The LCS is a long slit spectograph with a slit length of 2.5 arcmin. A slit width of 2 arcsec was used for the comet observations. Preliminary reduction proceeds in a typical manner: First, correct for bias and flat field. Next, fit a dispersion curve to the argon lamp observations taken intermixed with the comet data. The argon spectra are fit in two dimensions - both spatial and spectral. Standard stars and solar analog stars were observed to account for the response of the instrument and to replicate the solar continuum respectively. These spectra were extracted and summed using variance weighting. Finally, the cometary spectra were calibrated with the response function determined with standard stars. Once we had cometary spectral images with flux versus wavelength, the analysis could begin. First, a sky spectrum, obtained with the same instrument and generally on the same night, was subtracted from the cometary spectrum. This sky spectrum was weighted to remove the 5577A night sky line. Next, the color of the solar analogue was corrected to match the shape of the cometary continuum and removed. In theory, this should leave no continuum and one would just need to integrate the band. In practice, the color weighting is imperfect because of the limited number of non-emission regions in the cometary spectrum which could be matched. Thus, we fit a line between two continuum regions and integrated the band above that. For the band integrations the following bandpasses (in Angstroms) were used: Molecule Band Blue_Continuum Red_Continuum Bounds Bounds Bounds ____________________________________________________________ OH 3070 3108 3020 3060 3112 3128 NH 3325 3385 3200 3315 3400 3500 CN 3830 3905 3715 3770 4150 4175 C3 3975 4150 3715 3770 4150 4175 CH 4280 4340 4150 4175 4400 4460 C2DEL1 4460 4770 4400 4460 4770 4830 C2DEL0 4860 5185 4770 4830 5220 5270 Once we had the integrated intensities, they could be converted to column densities with the application of simple physics. This includes oscillator strengths and transition probabilities and may be subject to revision with revised lab values. For C3, CH, and C2 delta v=0 and 1, the column density can be computed from log N = const + log L + 2 log R - log Omega where N is the column density (in mol/cm**2), L = 4 pi Delta F, R is the heliocentric distance in AU and Omega is aperture size in steradians. For L, Delta is the geocentric distance in AU and F is the integrated band flux in ergs/(cm**2 sec). The constants used were: C3 12.42 CH 12.98 C2 delta v=1 12.620 C2 delta v=0 12.347 For CN, OH and NH the g factor was taken from various Swings calculations at the appropriate heliocentric radial velocity (given in the observing log). For OH, the g factors are from Dave Schleicher (personal communication) and are the unquenched (0,0) band factors at 1au. The values are tabulated at every 1 km/sec from -60 km/sec. The table of these values is: 6.50 ,6.39 ,6.29 ,6.19 ,6.06 ,5.88 ,5.65 ,5.36 ,5.00, 4.59 ,4.15 ,3.73 ,3.39 ,3.10 ,2.85 ,2.67 ,2.64 ,2.81, 3.09 ,3.37 ,3.53 ,3.56 ,3.48 ,3.37 ,3.34 ,3.42 ,3.51, 3.50 ,3.38 ,3.22 ,3.11 ,3.01 ,2.90 ,2.77 ,2.59 ,2.39, 2.17 ,1.98 ,1.85 ,1.74 ,1.64 ,1.52 ,1.44 ,1.46 ,1.65, 2.01 ,2.43 ,2.77 ,2.93 ,2.84 ,2.47 ,2.00 ,1.70 ,1.64, 1.70 ,1.74 ,1.73 ,1.70 ,1.64 ,1.54 ,1.49 ,1.56 ,1.76, 2.05 ,2.33 ,2.56 ,2.68 ,2.71 ,2.72 ,2.84 ,3.18 ,3.74, 4.47 ,5.19 ,5.74 ,5.93 ,5.70 ,5.24 ,4.91 ,4.92 ,5.26, 5.78 ,6.34 ,6.85 ,7.16 ,7.25 ,7.18 ,7.04 ,6.84 ,6.48, 5.89 ,5.22 ,4.67 ,4.24 ,3.80 ,3.33 ,2.91 ,2.59 ,2.30, 2.03 ,1.82 ,1.79 ,2.10 ,2.84 ,3.93 ,5.14 ,6.16 ,6.84, 7.21 ,7.43 ,7.64 ,7.90 ,8.13 ,8.26 ,8.22 ,7.99 ,7.62, 7.10 ,6.47 ,5.87 ,5.45 These values must be multiplied by 1e-15. They are in energy units. For the constant needed in the equation above const = log (4 pi/g) where g is interpolated from the table. The g factors for NH come from the Swings effect calculation of Kim, A'Hearn and Cochran (Icarus 77, 98-108, 1989). They tabulated the values every 2.5 km/sec starting at -70 km/sec. The table of values used is: 7.40e-14,6.47e-14,5.79e-14,6.58e-14,7.21e-14,7.26e-14, 7.43e-14,7.59e-14,7.81e-14,7.99e-14,6.15e-14,4.11e-14, 4.32e-14,5.70e-14,7.23e-14,7.56e-14,7.58e-14,7.36e-14, 7.02e-14,6.44e-14,6.41e-14,6.80e-14,6.71e-14,6.82e-14, 6.84e-14,7.31e-14,7.09e-14,6.28e-14,6.00e-14,7.08e-14, 8.13e-14,8.60e-14,8.19e-14,7.47e-14,7.19e-14,7.00e-14, 6.86e-14,6.80e-14,6.02e-14,6.45e-14,7.50e-14,7.96e-14, 8.10e-14,7.81e-14,7.27e-14,6.44e-14,6.20e-14,6.40e-14, 6.92e-14,7.35e-14,6.82e-14,4.90e-14,5.58e-14,6.79e-14, 6.77e-14,5.99e-14,5.04e-14 The values are in energy units and again, the const = log (4 pi/g). There are several Swings calculations for CN. We use one of the oldest, which is that of Tatum and Gillespie (Astrophysical Journal 218, 569-572, 1977). They computed L/N, which differs slightly from a g factor. The table of values used is 3.70,3.72,3.65,3.50,3.45,3.56,3.70,3.76, 3.90,4.06,4.00,3.79,3.45,3.14,3.05,3.10, 3.08,3.11,3.31,3.36,3.24,3.02,3.00,3.33, 3.68,3.87,3.97,4.02,4.07,4.06,4.90,3.48, 3.38,3.39,3.54,3.67,3.64,3.46,3.24,3.40, 3.89,3.94,3.90,3.83,3.74,3.55,3.36,3.09, 2.70,2.36,2.29,2.43,2.98,3.63,4.27,4.54, 4.70,4.75,4.67,4.72,4.79,4.74,4.43,4.01, 3.89,3.95,3.93,3.70,3.50,3.48,3.45,3.45, 3.44,3.54,3.66,3.67,3.54,3.35,3.09,2.98, 3.17,3.51,3.80,3.86,3.86,3.90,3.96,3.95, 3.80,3.72,3.82,3.90,3.96,3.97,3.98,3.93, 3.84,3.79,3.65,3.41,3.30 These values must be multiplied by 10**{-28}. If we denote these values as f, then, in this case, const = log (f).