Mycorrhiza and plant (seedling) survival
Should we think more about the role of arbuscular mycorrhiza in plant survival rather than about plant growth effects?
It is my impression — admittedly not verified by a dedicated effort to the review the literature — that research on arbuscular mycorrhizal effects on plants is skewed towards measuring plant performance traits, like growth, rather than plant survival.
If this is accurate (and please correct me if I’m wrong), perhaps there are two reasons for this. The first is that there has been a strong agricultural influence on mycorrhizal effects, and the interest there has been on yield and growth, and not so much on survival. The other reason is one that is related to the way mycorrhizal experiments are typically done. Let me explain.
When you do an experiment on the effects of mycorrhizal fungi on plants you actually don’t have very many good options. Typically you are comparing a plant with mycorrhizal fungi to one without, which means that you inoculate the plant that receives the treatment with some source of mycorrhizal fungus, while not inoculating the control (both soils would have need to have their native AM fungi eliminated, typically by steaming, or some other method). This means that at the time the mycorrhizal treatment is added, the mycorrhizal fungus itself (or the fungi, if a community was used to inoculate) is in a state of propagules. Those can be spores, hyphae, and colonized root pieces from the inoculum source. What now needs to happen, is that the plant needs to be colonized. Nothing wrong with that per se.
But when you’re studying plant survival, this is often done with seedlings (again, correct me if I’m wrong), likely because seedlings are the most sensitive life history stage of the plant, where most mortality happens. And so this means the fungus — in your normal experiment — does not have much time to establish the symbiosis before you start measuring the outcome of the experiment. This is a problem, because the symbiosis will barely have had a chance to interact with the host (colonize plant roots, produce mycelium in the soil supported by that colonization) at the time the effect is being assessed.
There are ways around this of course. For example, you can have a ‘nurse’ plant growing in your pot before the experiment, so that a mycorrhizal fungal network already has a chance of establishing before you start your actual experiment. Then you would need to remove that nurse plant, leaving the mycelium in the soil. It is more complicated, requires an extra step, and it is a bit awkward, because then you have a bunch of dead roots in your pots, and that nurse plant (in the control and in the later mycorrhizal treatment) will have already have interacted with the soil (for example, it will have extracted nutrients etc.).
The other way around, that is eliminating mycorrhiza in one treatment, rather than inoculating, woud also be a theoretical option, but this doesn’t work well, because we don’t know of anything that selectively removes just mycorrhizal fungi while not affecting other soil biota, such as other fungi. You could use mutant plants that are defective in their ability to form mycorrhiza, but there are only relatively few species for which this option is available. And you could work with rotated/ non-rotated core arrays, but it may be challenging to do this with a bunch of seedlings in these cores.
The other option is to study survival when the plants and the mycorrhiza have already established themselves, which means working with older plants. This seems perhaps the easiest option, maybe at the risk of missing some very important effects at the seedling stage. But it still seems a very worthwhile thing to do.
All of the above is for laboratory or greenhouse experiments; of course, the gold standard for ecological effects (not necessary for uncovering mechanisms) is a field approach. But this would be even more challenging, for all the reasons above — and additionally because of the difficulty of maintaining any treatments, such as non-inoculated controls. So let’s not go there for now….
So, fellow mycorrhizologists and other people with an interest in the symbiosis: do you agree? What do you think about all this? Are there any ideas out there on how to study seedling-stage mycorrhizal effects that I have missed?
All of the above is on arbuscular mycorrhiza; if the situation is different for ectomycorrhiza or other types, do let me know!
Looking forward to your comments!
Hi Matthias, thank you for this great content. In my opinion, it depends. From agricultural perspectives, plant (mainly crops) survival seems not a big issue in current agricultural practice. here I did not consider extreme conditons. People care more about quality and quantity of food. But in ecological restoration, particularly in post-mining sites where top soil was completely removed. Plant survival is a big issue which further compromise the successful of ecological restoration. There are few examples to use AMF to improve plant establishment in field. However, outcomes are not so compelling. As you know, filed conditions are alway sophisticated, and it is hard to uncover the single effect of AMF.
This is right, the establishment period can be long and very demanding for the plant during the early stage of the establishment and deployment of the network.
I remember during my PhD work, I was assessing the AMF inoculum on Prunus africana seedlings which is a tree endemic in some African countries, so I had to conduct the experiment in a nursery for almost 12 months in order to observe any changes, encompass the survival.