Taxacom: Collective groups and form(al) genera

[Ziv Lieberman]

Hi all,
I'm thinking about the differences between "form" or "formal genera" and
"collective groups" and how to best approach these entities when
establishing new nomenclature.
In ants, there are many groups of fossil taxa which have been established
for convenience, without implying that they form a natural group - mostly,
e.g., impressions of wings or incompletely preserved fossils which can be
associated based on one or more morphological features. Authors generally
state that these are "form genera" (or, perhaps confusingly, "formal
genera", especially in the Russian school of paleontology). However, they
are nomenclaturally treated as "real" genera, including in that they have
type species.
To my understanding, these "form genera" align with the Code's definition
of collective groups: "An assemblage of species. . .that cannot be
allocated with confidence to nominal genera. Names proposed or used for
collective groups are treated as genus-group names but special provisions
apply to them". These provisions are given in 42.2.1 ("The names
established expressly for certain assemblages of taxonomic convenience
known as "collective" groups. . .established at the genus-group level are
to be treated as genus-group names in the meaning of the Code." 42.3.1
states "Collective groups have no type species."
Therefore, I arrive at the conclusion that even though many "form(al)
genera" are "established expressly for assemblages of taxonomic
convenience", they are *not* collective groups, since they have type
species.
This leads to my direct question: If I want to establish a new collective
group, should I follow 42.3.1 and not fix the name to a type species? This
seems like it must be the right answer, but I am hesitant because I have
never seen an example of such a name in practice.
Further, if the type species of a form genus is transferred to a type-free
collective group, are there any additional complications to consider?
Thank you!
-Ziv