Tuomo Pekkanen's Ars Grammatica (a Latin grammar in Finnish) says that the second declension has three neuters ending in -us: vīrus, vulgus and pelagus.
They are only used in the singular, and accusative is exactly like the nominative (not -um).
I have no clue about the origin of these words.
I'm not sure if these words even have a similar history.
There is a small difference between the people and the tree:
Indicating long vowels with a bar and short ones with a cup as usual, the masculine word is pŏpŭlŭs and the feminine one pōpŭlŭs.
Long vowels are indicated in many dictionaries, whereas short vowels are indicated by a lack of macron (the bar).
Acc. pl. pelagē occurs in Lucretius, De Rerum Natura 6.619:
at pelage multa et late substrata videmus
This is simply the Greek plural: the word is a loan of Greek πέλαγος, whose nom./acc. pl. is πελάγη.
A Packard search finds no results for vira, virora, vulga, vulgora.
Here’s a summary of what most authoritative Latin grammars say on the genitive singular ending of –io stems (Weiss 2009/2011: 222-223; Leumann 1977: 424-425; Sihler ). For the sake of simplicity and consistency, in my answer I use the periodization of Latin as used in Weiss (which is different from, for instance, Clackson and Horrocks or Meiser). Weiss has ...
Edgar H. Sturtevant's dissertation "Contraction in the case forms of the Latin io- and ia stems, and of deus, is, and idem" (1902) seems to have some relevant info, although I don't know if more has been discovered since then.
Contraction in the genitive singular
Sturtevant starts out by summarizing the genitive singular forms: he says that in early Latin, ...
A summary based on a number of sources (Kühner and Holzweissig 1912, Leumann 1977, Tronskii 1960, Weiss 2009/2011)
humus, vannus (always feminine);
alvus, colus (these two words alternated between feminine and masculine). Weiss notes that even though alvus is regularly feminine in Classical Latin, examples of masculine gender are found in Old Latin (in suom ...
Kennedy's Revised Latin Primer gives a list of four nouns of the Second Declension ending in -us, which are: alvus, paunch; colus, distaff; humus, ground; vannus, winnowing-fan.
The same source points out that there are several others from the Greek, including arctus, the Bear constellation; carbasus, linen (but in pl. carbasa, n., sails).
[It may be of ...
A full table of "standard" (post-Augustan) -ius/-ium endings would be:
M SG M PL N SG N PL
NOM -ius -iī -ium -ia
GEN -iī -iōrum -iī -iōrum
DAT -iō -iīs -iō -iīs
ACC -ium -iōs -ium -ia
ABL -iō -iīs -iō -iīs
VOC -ī -iī -ium -ia
(O tempora! O mores! Why must upstanding citizens be ...
It seems there are many theories, among them viri, virora, and virus. (The point of the article is that it's pointless to try to figure it out.)
Here, however, is a refutation of the article linked above that contains an argument based on what seems to be solid linguistics ("seems" because it's been so long since I studied linguistics that I can't really ...
There surely is an etymological reason, but unfortunately we don’t know it. Or there is no consensus about it.
All grammars mention the three second-declension neuters in -us, pelagus, virus and vulgus, and many go on explaining that pelagus is a recent Greek loanword which retained its gender (possibly influenced by mare), while virus has Indo-european ...