Why can "bubo" ("owl") be feminine or masculine?

Question

Why the occurrence of "bubo" in the Virgilius text is an hapax?

This text is the only one listed in Lewis & Short with "bubo" being feminine.
Usually, it's a masculine noun. So, it is an hapax.

Why?
It is a mistake? The hint of a variant in the use of the word, existing among the Roman population, but not attested in other texts? Is there a logical reason for "bubo" having 2 possible genders? Did it have any consequences in the gender of descent languages?

Is it common for a noun to have 2 possible genders in Latin, or exceptional?

būbo, ōnis, masculine (feminine only once Vergilius A. 4, 462; cf. Serv. ad loc.;

I wonder that, because this word is listed as masculine and feminine in many dictionaries.

Answer 1

The relevant passage is this one, from Aeneid IV.462-3:

sōlaque culminibus fērālī carmine būbo
saepe quer' et longās in flētum dūcere vōcēs

And the lone owl on the rooftops would cry out its mournful song, drawing out its long calls into an elegy.

I can see a few possible reasons why Vergil chose to make this particular būbo a feminine one:

• Masculine wouldn't fit the meter—for these lines to scan, the second syllable has to be light, and sōlusque would be heavy. These sorts of changes are common in poetry: note also how there's only a single owl on multiple rooftops, even though logically one owl can be on only a single rooftop at a time.
• This whole passage is about Dido's madness, and Dido is female, so it's poetically appropriate to have the owl be female too.
• Many names of animals in Latin have "common gender", and can take either masculine or feminine adjectives: if it were a lone dog mourning on the rooftops, it could be either a canis sōlus or a canis sōla. Būbo is not one of these (it instead has "epicene gender"), but the idea of an animal name switching its grammatical gender based on the sex of its referent wasn't a foreign or uncommon one.
• Many other words for owls, such as strix, noctua, and ulula, are grammatically feminine.

Any or all of these might have weighed into his decision.

Answer 2

Since the Virgil's passage in question happens to start with 'Tum' I have to contribute something...

(This is not an answer but I see no way of putting this as a comment due to formatting issues.)

There appears to be no clarity on the issue, so the chances of getting a clear, authoritative answer are actually slim.

"The Royal Grammar, Commonly Called Lilie's Grammar, Explained ... By William WALKER (B.D., Schoolmaster of Grantham.)" has the following paragraph with regards to this usage:

P. Virgilii Maronis Æneis. The Æneïd of Virgil, with Engl. notes [&c.] by C. Anthon. Ed. with alterations by W. Trollope also mentions an opinion that 'sola bubo' is to be understood as 'sola avis bubo':

And, for what it's worth, there is a whole article called "Dido and the Owl" by Emily Gowers that offers, what I would call, a poetic rather than a scientific interpretation (this is just an excerpt, there are a few pages dedicated to the topic):