Given that they Numbersare unique, then
v1 <- c(t(df4))
v1[!duplicated(v1, fromLast = T)]
#[1] "1" "2" "A" "3" "B" "4" "5" "C"
Numbers, (, Numbers = c(1, 1, 3, 4, 5)), transform make.unique . sub make.unique
( unique, @Jaap , ),
sub('\\..*' ,'',unique(c(t(transform(df4, numbers = make.unique(as.character(numbers))))),
fromLast = TRUE))
#[1] "1" "1" "A" "3" "B" "4" "5" "C"