Your (not only your, of course) use of doubtful and/or incomplete homologies to demonstrate a given hypothesis is dangerous

Danger is my middle name by the way.

This post (and the others in the series) is not meant to demonstrate anything. It is meant to be an interesting story (I hope) about a particular body structure, the homology of which seems to be well established. Now, elucidating the adaptive scenario (if any) leading to the origin of this morphological novelty and it’s subsequent loss in a few derived lineages will always remain a difficult matter.

More realistic information on the metapleural gland: Crematogaster are arboreal ants ecologically successful at least as much as Camponotus and regularly equipped with a metapleural gland.

I thought that over 2,000 Camponotus and Polyrhachis species were without metapleural gland. Am I wrong, or are Camponotus and Polyrhachis not ants?

You are right of course, and this secondary loss probably represents one evolutionary event. However, I don’t think I need to tell you, out of all people, that synapomorphies are evidence of monophyly regardless of subsequent modification.

Please don’t tell me again the story of the secondary loss of the gland among these ants ‘because they live on trees…’
There are hundreds of arboreal ants with metapleural gland and hundreds of Camponotus nesting in soil without it.

Your point is valid, but this doesn’t changes the fact that the ‘because they live on trees’ explanation is consistent with the ecological success hypothesis that has been advanced in the scientific literature. In any event, testing the adaptive significance of traits at the macroevolutionary level will always be difficult.