[Oberon] Compiler behaviour for a32 := a8

[Hans Klaver]

Thanks to Chris and August it's becoming clearer to me now. 
The documents on Prof. Wirth's website are not always in sync with the reality of the compiler.

August wrote:

> As mentioned before in this mailing list, I asked Niklaus Wirth in 2017 if structural equivalence of arrays and procedures is to be seen as a language extension and he answered:
> 
> "Yes, I consider this type compatibility as an addition to the language definition. I think it makes sense, and it is easy to implement. Please look at the source of the compiler under Project Oberon."

I'm not really at home in the compiler source files.
Could you tell me where exactly in the source text one can read about this structural equivalence of arrays?

Regards,

Hans