The Parts-of-Speech file is 'partsofspeech.jlf', and it's just a normal text file.
This file is used to convert tiny little markers into readable explanations of a word (the ones seen in the 'info' part of the Result Window).
Here's an except:
#
# For EDICT
#
abbr abbreviation
adj adjective
adv adverb
adv-n adverbial noun
adj-na adjectival noun
Most EDICT entries have one of these tacked on before the entry's meanings really start. A sample entry for 'daiji' might start like this:
(kanji: daiji) [(kana: daiji)] /(adj-na) important/
Because the little part-of-speech marker is clearly in the English area, you can use English searches to find parts of speech. However, you cannot mix English and Japanese searches yet, so this feature isn't really worthwile.
The #s in the file signify comments. One MUST be the first character of the line, or else it will not be recognized as a comment marker.