Let $\mathbb F$ denote a field and let $V$ denote a vector space over $\mathbb F$ with finite positive dimension. We consider a pair of linear transformations $A:V\to V$ and $A^*:V\to V$ that satisfies the following conditions: (i) each of $A,A^*$ is diagonalizable; (ii) there exists an ordering $\lbrace V_i \rbrace_{i=0}^d$ of the eigenspaces of $A$ such that $A^* V_i \subseteq V_{i-1}+V_i+V_{i+1}$ for $0 \leq i \leq d$, where $V_{-1} = 0$ and $V_{d+1} = 0$; (iii) there exists an ordering $\lbrace V^*_i \rbrace_{i=0}^��$ of the eigenspaces of $A^*$ such that $AV^*_i \subseteq V^*_{i-1}+V^*_i+V^*_{i+1}$ for $0 \leq i \leq ��$, where $V^*_{-1} = 0$ and $V^*_{��+1} = 0$; (iv) there is no subspace $W$ of $V$ such that $AW\subseteq W$, $A^*W\subseteq W$, $W \neq 0, W \neq V$. We call such a pair a {\it tridiagonal pair} on $V$. It is known that $d = ��$ and that for $0 \leq i \leq d$ the dimensions of $V_i, V_{d-i}, V^*_i, V^*_{d-i}$ coincide; we denote this common value by $��_i$. The sequence $\lbrace ��_i\rbrace_{i=0}^d$ is called the {\it shape} of the pair. In this paper we assume the shape is $(1,2,1)$ and obtain the following results. We describe six bases for $V$; one diagonalizes $A$, another diagonalizes $A^*$, and the other four underlie the split decompositions for $A,A^*$. We give the action of $A$ and $A^*$ on each basis. For each ordered pair of bases among the six, we give the transition matrix. At the end we classify the tridiagonal pairs of shape $(1,2,1)$ in terms of a sequence of scalars called the parameter array.