The Forgotten Oil Millet of Taiwan

Question: so what is the world's most obscure crop? Browntop millet (Brachiaria ramosa), now very relict, but clearly important as a millet in Neolithic South India is normally one of my favorite candidates, or perhaps one of the minor Digitaria domesticates of West Africa or Assam. The poetically named sumpweed, Iva annua, now extinct from cultivation is another good candidate, although it is now well-known to archaeologists through the work of Bruce Smith and others (see, e.g. his recent article discussed below). I now think that this dubious distinction goes to a crop from Taiwan.So I should put on record my fascination with an obscure, and largely forgotten, millet, and an obscure article. This article is mainly in Japanese (but with an English abstract).

• Takei, Emiko (2008) Historical review of Spodiopogon formosanus Rendle, a minor grain crop in Taiwan. Bulletin of the Cultural and Natural Sciences in Osaka Gakuin University 57: 43-66.....Unfortunately neither this publication nor its abstract seems to be available freely on-line, although a bibliographic record is here. I reproduce the English abstract here.

Abstract: This paper re-examines botanical descriptions of Spodiopogon formosanus Rendle (syn. Eccoilopus formosanus (Rendl.) A. Camus), a cultivated millet endemic to Taiwan. In the early 20^th Century, this species was described as a new species, and its cultivation by Taiwan aboriginal people was noted. No detailed account of its cultivation was published, and the plant became an invisible crop in botanical and ethnographic literature.During the Japanese occupation period (1895-1945), classification of this and related grasses became confused and unclear. The genus was changed, and new species and varieties added. The fact of cultivation was given little attention, and the cultivation itself became less common over time. The existence of this crop is still not well known and documented. The most recent edition of the Flora of Taiwan does not indicate that the plant is cultivated, and gives no information about its distribution. The confusion in taxonomy and general lack of information have also caused difficulty in other fields. Non-botanists have documented this species under a different plant name, not only because they were not familiar with the plant, but also because no reliable botanical information was available.Spodoipogon formosanus has not been mentioned in ethnological or agricultural papers over the last 100 years, but there may be reports under different plant names. Further exploration is needed to locate reports concerning the cultivation of minor grain crops in Taiwan, and some may need re-interpretation.

Spodiopogon formosanus is described in Kew Garden's world grass flora, but here no mention is made of cultivation. In the new Flora of China, on which Missouri Botanical Garden is collaborating, it is noted that this species "has been cultivated in the uplands of Taiwan," and we can note that the species is reported to be endemic to Taiwan between 1000 and 2000 meters. It also is decribed as being 'not bearded' (i.e. awnless) with spikelets that do not disarticulate. In otherwords it has two clear traits of morphological domestication that are part of the syndrome relating to loss of natural seed dispersal.

This species was first described by Rendle (1904) in his contribution on the grasses of Taiwain within the larger study of Chinese plants by Forbes and Hemsley published in Journal of the Linnean Society 36.(p.351). Rendle classified this by comparison to Spodiopogon sibiricus, indicating that the comparison suggests “a cultivated form of that species”. Notable contrasts in S. formosanum includes the absence of sessile spikelet (typical of Spodiopogon), less hairy spikelets, “the almost complete disappearance of the awn on the fertile glume”. Most of these traits are plausible adaptations of the domestication syndrome: a reduction in dispersal aids (hairs and awns) and the development of non-shattering pedicel in place of a dehiscence scar, which is paralleled in the evolution of non-shattering in domesticated pearl millet (Pennisetum glaucum).

So this is a domesticated species, but what is its wild progenitor, and was it brought into cultivation only in Taiwan, or was there a former distribution or even origin on the mainland? The wild species S. cotulifer, also in Taiwan and on the mainland, is perhaps the best candidate according to Takei. Further botanical and ethnobotanical studies, which Dr. Takei is actively pursuing are important, but there is a role to be played (eventually) by archaeobotanists, in documenting the present and past use of this species in the past. It is possible that this a relatively recent secondary domesticate. To that end we of course need to know what it looks like. Here is an SEM of a grain taken at RIHN in Kyoto (where I have been visiting) from material collected by Takei.

In very general terms it resembles some of small-medium sized Panicum spp. (like Indian P. sumatrense and perhaps Panicum miliaceum var ruderale), but it is reassuringly different from Setaria italica or Panicum miliaceum var milicaeum, that when found in quantity it should stick out. Its cultivation is distinctive too, as it is a perennial with abundant secondary tillering—very clear on plants that Dr. Takei is growing in her garden in Kyoto.

So what do we call this millet? For the specialist, Spodiopogon formosanum, is obviously to be preferred, but a catchy handle might help to draw attention to our glaring knowledge gap on this species. (Pre)histories of Taiwan are usualy written with an emphasis on rice, Setaria italica, and sometimes sugarcane, as species that reconstruct to Proto-Austronesian, and have at least some archaeological presence. But archaeologists, linguists, botanists need to not ignore this species but seek its history; so perhaps some agreed upon common name might help. As far as I can discover this crop lacks a common name in English, which doubtless helps to keep it out of mind and largely forgotten. The USDA Agricultural Research service lists 'silver greybeard' or 'silver spike' as names for Spodiopogon sibiricus, a wild relative of northeast Asia. But given the awnless state of S. formosanus, and the fact that it sometimes has reddish leaves, does not recommend adapting one of those. The Chinese name applied to this genus of grasses, you mang, means 'oily Mang-grass' (with mang being the general Chinese name of the Miscanthus complex), and it is worth noting that the grains of Spodiopogon appear unusually oily for a grass or cereal, so I have tentatively suggested that we might refer to this species as the moutain oil millet, or perhaps Formosan Oil Millet?