Before the loop, struct2cell(Ribboned_100) . Then look at the fieldnames() of Ribboned_100, and match those against the names from XT, getting the respective indices. Extract those cells from the cell array.
Now you can loop over i, extracting the fields named by var{i} from those particular cells.
All of this is only likely to be more efficient than what you have if the maximum index is large.
If the var{i} between them cover all of the fields, then you could reorderfields() of those cell arrays according to var order, and then cell2mat(), after which you would be doing numeric indexing instead of struct reference.
