During the last decade, there has been a huge interest in understanding the role of reactive oxygen species (ROS) in plant signalling transduction pathways. This understanding requires precise quantification of ROS levels in each cell and each cellular compartment. However, the current methods of ROS detection and measuring are limited. This paper revisits the existing ROS detection methods and discuss general guidelines for applying them to specific cases.IntroductionAll plants require molecular oxygen for survival (Mittler, 2017). ROS formation naturally occurred during electron transport through all membranes which, in turn, regulate DNA repair systems, cell cycle, phytohormone-dependent signalling and pathogen integration (Huang et al., 2019). In the non-photosynthetic plant tissue, the mitochondrial electron transport system of oxidative phosphorylation is the major site for ROS generation (Dourmap et al., 2020). While in photosynthetic tissue, electron transport between stroma and thylakoid is the primary ROS source (Asada, 2006). On plasma membranes and on endoplasmic reticulum membranes, ROS is mainly produced via NADPH oxidases (Foreman et al., 2003).Cell wall peroxidases are another source of apoplastic ROS (Torres, 2010). In addition, peroxisomes can be considered as the major site of intracellular hydrogen peroxide (H2O2) production (Sandalio et al., 2021).Major ROS produced by cellular processes are superoxide (O2-), H2O2, and hydroxyl radical (\u2219OH). Superoxide is rapidly converted to H2O2 by superoxide dismutase enzymes (SODs; Cu\/Zn-SOD in chloroplasts and cytoplasm, Fe-SOD and Mn-SOD in mitochondria). Hydroxyl radicals are thus generated in the cell wall, plasma membrane, and intracellularly by a range of peroxidases, superoxide dismutases, NADPH oxidases, and transition metal catalysts (Richards et al., 2015). Because of the cellular and biochemical damage caused by oxidative stress (Huang et al., 2019), ROS levels should be precisely controlled in each subcellular compartment and each cell type. ROS are highly reactive molecules rapidly subjected to scavenging or degradation, in processes that are highly sensitive to any environmental change, therefore making ROS extremely unstable and difficult to directly detect. Transferring of the plants to buffers with non-physiological pH can be considered as an abiotic stress factor and it eventually might change endogenous ROS levels (Choudhury et al., 2017). However, many established protocols for ROS measurement (Dunand et al., 2007, Jambunathan, 2010, Rodr\u00edguez & Taleisnik, 2012) included the soaking of plant tissues on non-physiological buffers, which might alter steady-state ROS levels. Several methods have been used for ROS localization and they rely on histochemistry, fluorescent dyes, and spectrophotometric measurements (Mittler et al., 2011).HistochemistryHistochemical methods are based on the oxidation of dyes in the presence of ROS, resulting in the production of insoluble precipitates. For example, nitro blue tetrazolium (NBT) chloride reacted with O2- to generate water-insoluble di-formazan, while 3-3-diaminobenzidine (DAB) is oxidized by H2O2 in the presence of peroxidases with formation of a dark-brown precipitate (Jambunathan, 2010).Fluorescent dyesSome chemical dyes became fluorescent after oxidation by ROS, like H2DCFDA, DHE or Amplex red (Ortega-Villasante et al., 2016). These dyes can be used for direct ROS localization.Spectrophotometric methodsThey allow to quantitatively determine ROS level after tissue homogenization, such as the determination of H2O2 levels with 3,5-dichloro-2-hydroxybenzensulfonic acid (DCHBS)in conjunction with 4-aminoantipyrine (AAP) (Van Gestelen et al., 1998). There methods were summarised in the graphical abstracts.Here we provide several detailed protocols for ROS localization and quantification under physiological conditions, aimed to improve current methods and to minimize artefacts.