ConspectusDicarbon, the molecule formed from two carbon atoms, is among the most abundant molecules in the universe. Said by some to exhibit a quadruple bond, it is bound by more than 6 eV and supports a large number of valence electronic states. It thus has a rich spectroscopy, with 19 one-photon band systems, four of which were discovered by the author and co-workers. Its spectrum was among the first to be described: Wollaston reported the emission spectra from blue flames in 1802.C2 is observed in a variety of astronomical objects, including stars, circumstellar shells, nebulae, comets and the interstellar medium. It is responsible for the green color of cometary comae but is not observed in the comet tail. It can be observed in absorption and emission by optical spectroscopy in the infrared, visible, and ultraviolet regions of the spectrum, and because it has no electric-dipole-allowed vibrational or rotational transitions, its spectral signature is a sensitive probe of the local environment.Before the work described in this Account, models of C2 photophysics included the thitherto-unobserved c3Σu+ state and parametrized the strength of spin-forbidden intercombination transitions. Furthermore, they did not account for photodissociation of C2, even though it was identified in the 1930s as a key process. Inspired by the observation of C2 in the Red Rectangle nebula, the author was motivated to instill rigor into C2 models and embarked on a spectroscopic and computational journey that has lasted 15 years.We were the first to identify the c3Σu+ state through the d3Πg-c3Σu+ transitions, which were to become known as the "Duck" system. This minor partner to the well-known Swan bands is a key part of astrophysical C2 models and can now be included with rigor. We identified the e3Πg-c3Σu+ system, and the c3Σu+ state is now well-studied. Meanwhile others described the singlet-triplet and triplet-quintet interactions in exquisite detail, allowing rigorous modeling of the a-X and c-X intercombination transitions.The final piece of the C2 puzzle would be understanding how long it survives before being broken into carbon atom fragments. Though predicted by Herzberg, predissociation in the e3Πg state had never been observed. To find it would require the complicated ultraviolet spectroscopy of C2 to be disentangled. In so doing, we identified the 43Πg and 33Πg states of C2, thus uncovering two new band systems. The 43Πg state allowed the first accurate determination of the ionization energy of C2. With these new band systems secure, we extracted new levels of the D1Σu+ state (Mulliken bands) and the e3Πg state (Fox-Herzberg bands) from our spectra. Upon climbing the energy ladder in the e3Πg state to v = 12, we finally identified the route to predissociation of C2 via non-adiabatic coupling to the d3Πg state. This observation provided the first laboratory evidence for why C2 is observed in the coma of a comet but not the tail.