We focus on two Hubble Space Telescope/Space Telescope Imaging Spectrograph (HST/STIS) spectra of the Weigelt Blobs B&D, extending from 1640 to 10400Å; one recorded during the 1998 minimum (March 1998) and the other recorded in February 1999, early in the following broad maximum. The spatially-resolved spectra suggest two distinct ionization regions. One structure is the permanently low ionization cores of the Weigelt Blobs, B&D, located several hundred AU from the ionizing source. Their spectra are dominated by emission from H I, [N II], Fe II, [Fe II], Ni II, [Ni II], Cr II and Ti II. The second region, relatively diffuse in character and located between the ionizing source and the Weigelt Blobs, is more highly ionized with emission from [Fe III]Through photoionization modeling, we find that the radiation field from the more massive B-star companion supports the low ionization structure throughout the 5.54 year period.The radiation field of an evolved O-star is required to produce the higher ionization emission seen across the broad maximum. This emission region is identified with slow-moving condensations photoionized by the O star and located in the extended mass flow emanating from the B star primary. Comparison between the models and observations reveals that the high ionization region is physically distinct (n H ≈ 10 7 cm -3 and T e ~ 10 4 K) from the BD Blobs (n H ≈

We use the narrow-lined broad-line region ( BLR) of the Seyfert 1 galaxy, I Zw 1, as a laboratory for modeling the ultraviolet (UV ) Fe ii 2100Y3050 8 emission complex. We calculate a grid of Fe ii emission spectra representative of BLR clouds and compare them with the observed I Zw 1 spectrum. Our predicted spectrum for log n H /(cm À3 ) ½ ¼ 11:0, log È H /(cm À2 s À1 ) ½ ¼ 20:5, and /(1 km s À1 ) ¼ 20, using Cloudy and an 830 level model atom for Fe ii with energies up to 14.06 eV, gives a better fit to the UV Fe ii emission than models with fewer levels. Our analysis indicates (1) the observed UV Fe ii emission must be corrected for an underlying Fe ii pseudocontinuum; (2) Fe ii emission peaks can be misidentified as that of other ions in active galactic nuclei (AGNs) with narrow-lined BLRs possibly affecting deduced physical parameters; (3) the shape of 4200Y4700 8 Fe ii emission in I Zw 1 and other AGNs is a relative indicator of narrow-line region ( NLR) and BLR Fe ii emission; (4) predicted ratios of Ly , C iii], and Fe ii emission relative to Mg ii k2800 agree with extinction corrected observed I Zw 1 fluxes, except for C iv k1549; (5) the sensitivity of Fe ii emission strength to microturbulence casts doubt on existing relative Fe/ Mg abundances derived from Fe ii (UV)/Mg ii flux ratios. Our calculated Fe ii emission spectra, suitable for BLRs in AGNs, are available at