Original Translation (from the book)

The right-most character shown for p'o (above) is of course a valid character, but not for the Wade-Giles callout shown just below it. This character translates to 'gushing', which is po without the tone-mark punctuation. I believe this is not the character that Hobson called for.

Though my resource is based on older in-depth Chinese etymology teachings, this is just as easily confirmed with the present day references.

Under the Wade-Giles system, there are 13 different characters for p'o (punctuated) and at least 44 different characters for po (unpunctuated) adding to the confusion. Though Hobson called it out correctly, it's very obvious that an error was made in the choice of the character.

That simple tone mark in p'o being missed in the match up didn't seem too serious at the time. At least not to the English speaking world. Even without the tone mark, the word and it's pronunciation was still po, same look, same sound, with little attention being given to the mysterious associated Chinese character. But the worlds of Wade-Giles and Pinyin were about to collide, and collide they did.

During the massive Chinese romanization project to convert to Pinyin, with so many different characters being identified to the same word, the weight of the character itself took precedence if there was question. The difficulty and error-rate that the older Wade-Giles system offered, validated the character being the final deciding factor. Especially if, as in this case, the meaning of the character was not clear.

Conversion Table

The tone mark was dropped, making the correct p'o lost forever when Pinyin came into play.  The simple error propagated, now becoming a second and more serious error. The character previously (and incorrectly) identified as po, became bo. Hence the birth of su-ni-bo, a name starting to show up in literature as being that of imported cobalt during the Yuan and Ming dynasties.

So the origin of the mystery word is based on an error which, though small at the time, was further propagated with the changeover to Pinyin. Though I have my own idea, it would be presumptuous of me to guess at what Hobson's intended character for p'o was. We'll never know for sure. But obviously not the (po) given. 

In the Preface of his book, he thanks his many colleagues at the Louvre and the Victoria and Albert Museums, naming those that contributed, including the checking of the Chinese characters. This tells me Hobson himself made no mistake. The fact that there was not a single Chinese person on the list might explain the error.

It is a bit confusing. It helps to have a good understanding of the now obsolete Wade-Giles system. Luckily I'm more comfortable in that world. I guess that kind of makes me obsolete as well.