NOTES FOR TERNARY OLIVINE SUBROUTINES:	

	There are 2 FORTRAN files for use in calculating activities and 
chemical potentials of ternary olivine solutions here:
	NI_OLIVINE_CALC.FOR
	CA_OLIVINE_CALC.FOR.
They are meant to permit calculations for (Ni,Mg,Fe)2SiO4 and (Ca,Mg,Fe)2SiO4
olivines using the solution models of Hirschmann,M. 1991 Am. Min. 76:1232-1249.

	These subroutines were written for VAX FORTRAN, but should compile 
easily on most modern FORTRAN 77 compilers.  If you are running an 
IBM-compatible microcomputer, you may need to shorten some file names and 
subroutine names to conform to operating system limits 

NOTES:
	Both files consist of a main subroutine plus additional subroutines 
that are called by the main subroutine.  The additional subroutines are needed 
to calculate the equilibrium ordering state of the olivine at a given P,T and X 
and do a few other things (calculate standard state free energies, etc.).  
	In order to use these subroutines, you will need to write short 
programs that calculate the activities, chemical potentials, etc. for the
temperatures, pressures, and compositions that suite your needs.  Please
note that the subroutines read compositions in terms of the variables
p,q,r, as in the paper:
			p=2X Ca2SiO4
			q=2X Ni2SiO4-1
			r=2X Fe2SiO4-1

	BE VERY CAREFUL TO NOTE THE FORM OF ACTIVITY RETURNED.  In some
cases it is on a 1 atom basis and in some cases it is on a 2 atom basis.
This is clearly indicated in the documentation at the beginning of each
routine.

	These routines are rather primitive; when they are unable to calculate 
a given chemical potential, they return a value of zero.  For example, the
chemical potential of CaMgSiO4 in pure forsterite is returned as zero,
instead of its actual value, negative infinity.

	There are 2 errors in the Appendix of the paper.  In equation A2,
the term (1+r-2q)/2 should be (-1+r-2q)/2.  In equation A3, the term (r-q)/2
should be (r-q)/4.

If you have problems with or questions about these subroutines, contact me
via Email at 
		schmelz@max.u.washington.edu.

Marc Hirschmann