Using 3D infrared (IR) exchange spectroscopy, the ultrafast hydrogenbond forming and breaking (i.e., complexation) kinetics of phenol to benzene in a benzene/CCl 4 mixture is investigated. By introducing a third time point at which the hydrogen-bonding state of phenol is measured (in comparison with 2D IR exchange spectroscopy), the spectroscopic method can serve as a critical test of whether the spectroscopic coordinate used to observe the exchange process is a memory-free, or Markovian, coordinate. For the system under investigation, the 3D IR results suggest that this is not the case. This conclusion is reconfirmed by accompanying molecular dynamics simulations, which furthermore reveal that the non-Markovian kinetics is caused by the heterogeneous structure of the mixed solvent.3D IR spectroscopy | solvent dynamics | ultrafast dynamics W henever one writes a chemical reaction equation, such as in the simplest case of an equilibrium between two states of a molecular systemwith the corresponding Master equation for its kineticsone implicitly assumes Markovianity. That is, it is assumed that the probability to react per time unit, e.g., from C to F, scales linearly with concentration [C] with proportionality constant k CF , but does not depend on the history of how C has been formed. When the reaction is slow in comparison with the relaxation of its environment, then this assumption is well justified. One can however think of many situations where this is not the case. For example, proteins are heterogeneous on very long timescales due to their structural complexity, so enzymatic reaction catalyzed by them might not fulfill that condition (1). Other examples are found in glass-forming liquids close to the glass transition, when their relaxation covers many orders of magnitude in time (2). When the speed of a reaction approaches the ultrafast picosecond regime, which is the topic of the present paper, then even the solvation time of normal liquids far away from any glass transition may become time-limiting. For example, it has been shown that the distance between a Na + and a Cl − ion is not a good reaction coordinate to describe ion pair dissociation in aqueous solution (3), in the sense that the distance for which the free energy peaks does not correspond with the transition state of the reaction. Hidden coordinates exist, which must be solvent coordinates because there are no other degrees of freedom. Unless these solvent coordinates have not fully relaxed and lost their memory about a reaction event, the probability for a back reaction must depend on the history. Markovianity is related to a separation of timescales between the speed of a chemical reaction of a molecular system versus that of the relaxation of its solvent. Often, a separation of timescales goes hand-in-hand with a separation of solute and solvent, but such a system-bath approach fails for ultrafast reactions.It is important to recognize that the question of Markovianity is not an intrinsic property of a chemical process, but rather is relate...