Yes, the county number codes do create confusion for genealogists. I'm not sure when, but I suspect the number codes for today are not the same as they were when that marriage index was created. Geneva is not the same number code now that Ancestry is saying it was then. If I remember when I looked it up, Geneva is 34 now, but Ancestry is saying that 31 (where that marriage license was issued) was Geneva. (or vice versa--I'm trying to remember from when I looked at it originally.)
I was born and lived most of my life in Alabama. The codes were based on population--the county with the largest population, Jefferson, was/is number 1. I lived in Mobile County growing up, it was/is number 2. Later I lived in Baldwin County which was/is number 5. So, perhaps at some point, because of population shifts, someone had the bright idea of reassigning the county codes. Just a possibility...
I've been encountering this situation with FamilySearch with so many marriages in my extended family supposedly taking place in Etowah County. So I tried to research to find out why. There is a famous waterfall there, but I couldn't find anything about it being a "wedding destination" back in the day. That is when I concluded that it was probably an indexing problem. Oh, it would be so much better to see the actual image of the certificate instead of relying on the index of an index. But I suppose we have to take what we can get sometimes.