Theory of the Fireball

of about 200 from 4000 to 600o0, about 10 from 6000 to &looo except for
the lowest density, and 10 to 50 from 8000 to 12,000°. In most cases,
the dependence on wave length is slight until the rapid increase of
absorption in the ultraviolet which sets in at about 2.5 ev for kOOOO,
3 ev at 6000 and 8000°, and not at aU at 12,000'. Tne especially strong
0
increase for 4000 is due to the Schumam-Runge bands, which are not very
sensitive to temperature*" see Sec. 4c) .'
2
In Sec. 5d we shall need the mean free path in @/cm , suitably
averaged over the "transparent" region. From Table V it appears that a
reasonable estimate of this region is from hv = 1/2 to 2-3/4 ev. We
have averaged p/p as calculated from Table V, with the weighting factor
3 -u k X booo
ue Y u= hv (4.9)
which does not vary much (from 0.84 to 1.33) between 1-1/8 and 2-5/8 ev.
The result is given . i n Table VI.
*After completion of this paper, I received new absorption coefficients
by F . R . Gilmore , cf ref 23. Aside from including the absorption by
NO2 ( see Sec. ha), Gilmore includes the free-f ree absorption in the
field of neutrals, ba ed on ref . 18, and also new data on f nmbers for
the important bands.2f; The most important change is a reduction in the
f nmber of the IT2 first positive system from 0.02 to 0.0028, which w ill
substantially reduce the absorption in the visible . Unfortunately, this
will further raise the theoretical radiating temperature ( Sec . 5d) ,
which ,is already higher than observed.
'Some recent Avco-Everett experiments may indicate that the free-free
electron transitions in the field of the I? atom qe enhanced in the
visible as compared to the infrared. This may compensate to some extent
for the reduction of the N2 first positive bands (see footnote above)
39