I tried this idea and it works most excellently. It's super fast and I'm able to easily split the subanagrams up by length. I have a listbox for 3-letter, 4-letter, etc. words.

I did come across a problem. Initially, I created a new word list file that had each word paired with its "prime value". However, when the really larger number gets printed to the file, it's written in "E" notation and some of the numbers lose decimal places causing the values to not match the value of the letters I'm trying to anagram. I corrected this by creating the "prime values" on the fly as the word file is read into the array.

Now that "anagrams" is complete (except for entering "*" for blank scrabble tiles), I'll work on "words that end with", "words that star with", and "words that contain", which should all be pretty simple.