The flexural strength of ice is not a basic material property, but rather is an estimate of the maximum stress in the outermost fiber of an ice specimen when it fails in bending. Such conditions correspond to a number of important engineering applications, such as interactions between ice and a sloping structure or between ice and ships. Ice flexural strength is therefore highly important for calculating ice pressures and forces of interest for engineering design. While there has been considerable discussion in the literature regarding scale effects related to ice crushing against a vertical structure, scale effects in relation to bending failure have received much less attention. To this end, more flexural strength data for large, full-thickness sea ice beams are needed. To address these data gaps, a field data collection program was carried out in Pistolet Bay, Newfoundland over two field seasons (2017–2018). During this program, large sea ice beams were tested in-situ using a custom four-point bending apparatus, which was comprised of several main subsystems (e.g., the ram loading system, the platen, the ubrackets, and the hydraulic system). The sea ice beams were completely cut free from the ice cover and loaded at four points, such that the center load is parallel, but opposed to, the loads at the ends of the beam. All tests were done in-situ so that no brine drainage took place and the temperature gradient remained consistent. Tests were carried out for several combinations of beam geometry, which were scaled relative to the ice thickness. In addition to flexural strength, during the Pistolet Bay field program, the physical properties of the ice were measured (temperature, salinity, density). In this paper, a description of the field apparatus, test program and results from the full-thickness in-situ four-point beam bending tests are presented, along with a discussion of practical implications and future work.