Surface G2 matching produces different result to CATIA method

Part of a grabhandle. The U-shaped surface, when matched G2 with MatchSrf on both sides, pulls off the purple projection curve and the black tangency enforcement strip at the bottom. In CATIA, this does not happen (grey locked surface). What would be the correct Rhino method to match G2 without the jumping off effect?

Thanks!

Surface matching issue.3dm (114.0 KB)

Hello- MatchSrf the top and bottom edges to the curves for Position - use ‘Match by closest points’. Does that do it?
But, it is on the pile to match surfaces without upsetting the startiong control point locations so much - can you post the surface as it was before matching?
-Pascal

Thanks, I used MatchSrf G0 along the top and G1 along the bottom. G2 on the left and right was not destroyed. When measured (Rhino 7), it’s not too bad. Some CPs need to be UVN massaged, because visual inspection with Zebra shows internal flat-ish areas. UVM moving of CPs, however, means one could destroy one or all of the matchings, and one has to MatchSrf again.