The B800-850 LH2 antenna from the photosynthetic purple sulfur bacterium Allochromatium vinosum exhibits an unusual spectral splitting of the B800 absorption band; i.e., two bands are well-resolved at 5 K with maxima at 805 nm (B800(R)) and 792 nm (B800(B)). To provide more insight into the nature of the B800 bacteriochlorophyll (BChl) a molecules, high resolution hole-burning (HB) spectroscopy is employed. Both white light illumination and selective laser excitations into B800(R) or B800(B) lead to B800(R) -> B800(B) phototransformation. Selective excitation into B800(B) leads to uncorrelated excitation energy transfer (EET) to B800(R) and subsequent B800(R) -> B800(B) phototransformation. The B800(B) -> B800(R) EET time is 0.9 +/- 0.1 ps. Excitation at 808.4 nm (into the low-energy side of B800(R)) shows that the lower limit of B800(R) -> B850 EET is about 2 ps, as the B800(R) -> B800(B) phototransformation process could contribute to the corresponding zero-phonon hole width. The phototransformation of B800(R) leads to a similar to 200 cm(-1) average blue-shift of transition energies, i.e., B800(R) changes into B800(B). We argue that it is unlikely that B800-B850 excitonic interactions give rise to a splitting of the B800 band. We propose that the latter is caused by different protein conformations that can lead to both strong or weak hydrogen bond(s) between B800 pigments and the protein scaffolding. Temperature-dependent absorption spectra of B800, which revealed a well-defined isosbestic point, support a two site model, likely with strongly and weakly hydrogen-bonded B800 BChls. Thus, BChls contributing to B800(R) and B800(B) could differ in the position of the proton in the BChl carbonyl-protein hydrogen bond, i.e., proton dynamics along the hydrogen bond may well be the major mechanism of this phototransformation. However, the effective tunneling mass is likely larger than the proton mass.