Mike's right, and it's quite an interesting story (as are most language stories).
The Chinese characters were first imported into Japan with Buddhism back in the 5th Century or so (my memory fades). For some ideas (ideas of governance, Buddhism, history, etc) the Chinese words fit precisely or the words were imported with the letters.
When the Japanese tried to apply the Chinese characters to convey the complete scope of the Japanese language, they ended up shoe-horning the chinese characters into local useage by adding hiragana letters to start a word with a Chinese character and finish with a Japanese letters to complete the local pronunciation an useage. Most kanji have both on (Chinese origin) and kun (Japanese origin) pronunciations depending upon their use. Similar appropriation took place when the Akkadians took over cuneiform from the Sumerians: a lot of the glyph designs remained the same, but the pronunciation changed and supplementary glyphs came into being.
Moreover, since the Meiji restoration in the 19th century, there have been a couple major revisions for simplification of the Chinese characters used in Japan as reading became less about status and more about education of the general population.
Then there is the Meiji construction and names. A whole lotta Japanese serfs didn't have official family names until the end of the feudal system of governance. A lot of "new" family names used semi-geographically descriptive. Okamoto? I'm betting your great-granddad lived at the base of the hill (not a hard trick, since there lots of valleys in Japan). Nakamura? I bet your family used to live in the village... So given the relatively short amount of time between the Meiji and the end of WW2, I'd bet that Okamoto was probably unchanged.
The most interesting thing that I noticed when comparing Japanese and English root words (in a completely unscientific manner in my own mind) is that often in English we have a basic understanding of what a word is about by how it sounds. For example, "hydro-power, hydro-dynamic, hydro-X) is going to involve water somehow and 90% of the time we're right. We might spell hydro very creatively, but the idea remains the same when conveyed aurally. In Japan, there are lots of words that sound similar and their ultimate meaning might remain vague until the confused party can see the kanji. Homophones abound. Now placenames? You could show any well-educated, erudite Japanese the name of a random unknown village written in kanji and if pressed for a pronunciation, they would probably say "I"m not sure, but... it could be...." b/c local dialects gave towns all sorts of odd, one-off pronunciations.
cliff notes: names usually take Japanese pronunciation. Other words, maybe.
edit: to keep this on topic, descriptions of gun parts and terms involving firearms are usually Chinese kun pronunciations
I do that often. You should see what I write if I'm researching two different guns at the same time.