Patronymics are generally Greek-derived Latin nouns with special endings, appearing frequently in epic poetry and rarely elsewhere. They may be masculine or feminine, but always one or the other, depending on the specific ending:
-adēs, idēs, īdēs and -eus are masculine
–ās, is, and -ēis are feminine
In Greek, these are usually adjectives; in Latin, they are usually nouns.
Tydareus, -ī (the Spartan king) —> Tyndaridēs, -ae (either Castor or Pollux, the twin sons of Tyndareus)
Tydareus, -ī (the Spartan king) —> Tyndaris, -idis (Helen, the daughter of Tyndareus)
Anchīsēs, -ae (Anchises, the Dardanian prince) —> Anchīsiadēs, -ae (Aeneas, the son of Anchises)
Tydeus, -ī, (Tydeus, the Aeolian hero) —> Tydīdēs, -ae (Domedes, the son of Tydeus)
Oīleus, -ī (Oileus, the Locrian king) —> Aiāx Oīleus, Aiācis Oīleī (Ajax Minor, the son of Oileus)
Hesperus, -ī (the Evening Star) —> Hesperis, -idis sg. (a daughter of Hesperus) —> Hesperides, -um pl. the Hesperides (not Hesperidēs, -ae, which would be masculine)
A number of adjectival endings denote what Allen and Greenough refer to as a ‘gentile’ relationship—demonstrating ‘relation to’ or ‘belonging to’ the corresponding class of nouns. One of these is -ānus, -a, -um.
montānus, -a, -um, of mountains (mōns, montis, mountain)
veterānus, -a, -um, of veterans (vetus, veteris (adj), old)
antelūcānus, -a, -um, before daylight (ante lūcem, before light)
Rōmānus, -a, -um, Roman (Rōma, -ae, Rome)
Sullānī, -ōrum, of Sulla’s veterans (Sulla, -ae, Sulla)
Some of these derived adjectives have furthermore been transformed to new nouns.
Silvānus, -ī, Silvanus, a woodlands deity (silva, -ae, the wood)
Recall the standard stem for Latin comparatives is -issimus. However, most student of Latin are familiar with a variety of alternative, irregular forms. For instance:
Bonus, melior, optimus
Malus, peior, pessimus
Magnus, maior, maximus
These are all more archaic forms of the superlatives. (Hence there appearance in very basic, common adverbs, which we can predict would be more resistant to phonological change due to frequency of use.)
Furthermore, A&G note that certain superlative adjectives are derived from their comparative forms, not from their positives. They aren’t explicit about how this works, but the example they offer is extrēmus, which might go exterior –> *exterīmus –> extrēmus.
A&G compare this to the derivative development of ‘childish’ superlatives like the English furtherer and furtherest. Again, this is all a little mysterious to me, as a non-phonologist. If anyone has thoughts, I would love to hear them.