated using Extended R-value (ERV) model based on pollen analysis and vegetation survey of 30 surface soil samples from typical steppe area of northern China. Artemisia, Chenopodiaceae, Poaceae, Cyperaceae, and Asteraceae are the dominant pollen types in pollen assemblages, reflecting the typical steppe communities well. The five dominant pollen types and six common types (Thalictrum, Iridaceae, Potentilla, Ephedra, Brassicaceae, and Ulmus) have strong wind transport abilities; the estimated Relevant Source Area of Pollen (RSAP) is ca. 1000 m when the sediment basin radius is set at 0.5 m. Ulmus, Artemisia, Brassicaceae, Chenopodiaceae, and Thalictrum have relative high RPPs; Poaceae, Cyperaceae, Potentilla, and Ephedra pollen have moderate RPPs; Asteraceae and Iridaceae have low RPPs. The reliability test of RPPs revealed that most of the RPPs are reliable in past vegetation reconstruction. However, the RPPs of Asteraceae and Iridaceae are obviously underestimated, and those of Poaceae, Chenopodiaceae, and Ephedra are either slightly underestimated or slightly overestimated, suggesting that those RPPs should be considered with caution. These RPPs were applied to estimating plant abundances for two fossil pollen spectra (from the Lake Bayanchagan and Lake Haoluku) covering the Holocene in typical steppe area, using the “Regional Estimates of Vegetation Abundance from Large Sites” (REVEALS) model. The RPPs-based vegetation reconstruction revealed that meadow-steppe dominated by Poaceae, Cyperaceae, and Artemisia plants flourished in this area before 6500-5600 cal yr BP, and then was replaced by present typical steppe.

Palynological perspectives on vegetation survey: a critical step for model-based reconstruction of Quaternary land cover

1. Quantitative reconstruction of past vegetation distribution and abundance from sedimentary pollen records provides an important baseline for understanding long term ecosystem dynamics and for the calibration of earth system process models such as regional-scale climate models, widely used to predict future environmental change. Most current approaches assume that the amount of pollen produced by each vegetation type, usually expressed as a relative pollen productivity term, is constant in space and time.2. Estimates of relative pollen productivity can be extracted from extended R-value analysis (Parsons and Prentice, 1981) using comparisons between pollen assemblages deposited into sedimentary contexts, such as moss polsters, and measurements of the present day vegetation cover around the sampled location. Vegetation survey method has been shown to have a profound effect on estimates of model parameters (Bunting and Hjelle, 2010), therefore a standard method is an essential pre-requisite for testing some of the key assumptions of pollen-based reconstruction of past vegetation; such as the assumption that relative pollen productivity is effectively constant in space and time within a region or biome.3. This paper systematically reviews the assumptions and methodology underlying current models of pollen dispersal and deposition, and thereby identifies the key characteristics of an effective vegetation survey method for estimating relative pollen productivity in a range of landscape contexts.4. It then presents the methodology used in a current research project, developed during a practitioner workshop. The method selected is pragmatic, designed to be replicable by different research groups, usable in a wide range of habitats, and requiring minimum effort to collect adequate data for model calibration rather than representing some ideal or required approach. Using this common methodology will allow project members to collect multiple measurements of relative pollen productivity for major plant taxa from several northern European locations in order to test the assumption of uniformity of these values within the climatic range of the main taxa recorded in pollen records from the region.

Analysis of the Effects of Polymorphism on Pollen Profilin Structural Functionality and the Generation of Conformational, T- and B-Cell Epitopes

An extensive polymorphism analysis of pollen profilin, a fundamental regulator of the actin cytoskeleton dynamics, has been performed with a major focus in 3D-folding maintenance, changes in the 2-D structural elements, surface residues involved in ligands-profilin interactions and functionality, and the generation of conformational and lineal B- and T-cell epitopes variability.Our results revealed that while the general fold is conserved among profilins, substantial structural differences were found, particularly affecting the special distribution and length of different 2-D structural elements (i.e. cysteine residues), characteristic loops and coils, and numerous micro-heterogeneities present in fundamental residues directly involved in the interacting motifs, and to some extension these residues nearby to the ligand-interacting areas. Differential changes as result of polymorphism might contribute to generate functional variability among the plethora of profilin isoforms present in the olive pollen from different genetic background (olive cultivars), and between plant species, since biochemical interacting properties and binding affinities to natural ligands may be affected, particularly the interactions with different actin isoforms and phosphoinositides lipids species.Furthermore, conspicuous variability in lineal and conformational epitopes was found between profilins belonging to the same olive cultivar, and among different cultivars as direct implication of sequences polymorphism. The variability of the residues taking part of IgE-binding epitopes might be the final responsible of the differences in cross-reactivity among olive pollen cultivars, among pollen and plant-derived food allergens, as well as between distantly related pollen species, leading to a variable range of allergy reactions among atopic patients. Identification and analysis of commonly shared and specific epitopes in profilin isoforms is essential to gain knowledge about the interacting surface of these epitopes, and for a better understanding of immune responses, helping design and development of rational and effective immunotherapy strategies for the treatment of allergy diseases.

Pollen flow in fragmented landscapes maintains genetic diversity following stand-replacing disturbance in a neotropical pioneer tree, Vochysia ferruginea Mart

In forests with gap disturbance regimes, pioneer tree regeneration is typically abundant following stand-replacing disturbances, whether natural or anthropogenic. Differences in pioneer tree density linked to disturbance regime can influence pollinator behaviour and impact on mating patterns and genetic diversity of pioneer populations. Such mating pattern shifts can manifest as higher selfing rates and lower pollen diversity in old growth forest populations. In secondary forest, where more closely related pollen donors occur, an increase in biparental inbreeding is a potential problem. Here, we investigate the consequences of secondary forest colonisation on the mating patterns and genetic diversity of open-pollinated progeny arrays for the long-lived, self-compatible pioneer tree, Vochysia ferruginea, at two Costa Rican sites. Five microsatellite loci were screened across adult and seed cohorts from old growth forest with lower density, secondary forest with higher density, and isolated individual trees in pasture. Progeny from both old growth and secondary forest contexts were predominantly outcrossed (tm=1.00) and experienced low levels of biparental inbreeding (tm-ts=0.00-0.04). In contrast to predictions, our results indicated that the mating patterns of V. ferruginea are relatively robust to density differences between old growth and secondary forest stands. In addition, we observed that pollen-mediated gene flow possibly maintained the genetic diversity of open-pollinated progeny arrays in stands of secondary forest adults. As part of a natural resource management strategy, we suggest that primary forest remnants should be prioritised for conservation to promote restoration of genetic diversity during forest regeneration.

Identification of potential sources of airborne Olea pollen in the Southwest Iberian Peninsula

This study aims to determine the potential origin of Olea pollen recorded in Badajoz in the Southwest of the Iberian Peninsula during 2009-2011. This was achieved using a combination of daily average and diurnal (hourly) airborne Olea pollen counts recorded at Badajoz (south-western Spain) and Évora (south-eastern Portugal), an inventory of olive groves in the studied area and air mass trajectory calculations computed using the HYSPLIT model. Examining olive pollen episodes at Badajoz that had distinctly different diurnal cycles in olive pollen in relation to the mean, allowed us to identify three different scenarios where olive pollen can be transported to the city from either distant or nearby sources during conditions with slow air mass movements. Back trajectory analysis showed that olive pollen can be transported to Badajoz from the West on prevailing winds, either directly or on slow moving air masses, and from high densities of olive groves situated to the Southeast (e.g. Andalucía). Regional scale transport of olive pollen can result in increased nighttime concentrations of this important aeroallergen. This could be particularly important in Mediterranean countries where people can be outdoors during this time due to climate and lifestyle. Such studies that examine sources and the atmospheric transport of pollen are valuable for allergy sufferers and health care professionals because the information can be incorporated into forecasts, the outputs of which are used for avoiding exposure to aeroallergens and planning medication. The results of studies of this nature can also be used for examining gene flow in this important agricultural crop.

Quantitative Assessment of Wheat Pollen Shed by Digital Image Analysis of Trapped Airborne Pollen Grains

The objective of the present study was to develop a technique for quantifying the dynamics of wheat pollen shed under field conditions. Pollen traps with an adhesive film were used to assess the relative pollen shed of 12 winter wheat lines. Quantitative measurements were performed in 2012 and 2013 over a period of up to 20 days. The amounts of trapped pollen were automatically determined using a customized image analysis program. We demonstrated that this method is suitable for the assessment of wheat pollen shed. The possible impact of the technical advances revealed in this study for the selection of pollinators in routine wheat breeding programs is discussed.

Estimates of relative pollen productivity (RPP) for selected taxa from southern Greenland: A pragmatic solution

This paper presents the reanalysis of a modern surface sample pollen–vegetation dataset from southern Greenland aimed at extracting estimates of relative pollen productivity (RPP). RPP is a key parameter required in order to use models of pollen dispersal and deposition for hypothesis-testing and quantitative reconstruction of past vegetation cover from pollen assemblages. The data were originally collected for ordination analysis, but as sample points were located in relatively homogenous stands of representative vegetation it was possible to reprocess the vegetation data into a suitable format for obtaining initial estimates of RPP using the extended R-value approach. Although it is not possible to accurately estimate the relevant source area of pollen (RSAP) and thus obtain the best available estimates of RPP, preliminary estimates of RPP relative to Poaceae were obtainable for 17 pollen types.

Is relative pollen productivity constant across Europe?

Is relative pollen productivity constant across Europe?

The interpretation of Holocene pollen assemblages from coastal lowlands: differentiating natural and anth-ropogenic causal processes and the problems of equifinality

Vegetation reconstructions from palaeoecological records depend on adequate understanding of relationships between modern pollen, vegetation and climate. A key parameter for quantitative vegetation reconstructions is the Relative Pollen Productivity (RPP). Differences in both environmental and methodological factors are known to alter the RPP estimated significantly, making it difficult to determine whether the underlying pollen productivity does actually vary, and if so, why. In this paper, we present the results of a replication study for the Bashang steppe region, a typical steppe area in northern China, carried out in 2013 and 2014. In each year, 30 surface samples were collected for pollen analysis, with accompanying vegetation survey using the “Crackles Bequest Project” methodology. Sampling designs differed slightly between the two years: in 2013, sites were located completely randomly, whilst in 2014 sampling locations were constrained to be within a few km of roads. There is a strong inter-annual variability in both the pollen and the vegetation spectra therefore in RPPs, and annual precipitation may be a key influence on these variations. The pollen assemblages in both years are dominated by herbaceous taxa such as Artemisia, Amaranthaceae, Poaceae, Asteraceae, Cyperaceae, Fabaceae and Allium. Artemisia and Amaranthaceae pollen are significantly over-represented for their vegetation abundance. Poaceae, Cyperaceae and Fabaceae seem to have under-represented pollen for vegetation with correspondingly lower RPPs. Asteraceae seems to be well-represented, with moderate RPPs and less annual variation. Estimated Relevant Source Area of Pollen (RSAP) ranges from 2000 to 3000 m. Different sampling designs have an effect both on RSAP and RPPs and random sample selection may be the best strategy for obtaining robust estimates. Our results have implications for further pollen-vegetation relationship and quantitative vegetation reconstruction research in typical steppe areas and in other open habitats with strong inter-annual variation.

Relative pollen productivity estimates for herbaceous taxa in the Hulunbeier steppe, Inner Mongolia

We estimated relative pollen productivity estimates (PPE), key parameters for the quantitative interpretation of pollen data, for 13 taxa using modern pollen assemblages from 54 sites and recent vegetation data. Vegetation mapping in the area covered a minimum radius of 2 km around each sampling site. Vegetation data were weighted by the Prentice model, i.e. weighting by distance and by the dispersal–deposition parameters of different pollen types. PPE values were calculated by three submodels of the Extended R-value model. ERV 1 produced the best goodness of fit. The PPEs for Urtica and Sambucus nigra are published here for the first time, and the PPE for the Chenopodiaceae represents the first estimate for Europe. Values for the other ten taxa (Poaceae, Pinus, Salix, Fraxinus, Quercus, Tilia, Artemisia, Plantago lanceolata, Alnus and Cerealia) are comparable with or fall within the ranges of values published in previous studies. Herb taxa produce ca 3–11 times more pollen than the Poaceae. Herbs produce even more pollen than trees, whose production is 1–6 times higher than that of the Poaceae. The lowest pollen producers are the Cerealia, producing 20 times less pollen than the Poaceae. Our estimate of the relevant source area of pollen (RSAP) of 1050 m is relatively high compared to other studies in semi-open landscapes. This is possibly caused by the uneven pattern of some taxa in the vegetation mosaic (Pinus, P. lanceolata, Salix and Alnus). The distance of 1100 m, at which all taxa are present around each site, is similar to the RSAP distance (1050 m).

Role of food quality in bee susceptibility to fipronil and clothianidin

Pollen is the honey bee’s main protein supply. Newly emerged bees need pollen alimentation to guarantee correct development of physiological conditions and breeding potential. To ensure functional and efficient adult bees not only the quantity but also the quality of pollen is important: pollen of different plant species vary in nutritional quality for honey bees. Previous studies showed that a protein alimentation with pollen mixture is more adequate for bees than monofloral pollen. The kind of pollen collected by bees is in close relationship with the vegetation spectrum of the hive surrounding area. Thus, if the bee colony is surrounded by areas characterised by intensive agriculture e.g. the Po Valley in Italy it may mainly collect monofloral pollen. In these areas, use of pesticides is generally widespread. The present study is based on the hypothesis that the quality of pollen available for the honey bee colony may influence the bee susceptibility to the intoxication by pesticides. For this reason, the same pesticide treatment could cause negligible or significant damages in relation to the availability of high-quality pollen (high amino acid diversity and high protein content). In the experimental apiary, in order to obtain newly emerged bees of the same age, the queen bee was isolated on a comb in a queen-excluding cage for about 30 hours. Subsequently the queen bee was removed from the cage and the comb was left isolated for another three days, to avoid further egg laying. The comb was incubated inside the beehive for 20 days in order to guarantee the most natural conditions, then it was moved to an emerging cage and kept at 34,5°C. The bees were incubated at this temperature until the end of the test. At the beginning of the emergence the bees were fed ad libitum with water, organic Robinia honey and a kind of pollen in relation to the thesis: Zea mays, Papaver, Cruciferae, Trifolium, Taraxacum or a mixture of pollen. Thus groups of bees provided with different pollen diets were obtained. On the 9 day, the bees were divided into groups of 20 individuals in small test cages in which clothianidin or fipronil pesticides considered as one of the possible causes of the recent colony losses in sucrose syrup (50%) were administered. Once the bees consumed completely the test solution, sucrose syrup (50%) was supplied ad libitum. Bee mortality was recorded and the LD50 of the active ingredients was calculated in relationship to the alimentation quality. In conclusion, the influence of the quality of protein nourishment provided during the first days of adult life on the response to intoxication by clothianidin and fipronil was shown as evidence. Bees fed with low-quality pollen (low amino acid diversity and low protein content) seem to be more susceptible to pesticide intoxication and other stress factors than bees fed with high-quality pollen.

Pollen Biology and Dispersal Dynamics in Waterhemp (Amaranthustuberculatus)

Waterhemp is a major weed of field crops in the Midwestern United States. Its prevalence is at least partly due to its rapid evolution of resistance to many groups of herbicides over the last two decades. In light of its dioecy and anemophily, pollen movement in waterhemp is considered to be an important factor in the spread of herbicide resistance genes. Thus, the biology and dispersal profile of waterhemp pollen are critical determinants for understanding and predicting the spatial population dynamics of herbicide resistance in this species. In this study, pollen longevity was investigated with greenhouse experiments, and pollen dispersal and the effect of pollen competition were investigated in field plots. Pollen dispersal was determined by measuring the frequency of seeds produced on receptor plants positioned at various distances from a pollen source, which flowered in synchrony with the receptor plants. Results indicated that waterhemp pollen can remain viable up to 120 h, implying a low limitation of pollen dispersal by its longevity. Effective pollen dispersal declined exponentially with distance, with most pollen fertilizing recipient plants within 50 m of the pollen source. However, long distance pollen dispersal (800 m) was also observed. We also saw evidence for pollen swamping in this species. Under conditions of pollen competition among distinct genotypes, pollination success was inversely related to distance between pollen donors and receptors. However, relative pollen density may also play an important role in determining the rate of long distance gene flow. This study confirmed the potential of waterhemp pollen to effect long distance gene flow and provides supporting data for quantitative spatial modeling of waterhemp resistance dynamics.

Predicting maize pollen production using tassel morphological characteristics

In the European Union, it has been decided that farmers and consumers should be free to choose between conventional, organic, and genetically modified (GM) crops/products. This concept of coexistence implies the necessity of developing coexistence strategies which curb the risk of mixing GM and non-GM material along agricultural supply chains. Among many things, this requires understanding and predicting the process of cross-pollination between the different varieties cultivated on a given landscape. In order to provide a prediction accurate enough to cope with legal thresholds, it is necessary to be able to rank varieties according to their pollen production. However, this parameter is rarely known, partly because it is difficult and expensive to measure. We investigated how easy-to-measure tassel morphological characteristics can be used to predict the quantity of pollen produced by hybrid maize varieties. Thirteen varieties were grown in France on experimental plots during three years with two varieties added the last year. For each variety, we measured the pollen production and variables that describe the tassel architecture. We found that the number of ramifications, the length of the tassel, the length of the emitting section, the number of spikelets and the TAI (the Tassel Area Index which aims at estimating the pollen emitting area) were correlated with pollen production. Using linear models, we identified the number of spikelets as the best architectural variable related to pollen production. In order to evaluate the ability of tassel characteristics to predict pollen production, we analysed how the relationships established in the case of any two years made it possible to predict the pollen production of the third year. The model that combined the total length of the tassel and the total number of spikelets was found to be the best predictive model of the relative pollen production between varieties.