How were names ending in -ιον used in Greek?

Question

In the plays of Plautus, there are some names ending in -um. They are generally formed as Greek names (whether genuine or pseudo-Greek), and the Latin ending -um here seems to correspond to the Greek ending -ον, as is usual.

Although some of these names ending in -um are male, most of them are female. According to one section of "Some Roman Slave-names" by F. F. Bruce:

Greek diminutives in -ιον, mainly names of women [...] appear in Latin with the ending -ium as well [as the ending -io]. Thus Φιλημάτιον takes in Latin the forms Philematium and (as here) Philematio.

When discussing Latin -ium names for women, Henry John Roby interestingly remarks that although these names belong to the feminine grammatical gender in Latin, they belong to the neuter grammatical gender in Greek. (Roby's specific examples are Planesium, Glycerium, Phronesium, Stephanium, Delphium.) (Grammar of the Latin Language from Plautus to Suetonius, Part 1, p. 106)

In Greek Personal Names and Linguistic Continuity, by Anna Morpurgo Davies, I found the following footnote:

13 Some women's names in Greek are neuter, but this need not count as an exception to the general need for marking sex differences. Men's names are never neuter. (p. 21)

Could anyone tell me more about the use of names ending in -ιον in Greek-language contexts? Did they trigger neuter agreement without exception, or might they show varying or mixed patterns of agreement (for instance, I wonder whether perhaps they could have triggered neuter agreement within the noun phrase—i.e. on the article and on attributive adjectives—but had natural gender agreement as a possibility outside of the noun phrase—e.g. on predicate adjectives)?

Answer 1

Well, I don't have an answer, but I found some additional information that I will post here because I don't want to make the question too long. It might be considered the rudiments of an answer.

I haven't found any evidence to support the idea that these names belonged to the neuter gender in Greek any more than -ium names belonged to the neuter gender in Latin. Even though diminutive nouns formed with the suffix -ιον are neuter as a rule (e.g. τὸ παιδίον or τὸ γύναιον¹), there seem to be examples of womens' names ending in -ιον taking the feminine article, which implies that these names are not neuter but feminine. In contrast, I haven't yet found an example of a woman's name being used with any kind of neuter agreement in Greek. (From a theoretical viewpoint, neuter agreement doesn't seem too implausible. As far as I can tell, it is possible to use a neuter article in German with the diminutive name "Gretchen", as in "das arme Gretchen", although Otto Jespersen indicates that feminine agreement may also be an option—I asked a question to check about that situation on German SE.)

Grammars that indicate that -ιον names take feminine agreement

Apparently Priscian, in his Institutiones Grammaticae,

identifies some female names from Greek that end in –on and take a feminine article (i.e. demonstrative) but decline like a neuter noun in –um e.g. ἡ Δόρκιον haec Dorcium, ἡ Γλυκέριον haec Glycerium, ἡ Χρύσιον haec Chrysium, etc. (VI: 24).

(The Western Classical Tradition in Linguistics, 2nd ed., by Keith Allan, 2009, p. 116)

The example of ἡ Γλυκέριον is also given in Smyth's Greek Grammar (1st ed.), 199d.

I found a discussion here that indicates that Smyth may not be the best source for information about this kind of thing. There are also some pointers to other sources (namely, a suggestion by Timothée to consult "Schwyzer—Debrunner, p. 37.").

Specific quotations that show feminine grammatical agreement

TKR pointed out in a comment that the first entry in the linked discussion cites an occurrence of "ἡ Λυκαίνιον" in Daphnis & Chloe by Longus.

It's hard for me to figure out what this example tells us. A discussion on the talk page of the Wikipedia entry for Daphnis and Chloe indicates that it may have been written anywhere from 100 - 400 AD, possibly by someone who did not "write as a native ancient Greek speaker" (I'm not sure whether the suggested alternative native language for Longus would be non-ancient Greek, or Latin).

Citations

1. Smyth's Greek Grammar (1st ed.), 197b