pcplantdb

[Richard Morris <webmaster@pfaf.org>]

Dear all,

In the developing permaculture info project where we are trying to 
create a knowledge base for permaculture information
see the mailing list at
http://lists.ibiblio.org/mailman/listinfo/permacultureinfo
for more info.

As part of the development of that we have been working on the structure 
of a plant database and how to represent plants and their 
characteristics systematically.

I've been reading a book on XML topic maps which describes a way of 
constructing diagrams of related concepts (a topic map is very like a
touchgraph graph). In one of the examples they devise a hierarchy of 
plants where "an oak tree is a type of tree". At first sight this seems 
a fairly uncontroversial statement.

But think for a moment how this matches with the botanical plant
hierarchy. Trees are liberally dispersed over a wide range of families
from Tree Ferns to Conifers and Ginkgos to Monocots (Yucca) and 
liberally dispersed among the dicots. However trees have a patchy 
coverage with some families having a wide range of trees and non-trees.

The pea family (Leguminosae) is a particularly interesting example which
has 69 annuals (eg peas), 133 perennials (clover), 96 shrubs (gorse) and 
39 trees (acacia) mentioned in the Plants For A Future database.
How come a family can have woody plants like gorse and acacia
and other like peas and clover which are very far from treeness?

So it seems that treeness does not really tie well with the typical
plant systematics. Which leeds to a question "what makes a tree a tree?"
Is there a tree gene? How come some plants switch it on and other plants
don't?

Treeness is a very different sort of thing to "has a backbone"
a feature which is is inherited though a large part of the animal
kingdom, there was some root animal which had a backbone and all its
descendants: fishes, amphibians, reptiles, birds, mammals all share this
property. Flowering plants also fit well with the botanical systematics, 
all flowering plants (grasses, lilies, orchids, brasicas, roses, oaks
willows, the pea family, ...) belong to the grouping Angiosperm a well 
define section of the plant kingdom.

For trees there may have been some very ancient plant which was a tree
(probably a tree fern). But it's descendants are not all trees, some are
some are not. (Its possible that all its descendants may share some deeper
 characteristic such as a particular cell structure).

Perhaps treeness is more akin to having an eye. This is a property which
has developed independently several times through history, vertebra have
them, insects have them, squid have them, but none of these share a
common root which had an eye, they all developed independently (I 
believe its developed twelve different times).

Or is it more like a true habit like swims in water? Fish do, dolphins
do it, water snakes do it. All inhabit the same ecological niche
but have arrived at different ways to fill the niche.

I don't think the independent evolution idea fits well for trees though. 
If you compare the structure of an oak tree
and a pine tree they are very similar in terms of cell structure.
There does seem to be a stronger link between the two than just
the fact that they inhabit a similar ecological niche. Woody growth
is a strong shared characteristic of all trees.

So you could take the radical proposition that all plants (from tree
ferns onwards) are really trees, just that they have not decided to
become woody and branch at the base. Or more scientifically their might 
be a "tree gene" shared by a good part of the plant kingdom, but certain 
plants decide to switch this gene on and others don't.

So in conclusion the best answer might be that the concept of a tree is
a really complicated thing it means
        perennial
        woody
        branches higher up
        inhabits a particular ecological niche (keep leaves high up).
In other words a tree is not an atomic thing, unlike flowering plant
And should not form the base of any serious hierarchy of plant. The 
concept of a tree should be treated with caution.

        ttfn

                Rich

Object oriented programmers note: A tree is a poor candidate for a base 
class for trees. Perhaps better though of as an interface which certain 
plants can implement. Or maybe don't use the concept tree at all, stick 
to the constituent parts, such a woody, perennial, not branching at base 
 and fits a particular ecological niche.

--
Plants for a Future: 7000 useful plants
Web:   http://www.pfaf.org/ same as http://www.comp.leeds.ac.uk/pfaf/
Post:  1 Lerryn View, Lerryn, Lostwithiel, Cornwall, PL22 0QJ
Tel:     01208 872 963 / 0845 458 4719  
Email: webmaster@pfaf.org
PFAF electronic mailing list http://groups.yahoo.com/group/pfaf