Laser induced fluorescence from a dye contained in Unocal RF-A gasoline was excited using 355nm light and the resulting fluorescence imaged (λ>420nm). In order to minimize the changes to the intake geometry the fluorescence was collected by a fiberoptic probe with an articulatible tip. The collected light was imaged onto an intensified CCD camera synchronized with the laser, which was timed to illuminate the intake port after the completion of injection. Cold-starts from 20°C were conducted on an engine dynamometer test stand with two fuel systems: pintle-type port fuel injection, and air-forced port fuel injection. When the injection timing and initial enrichment were optimized the transient emissions from the air-forced system were significantly reduced compared with the conventional system. These two systems enable the comparison of the wall wetting effects from a coarse spray typical of those injectors in current use with that from the extremely fine spray produced by an air-forced injector.In S.I. engines with port injection systems, deposition of liquid fuel films in the intake port is expected to have significant effects on UHC emissions particularly during cold-start . On the FTP cycle most of the UHC emission occurs during the first 5 minutes prior to catalyst ‘light off’ e.g. . Due to the importance of this process, there have been many studies of injection in transparent manifolds, e.g. [3, 4, 5, 6, 7, 8 and 9], using stroboscopic illumination and filming, PDA, and sampling of the manifold contents. Some studies have been carried out using endoscopic observation of the spray in the manifold of a practical engine  but using such systems to observe the behavior of the film after the end of the spray is difficult because of the problem of distinguishing between the film and flare of the incident light from the manifold itself. Laser induced fluorescence allows the signal from the fuel to be separated from the scattering of the incident light due to the shift in wavelength and therefore should be a suitable diagnostic for visualization of liquid films in the manifold. Various fluorescent dopants have been added to fuels for two-dimensional imaging of in-cylinder fuel distribution using laser sheet illumination [11, 12, 13 and 14]. Johnen and Haug have also used this technique for imaging the sprays in a transparent intake manifold and used fiber optic probes to make point measurements of film thickness at the walls using L.I.F. . The dopants used in those studies are not appropriate when the distribution of liquid fuel over a surface is to be visualized as the low incident light intensity results in an unacceptably low signal; dopants with a much higher fluorescence yield are required.In a related study, Fulcher et al.  have studied the effect of different types of fuel preparation on the UHC emissions during cold-start. Using the same engine we investigated the formation and subsequent behavior of fuel films in the intake during cold-start. The method used was to image the liquid fuel in the manifold using L.I.F. from a dye in the fuel via a fiberoptic probe inserted through the manifold wall.