The origin of lithium (Li) and its production process have long been an unsettled question in cosmology and astrophysics. Candidates environments of Li production events or sites suggested by previous studies include big bang nucleosynthesis, interactions of energetic cosmic rays with interstellar matter, evolved low mass stars, novae, and supernova explosions.Chemical evolution models and observed stellar Li abundances suggest that at least half of the present Li abundance may have been produced in red giants, asymptotic giant branch (AGB) stars, and novae [1][2][3] . However, no direct evidence for the supply of Li from stellar ob- High-resolution spectra (R = 90, 000-60, 000) of V339 Del were obtained at four epochs after its outburst (+38, +47, +48, and +52 d). These spectra contain a series of broad emission lines originating from neutral hydrogen (H I, Balmer series) and other permitted transitions of neutral or singly ionized species (e.g., Fe II, He I, Ca II). These emission lines are usually seen in post-outburst spectra of classical novae. Most of these broad emission lines are accompanied by sharp and blue-shifted multiple absorption lines at their blue edges. The typical radial velocity (v rad ) of these highly blue-shifted absorption lines is ∼ −1, 000 km s −1 . Figure 1- There are no Na I D doublet lines, which are often found to be the strongest absorption features in novae within a few weeks after their outbursts 8,10 . We interpret this as indicating that the ionization state of the ejected gas has evolved into a higher stage of excitation before our observing epochs (5-7 weeks). The observed spectral energy distribution of this nova indicates that the shape of the continuous radiation had entered a very hot stage (effective temperature >100,000 K) within 5 weeks after the explosion 11 . Other observed characteristics of this nova (e.g., light curves, optical and UV emission lines) show that it is a typical Fe II nova with a CO white dwarf (WD) 12,13 .Among these absorption line systems, we have noticed two remarkable pairs of absorption features near 312 nm. These correspond to the absorption components originating from transitions 3 at ∼313 nm. These pairs are marked as A, B and C, D, respectively, in Figure 1- The transition probability of the 7 Be II line at 313.0583 nm (log gf = −0.178) is twice as large as that of the 7 Be II at 313.1228 nm (log gf = −0.479) 14 . Due to saturation effects, the ratio of their equivalent widths is expected to be in the range between 2 (no saturation) to 1 (complete saturation). The measured ratios are 1.1 ± 0.3 and 1.6 ± 0.4 for the components at v rad = −1, 268and −1, 103 km s −1 , respectively. These are within the range expected for the doublet, although the values contain some errors ( ∼ < ±25%) due mainly to the uncertainty in the continuum placement.The weaker component at v rad = −1, 268 km s −1 has a ratio closer to complete saturation. This can 4 be interpreted as resulting from the fact that the absorbing gas cloud moving with v rad = −1, 268 km s...