Pedogenesis along chronosequences has received attention from soil scientists for decades, following the pioneering work of Walker and Syers (1976), who showed a general pattern of the availability of nitrogen (N) and phosphorus (P) subsequently confirmed for a range of chronosequences (Huggett 1998; Turner & Condron 2013; Vitousek & Farrington 1997). Much more recently, changes of plant nutrient-acquisition strategies with soil development have drawn attention, first with a conceptual model (Lambers et al. 2008), which was subsequently confirmed with solid experimental data collected along the Jurien Bay chronosequence in south-western Australia (Zemunik et al. 2015). The original conceptual model has since been amended, to explain why presumably less-effective mycorrhizal P-acquisition strategies coexist with more-effective strategies based on the release carboxylates (organic anions) (Lambers et al. 2018). Mycorrhizas may not be effective at acquiring P from severely P-impoverished soils, but they do boost the defence against pathogens, which strongly affect the growth of non-mycorrhizal carboxylate-releasing Proteaceae (Albornoz et al. 2017; Laliberté et al. 2015). What has received little or no attention is the P-acquisition strategies of species that are the first to colonise the bare regolith of the earliest stage of chronosequences (Figs 1 and 2). When the soil P availability is relatively high such as along the Jurien Bay chronosequence (Turner, Laliberté & Hayes 2018), there is little point searching for specialised strategies, but along many chronosequences, that is not the case (Zhou et al. 2019; Richardson et al. 2004; Egli et al. 2012). We propose an experimental design to test which species at very early chronosequence stages acquire P based on accumulation of carboxylates in their rhizosphere (Fig. 3). What is needed are both positive and negative controls (Zhong et al. 2020). A positive control is a plant i.e. known to release large amounts of carboxylates, e.g. a Proteaceae or Cyperaceae species (Delgado et al. 2014; Shane et al. 2004, 2006) or some Fabaceae, e.g. Lupinus (Lambers et al. 2013) or Astragalus (Lan et al. 2012; Zhang et al. 2019) species. Ideally, a mycorrhizal plants i.e. not being facilitated is used as a negative control, e.g. Xanthorrhoea preissii (Zhong et al. 2020). If this is not possible, then the youngest leaves of a carboxylate-releasing plant might be a good negative control, as the leaf [Mn] is typically high only in mature leaves, and not in young expanding leaves (Shane and Lambers 2005), but this would need to be verified. Plants with leaf [Mn] in between the two contrasting controls are likely facilitated (Fig. 4).