As a result of the re-analysis of the Phase 1 data with the improved model of the diffuse gamma radiation, a number of the sources which appeared in the first EGRET catalog ([Fichtel et al. 1994b]) do not appear in this one (and conversely, some in the present catalog would have appeared in the first catalog using the improved model). The sources from the first EGRET catalog which were affected enough by the re-analysis to be dropped from the second catalog fall into two categories:
1. Those sources whose statistical significance dropped from just above the catalog threshold to just below it. These sources are scattered throughout the sky; and
2. Those sources which lie in the region which was explicitly noted by [Fichtel et al. 1994b] as having significant uncertainties due to the diffuse model. In this region, the improved diffuse model provides a much better fit to the total radiation, so that fewer excesses (sources) are found.
The ``lost'' sources are listed in Table 5, showing the maximum 
found for this source in the current analysis. These include 5 of the AGN listed in the first catalog. In all 5 cases, these sources are still seen, but with significance just below the threshold for the current catalog.