ARE LINEAR MOLECULES REALLY LINEAR?    II. RE-INTERPRETATION OF EXPERIMENTAL B0-VALUES.

As discussed in the preceding talk, any linear triatomic molecule will be observed as being ``bent'' on ro-vibronic average in any ro-vibronic state.\footnote{T.~Hirano, U.~Nagashima, {\it J. Mol. Spectrosc.}, {\bf 314}, 35--47 (2015)}$^{,}$% \footnote{T.~Hirano, U.~Nagashima, P.~Jensen, {\it J. Mol. Spectrosc.} {\bf 343}, 54--61 (2018).}$^{,}$\footnote{T.~Hirano, U.~Nagashima, P.~Jensen, {\it J. Mol. Spectrosc.} (2018), https://doi.org/10.1016/j.jms.2017.12.011; and references therein.} Experimentally derived $B_{0}$ constants are the results of the ``observation'' of Nature. This suggests that the observed $B_{0}$ values are in fact those for the ro-vibrationally averaged bent structures. The easiest way to check this proposition is to interpret the set of $B_{0}$ values of isotopologues taking the bond-angle as a ``variable,'' discarding the preconceived, conventional notion that the ro-vibrationally averaged bond angle of a linear molecule is 180$^{\circ}$. We have shown in previous publications$^a$ that bond length values derived from a set of experimental $B_{0}$ values under the assumption of a linear $r_0$ structure, is not the ro-vibrationally averaged bond lengths, but their projections onto the molecular axis. Therefore, when the projection angle is not accounted for, the bond length values obtained from the $B_0$ values may differ significantly from the averaged bond lengths. We will show how we can derive physically sound ro-vibrational structures from the experimentally reported $B_{0}$ values, taking the FeCO, NCS, HCO$^{+}$, HCN, and C$_{3}$ molecules as examples. The averaged bond-angle deviations from the linearity, derived from experimentally reported $B_{0}$ values of multiple isotopologues, are 7.8$^\circ$, 9.5$^\circ$, 12.5$^\circ$, 14.3$^\circ$, and 23.4$^\circ$, respectively, for NCS, FeCO, HCO$^{+}$, HCN, and C$_{3}$ in their respective vibrational ground states. Thus, we can conclude that both theoretically (as described in the preceding talk) and experimentally (as shown here), the ro-vibrationally averaged structure of a linear molecule is observed as being bent