- models first posted on August 15, 2000
 
 - wrong threshold energies for photoionizations that end on a level
	different from the ground level of the target ion fixed on
	February 12, 2003 (Thanks to Philip L. Dufton).
	Affected ions: Al I, C I, C II, F II, N II, O I, S I, Si I, and Si II.
	
	
 - model ion for Si III posted on November 11, 2004
 
 - model ion for Si II modified	to include improved collisional
   rates calculated by Dufton, P.L., Kingston, A.E. (MNRAS vol. 248, Feb. 15, 
  	1991, p. 827, 828) posted on November 11, 2004 (Thanks to P. Dufton
  	and R. Ryans).

 - models modified including a change in the way to code the hydrogenic 
   cross-section for levels with no photoionization information in topbase 
   and corrected a mistake in the sigma_0 for Seaton's (coll. ioniz.)
   formula (It is not the photoion. x-section at the threshold (s0)
   as claimed in the Tlusty manual, but s0*gbar, where gbar=0.1 for
   neutral species and gbar=0.3 for ions.  May, 2006
   
 - The photoionization cross-sections provided in separate files are 
   reformatted  to follow TOPBASE format. May, 2006
   
 - change of format for the energy levels: besides the
   usual information (ENION, G, NQUANT, TYPLEV, IFWOP, FRODF, IMODL)
   we also print: min(ENION) for superlevels, lowest val. included 
		  max(ENION) for superlevels, highest energy incl.
		  min(2*S+1), max(2*S+1) range of vals. for superl.
		  min(L), max(L) range of vals. for superl.
		  min(P), max(P) range of vals. for superl.
		  
  As all the t-files only include individual states (not superlevels)
  ENION~min(ENION)~max(ENION) (where the differences account for the 
  energy range of the fine structure splitting)
  min(2*S+1)=max(2*S+1)
  min(L)=max(L), and
  min(P)=max(P). 
  The change of format only adds extra columns, but the first 7 columns
  are unchanged, so they are backwards compatible with previous Tlusty/Synspec
  versions which will ignore the extra information. May 2006
  
 - electron collisional excitation rates for OI updated for the levels in 
   Zatsarinny & Tayal (2003, ApJS, 148, 575).  Courtesy of I. Ramirez. 
   May 2006
   
 - Yet another change for the hydrogenic collisional ionization rates. 
    (see correction above on May 2006). gbar=0.1 for neutral species,
    0.2 for sinly-ionized species, and 0.3 for higher ionization species.
    December 2006
    
 - A modification of the values used with ICOL for permitted bound-bound
    transitions. We still use the same approximations (Van Regermorter's)
    described in Allende Prieto, Lambert, Hubeny & Lanz 2003 (ApJS, 147, 363)
    but they are no longer coded through negative values of ICOL, sending
    the control to a specific form of the routine CSPEC
    (http://hebe.as.utexas.edu/at/cspec.f), but now we use the standard
    ICOL=1 and a newly implemented (in Tlusty) ICOL=5:
    for neutral species: ICOL=-1 -> ICOL=5
    for ions:		 ICOL=-2 -> ICOL=2
    December, 2006

 - Updated atomic linelist based on data from Kurucz's web site. The
	format is now expanded to include quantum numbers, which is     
	quite useful to handle NLTE tlusty model atmospheres with
	complex model ions avoiding confusion among levels. A new
	version of synspec is able to handle the extra columns, but
	the new format is compatible with older versions, as the first
	80 columns remain unchanged.
    November 2007

 - Updated molecular linelist for synspec created by merging the files from
	Kurucz's web site for different isotopologues, using terrestrial
	proportions for the stable isotopes involved
    March 2008

 - Slight modifications in the atomic linelist, changing the damping
  param. for the HeI line at 402.6210 nm to avoid a crash in the iq
  package (due to a negative opacity). Thanks to Lars Koesterke for
  locating this mysterious black spot. The update list is named 
  gfall_bpox2.19
	February 2009

- Chagned perl cgi scripts to static html
	June 2015
 
   
	C. Allende Prieto