Regulatory tests assess the ‘active substance’ in a pesticide, but farmers use products with lots of extra ingredients (surfactants, adjuvants) that can amplify its toxicity. Often, the product used by farmers is not evaluated.

See here for a French court ruling that anti-glyphosate protest about the harm of the ‘cocktail effect’ was necessary and proportional to protect public health on this very issue.

Tests to ascertain how deadly new pesticides are for humans and wildlife are often done in-house by the companies seeking approval. This research is rarely made public as it is considered commercially sensitive.

• An example of this was outlined by ABC Australia, who found that the science underpinning APVMA’s advice on paraquat’s connection to Parkinson’s was based on an unpublished paper funded by the maker of the chemical, Syngenta.

Pesticide tests focus on the death of an animal and ignore any important “sublethal” effects.

• If a creature, such as a bee, is alive 48 hours after exposure, then it is deemed that all is well, and the chemical may be approved for use. The bee may be unable to fly or navigate, or its immune system may no longer function, but that is not recorded.

Pesticide cocktails are not measured in MRLs: When you eat a strawberry or a grape, it will likely have multiple pesticides on it, not just one. However, each MRL is only for each particular pesticide active ingredient.

Also, a MRL looks at one exposure of one chemical – eg. one residue on one apple. But we may eat an apple with multiple different residues, or a fruit salad, then drink water with pesticide residues, then swim in contaminated beaches, then eat contaminated oats – we are exposed to repeated low level exposure to a cocktail of pesticides. This is not measured.

APVMA are regularly changing the MRLs, often changing and increasing the MRL that is considered ‘safe’.

You can find the international MRLs here: https://www.agriculture.gov.au/agriculture-land/farm-food-drought/food/nrs/databases