U in Junction and in Derivation

The short labial vowel behaves in the rules of junction exactly as i does in its own series. Before a dissimilar vowel it becomes the semivowel va, so a word ending in u and a word beginning in a run together with the vowel rewritten as a consonant, and the seam is invisible until the reader undoes it. Raised one grade it becomes o and raised a second it becomes au, and a family of Sanskrit words are the same root taken at different points on that ladder.

The vowel is also the ending of an important class of nominal stems and the vowel of a set of common suffixes, so it is met constantly in ordinary vocabulary. In the Prakrit of the Ashokan edicts it is common as well, and it is one of the vowels that the older syllabic r fell together with in some dialects, so a Prakrit word may show u where the Sanskrit cognate shows the syllabic liquid. That kind of correspondence is one of the tools by which the dialect of a given edict is identified, since the Mauryan inscriptions were cut in several different Middle Indo-Aryan dialects across the empire and the vowel treatments differ from one to another. A reader working from Sanskrit therefore has to read a Brahmi inscription with the language in mind rather than simply transliterating it.