Misdiagnosis and uncritical use of plant mycorrhizal data are not the only elephants in the room

There is increasing interest in using plant mycorrhizal traits – characteristics related to a plant’s ability to form mycorrhizal symbiosis – to understand the role ofmycorrhizas within and across communities, ecosystems and biogeographical regions (Moora, 2014; Tedersoo, 2017). Recent studies incorporating plant mycorrhizal traits, mainly mycorrhizal type (e.g. ecto-mycorrhizal (ECM), ericoid-mycorrhizal (ERM), arbuscular-mycorrhizal (AM), and non-mycorrhizal (NM)) and mycorrhizal status (obligately mycorrhizal (OM) and facultatively mycorrhizal (FM); e.g. Correia et al., 2018; Gerz et al., 2018), have assigned trait values from published empirical data for plant species of interest. Dedicated efforts to build and improve databases of empirical plant mycorrhizal traits (e.g. Chaudhary et al., 2016; Bueno et al., 2017) aremaking these resources increasingly comprehensive, transparent and accessible. A recent Viewpoint article by Brundrett & Tedersoo (2019) highlights several challenges connected with assigning mycorrhizal traits to plant species and criticizes the use of mycorrhizal trait databases. The authors argue that the use of such trait databases (which they refer to as ‘recycled data’) is ‘inappropriate’ due to two underappreciated problems: the databases (1) may include ‘mycorrhizal trait allocation errors’ due to misidentification of root mycorrhizal structures; or (2) may contain data derived using diagnostic criteria that are flawed. They propose instead (1) diagnostic criteria for defining mycorrhizas that make specific assumptions about the biology of the interaction, and (2) checking databases against their curated list, derived from prior experience of assigning mycorrhizal traits to plant species (hereafter referred to as a ‘standard reference’, Brundrett, 2009; Tedersoo, 2017). Although we fully agree that standard protocols are required, there are several questions raised by the Brundrett & Tedersoo’s Viewpoint that deserve critical consideration if consensus is to be reached among researchers in the field: (1) should the presence of minimum number of arbuscules define the AM symbiosis; (2) does plant taxonomy accurately predict plant mycorrhizal traits (i.e. high phylogenetic trait conservatism); (3) should the results of published studies be considered ‘incorrect’ when they do not match this, or any proposed, ‘standard reference’? Later, we discuss these three questions and address an additional topic that arises from consideration of Brundrett & Tedersoo’s criticisms of prior research: (4) the level of transparency involved in building plant mycorrhizal trait databases and directions for future research.