While working in the Taichung Museum of Natural History in Taiwan during June 2010, I was shown some dried specimens of a newly discovered yellow-flowered banana from Ilan, north-western Taiwan, and was told that Dr. Hui-Lung Chiu from the Taiwan Agricultural Research Institute was very keen to meet with me to discuss it. A meeting was arranged for the 29th June 2010 after returning from my work on Lanyu Island.
I could learn very little from the dried specimens other than the seeds looked exactly the same as Musa itinerans var. formosana (Syn. Musa formosana).
The meeting with Dr. Chiu was a long and interesting one. He told me that he discovered the yellow-flowered banana in Ilan 3 years earlier whilst looking for M. itinerans var. formosana there. He has since located a second population 2 km further up the road, making a total of two, possibly three locations for the form. He had provisionally decided to name it Musa itinerans var. gamalamesis after the location where he found it. The species was finally published as Musa itinerans var. kavalanensis H.L.Chiu, C.T.Shii & T.Y.A.Yang, Novon 21: 410 (2011). The epithet honours the Kavalan aboriginal people of Ilan county. The two populations Dr. Chiu identified are on the river side of Bei-bu Road, Cross-Island Highway No. 7. at 212 m and 276 m elevation. He thought he may have found a third on the other side of the road, but as they were not flowering he couldn’t be sure. He had not searched the area for more.
I put it to him that as they are directly by the roadside and only in this area that they could be planted by humans, and that I would be willing to go with him on a field trip to thoroughly search the location to possibly find more and establish what the total area of the species is. We arranged the field trip for 11th July 2010 after my return from Japan.
We set off at dawn from Taichung to reach Ilan before midday. The first population was easily seen beside the road. A large population of what were very large plants, much taller than any M. itinerans var. formosana I had ever seen. We were later to discover that the size of these plants was due to the rich soil beside the riverbed, and less nourished plants do not grow so large.
Other than the size of the plants they looked very similar to M. itinerans var. formosana. Certainly in leaf and stem detail, but also in the clumping habit.
The obvious differences are the male bud being a pale greenish/yellow colour, the same basal colour as the M. itinerans var. formosana male bud but without any of the M. itinerans var. formosana reddish streaks and markings. The fruits also have no reddish/purple colouration leaving them a very clean looking pale bluish/green colour. The pseudostem shows very minimal or zero red blotches or markings unlike M. itinerans var. formosana.
The size of the male bud and overall bunch size is variable even within these larger plants on this fertile riverbed. The male bud size can vary from small, 8 cm long to over 25 cm long.
From the location of this first clump by the roadside I could see that there were no other bananas within sight. The second clump was a further 2 km up the road. The clump was of a similar size to the first clump, and again there were no other bananas around that were visible with the naked eye. I used my field glasses to scour the hillsides. On the other side of the riverbed across the valley, about 0.5 km away. I could see more bananas, so we decided to trek across the mainly dry riverbed to get a closer look. These bananas proved to be M. itinerans var. formosana. The fruits had red blotches and streaks and there were visible red blotches on the pseudostem. Now, from the other side of the valley it was very clear that the yellow-flowered bananas were isolated.
We then decided it was time we started a thorough search of the hillsides to see if there were more populations. The hillsides were very steep and inaccessible near the two clumps so we chose a spot a little further up the road where we could get access to the hillside. Less than 20 m up the hill from the road we found our first new specimen with a yellow bud. A few metres further on from there we found the next and the next. Within minutes and climbing only about 100 m we were finding several small clumps.
At about 300 m elevation we stumbled upon an isolated cemetery grave in the forest. Planted beside it was a cultivated banana with fruits. Directly behind the grave were many yellow flowered bananas. There was no evidence of any hybridisation taking place.
At 400 m elevation we were finding many huge clumps of the yellow-flowered banana, and we were now loosing count of what we’d seen.
At 500 m elevation we came across a large clump, and looking through it it looked more like M. itinerans var. formosana with many dark red blotched on the pseudostem. We had to search through all the stems until we found flowers and fruit. Sure enough this was Musa itinerans var. formosana and not the yellow-flowered banana, Musa itinerans var. kavalanensis.
A few more metres up the hill and we found a large clump of yellow-flowered banana standing alongside a large clump of M. itinerans var. formosana. At no time did we find any evidence of intermediate hybridisation between the two forms, and conclude that the two forms do not hybridise.
At 750 m elevation the forest opened up to reveal a much more level landscape, well much less steep, which was being cultivated. We stopped climbing at 800 m, however, bananas were still visible further up the hillside.
Further information and complete description can be found on http://www.trebrown.com/plant_info.php?species=Musa+itinerans+var.+kavalanensis Use the interactive distribution map at the bottom of that page. The map may not show the complete distribution of the species, only to the extent that we searched. We do know the bounds to the east and the bounds to the south (the riverbed).