That does make sense. Is the real problem the sorting issue then? What it seems we have is a mix of numbers and text strings. As noted above, numbers are all smaller than text, so it appears that Excel is sorting all the numbers in increasing numeric order, then sorting all the text in increasing text order.

The usual solution I see to this is to parse the text so that each element is in its own cell/column/field as a number. Is that allowed? Text to columns can do this quite readily. One time through using "-" as the delimiter, and a second pass through using "." as the delimiter. Then, each element is in its own field as a number. Then call up the sort algorithm and tell it to sort first by the first column, then add level to tell it to sort by the second column, and so on for as many elements/columns/fields you have.

In order to get the correct sort, my version of Excel wanted 0s instead of blanks in the empty element spaces. On this small sample, that was not difficult to do manually, but, on a larger sample, that may get tedious to do manually. You might see if this is also going to be necessary in your version, and consider how to handle this issue.