The Four Maxwell's equation that are given by $$\nabla . \mathbf{E}=\frac{\rho}{\epsilon_0}$$ $$\nabla.\mathbf{B}=0$$ $$\nabla\times \mathbf{E}+\frac{\partial \mathbf{B}}{\partial t}=0$$ $$\nabla \times \mathbf{B}-\frac{1}{c^2}\frac{\partial \mathbf{E}}{\partial t}=\mu_0 \mathbf{J}$$ For an vector $\mathbf{V}$ , you can define a matrix $$\partial _i \mathbf{V}_j = \left( \begin{array}[ccc] \partial \mathbf{V_1} /\partial x_1 & \partial \mathbf{V_1} /\partial x_2 &\partial \mathbf{V_1} /\partial x_3 \\ \partial \mathbf{V_2} /\partial x_1 & \partial \mathbf{V_2} /\partial x_2 & \partial \mathbf{V_2} /\partial x_3 \\ \partial \mathbf{V_3} /\partial x_1 & \partial \mathbf{V_3} /\partial x_2 & \partial \mathbf{V_3} /\partial x_3 \\ \end{array} \right)$$
In Maxwell equation He take some special combination from this matrix , Why it is so? How do you know that How do you know that only these special combination are needed? Is there any deep meaning ?