Fungicides and inoculants: what actually kills what

It is one of the most common questions we are asked, and one of the most poorly answered online: if I use a microbial inoculant, can I still use a fungicide? Anyone who answers with a flat yes or a flat no is not answering the question. The honest answer is that it depends on two things, and once both are known, most cases resolve themselves.

The two variables are what is actually alive in your inoculant, and how the fungicide kills. Set them against each other and it becomes possible to reason about a combination never seen before. That is more useful than any list, because no list will contain the exact product in hand.

Variable one: what is alive in the bag

Biologicals divide first into bacteria and fungi. Bacterial products are built on genera such as Bacillus, Brevibacillus, Paenibacillus, Pseudomonas, Azospirillum and the rhizobia; fungal products on Trichoderma, Purpureocillium, Pochonia and similar genera. That distinction matters enormously, because a fungicide is designed to kill fungi. From the fungicide’s point of view, a fungal inoculant is a target.

Among bacteria there is a second distinction that matters almost as much and is discussed far less: whether the organism forms endospores. A spore-forming bacterium survives conditions that kill a vegetative cell outright. The spore is dehydrated, its core is packed with calcium dipicolinate, its DNA is wrapped in small acid-soluble proteins, and it is protected by a thick cortex. That is why the same tank can be survivable for one bacterial product and lethal for another.

The table contains a counter-intuitive point worth dwelling on. The product most at risk from a seed treatment is not the one carrying the most sophisticated biology. It is Nodu™, the simple rhizobial inoculant. Rhizobia are Gram-negative, non-spore-forming and thin-walled, and they begin dying on dry seed within hours, even before any chemistry is added. Their loss is also the most costly, because no rhizobia means no nodules and no fixed nitrogen at all. Orvan™, by contrast, consists of three spore-formers and has a margin that Nodu™ lacks.

Variable two: how the fungicide kills

Fungicides fall into two broad classes by mechanism. Single-site fungicides bind one specific enzyme or protein in the fungus. Multi-site fungicides do not have a target in that sense at all: they oxidise, chelate, denature proteins, or dissolve membranes. Single-site chemistry can be selective. Multi-site chemistry, by definition, cannot be.

The never-mix list

These are the multi-site and oxidising materials, and no inoculant of any kind belongs in a tank with them. They kill by destroying structure rather than by occupying a binding site, and a bacterium has the same kinds of membranes and proteins as a fungus.

• Copper in any form — hydroxide, oxychloride, sulphate, Bordeaux mixture. Copper ions are as antibacterial as they are antifungal; this is the single most common way an inoculant is wasted. Copper should also not be combined with strong acids or phosphate fertilisers, and an over-concentrated coating depresses germination in its own right.
• Dithiocarbamates — thiram, mancozeb, ziram. Broad protein poisons.
• Chlorine- and bromine-releasing disinfectants: chloroisobromine cyanuric acid, dichloroisocyanurates, hypochlorite.
• Bronopol, ethylicin, octylamine acetate, quaternary ammonium disinfectants.
• Sulphur, and lime sulphur, at seed-treatment concentrations.

There is a second, often overlooked cost. The reaction runs both ways: while the copper is killing the bacteria, the bacteria (and the peat, the carrier, and the organic matter in the slurry) are consuming the copper. The inoculant is lost and the fungicide is weakened. Neither does its job properly.

Where the mode of action allows a margin, and why the usual explanation is too simple

Single-site fungicides are a different matter, and here the explanation commonly found online is worth correcting, because getting the reasoning right shows how far the margin can be trusted.

The QoI or strobilurin group (azoxystrobin, pyraclostrobin, kresoxim-methyl) binds the Qo site of cytochrome b and shuts down the respiratory chain. The usual explanation is that bacteria have no mitochondria, so they are safe. That is not quite true. Bacteria run their electron transport chain in the plasma membrane and many of them carry a cytochrome bc1 complex that is genuinely homologous to the fungal one. The real reason for the selectivity is narrower: the Qo binding pocket of fungal cytochrome b differs enough from its bacterial counterparts that practical activity against bacteria is low. The distinction matters, because it means the margin is empirical rather than structural. It holds in practice, but it is not a law of biology, and it is no substitute for testing.

The same correction applies to the SDHI group (fluopyram, boscalid, fluxapyroxad), which inhibits succinate dehydrogenase, complex II of the respiratory chain. Bacteria have succinate dehydrogenase too. Again, the selectivity is real, but practical rather than structural.

There is one group for which the reason genuinely is structural, and it deserves to be better known. The DMI or triazole fungicides (tebuconazole, difenoconazole, prothioconazole and the rest) inhibit CYP51, the enzyme that demethylates sterols in fungal membranes. Bacteria do not make sterols at all, so there is no target to hit. Of the widely used seed-treatment chemistries, the triazoles are among the best tolerated by a bacterial inoculant, and that is a structural fact rather than an empirical observation. Phenylamides such as metalaxyl are similarly narrow, as they target oomycete RNA polymerase.

The case that is often misread

Bactericides invert every rule above. Streptomycin, kasugamycin, zhongshengmycin and oxytetracycline act on the bacterial ribosome, a structure fungi do not have; zinc thiazole, benziothiazolinone, oxine-copper and bismerthiazol are registered for bacterial diseases, act by other mechanisms, and belong in the same category. They are therefore comparatively safe to mix with a Trichoderma product, and lethal to Orvan™, to Vigor™, and above all to Nodu™. We raise this because the mistake follows a predictable pattern: a grower treating for a bacterial disease reasons that a bactericide has nothing to do with a fungicide problem, and applies it over a freshly inoculated crop. It is the one combination where the label and the inoculant are in direct opposition.

The active ingredient is not the whole product

This is the point that catches out people who have done everything else right, and it is the reason we do not publish a simple green-light list. A commercial fungicide is not its active ingredient. It is the active ingredient plus solvents, emulsifiers, wetters, dispersants, antifoams, dyes, stickers, and in many liquid formulations an in-can preservative whose entire job is to stop microorganisms growing in the drum.

An EC formulation carries aromatic solvents that dissolve lipid membranes. Non-ionic surfactants at slurry concentrations do the same. An isothiazolinone preservative is a broad-spectrum biocide by design. Any of these can sterilise a slurry that the mode-of-action table indicated was safe. This is why a jar test with the actual commercial product in hand is worth more than any amount of reasoning about active ingredients, and why two products with the same active ingredient from two manufacturers can behave completely differently.

What we recommend, in order of preference

• Work from the tested list first. For our own seed-applied formulations, the co-application list is published on the Orvan™ page, and a combination on it needs no separation. For anything outside a tested list, the reasoning above applies, and separation in time is the safe default.
• Where the two are separated: treat the seed with fungicide, let it dry completely, then inoculate as a separate operation. For a soil-applied or foliar fungicide, leave 7–14 days either side of the inoculation.
• If both must go on the same seed, keep to a fixed order: fungicide first, then drying, then the inoculant last. Sow within 24 hours. Contact time determines the dose, and the clock starts when the two meet.
• Never pre-mix and store. A slurry held overnight is a completely different exposure from one sown within the hour.
• Check the water as carefully as the chemistry: free chlorine below 2 ppm, pH 6–7.5, temperature below 30 °C, and no galvanised or copper tanks or fittings.
• When the seed treatment is heavy, take the biology off the seed entirely. An in-furrow granule or liquid never touches the treated seed coat, carries a far higher dose, and removes the compatibility question rather than managing it.

Test it yourself — it takes one afternoon

You do not need a laboratory to answer this for your own tank. Prepare the slurry exactly as you would in the field, at field rate, with the water you will use. Take a sample at mixing, at two hours, and at twenty-four hours. If you have access to a lab, plate each sample and count; a drop of more than about one log (a tenfold loss) at the contact time you actually intend to use means the combination is not viable. If you have no lab, run the simpler version, which still answers the question: four short strips at sowing (untreated, inoculant alone, fungicide alone, and the mix) and dig roots at flowering to count nodules. It is slower, but it measures what actually matters.

A further point: mixing inoculants with each other

Orvan™, Nodu™ and Vigor™ are designed as a programme and can be co-applied without difficulty; that compatibility was built in. Mixing biologicals from different sources is a separate question, and the answer is often no. One organism’s metabolite can be another’s toxin, and the effect may be specific and severe.

Our own products illustrate the point rather than being exempt from it. Bacillus velezensis, one of the three strains in Orvan™, is actively antagonistic to filamentous fungi, and that antagonism is part of how the consortium shifts the balance against Aspergillus in the root zone. It follows directly that tank-mixing Orvan™ with a Trichoderma or Purpureocillium product is self-defeating: you would be paying for a fungal biocontrol agent and then applying something designed, in part, to suppress fungi. Keep them separate in time, or choose one.