This is a synthetic saltdome velocity model issued by EAEG/SEG velocity modeling committee. The size of the model is grids and the geometric size of each cube is .First, we smooth the model using the same smoothing filter we used in the last model. Figure 5 shows the portion where the raypaths hit the flank of the saltdome. Since the saltdome velocity is much higher than the other parts. Most of the rays are ``bounced'' back by the flank except those rays hitting the flank with small incident angle. We also show a small portion of the corresponding wavefront. The color of the wavefront represents the magnitude of the amplitude. The amplitude is homogeneous before the wavefront hits the saltdome. When the wavefront hits the flank of saltdome, the amplitude changes quickly because of the wave resistance. Figure 6 is generated using AVS's clipping function. We use one model slice and one clipping plane (transparent) to extricate a ring of wavefront. The model slice shows the saltdome's structure clearly. We can see several triplications on the ring. This phenomenon is caused by the complicated velocity model. Therefore it is obvious that first arrival approach cannot image such a complicated structure. We need to do some interpolation to get the maximum energy traveltime.