Please note – what follows is a pre-publication paper.

Biofertilizers and biostimulants: Is the bubble about to burst?

Ken Giller

Ken Giller was recognised as the TAAI “Development Agriculturalist of the Year” in 2024. He has conducted research on nitrogen fixation in the tropics for over 40 years and is author of the book “Nitrogen Fixation in Tropical Cropping Systems” published by CABI in second edition in 2001. He has led a number of initiatives such as the Bill & Melinda Gates Foundation funded “N2Africa – Putting Nitrogen Fixation to Work for Smallholder Farmers in Africa – https://www.n2africa.org/” which worked in 11 countries over ten years from 2009-2019. Ken has >500 peer-reviewed papers and has been on the Web of Science list of Highly cited researchers (top 1%) since 2016. He has supervised over 100 PhDs successfully – more than half from African countries. Ken joined Wageningen University as Chair of Plant Production Systems in 2001 after holding professorships at Wye College, University of London, and the University of Zimbabwe.

ken.giller@wur.nl

Abstract

Inoculants of bacteria that fix nitrogen with cereals and other non-legume crops have developed into a multi-million dollar industry. The rise of these microbial inoculants has been driven by exaggerated claims by scientists and vast amounts of venture capital. An increasing number of scientific reports indicate no response in crop growth or yield to inoculation with these products in the field, A review of the scientific literature indicates that, when observed, crop responses are due to the effects of plant hormones produced by the bacteria and not N2-fixation, As increasing numbers of farmers and scientists become disillusioned with the false promise of microbial N2-fixation in non-legumes, is the bubble about to burst?

Introduction

One of the dangers of having been around for a long time is that you get a feeling you’ve seen it all before. That’s certainly the case with the promise of microbial nitrogen fixation with cereals. My first job after completing my PhD in 1982 was to measure nitrogen fixation in the rhizospheres of sorghum and millet. I was employed at Rothamsted Experimental Station, working in collaboration with the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Hyderabad, India. The research was funded by a grant from the UK Overseas Development Administration, and the aim was to gather sufficient scientific evidence to determine whether inoculants of free-living nitrogen-fixing bacteria warranted promotion and commercialisation.

The background to my project was a long-standing collaboration between John Day at Rothamsted and Peter Dart at ICRISAT. They were both authors of a seminal paper which showed high rates of nitrogen fixation associated with the roots of tropical grasses in Brazil using the acetylene reduction assay (Döbereiner et al., 1972). Dinitrogen gas, N2, which comprises almost 80% of the atmosphere, is a triple-bonded molecule that is reduced to ammonia by the enzyme nitrogenase. Acetylene is also a triple-bonded molecule which is reduced to ethylene by nitrogenase, so the rate of N2-fixation could be measured by incubating roots of tropical grasses in chambers enriched with acetylene and measuring the accumulation of ethylene. The publication of their paper aroused a lot of interest and, when coupled with reports of responses of crop growth to inoculation with N2-fixing bacteria (Smith et al., 1976), the prospect of N2-fixing cereals seemed a reality. Although a paper published in 1979 showed that these microbial inoculants stimulated root branching and root hair growth through the production of plant hormones (Tien et al., 1979), the enthusiasm for nitrogen fixation in cereals was not quelled.

Peter Dart was leading a team at ICRISAT investigating nitrogen fixation associated with the roots of millet and sorghum. Much of the work involved use of the acetylene reduction assay to measure nitrogen fixation. Observations of different rates of N2-fixation associated with different genotypes of millet and sorghum led to crossing experiments to see if this was a heritable trait (Fig. 1).

Fig 1. Paper bags being placed over the flowering spikes of millet plants grown in the glasshouse at ICRISAT to allow specific crosses to be made. The plants are growing in large pots that can be sealed to allow injection of acetylene to assess rates of N2-fixation. Photo: Ken Giller

Many glasshouse and field experiments to test the growth and yield response of millet and sorghum to inoculation with N2-fixing bacteria were also being conducted.

My research involved the use of the stable isotope of nitrogen, 15N, to trace microbial N2-fixation. I conducted experiments in which intact roots of small sorghum plants were exposed to 15N2 gas (Fig. 2).

Fig. 2. Experiments conducted in the glasshouse exposing the roots of sorghum plants to 15N2 gas at ICRISAT, India. a) the plumbing to allow the growth chambers to be flushed and 15N2 gas introduced. b) young sorghum plants sealed to all exposure of the root systems to 15N2 gas. See Giller et al. (1984) for details of the method. Photos: Ken Giller.

Using this very sensitive method, we could demonstrate that N2 was fixed in association with the roots and taken up by the sorghum plants (Giller et al., 1984). But the amounts of N2-fixed were very small and only detectable when we added farmyard manure to the growing medium. I also conducted studies using isotope dilution and nitrogen balance methods which indicated no significant amounts of N2-fixation in millet and sorghum. As time went on, I became more suspicious of the results being reported. Crop response to inoculation in field experiments was very erratic, but there seemed to be selective reporting of the positive responses. Further, although the acetylene reduction assay is highly sensitive, it is subject to a number of problems. Long incubation times were used, which gave inflated N2-fixation rates. Notably acetylene blocks the normal oxidation of acetylene in the soil, so ethylene accumulates during incubations that has nothing to do with N2-fixation. This problem was already highlighted in 1976 and yet the acetylene reduction assay was still routinely employed (Giller, 1987). Based on my experiences in India and on an extensive literature review, I concluded that the amounts of N2-fixed by bacteria associated with the roots of cereals were of little significance in agriculture (Giller & Day, 1985). To my surprise, a year later, ICRISAT produced a workshop proceedings entitled “Cereal Nitrogen Fixation” (Wani, 1986), which concluded the opposite, as did another review article (Dart, 1986). This was my first experience that scientists furnished with the same information could reach diametrically opposite conclusions.

Microbial inoculants offering the promise of N2-fixation with cereals and other non-legume crops had been around for a long time: Azotobacterin was already used on 10 Mha in Russia in the 1930s (Brown, 1974). Researchers had concluded that the (often erratic) responses of cereals and grasses to inoculation with free-living N2-fixing bacteria were due to hormonal effects and not N2-fixation (Brown, 1974, Giller & Day, 1985). In 2002, almost 20 years after the publication of our 1985 review, I was invited to give the opening talk at the 9th International Symposium on Nitrogen Fixation with Non-legumes in Leuven. In the intervening period, a number of papers had appeared claiming substantial N2-fixation and crop growth responses in non-legumes due to the bacterium Gluconacetobacter diazotrophicus, but yet again these had been shown to be due to production of plant hormones (Sevilla & Kennedy, 2000). In my talk I referred to three overlapping cycles of discovery and disappointment related to microbial N2-fixation in non-legumes, where in all cases it had been concluded that crop responses were due to hormonal effects and not N2-fixation (Giller & Merckx, 2003). This was also the conclusion of the symposium summary (Kennedy, 2003), and other authors in the same year (Dobbelaere et al., 2003).

What is happening now?

Fast forward another 20 years to 2023; I was invited to a convening at the Bill & Melinda Gates Foundation in Seattle to explore the possibilities of microbial N2-fixation in non-legumes. Having not been directly involved in research on this topic for some time, I was astounded to see how many microbial inoculants were being sold to farmers, particularly in North America, for use on cereals. Gluconacetobacter diazotrophicus, the bacterium promoted and sold by Azotic, is actively promoted in Africa by the so-called “Sustainable Nitrogen Foundation”. Around the same time, I was contacted by a journalist, Joost van Kasteren, who had become aware of reports emerging from American universities that reported no crop response to such products (e.g. Franzen et al., 2023). I shared my experiences with Joost, who also interviewed Julie Ardley of Murdoch University, Perth, Australia. She could clearly show that that the strain of Methylobacterium symbioticum used in the product BlueN, which is marketed as UtriciaN in North America, cannot make nitrogenase – it does not have the HDKENX gene cluster found in all N2-fixing bacteria (Fig 3).

Fig. 3. The bacterium used in UtrishaN/ BlueN (Methylobacterium symbioticum strain SB0023/3T) has genes homologous to nifH, the gene encoding the Fe protein of nitrogenase, but lacks the genes required for nitrogenase biosynthesis. The two nifH-like homologues found in Methylobacterium symbioticum strain SB0023/3T were identified as genes encoding proteins present in two bacteriochlorophyll biosynthesis clusters. By contrast, the N2-fixing rhizobial strains Methylobacterium sp. WSM2598, Bradyrhizobium sp. ORS285 and Methylobacterium nodulans ORS2060 all possess the conserved nitrogenase biosynthesis nifHDKENX cluster. The two nifH-like homologues found in Methylobacterium symbioticum strain SB0023/3T were identified as genes encoding proteins present in two bacteriochlorophyll biosynthesis clusters (see Giller et al., 2025 for further explanation). I thank Julie Ardley for preparing the figure and for permission to reproduce it.

Joost reported these findings in a popular article, in which he talked of the “fourth wave” of excitement about microbial N2-fixation with non-legumes (Van Kasteren, 2023). On publication of the article, we were contacted by the manufacturer of BlueN, Corteva Agriscience, and Julie Ardley, and I shared our findings with them in detail in August 2023. Together with Euan James, an expert on microbial colonization of plants, and Murray Unkovich, a leading expert on measurement of N2-fixation, we published a detailed analysis of the current state of knowledge on microbial N2-fixation with cereals and other non-legumes (Giller et al., 2025). In this review, we call for stringent regulations to prevent the sale of microbial inoculants with spurious claims of significant N inputs from N2-fixation with non-legumes. Before products are registered for sale, we argue that independently verified evidence should be provided that the inoculant bacterium can fix N2 and that significant inputs of N are derived from N2-fixation by non-legumes in the field, confirmed by more than one method (Giller et al., 2025).

What is the theoretical limit for how much N2 could be fixed with non-legumes?

Andrew Hanson recommends that “Fermi calculations” – which are basically back-of-the-envelope calculations, or order-of-magnitude estimates (Oliveira-Filho et al., 2025) – should be used to scope whether proposed solutions to important problems are realistic. Such calculations, based on the efficiency with which carbon can be used to fix N2, by bacteria, the ability of N2-fixing bacteria to compete for carbon released by plants, and the amounts of plant carbon available for use by bacteria have been made several times (Table 1). These estimates are fairly similar, suggesting that the maximum rates of N2-fixation under ideal conditions are 4-20 kg N ha-1. These amounts are unlikely to be achieved in the field, and are far less than those claimed by the manufacturers of microbial inoculants.

Cereal
Maximum rates of N2-fixation that could be supported (kg N ha-1)Reference
Rhizospheres of wheat 15(Giller & Day, 1985)
General model for cereals4(Jones et al., 2003)
General model for cereals if the populations of N2-fixing bacteria could be enhanced20(Jones et al., 2003)
Aerial roots of maize15(Gelder et al., 2023)

Table 1. Fermi calculations of the theoretical maximum rates of N2-fixation that could be supported by a crop of cereals

A market out of control?

Despite the overwhelming scientific evidence summarised above, the number of products claiming to fix N2 with non-legumes increases year by year. Reports suggest that US$1.5 billion of venture capital has been invested in what are termed “crop agnostic nitrogen fixing or availability enhancing products”, which effectively means bacteria that will fix nitrogen with any crop. This is not just money invested in research; it is also building a massive industry in inoculants of N2-fixing bacteria. For example, Azotic’s products were used on 0.8 Mhai in North America, and PivotBio’s on 1.4 Mhai. Shoham (2026) estimate that the market for biological products was growing by 15% a year until recently. Although biofertilizers currently comprise only a small proportion of the total, this amounted to US$1.3 billion in 2023 and is estimated to rise to US$2.14 billion by 2028 (Shoham, 2026). Among the companies, Corteva Agriscience, the manufacturer of BlueN, is said to be leading the sales with 12.5% of the global marketii. As indicated above, we shared our findings with Corteva in 2023, yet they continue to use the same publicity in which they claim that BlueN fixes N2 in cereals and a whole range of non-legume crops. This is misleading and surprising for a company that supposedly prides itself on its “Purpose and Values” on its website.

Accumulating disillusionment in product performance

Since the publication based on ten American universities in the USA, which found no response to microbial inoculants of N2-fixing bacteria in 59 out of 61 trials (Franzen et al., 2023), other reports have also emerged with rare or no response to such microbial inoculants. Independent experiments at eight locations in Canada found no effects of UtrishaN or EnvitaN on wheat grain yield or protein concentration, nor on N status in canola, despite an expected response to N fertilizeriii,iv. Experiments in Denmark showed no effect of Vixeran (Syngenta) or BlueN on growth or yields of winter wheatv. Six trials across three locations in Sweden over two years testing a range of biostimulants, including BlueN, showed no effect on spring barley yields. These accumulating results demonstrate that stringent regulations are needed to prevent such products from being marketed based on claims they cannot fulfil (Giller et al., 2025, Termorshuizen et al., 2026).

Conclusion

So what has led to this current popularity of microbial products which appear not to live up to the claims of the manufacturers and the expectations of the farmers? The rise of Regenerative Agriculture and the pressure to reduce the amounts of mineral N fertilizers used in agriculture no doubt plays a major role, as does the profitability of microbial inoculants. Perhaps the blame does not solely lie with companies running ahead of the evidence to market products and profit from their sale. Undoubtedly the exaggerated claims made by scientists to gain attention for their work and garner funding are also to blame.

Are there fairies at the bottom of the garden? Probably not – but we keep telling stories about them generation after generation! That’s how I started my talk entitled “Microbial N₂-Fixation in Cereals: Hype or Reality?” at the PlanteKongres in Herning, Denmark in January, 2026. It’s such a nice idea that cereals and other non-legumes can fix nitrogen through inoculation with bacteria, that we’ve been telling this story in repeated cycles for almost 100 years – although the evidence points to any increases in crop growth are due to plant hormone effects and not N₂-fixation. My conclusion is that, although there probably are no fairies at the bottom of the garden, the stories will run and run into the future. Unfortunately, the same goes for the fantasy of microbial N2-fixation with cereals!

References

Brown, ME 1974. Seed and root bacterization. Annual Review of Phytopathology, 12, 181-197.

Dart, PJ 1986. Nitrogen-fixation associated with non-legumes in agriculture. Plant And Soil, 90, 303-334.

Dobbelaere, S, Vanderleyden, J, Okon, Y 2003. Plant Growth-Promoting Effects of Diazotrophs in the Rhizosphere. Critical Reviews in Plant Sciences, 22, 107-149.

Döbereiner, J, Day, JM, Dart, PJ 1972. Nitrogenase activity and oxygen sensitivity of the Paspalum notatum – Azotobacter paspali association. Journal of General Microbiology, 71, 103-116.

Franzen, D, Camberato, J, Nafziger, E, Kaiser, D, Nelson, K, Singh, G, Ruiz-Diaz, D, Lentz, E, Steinke, K, Grove, J, Ritchie, E, Bortolon, L, Rosen, C, Maharjan, B, Thompson, L 2023. Performance of Selected Commercially Available Asymbiotic N-fixing Products in the North Central Region. North Dakota State University. https://www.ndsu.edu/agriculture/extension/publications/performance-selected-commercially-available-asymbiotic-n-fixing-products.

Gelder, KV, Oliveira-Filho, ER, Messina, CD, Venado, RE, Wilker, J, Rajasekar, S, Ané, J-M, Amthor, JS, Hanson, AD 2023. Running the numbers on plant synthetic biology solutions to global problems. Plant Science, 335, 111815.

Giller, KE 1987. Use and abuse of the acetylene-reduction assay for measurement of associative nitrogen-fixation. Soil Biology and Biochemistry, 19, 783-784.

Giller, KE, Day, JM 1985. Nitrogen fixation in the rhizosphere: Significance in natural and agricultural systems. In: Fitter, AH, Atkinson, D, Read, DJ,

Usher, MB (eds.) Ecological Interactions in Soil: Plants, Microbes and Animals. Oxford, UK: Blackwell Scientific.

Giller, KE, Day, JM, Dart, PJ, Wani, SP 1984. A method for measuring the transfer of fixed nitrogen from free-living bacteria to higher plants using 15N2. Journal of Microbiological Methods, 2, 307-316.

Giller, KE, James, EK, Ardley, J, Unkovich, MJ 2025. Science losing its way: examples from the realm of microbial N2-fixation in cereals and other non-legumes. Plant and Soil, 511, 1-24.

Giller, KE, Merckx, R 2003. Exploring the boundaries of N2-fixation in non-legumes: An hypothetical and experimental framework. Symbiosis, 35, 3-17.

Jones, DL, Farrer, JF, Giller, KE 2003. Associative nitrogen fixation and root exudation – what is theoretically possible in the rhizosphere? Symbiosis, 35, 19-38.

Kennedy, IR 2003. Symposium summary and future perspectives Symbiosis, 35, 271-278.

Oliveira-Filho, ER, Campos-Silva, R, Hanson, AD 2025. Running Fermi calculations as a superpower to gauge reality. Plant Physiology, 198, kiae347.

Sevilla, M, Kennedy, C 2000. Genetic analysis of nitrogen fixation and plant-growth stimulating properties of Acetobacter diazotrophicus, an endophyte of sugarcane. In: Triplett, E (ed.) Prokaryotic Nitrogen Fixation. Wyndham, UK: Horizon Scientific Press.

Shoham, J 2026. Biologicals – commercial reality – current and future potential market. Agriculture for Development, in press.

Smith, RL, Bouton, JH, Schank, SC, Quesenbury, KH, Tyler, ME, Milam, JR, Gaskins, MH, Littell, RC 1976. Nitrogen fixation in grasses inoculated with Spirillum lipoferum. Science, 193, 1003-1005.

Termorshuizen, AJ, Kuyper, TW, Struik, PC, Giller, KE 2026. Proper efficacy testing should be mandatory for registration of biostimulants. Outlook on Agriculture, in press.

Tien, TH, Gaskins, MH, Hubbell, DH 1979. Plant growth substances produced by Azospirillum brasilense and their effect on the growth of pearl millet (Pennisetum americanum L.). Applied and Environmental Microbiology, 37, 1016-1024.

Van Kasteren, J 2023. Serious doubts about alternatives to fertiliser (original in Dutch: Twijfels over biologische alternatieven voor kunstmest). Vork, 2, https://www.joostvankasteren.nl/serious-doubts-about-alternatives-to-fertiliser/.

Wani, SP 1986. Cereal Nitrogen Fixation. Proceedings of a Working Group Meeting, 9-12 Oct 1984., Patancheru, India, ICRISAT.