We have studied ∼0.3 to >100 MeV nucleon −1 H, He, O, and Fe in 17 large western hemisphere solar energetic particle events (SEP) to examine whether the often observed decrease of Fe/O during the rise phase is due to mixing of separate SEP particle populations, or is an interplanetary transport effect. Our earlier study showed that the decrease in Fe/O nearly disappeared if Fe and O were compared at energies where the two species interplanetary diffusion coefficient were equal, and therefore their kinetic energy nucleon −1 was different by typically a factor ∼2 ("energy scaling"). Using an interplanetary transport model that includes effects of focusing, convection, adiabatic deceleration, and pitch angle scattering we have fit the particle spectral forms and intensity profiles over a broad range of conditions where the 1 AU intensities were reasonably well connected to the source and not obviously dominated by local shock effects. The transport parameters we derive are similar to earlier studies. Our model follows individual particles with a Monte Carlo calculation, making it possible to determine many properties and effects of the transport. We find that the energy scaling feature is preserved, and that the model is reasonably successful at fitting the magnitude and duration of the Fe/O ratio decrease. This along with successfully fitting the observed decrease of the O/He ratio leads us to conclude that this feature is best understood as a transport effect. Although the effects of transport, in particular adiabatic deceleration, are very significant below a few MeV nucleon −1 , the spectral break observed in these events at 1 AU is only somewhat modified by transport, and so the commonly observed spectral breaks must be present at injection. For scattering mean free paths of the order of 0.1 AU adiabatic deceleration is so large below ∼200 keV nucleon −1 that ions starting with such energies at injection are cooled sufficiently as to be unobservable at 1 AU. Because of the complicating factors of different spectral break energies for different elements, it appears that SEP abundances determined below the break are least susceptible to systematic distortions.