In looking up the etymology of the word はと【鳩】 hato "dove, pigeon", I came to the conclusion that there is very little dialectal variation in the Japonic languages and that all regional variants point to the Proto-Japonic form *pato. On the opposite side of the Sea of Japan, the equivalent in modern Korean is 비둘기 bidulgi /piduɭɡi/, which has the obsolete alternative forms 비돌기 bidolgi, 비도리 bidori, 비두로기 bidurogi, and 비두리 biduri. According to Sergei Starostin's Altaic etymology database , the Proto-Korean form would have been *pìtùrí.
What piqued my curiosity with the Korean form is that the two languages both exhibit a plosive in the first syllable and an alveolar stop and rounded vowel in the second (*pVto-). The two last syllables in the Korean form are also not all that unlike the Japanese word とり【鳥】 tori "bird". Considering the two, it makes me wonder if the original Proto-Japonic word was not something like *patori instead, with the ending *-ri having been lost in time.
In a previous topic , I noted that the Japanese words *pa(p) "leaf" and *pa "tooth" shared an interesting parallel with the Korean equivalent *nip "leaf" and *ni "tooth". So perhaps there exists some sort of parallel between the vowels /a/ in Japonic and /i/ in Korean that would have caused the divergence we see today. More data would be needed to confirm this idea. As for the velar consonant in Korean, perhaps it's a suffix of some sort, similar to what see in the words for "crab" (Korean ge :: Jeju ging-i :?: Jpn kane/gane) and "bird" (Korean sae :: Jeju saeng-i :?: Jpn sagi 'heron') .
All of this, of course, is just conjectures, but I thought I'd share anyways.