High resolution infrared spectroscopy is the key technique to look at the inner regions of protoplanetary disks. As molecular hydrogen is an inefficient emitter, CO gas is the single most important molecular probe of the disk. The energy gaps of the vibrationally excited levels (ΔE > 3000 K) and the critical density required to keep the molecules in the excited state (nc ~ 1010cm−3) match well to the physical condition of the inner regions of protoplanetary disks. In order to resolve the vibrational lines of different rotational states, a spectral resolving power of λ/Δλ > 10000 is necessary; or even higher (> 30000 –100000), if we would like to fully resolve the gas kinematics. Scoville et al. (1980) provided the fundamentals of the excitation mechanisms, which is essential for the interpretation of the vibrational transitions of CO, and pioneered the study of the circumstellar environment with infrared CO lines in the observation of BN (Scoville et al. (1983)). The bandhead emission of CO at 2.3 μm from young stars was unambiguously attributed to the circumstellar disks by Carr (1989) and Najita et al. (1996), because the gas kinematics matches well to what is expected from Keplerian rotation. Since then, the gas kinematics have been extensively used to shed light on peculiar disk structures, such as the inner truncation (Brittain et al. 2003), the outer truncation (Najita et al. 2008), and the gap (van der Plas et al. 2008; though this is an oxygen forbidden line).