Computing every derivative directly from the limit definition ("from first principle") is correct but impractical for anything beyond the simplest functions. The five rules below, proved once from the definition, make differentiation a mechanical, algebraic process for any combination of known derivatives — no further limit computation is needed once they are established. Throughout, u=u(x), v=v(x) are differentiable functions of x, and k is a constant.
Sum Rule (Theorem 10.2). dxd(u+v)=dxdu+dxdv — differentiate term by term. Extends to any finite sum.
Product Rule (Theorem 10.3). dxd(uv)=udxdv+vdxdu, i.e. (uv)′=uv′+vu′ — "first times derivative of second, plus second times derivative of first." Extends to three or more factors, e.g. (uvw)′=u′vw+uv′w+uvw′: differentiate one factor at a time, summing over which factor is differentiated.
Quotient Rule (Theorem 10.4). For v=0, dxd(vu)=v2vu′−uv′ — "bottom times derivative of top, minus top times derivative of bottom, all over bottom squared."
Chain Rule (Theorem 10.5). If y=f(u) and u=g(x) are both differentiable, so y=f(g(x)), then
dxdy=dudy⋅dxdu=f′(g(x))g′(x)
— differentiate the outer function f with respect to its argument u=g(x) (the inner function), then multiply by the derivative of the inner function. This is the single most-used rule in the chapter: every composite function (a power of an expression, sin of an expression, eexpression, and so on) needs it.
Constant Multiple Rule (Theorem 10.6). dxd[kf(x)]=kf′(x) — a constant factor simply carries through differentiation unchanged. …