View Single Post
Old 07-17-2007, 02:25 PM   #6 (permalink)

Gabe15's Avatar
Location: Oahu, Hawaii
Zone: 12
Name: Gabe
Join Date: Jul 2005
Posts: 3,757
BananaBucks : 13,246,577
Feedback: 5 / 100%
Said "Thanks" 1 Times
Was Thanked 7,654 Times in 2,050 Posts
Said "Welcome to Bananas" 8 Times
Default Re: Banana Flower Photo Database

To elaborate on my eariler thoughts, you really need to know a lot more information than just a single close up of the female flowers. That would be like trying to identify a person simply by thier eyes, it may help sometimes but will surely not work everytime. Another problem I forsee is inexperience with floral parts, some people may think the flowers match with a quick side-by-side comparison but further investigation may show small yet important differences.

In my opinion, identifying edible bananas requires lots of detail (because many of them are so similar and of the same origins) and I think it would be unwise to take the approach of indentifying the plant straight to the cultivar because if you mess up on the one and only descriptor, you will get the wrong ID. What I mean by this is that although some bananas can be ID'd right off the bat, when looking at all the varieties we grow, we need to broaden the scope and start with the big picture. There are some very easy descriptors that can be seen, such as erect or pendent bunches, from there you can start to narrow down different distinction between groups. Once you figure out what group it is (whether it be a genome of edible bananas or a subgenera of wild species), you can start to then narrow it down to the specific plant.

For example, say I had a Dwarf Cavendish that I was trying to ID by myself with our system, by looking at the female flowers alone I would get matches for about 5 different varieties if I managed to get the ID right the first time, although I now may have an idea of the type of plant I have, I still wouldn't know its name. But if we take a more detailed approach such as first trying to determine its ploidy level, this is a fairly simple task if you know what to look for and you would find that your plant (the Dwarf Cavendish) is a triploid. From there we have a few options, it could be AAA, AAB or ABB. Again, by looking for the specific descriptors, you could then figure out that the plant you have is a AAA, now you have it narrowed down to the genome. You then look for the descriptors at subgroup level, and you would match your plant to the descriptors for the Cavendish subgroup and you would know that it is a Cavendish of some sorts. Now that you know it is a Cavendish, you would see that your plant is about 5-6ft tall and you could safely assume you have a Dwarf Cavendish.

If this all seems very tedious and confusing, it is. But this is the same problem banana scientists have been working on for over a hundred years, how to classify and distinguish edible bananas (and for us, we would also need to incorporate wild species and hybrid seeded varieties into the mix because we are hobby growers). I'm not trying to bring down anyones ideas or anything, Im just trying to inform you what has already been done, and what methods have proven to work.
Growing bananas in Colorado, Hawaii and Washington since 2004.
Gabe15 is offline   Reply With Quote Send A Private Message To Gabe15