This is the emerging shoot of yabumame (Amhicarpaea edgeworthii), whose Japanese name translates as something like bush soy. The more typical trifoliate leaves will follow shortly. It looks very similar to Amphicarpaea bracteata, talet or hog peanut, about which I have been known to hold forth for hours on end to interested parties. They were the ones who didn't nod off, or stare out the window at some particularly fascinating cloud formations. But seriously, just like talet, this is a tasty plant, with the same sort of potential.
It is sometimes classified as Amphicarapa bracteata subsp. edgeworthii var japonica, which suggests that the resemblance to talet is more than superficial. Unlike its New World cousin, it has been subjected to some scrutiny as a potential food crop by the Japanese. If you fancy practicing your hiragana, katakana and kanji comprehension, then here's a good place to start. I tend to skip over the text and look at the pretty pictures. Nevertheless, it's probably a safe bet that these research findings would be applicable to talet as well. Scroll to the bottom for the English abstract. For other hard core Amphicarpaea edgeworthii fans there's more where that one came from. Or maybe I'm the only one.
One of the advantages of the amphicarpic habit is that the plant produces two types of seeds, the cleistogamous, self pollinated ones (underground and overground, just like the Wombles) and the open pollinated chasmogamous seeds from the flowers at the top of the plant. It should be possible to allow or assist the plants in crossing using the chasmogamous route, then sow the resulting seeds and maintain any favourable ones via the seeds produced by the handily self-fertilising cleistogamous flowers. That's my theoretical take on it anyway, which I've had no opportunity to verify or disprove so far.
I'm also wondering whether the two species can be crossed to introduce a bit of extra genetic variation. I haven't tried it and don't know whether it's possible, but seeing as the two are so closely related, it must be an avenue worth exploring.
As to differences, well, A. bracteata and A. edgeworthii are known to be colonised by different varieties of rhizobial bacteria, with A. edgeworthii unable to use the strains found on A. bracteata; A. bracteata, however, can use A. edgeworthii's bacteria. I shall have to see whether any nodules form on my yabumame's roots later on. Not much point having a plant with the ability to fix nitrogen without providing it with the means to do so.
Then there's A. africana - yes, an African hog peanut about which I know precious little, other than that it grows in montane Africa at elevations of around 2000 metres or so. For the sake of completeness, I really ought to try and grow it.
Delving deep into the Gormenghastian recesses of the internet in search of all this stuff is fun, but can't quite compete with the pleasure of devouring a plate of cooked talets, steaming majestically and crowned with a knob of butter. The former I can do any day; the latter, regrettably, will have to wait until the autumn.