Using the Goddard High-Resolution Spectrograph on the Hubble Space Tel
escope, we measured the flux of [Fe IV] (3d(5 4)P(5/2) --> 3d(5 ?6S?(5
/2)) lambda(rac) = 2836.56 Angstrom in the Orion Nebula, the first det
ection of an [Fe IV] line in an H II region. A useful upper limit is s
et on the sum of fluxes of [Fe IV] (3d(5) D-4(5/2,3/2) --> 3d(5) S-6(5
/2)) lambda(rac) = 2568.4, 2568.2 Angstrom. By comparing these observa
tions with predicted fluxes from simply ''retrofitting'' our two previ
ous photoionization models, we are able to derive (or set an upper lim
it on) the Fe/H abundance ratio: 70, 200 times lower than solar from t
he 2837 Angstrom line, and greater than or equal to 38, greater than o
r equal to 120 times lower than solar from the 2568 Angstrom line limi
t. If collisional excitation from the ground state were indeed the dom
inant mechanism for populating the respective upper levels of these li
nes, then the inferred Fe/H from the 2837 Angstrom line and Limit from
the 2568 Angstrom line would be similar to 3.0 and similar to 3.4 tim
es larger than above. All these ratios are much lower than several rec
ent determinations of gas-phase Fe/H similar to 3 X 10(-6) in Orion, w
hich themselves are a factor similar to 10 depleted relative to solar.
Because the inferred Fe/H should be at least as high in the Fe+3 zone
as in the Fe+ and Fe+2 zones, a reexamination of the Fe+3 atomic data
and improved modeling would be valuable.