It has been suggested that there may be a fundamental difference between substorms that occur during magnetic storms and those that occur at other times. Baumjohann et al. <1996> presented evidence that there is no obvious change in lobe field in quiet time substorms but that storm time substorms exhibit the classic pattern of storage and release of lobe field energy. This result led them to speculate that the former are caused by current sheet disruption, while the latter are caused by reconnection of lobe flux. In this paper we examine their hypothesis with a much larger data set using definitions of the two types of substorms similar to theirs, as well as additional more restrictive definitions of these classes of events. Our results show that the only differences between the various classes are the absolute value of the lobe field and the size of the changes. When the data are normalized to unit field amplitude, we find that the percent change during storm time and non-storm time substorms is nearly the same. The above conclusions are demonstrated with superposed epoch analysis of lobe field (Bt and Bz) for four classes of substorms: active times (Dst -25 nT), and quiet time substorms (no evidence of storm in Dst). Epoch zero for the analysis was taken as the main substorm onset (Pi2 onset closest to sharp break in AL index). Our results suggest that there is no qualitative distinction between the various classes of substorms, and so they are all likely to be caused by the same mechanism.