********> bugfix.41
Author:   Dave Case
Date:     9/18/95

Programs: Sander

Severity: Minor

Problem:  NMR chemical shift options: ring current contributions to 
          chemical shifts may be in error by a small amount: a 5% error 
          is the largest we have seen; most are considerably smaller than 
          this.

Cause:    Programming typo.


Fix:  Make the following change to cshf.f (the 2nd 'd1' becomes 'd2'):
--------------------------------------------------------------------------
*** OLD cshf.f
--- NEW cshf.f
***************
*** 473,479
              d1 = sqrt(r1sq)
              r2sq = r2(1)*r2(1) + r2(2)*r2(2) + r2(3)*r2(3)
              d2 = sqrt(r2sq)
!             temp = 0.5*(1./(r1sq*d1) + 1./(r2sq*d1))
              shift = s12*temp
  c
              do 250 n=1,9

--- 473,479 -----
              d1 = sqrt(r1sq)
              r2sq = r2(1)*r2(1) + r2(2)*r2(2) + r2(3)*r2(3)
              d2 = sqrt(r2sq)
!             temp = 0.5*(1./(r1sq*d1) + 1./(r2sq*d2))
              shift = s12*temp
  c
              do 250 n=1,9
--------------------------------------------------------------------------

Temporary workarounds: none

--