Many metallic elements and their oxides are named in pairs where the metal itself ends in “um” but that’s replaced by “a” for the oxide. For example, there are aluminum and alumina, titanium and titania, or hafnium and hafnia. How did this come about?
According to the Wikipedia article for magnesium,
“The name magnesium originates from the Greek word for a district in Thessaly called Magnesia.”
Ah, so the word for the oxide happened first and the metal itself was derived from it.
But I don’t think all the oxide names are also words or names unto themselves. Was magnesium the prototype for this? Is there maybe a false etymology or something?
I would suggest that it’s from the 2nd declension of neuter nouns in Latin, which are -um in the singular and -a in the plural. Just the first thing that came off the top of my head.
I take this to indicate that the appearance of a pattern here is wrong, and that the similarity of forms for alumina and magnesia is only coincidence? How about all the others?
I don’t think you read it right. The way I read it, metallic elements ending with -ium are named in the pattern established by sodium and magnesium. Somewhere I read that that ending is now required by the IUPAC in naming metallic elements while non-metallic elements get other endings.
Scientists established a lot of nomenclaure rules a long time ago, to make consistent sense of the language of science. Another example is the -ic and -ous suffixes for metals in compounds, depending on the valence. Ferric, Mercuric, cupric, etc. refer to compounds of the higher valence,and ferrous, mercurous, cuprous for the lower valence. Similarly for acids, such as sulfuric and sulfurous.
For examples like those described by the OP, the ending forms of the names of chemicals are formed exactly for the purpose of showing the relationshp between the two.