1: \begin{definition}

2: For $T \in {\mathcal T}_n$, let $v(T) = (d_{1,2}^T, d_{1,3}^T, \ldots, d_{n-1,n}^T) \in {\mathbb R}^{n \choose 2}$ be the vector of pairwise distances $d_{i,j}^T$ between leaves $i$ and $j$ in $T$.  The {\em dissimilarity map distance} is

3: $$d(T_1,T_2) = ||v(T_1) -  v({T_2})|| = \sqrt{(d_{1,2}^{T_1} - d_{1,2}^{T_2})^2 + \ldots + (d_{n-1,n}^{T_1} - d_{n-1,n}^{T_2})^2}$$

4: where $|| \cdot ||$ represents $L_2$ norm (Euclidean length).  

5: %$||v(T_1) -  v({T_2})||$ is the Euclidean length of the vector $v(T_1) -  v({T_2})$, and it is calculated by the root-square of sum of squares in each coordinate of the vector based on the Pythagorean Theorem.

6: \end{definition}

7: