"1. It returns a reference to the following variable". I read the reference and understand; now I need to think of what it means in the program.

"2. They operate on the array referenced by the scalar". Again read your reference and have concluded I need to read more about the use of references.

"3. After the first call returns". In thinking about this I believe what is happening is like "nesting". I have been going through this using a single IP range and printing to try to understand. After reading more on the use of references will continue that effort.

Thanks, will likely be back

now I need to think of what it means in the program.

You can't pass arrays to a sub. The only thing that can be passed to a sub is a list of scalars. The solution is to pass references to arrays, references being scalars.

In thinking about this I believe what is happening is like "nesting".

I didn't look at how recursion is used here, but it's usually used to divide and conquer. Let's take merge sort, for example.

Sorting

+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
|   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |   |
+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
[download]

is done by sorting

+---+---+---+---+---+---+---+---+-------------------------------+
|   |   |   |   |   |   |   |   |                               |
+---+---+---+---+---+---+---+---+-------------------------------+
[download]

and

+-------------------------------+---+---+---+---+---+---+---+---+
|                               |   |   |   |   |   |   |   |   |
+-------------------------------+---+---+---+---+---+---+---+---+
[download]

independently then merging the results. How does it sort the two halves? Well by breaking each down into two halves and sorting those. Etc.