We have analyzed isolated and storm time ion injections using geosynchronous particles, energetic neutral atom (ENA) data, and Dst. There are both surprising similarities between the two classes of events as well as important differences that bear directly on the relationship between storms and substorms. The average geosynchronous ion responses during the growth phase, at onset, and in the ≈15 min following onset are nearly identical in intensity, spectral hardness, and temporal profile. ENA observations confirm that similarity and additionally show that the two classes of injections span nearly the same extent in local time. The two classes of injections differ primarily in the subsequent behavior of the ion fluxes. For the isolated injections the fluxes return to preevent levels within about an hour, and exhibit the expected westward drift and dispersion. For the storm time injections the fluxes remain elevated for at least several hours following the initial injection. Additionally, the ENA observations show new evidence that the region of new particle injections expands eastward (opposite to the ion drift direction) to encompass most of the nightside. Within 3 hours, ENA emissions are observed coming from most of the inner magnetosphere but have still not formed a symmetric, trapped distribution. Within those same 3 hours Dst decreased an average of 40 nT with the initial decrease observed in the same hour as the initial injection. The isolated injections did not produce a measurable Dst signature. These results show that despite many remarkable similarities, storm time ion injection events are different from isolated injection events. ¿ 2001 American Geophysical Union