Normal Submagma? Is there a definition of normal submagma? visit https://en.wikipedia.org/wiki/Magma_(algebra)
For normal sub-quasi-group I found two:


*

*A sub-quasi-group $H$ is called normal if there exists a normal congruence $\theta$ such that $H$ coincides with one of the congruence classes.

*A sub-quasi-group $H$ of a quasi-group $(Q,.)$ is called normal if and only if $x.H=H.x$, $(x.y).H=x.(y.H)$ and $H.(x.y)=(H.x).y$, for all $x,y\in Q$.


If this is true and the two definitions are equivalent, then how to prove that the first definition gives the second one?
 A: There is a definition of normal subset (of a semigroup) at page 24 of A Survey of Binary Systems by R.H. Bruck:

If $\theta$ is a homomorphism of a semigroup $S$ into a groupoid, the
  image $S\theta=S'$ is also a semigroup. If $s'$ is an element of $S'$,
  let $K=s'\theta^{-1}$ be the inverse image of $s'$; that is,
  $K$ is the set of all $s$ in $S$ such that $s\theta=s'$. Clearly $K$
  has the following properties:
  
  
*
  
*if $k\in K$, $x\in S$ and $xk\in K$, then $xK\subset K$;
  
*if $k\in K$, $x\in S$ and $kx\in K$, then $Kx\subset K$;
  
*if $k\in K$, $x,y\in S$ and $xky\in K$, then $xKy\subset K$;
A non-empty subset $K$ of $S$ with properties 1., 2., 3. is called a normal subset.

This is followed by a reflection about why the study of normal subsets (subgroupoids) is less interesting than normal subsemigroups (or subgroups):

To a given normal subset $K$ of $S$ there may correspond two (or more) distinct homomorphisms of $S$ with $K$ as an inverse image; herein resides, perhaps, the relative poverty of the theory.

