How does i.cluster work?

[Harini Nagendra]

Hello everybody,
I had posted this question to grassu about 6 months ago, but didn't get 
any replies. Does anyone know the algorithm used by i.cluster in Grass? I 
need to know 
1] the name/a ref (i.e. is it a hirarchical divisive or agglomerative etc)
2] When you run i.cluster, there is a "distance" input which the user has 
to specify, and in the Users Manual it recommends that the distance be 
between 0 and 1.5. What units are these, and what do they mean?
Please please help, 'cos I have a paper I am writing up and I need to 
know this info for that, urgent!
Thanks a lot,
Harini
------------------------------------------------------------------------
| Harini Nagendra               E-MAIL : harini at ces.iisc.ernet.in      |
| Centre for Ecological         PHONE  : 91 80 309 2639 (Hostel: LR-22)|
| Sciences                             : 91 80 309 2506 (Department)   |
| Indian Institute of Science   FAX    : 91 80 334 1683                |
| Bangalore 560 012 India       TELEX  : 845 8349 IISc IN              |
------------------------------------------------------------------------