CCL: ONIOM input using GaussView for Gaussian03

 Sent to CCL by: "Georgios  Dimitrakakis" [giwrgis|*|chemistry.uoc.gr]
 Hello to all the members of this list.
 I am having a problem that has to do with an ONIOM implementation via GaussView.
 The problem is that I implemented ONIOM through GaussView with two different
 ways and the results were not the same.
 I wanted to optimize a structure using DFT->6-311++G** for the High Level and
 UFF for the Low Level.
 GaussView has its own way to introduce UFF by selecting Mechanics -> UFF. I
 did that (the one calculation) and I noticed that the Z-Mat had a strange
 structure. You can see a part here :
 C-C_R            0    0.000000    0.000000    0.000000 H
 C-C_R            0    0.000000    0.000000    1.457200 H
 C-C_R            0    1.378600    0.000000    1.790600 H
 ........................................................
 .............................(etc)......................
 So I tried this calculation and I tried another one by putting Custom->UFF.
 In this second calculation the input line was the same as previously but instead
 this time the Z-Mat was looking like this part:
 C                0    0.000000    0.000000    0.000000 H
 C                0    0.000000    0.000000    1.457200 H
 C                0    1.378600    0.000000    1.790600 H
 ........................................................
 .........................(etc)..........................
 The two calculations came out with different results (for the Energy) both at
 the ONIOM and at the High Level. The difference was for the ONIOM  on the second
 decimal digit while for the High on the third.
 Could anyone tell me why is there such a difference in the results??
 And another question that may sound silly. Which of the above ways is the
 correct?? I mean that if I was going to implement ONIOM by hand without the use
 of GaussView I was never going to create I Z-Mat like the first one.
 Any ideas?
 Best regards,
 Georgios Dimitrakakis
 giwrgis(at)chemistry.uoc.gr