Reply to Tedersoo et al.: Plant species within the same family or genus can have different mycorrhizal types?

There are two main methods used to assign plant mycorrhizal type to plant species: empirical vs. phylogenetic–taxonomic (hereafter taxonomic) methods (1). The empirical method uses only available collated empirical data based on direct observation of mycorrhizal type, while the taxonomic approach extrapolates plant mycorrhizal traits to complete taxonomic groups, such as at the genus or family level, based on information about mycorrhizal type from relatives from the same group. In Sun et al. (2), according to the procedure described by Brundrett (3) and Guo et al. (4), we perform our own empirical observations to assign plant mycorrhizal traits for 35 Chinese temperate woody species. Tedersoo et al. (5) instead use the taxonomic approach (most probably based on the background information from European plant species) and argue that Sun et al. misallocate the mycorrhizal types for many of the studied woody species.

While we greatly appreciate these concerns, we believe that using the genus or family as the taxonomic unit for extrapolating either mycorrhizal type or status may result in inaccuracies in assigning mycorrhizal type in some cases. We emphasize that empirical evidence has shown that plant species within the same family or genus perhaps can have different mycorrhizal types and status. For example, species in Tilia have long been well demonstrated to be associated with ectomycorrhizal fungi, mostly in Europe and North America, while our study (2) and 2 other published studies from the same study area in Northeast China (4, 6) observed that Tilia mandshurica and Tilia amurensis species are colonized by arbuscular mycorrhizal fungi. These 3 studies also found, unexpectedly, that Ulmus japonica and Ulmus laciniata are predominately or entirely associated with EM in the study area. In addition, another recent study (7) using 2 sister Howea palm species from a common plant ancestor reported that the mycorrhizal symbiosis was only preserved in the descendant species grown in the same soil environment as the ancestral species. The diversity of mycorrhizal types or status within a plant genus or family implies that adaptation of plant mycorrhizal associations to local conditions may be more common than currently understood.

In fact, a recent study (1) already compared empirical and taxonomic methods to assigning mycorrhizal types to species lists using available information in Europe and found that mismatches between the two methods occurred in most plant families. Based on empirical observations, plant species within families shared a common mycorrhizal type in only 46% of plant families, highlighting the diversity of mycorrhizal statuses within the plant families or genera.

We appreciate Tedersoo et al.’s (5) discussion of this timely and important topic. The evidence that plant species within a genus or family perhaps can have different mycorrhizal types highlights the need for high precision in assigning mycorrhizal traits in future research; we fully support that more direct observations are required to avoid errors in assigning mycorrhizal type and status, as emphasized by Tedersoo et al.