I have read a book written by C.V Durell on Geometry. In this book I have found a lemma which states that there is one and only one circle that passes through three vertices of a triangle.
I thought of proving by contradiction, but I just don't know how to go about.
Relax the definition of parallel so that a line may be said to be parallel to itself.
Prove that if lines $p_1$ and $p_2$ are parallel, and line $q_1$ is perpendicular to $p_1$ while line $q_2$ is perpendicular to $p_2,$ then $q_1$ is parallel to $q_2.$
Equivalent to 2.:If $q_1$ is not parallel to $q_2,$ and $p_1$ is perpendicular to $q_1,$ and $p_2$ is perpendicular to $q_2,$ then $p_1$ is not parallel to $p_2.$
Existence: Apply 3. to triangle ABC: The line thru $A,B$ is not parallel to the line thru $B,C$, so the right bisector $R_{A B}$ of $AB$ and the right bisector $R_{B C}$ of $BC$ are not parallel, so $R_{A B}$ and $R_{B C}$ intersect at some point $D.$ Since any point on $R_{A B}$ is equidistant from $A ,B,$ we have $AD=BD.$ Since any point on $R_{B C}$ is equidistant from $B, C$ we have $B D=C D.$ Hence $A D=B D= C D$ so $D$ is the center of a circle passing thru $A,B,C.$
Uniqueness: Any point $E$ which is equidistant from $A,B$ lies on $R_{A,B}$ because either (a) $E$ is the mid-point of $A B$, or (b) the altitude-line from $E$ to $AB$ in the isosceles triangle $AEB$ is $R_{A B}.$ Similarly, any point equidistant from $B,C$ lies on $R_{B C}.$ So if $E$ is equidistant from $A,B,C$ it must lie in the intersection of $R_{A B}$ with $R_{B C}.$ Since $R_{AB}$ and $R_{BC}$ are not parallel, they are not equal, so their only point of intersection is $D.$ So $E=D.$