IUBio GIL .. BIOSCI/Bionet News .. Biosequences .. Software .. FTP

Plant Identification

Mary Barkworth stipoid at CC.USU.EDU
Sat Dec 9 11:10:05 EST 1995


With respect to using relational databases for plant identification:

I am not aware of anyone using relational databases in this regard, but Mike 
Dallwitz in Australia has developed a very powerful program that, among 
other things, allows for online plant identification.  The program is DELTA 
and for information about it, consult 
http://muse.bio.cornell.edu/

One has to go a few subdirectories down, but that will get you started and I 
do not have the rest of the address.  The primary problem that I see iwth 
using DELTA to teach plant identification is that one wants a information 
tailored to one's own locality (state or region).  The chances are that it 
is not yet available.  There are DELTA databases for identifying grasses of 
thw orld or plant families of the world, but because they have to use 
exactly the same descriptive language for plants from throughout the world, 
the language and terminology used is more than my students would be able to 
use.  {To be honest, some of it would require me to go back to a botanical 
dictionary if I were seriously interested in using these programs for plant 
identification; I tend to use almost any other resource first).  

On the other hand, there is a way in which I plan to use DELTA in my class 
this spring:  I shall ask students to use it to develop a key and 
descriptions to a small group of plants, possibly 10 taxa.  I think that it 
will make the students more appreciative of the work involved in developing 
a key that encompasses the variability, the amount of work that needs to go 
one before one starts writing a key (e.g., one has to know what one has), 
and the difficulty in defining characters and character states in a way that 
can be used across even a moderately diverse group of organisms.  

Last year Harvey Ballard (a grad student at U. Wisc.) and I developed a 
quick and dirty key to the violets of Utah - which Harvey convinced me have 
been messed up by too much lumping.  My students thought it was the the 
cat's whiskers - and asked when the rest of the Utah Flora would be 
available in the same manner.  I laughed.  Harvey has studied violets for 
years, even before starting graduate school.  I had some familiarity with 
DELTA, although not the Windows version which allows one to incorporate 
pictures (which were a definite plus so far as my students were concerned).  
Mike Dallwitz is fantastic about responding to questions.  Even so, it 
probably took about 3 days to develop the key.  It needs some correction now 
and Harvey is busy doing the molecular thing, writing his dissertation, 
etc., so I am not sure when our effort will be "publishable" (i.e., made 
available through the MUSE site).  I do use it as a demo here on occasion.  
You would need DELTA to use it, so check out the MUSE site to find out more 
about DELTA.

As an aside, I am - in my spare time - pursuing a slightly different aspect 
of plant id, providing everyone with a means of checking an identification.  
Being a grass person, I have started with grasses.  One day (in the not too 
distant future), I want to incorporate a self-instructiona unit (that could 
also be incoroporated into a formal class) on grass identification.  For 
now, all I have is a few sample files on the WWW - and in a very low tech 
format.
Check http://www.biology.usu.edu/herbarium/vgh/grasses/stip/stipeae.html
Then feel free to look at some of the files in //..../herbarium/ which may 
give you a some idea of what I have in mind.  I suspect that by the time I 
can make this an official project, there will be far better ways of 
accomplishing my goals.  Indeed, about a week ago someone made a WWW 
compatible interface for DELTA.  Life races on.

Also, the Flora North America folks have come out with a CD ROM version of 
the second volume of that flora, i.e., the one treating ferns and 
gymnopserms.  I have not seen it advertised yet, so it may still be only in 
a test phase - and I have a feeling it will not be cheap.  

Another effort along these lines, one I was really impressed by, is being 
developed at MacDonald College in Quebec, Canada.  This is an English 
language institution in a French language province.  What the folks there 
are producing is what I really want to see:  a CD that students can use to 
help them learn.  The feature that particularly impressed me was the review 
mode.  A picture of a flower or plant would be shown on the screen and the 
user was asked to state what family it was in.  If the wrong answer is 
entered, the program not only says wrong (without ringing loud bells) but 
says what features IN THAT PICTURE would let you know that it would not be 
the family entered.  According to Marsha Waterway, writing these responses 
is the most demanding aspect of the program because one has to think of all 
the different reasons users would select a given answer.  The aspect I like 
about the program is that it concentrates on helping students learn rather 
than being something one uses, does, and puts away.  It was stimulated by 
the need to provide French speaking students with an opportunity to spend 
longer with the information being presented in lecture and to do so in an 
active manner, not just watching a video of the lecture again but it really 
does provide what we would like all students to have available.  I have no 
idea of the status of the program.  Discussions about copyright, 
publication, etc. were in progress this summer (Marsha had a demo version at 
the Plant Systematics meetings this summer).

Sorry about the length of this answer.  Hope some of it is of interest.





More information about the Plant-ed mailing list

Send comments to us at archive@iubioarchive.bio.net