Here is the partial differential equation for which the solutions Y are are the wave functions for the rigid-rotator (Equation 5.5 in McQuarrie and Simon's Physical Chemistry).
$$\sin\theta\frac{\partial}{\partial \theta}\left(\sin\theta\frac{\partial Y}{\partial \theta}\right) + \frac{\partial^2 Y}{\partial \phi^2} + \left(\beta \sin^2\theta\right)Y = 0$$
In McQuarrie and Simon's Physical Chemistry, it is said that
When we solve Equation 5.5, it turns out naturally that $\beta$ ... must obey the condition: $$\beta = J(J+1)$$
My question is why must $\boldsymbol{\beta}$ take the form $\boldsymbol{J(J+1)}$ (where $\boldsymbol{J}$ is a natural number))? Since $\beta$ is essentially a constant, this implies that using any value of $\beta$ (e.g., 1.5) would lead to a partial differential equation with no solutions. (1) is this assumption correct and (2) if so, what is the mathematical proof that $\beta$ in any other form would lead to no solutions?