Phonological search may need to be run only on a subset of the corpus. For example, a user (e.g., myself) may want to find /l/-ending words only in a sub-lexicon (e.g., Sino-Korean).
I envision an option like the following. Selecting 'Sino' there, for example, do the search only on the sino stratum.
