Grasses produce large amounts of pollen and are among the main causes of pollen allergy worldwide. Quantification of the roles of individual grass species in airborne pollen is an important task, because morphologically indistinguishable pollen grains of different species may differ in allergenicity. This requires knowledge of the pollen production of individual grass species; however, accumulated data are insufficient in this respect. Attempting to fill this gap, we studied pollen production per inflorescence in 29 grass species which are widespread in Middle Russia and India. Pollen production per inflorescence is determined by the number of grains per anther, the number of flowers in a spikelet and the number of spikelets per inflorescence, with the latter parameter being the most variable. We support the hypothesis that pollen production per inflorescence differs significantly between annual and perennial grasses. The greater pollen production of perennials can be interpreted as a tendency to guarantee cross-fertilization of species with self-incompatibility. The inferred pollen/ovule (P/O) ratios suggest the occurrence of facultative xenogamy in all annuals and obligate xenogamy in most perennials in the present dataset, though some self-incompatible annuals exist outside our sampling. Earlier data indicated that the P/O ratio of the annual cereal crop rye (Secale cereale) is higher than in any annual or perennial species sampled here. A ratio of pollen production to seed set (P/S ratio) is suggested to be another efficient parameter in reproductive biology of grasses. We highlight a need for detailed studies of reproductive biology in grasses that include both pollen and seed production. We found a correlation between pollen production per anther and anther length. A rough approximation of c. 1000 pollen grains per 1 mm of the length of an anther provides an instrument for estimates of pollen production in plant communities.

Grass pollen is one of the leading causes of pollinosis, affecting 10–30% of the world’s population. The allergenicity of pollen from different Poaceae species is not the same and is estimated from moderate to high. Aerobiological monitoring is a standard method that allows one to track and predict the dynamics of allergen concentration in the air. Poaceae is a stenopalynous family, and thus grass pollen can usually be identified only at the family level with optical microscopy. Molecular methods, in particular the DNA barcoding technique, can be used to conduct a more accurate analysis of aerobiological samples containing the DNA of various plant species. This study aimed to test the possibility of using the ITS1 and ITS2 nuclear loci for determining the presence of grass pollen from air samples via metabarcoding and to compare the analysis results with the results of phenological observations. Based on the high-throughput sequencing data, we analyzed the changes in the composition of aerobiological samples taken in the Moscow and Ryazan regions for three years during the period of active flowering of grasses. Ten genera of the Poaceae family were detected in airborne pollen samples. The representation for most of them for ITS1 and ITS2 barcodes was similar. At the same time, in some samples, the presence of specific genera was characterized by only one sequence: either ITS1 or ITS2. Based on the analysis of the abundance of both barcode reads in the samples, the following order could describe the change with time in the dominant species in the air: Poa, Alopecurus, and Arrhenatherum in early mid-June, Lolium, Bromus, Dactylis, and Briza in mid-late June, Phleum, Elymus in late June to early July, and Calamagrostis in early mid-July. In most samples, the number of taxa found via metabarcoding analysis was higher compared to that in the phenological observations. The semi-quantitative analysis of high-throughput sequencing data well reflects the abundance of only major grass species at the flowering stage.

Grasses (Poaceae) produce large amounts of pollen and are among the main causes of pollinosis worldwide. Despite their morphological similarity, pollen grains of different grass species may have different allergenicities. Therefore, quantification of the roles of individual species in airborne pollen is an important task. There are very few studies on this topic, and none of them have been conducted in a temperate continental climate. Our study was carried out for three years (2020–2022) in the urban territory of Moscow (Russia) and aimed to understand what grass species contribute the most to the total pollen load of the atmosphere. The comparison of aerobiological and phenological data was based on calculating the phenological index, which is a combination of phenological parameters, pollen productivity of individual species, and their abundance. Our data showed that the decomposition of pollination curves based on the phenological index was sometimes very efficient but not always possible in temperate continental climates. The main reasons for disagreement between aerobiological and phenological data were weather conditions and lawn mowing. Not all grasses were equally important as sources of allergenic pollen. The greatest contribution to the pollen load at the beginning of the season in Moscow was made by Dactylis glomerata, and to a lesser extent by Phleum pratense and Festuca pratensis. These are the most common species, which are widespread throughout Europe. The contribution of minor components is insignificant.