Philomen Probert (Wolfson College, Oxford) writes that
"[A] nominative/accusative dual ending in ω always has an acute, never a circumflex, if accented on the final syllable, regardless of contraction:
νόω > νώ (not *νῶ); ὀστέω > ὀστώ (not *ὀστῶ)
(Probert 2003, §112). This is a synchronic observation.
Other similar exceptions (to the normal rules ...
Three considerations are helpful to distinguish between the infinitive and the alternative 3rd person plural perfect ending:
-ēre is added to the perfect (3rd) stem, while the present infinitive is formed on the present (1st) stem. Although the perfect infinitive is also formed on the 3rd stem, its form (isse) is easily distinguishable.
-ēre has a long ē ...
I don't have precise figures for frequency, but G&L §1301b say that it is very common in earlier prose, less common in later:
In the second Singular, passive, in all tenses of the Present stem, the ending -re is much more common in early Latin than -ris, and is regular in Cic. except in the Pr. Indic., where he prefers -ris on account of confusion ...
As I noted in comment above, Kühner–Blass did not provide an explanation for the acute in νώ. For κανοῦν, on the other hand, they do note that the variable accentuation ὄστεον ~ *ὀστέον (> ὀστοῦν) was noted by Herodian—though they add that penult accentuation on -εον is not otherwise attested in Greek. They admit the operation of analogy from ἁπλόος > ἁπλοῦς ...
Manu Leumann's Lateinische Laut- und Formenlehre (Munich 1977) groups the contraction of -āvis- to -ās- in forms like amāstī and amāssem together with the contraction of -āver- to -ār- in forms like amāram and amāro. Leumann says that "Bei Plautus sind noch die Kurzformen seltener als die Vollformen" ("in Plautus, the short forms are still rarer than the ...
Weiss writes that
"The interrogative enclitic particle -ne becomes -n in Plautus when apocope produces an acceptable coda" (p. 147, footnote 79), i.e. *-Vsn- > *-V ̅n- (I.B.8.b, p. 169).
He gives the following examples:
scivin (Pseud. 977), vin (Curc. 313).
He also says that "[v]erbal forms ending in -s often lose the s with compensatory lengthening" ...
The verb κεῖμαι isn't a contract verb like θεάομαι or ἡγέομαι (or a 'regular' verb like λύω); it's an athematic verb like τίθημι, δίδωμι, or ἵημι, but deponent. So, the circumflex isn't showing contraction as it is in θεῶμαι and ἡγοῦμαι; it's used simply because the accent is on the penult, the penult is a long syllable, and the ultima is short, per the ...
It is very rarely the case that the alternative form of the perfect in third person plural really looks like present infinitive.
If you know how to conjugate your verbs, there is rarely a problem.
In the second conjugation this happens with the following words (I have only listed the relevant forms):
moveō, mōvī (with possible prefixes)
There are two main ways. The first is context; if there's no other main verb, by process of elimination, the word in question has to be. This is typically how I figure it out; I'll read the sentence taking it as an infinitive, realize there's no main verb, and go back to look at all of the candidates.
The second is that this specific kind is an ...
V can be lost when it acts as a perfect "suffix" (not found in the present stem)
Cser 2016 gives the rule that -v- cannot be lost when it is "part of the lexical makeup of the verb" ("Aspects of the Phonology and Morphology of Classical Latin", p. 117), agreeing with TKR's suggestion in the comments. In other words, -v- usually ...
There are certainly verbs whose stems end (or used to end) in -i- and -u-, but what would contraction with a following stem vowel mean? "Contraction" here should be expected to result in a rising diphthong: -ye/o- or -we/o-. but these would be transformed according to well-known phonological principles within Greek:
For /u/, consider on one hand the form ...
I know of 2 others off the top of my head.
satin from satisne
Plautus, Amphitruo 604:
quas, malum, nugas? satin tu sanus es?
Cicero, De officiis 3.73:
quid ergo? satin est hoc, ut non deliquisse videantur?
scin from scisne, mentioned by Rafael in a comment
Plautus, Amphitruo 671:
scin quam bono animo sim?
Cicero, Epistulae ad Atticum 1.13.4:
I found two examples of it followed by a vowel but scanning long in all of Catullus, Ovid, and Vergil.
Here are examples with context to show that perfect is likely:
Cuspis Echionio primum contorta lacerto
vana fuit truncoque dedit leve vulnus acerno;
proxima, si nimiis mittentis viribus usa
non foret, in tergo visa est haesura petito:
longius it; auctor ...
To follow up on my remark to @alex, the answer may be as simple as the -ώ being carried over from uncontracted -ο stems ending in -ώ. The fact that the dual was less used than other forms might make the carry-over more likely. (That may be a bit weak, but I haven't seen a better explanation.)
I think this is not normally done, except if the perfect stem ends on -ev/iv/av-, because only then does one get the familiar contracted sequences -esti/isti/asti. The sequence -usti would look too unfamiliar as a perfect contraction, I think, not to mention -urunt.
I believe irregular perfecta on -evi/ivi/avi are mostly irregular because the verb has a ...
The other answers explain the "what" very well; I wanted to add a bit on the "why".
According to Brent Vine (in Klein's Handbook of Comparative and Historical Indo-European Linguistics):
These [forms] have their sources in phonological reductions (thus audīstī and the like with regular glide loss and contraction for /u̯/ between like vowels), but then ...