I saw your questions, and I have had my hands very full…and have experienced some difficult times that have taken a lot out of me.
To respond to some of these questions is not a simple matter. I’m going to make a try (I’ve been writing this for nearly two hours already), but I cannot go on with this topic beyond this one response.
At the moment, I know much less about the anatomy of “var. guessowii” than I do about “subsp. flavivolvata.” I’m pretty sure that the yellow-capped and red-capped populations of the provisional “amerimuscaria” are currently capable of exchanging genes. Dr. Geml did feel there was evidence for “recent” gene exchange.
The following is my attempt to briefly describe my understanding of a very large topic area…with the emphasis on briefly.
A phylogenetic tree is a hypothesis. It is computer assisted, and it is a hypothesis. Underlying the tree are multiple hypotheses of the form “this statistical method applied in this way with this model of how evolution works (on an amino acid-by-amino acid level in genes) can be applied using these programming methods to this data and create a hypothetical evolutionary history of this set of existing organisms in a more reliable way than previous tools or previous versions of this tool.” [Note the appropriate and complete absence of a claim to the tools’ producing something like absolute truth.]
Certainly very capable people are proposing these hypotheses…very many of these hypothesis. Certainly such people are proposing new hypotheses at a significant rate. The flurry of new computer-based tree-estimating tools is quite amazing. Any hypothesis based on an “old” tool will appear outdated to some critics in a matter of a very few years…maybe less.
A morphologically defined taxon is also a hypothesis.
Both types of hypotheses are created within moving frames of knowledge.
It is absolutely to be expected that, given we are taking the best shot NOW at understanding nature, our hypotheses will be forced to change over time by such things as more and better methods, more and better observations, elimination of erroneous theories, etc. The rate of change at present is quite high.
A decade or so ago, people made trees based on single genes and used them in development of evolutionary hypotheses. Now, many phylogeneticists (based on personal communications to me) will say that a one gene tree should be considered not as an evolutionary hypothesis concerning organisms but as a hypothetical history of the gene in the sampled organism or organisms. In other words, the interpretation of the meaning of a single-gene tree has become quite restricted.
With few fossils, the statistical methods of phylogeny are trying to produce a history behind a set of existing organisms. For Amanita that history might extend backward over 100,000,000+ years. It is amazing that people are trying to build tools to do that for organisms the history of which is entirely unsupported by fossil evidence. The point of these comments is to stress the magnitude of the problem to which computer programs are being applied. Another point to be made is that only another computer program can act as a test of any given computer program. No human experience or examination of specimens can test the hypothesis (tree) generated by a program that generates millions of hypothetical trees in its search for one that is best according to the hypothesis governing the concept “best.”
Finally, to repeat, there can be no ultimate guarantee that any tree is “correct” in an absolute sense. On the other hand, humans can’t operate in the realm of absolute senses. Consider the history of particle physics (Heisenberg Uncertainty Principle) and mathematical logic (Gödel’s Proof—-that there are true propositions that cannot be proven from any set of axioms sufficiently strong to allow derivation of elementary Arithimetic).
If the above isn’t sufficient, I beg you to excuse me. I cannot continue on this topic due to lack of knowledge, lack of time, and lack of energy.