In chapter 1 problem 7 of the 3rd edition of Goldstein the reader is asked to prove that the Nielsen's form of the Lagrange's equations is equivalent to Lagrange's equations.

Lagrange's equations:

$$\frac{d}{dt}\frac{\partial T}{\partial \dot{q}_j} + \frac{\partial T}{\partial q_j} = Q_j.$$

Nielsen's form:

$$\frac{\partial \dot{T}}{\partial \dot{q}_j} - 2 \frac{\partial T}{\partial q_j} = Q_j.$$

What application does this form of the equations have? It seems like Nielsen's form form would be just as difficult to calculate as Lagrange's equations. Is there a context in which Nielsen's form form is easier to use?


1 Answer 1


No, its a form of naive algebra applied by mathematicians of the Euler age; sometimes right, sometimes obscure.

Generally, the kinetic energy is a quadratic local bilinear form on the tangent space, expressed by the vector coordinates, the velocities

$$T(\dot q, q) = \frac{1}{2} \sum g_{ik}(q) \dot q^i \dot q^k$$


$$\frac{d}{dt} T = \sum g_{ik}(q) \dot q^i \ddot q^k + \frac{1}{2}\sum_j\partial_j g(q) \dot q^i \dot q^k \dot q^j $$

Applying the rules of diffential algebra, a partial derivative with respect to $dot q_j$ seems to assume $\ddot q_i$ to be a constant.

So this differential rule is more of a trick to determine the coefficient in the equation, where $\ddot q$ is merely seen as another independent variable.

By the central identity of Euler-Langrange equations, the second derivatives will be quadratic forms over the velocity; so this attempt is more confusing than practicable, since the notations seems to prescribe a directional derivative on the sphere of constant velocity squared.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.