We present new experimental results from a radiation portal monitor based on the use of organic liquid scintillators. The system was tested as part of a 3He-free radiation portal monitor testing campaign at the European Commission’s Joint Research Centre in Ispra, Italy, in February 2014. The radiation portal monitor was subjected to a wide range of test conditions described in ANSI N42.35, including a variety of gamma-ray sources and a 20,000n/s 252Cf source. A false alarm test tested whether radiation portal monitors ever alarmed in the presence of only natural background. The University of Michigan Detection for Nuclear Nonproliferation Group’s system triggered zero false alarms in 2739 trials. It consistently alarmed on a variety of gamma-ray sources travelling at 1.2m/s at a 70cm source to detector distance. The neutron source was detected at speeds up to 3m/s and in configurations with up to 8cm of high density polyethylene shielding. The success of on-the-fly radionuclide identification varied with the gamma-ray source measured as well as with which of two radionuclide identification methods was used. Both methods used a least squares comparison between the measured pulse height distributions to library spectra to pick the best match. The methods varied in how the pulse height distributions were modified prior to the least squares comparison. Correct identification rates were as high as 100% for highly enriched uranium, but as low as 50% for 241Am. Both radionuclide identification algorithms produced mixed results, but the concept of using liquid scintillation detectors for gamma-ray and neutron alarming in radiation portal monitor was validated.