If the question was worth asking and the data worth having, then having 18 tables is just the way things are. If the journal has page limits then try a different journal. Not all science projects can fit into a 2000 word format and forcing them into that is bad science. That said, sometimes large projects have a few key points and a great deal of supporting material. In that case, try publishing in a journal that includes supplemental material. The supplemental material may be available only online, but that should work. As a last resort, publish the main findings and then include an unpublished supplement that you include on your RG account.

Is it absolutely necessary to report all F-statistics and relevant results? Perhaps you can omit some of them. The attached file, especially, table 1, has a relevance to the context of the main question of the thread and may be useful.

Regards

For descriptive purposes and being pragmatic about page limits, online supplementary material sections are allowed by many journals, but you can also post these places like ResearchGate. My question would be whether each of these is such is such a bold conjecture to require a hypothesis test (and make sure you adjust the p values appropriately). Some authors seem to report Anovas just because they can be done.

please How did you calculated the p value I used poptree2 to calculate the Fsts and drw the tree but I didn' get the p value.

any suggestion please

Yes. I did a similar analysis in my dissertation (1988)......