Iris Cohen wrote

I was curious to see what other families belong to Myrtales, & found
what I thought was a disparate collection. Besides the Myrtaceae &
Combretaceae, it includes Lythraceae, Punicaceae, Onagraceae, and some very
exotic families I never heard of. What they have in common, besides some
chemical & microscopic features, is flaky bark and clawed petals (when they
have petals at all). That I can relate to. However, they are also supposed
to have opposite leaves, which Conocarpus does not.
Iris

+ + +

Myrtales have been redefined several times over the past decades. The last
major reshuffling took place in the early eighties, when Lecythidaceae and
Rhizophoraceae were excluded. At an earlier time the Lecythidaceae were even
reckoned to be part of the family Myrtaceae.

Apparently the only consistently reliable typical characteristic of Myrtales
are internal phloem ('intraxylary') and vestured intervessel pits in the
wood. Neither character is easily perceptible.
The APG made a fairly significant addition to the order in joining
in Vochysiaceae, which indeed has those two characters.

I would not know about the position of leafs ('phyllotaxy') in Myrtales.
However I looked into Lauraceae the other day and there this character of
leaf position is quite variable and of very limited taxonomic value.

Taxonomy cannot always use the characters that are most convenient. Plants
were not designed to be easily classified ;-). Another odd fact is that
Vochysiaceae are known to accumulate aluminium, as are Crypteroniaceae and
maybe some other Myrtales?

I must admit I had never heard of Rhynchocalycaceae either (a segregate of
Crypteroniacae).
PvR