Orientation and Potential Differences

Why Orientation Matters

A curve describes a route, but a line integral also records the direction in which that route is traversed. The same geometric path can therefore have different integrals when followed in opposite directions. For a conservative vector field, the change is especially systematic: reversing the direction changes the sign of the integral.

This sign change is closely tied to how a potential is evaluated. The integral from a starting point $A$ to a finishing point $B$ uses the potential at $B$ minus the potential at $A$. Traversing the curve backward switches which point is the start and which is the finish. Keeping that order explicit prevents a frequent sign error.

Curves, Orientation, and Line Integrals

Let $\mathbf{r}:[a,b]\to\mathbb{R}^n$ be a continuously differentiable parameterization of a curve $C$, and let $\mathbf{F}$ be a continuous vector field defined along the curve. The line integral of $\mathbf{F}$ along $C$, in the direction given by increasing parameter $t$, is

\int_C \mathbf{F}\cdot d\mathbf{r} = \int_a^b \mathbf{F}(\mathbf{r}(t))\cdot\mathbf{r}'(t)\,dt.

The derivative $\mathbf{r}'(t)$ gives the direction of motion along the curve. Changing that direction changes the tangent vector in the integrand, which is why orientation affects the integral.

Definition: The reversed curve, denoted $-C$, is the same geometric path traversed from the original endpoint back to the original starting point. For the parameterization above, one parameterization of $-C$ is $\widetilde{\mathbf{r}}(t)=\mathbf{r}(a+b-t)$, for $a\le t\le b$.

Indeed, $\widetilde{\mathbf{r}}(a)=\mathbf{r}(b)$, while $\widetilde{\mathbf{r}}(b)=\mathbf{r}(a)$. The reversed parameterization changes the order of the endpoints; it does not change the geometric path.

The Sign Change Under Reversal

The reversal rule holds for any continuous vector field along a continuously differentiable curve. It does not require the field to be conservative.

Theorem: If $C$ is parameterized by $\mathbf{r}:[a,b]\to\mathbb{R}^n$ and $-C$ is the same curve with the opposite orientation, then $$ \int_{-C}\mathbf{F}\cdot d\mathbf{r} = -\int_C\mathbf{F}\cdot d\mathbf{r}. $$

To see why, write $g(u)=\mathbf{F}(\mathbf{r}(u))\cdot\mathbf{r}'(u)$. The derivative of the reversed parameterization is $\widetilde{\mathbf{r}}'(t)=-\mathbf{r}'(a+b-t)$. Substitute the reversed parameterization into the definition of the line integral, then use $u=a+b-t$. The limits change from $a,b$ to $b,a$.

\begin{aligned} \int_{-C}\mathbf{F}\cdot d\mathbf{r} &=\int_a^b \mathbf{F}(\mathbf{r}(a+b-t))\cdot[-\mathbf{r}'(a+b-t)]\,dt\\ &=\int_b^a g(u)\,du\\ &=-\int_a^b g(u)\,du\\ &=-\int_C\mathbf{F}\cdot d\mathbf{r}. \end{aligned}

The negative sign in the last equality comes from reversing the limits of integration: $\int_b^a g(u)\,du=-\int_a^b g(u)\,du$. This distinction matters. The substitution gives $\int_b^a g(u)\,du$, not $-\int_b^a g(u)\,du$.

Using a Potential: Preserve the Endpoint Order

A vector field $\mathbf{F}$ is conservative on a domain if it is the gradient of a scalar potential $\phi$ there: $\mathbf{F}=\nabla\phi$. Suppose the curve lies in a domain where $\phi$ is differentiable and the potential relation holds. If $C$ starts at $A$ and ends at $B$, the chain rule gives

\frac{d}{dt}\phi(\mathbf{r}(t)) = \nabla\phi(\mathbf{r}(t))\cdot\mathbf{r}'(t) = \mathbf{F}(\mathbf{r}(t))\cdot\mathbf{r}'(t).

Integrating this identity from $a$ to $b$ and applying the one-variable Fundamental Theorem of Calculus yields the endpoint formula. The proof also makes clear why endpoint order matters.

\int_C\mathbf{F}\cdot d\mathbf{r} = \phi(B)-\phi(A).

For the reversed curve, the starting point is $B$ and the finishing point is $A$. Therefore, its integral is $\phi(A)-\phi(B)$. In other words, the reversal rule and the endpoint formula agree exactly.

\int_{-C}\mathbf{F}\cdot d\mathbf{r} = \phi(A)-\phi(B) = -\bigl(\phi(B)-\phi(A)\bigr) = -\int_C\mathbf{F}\cdot d\mathbf{r}.

The potential itself is not required to have a particular reference value. Adding a constant to $\phi$ changes both endpoint values by the same amount, so their difference remains unchanged.

Worked Examples

Worked Example: Compute Both Orientations by Parameterization

Let $\mathbf{F}(x,y)=(2x,2y)$, and let $C$ go from $A=(1,0)$ to $B=(0,2)$ along the line segment $\mathbf{r}(t)=(1-t,2t)$, $0\le t\le1$. First, a potential is $\phi(x,y)=x^2+y^2$, since $\nabla\phi=(2x,2y)$. The ordered endpoint values give

\phi(A)=1^2+0^2=1,\qquad \phi(B)=0^2+2^2=4,\qquad \int_C\mathbf{F}\cdot d\mathbf{r}=4-1=3.

Check the forward integral directly. The velocity is $\mathbf{r}'(t)=(-1,2)$, and the field along the curve is $\mathbf{F}(\mathbf{r}(t))=(2-2t,4t)$. Thus

\begin{aligned} \int_C\mathbf{F}\cdot d\mathbf{r} &=\int_0^1 (2-2t,4t)\cdot(-1,2)\,dt\\ &=\int_0^1\bigl(-2+2t+8t\bigr)\,dt\\ &=\int_0^1(-2+10t)\,dt\\ &=\left[-2t+5t^2\right]_0^1\\ &=(-2+5)-0=3. \end{aligned}

For the reversed curve, the ordered endpoints are now $B$ and then $A$, so the endpoint calculation gives $1-4=-3$. Directly, use $\widetilde{\mathbf{r}}(t)=(t,2-2t)$. Then $\widetilde{\mathbf{r}}'(t)=(1,-2)$ and $\mathbf{F}(\widetilde{\mathbf{r}}(t))=(2t,4-4t)$. Therefore,

\begin{aligned} \int_{-C}\mathbf{F}\cdot d\mathbf{r} &=\int_0^1(2t,4-4t)\cdot(1,-2)\,dt\\ &=\int_0^1\bigl(2t-8+8t\bigr)\,dt\\ &=\int_0^1(10t-8)\,dt\\ &=\left[5t^2-8t\right]_0^1\\ &=(5-8)-0=-3. \end{aligned}

Both computations check the same result: reversing this curve changes the integral from $3$ to $-3$.

Worked Example: Reverse a Three-Dimensional Potential Difference

Consider $\mathbf{F}(x,y,z)=(yz,xz,xy)$. The scalar function $\phi(x,y,z)=xyz$ is a potential because

\nabla\phi = \left( \frac{\partial}{\partial x}(xyz), \frac{\partial}{\partial y}(xyz), \frac{\partial}{\partial z}(xyz) \right) = (yz,xz,xy) = \mathbf{F}.

Let $C$ start at $P=(1,2,3)$ and end at $Q=(2,-1,4)$. Evaluate the potential at each endpoint:

\phi(P)=1\cdot2\cdot3=6,\qquad \phi(Q)=2\cdot(-1)\cdot4=-8.

In the stated direction, the line integral is finish value minus start value:

\int_C\mathbf{F}\cdot d\mathbf{r} = \phi(Q)-\phi(P) = -8-6 = -14.

On the reversed curve, $Q$ is the start and $P$ is the finish. Thus

\int_{-C}\mathbf{F}\cdot d\mathbf{r} = \phi(P)-\phi(Q) = 6-(-8) = 14 = -(-14).

This verifies the sign change using the ordered endpoints. The negative value in the forward direction is not an error: the line integral is the change in the potential along the chosen direction, and here that change is from $6$ down to $-8$.

Worked Example: Reversal When the Potential Difference Is Zero

Let $\mathbf{F}(x,y)=(2x,2y)$, with potential $\phi(x,y)=x^2+y^2$, and take the line segment from $A=(1,0)$ to $B=(0,1)$. Since $\phi(A)=1$ and $\phi(B)=1$, the endpoint formula gives

\int_C\mathbf{F}\cdot d\mathbf{r} = \phi(B)-\phi(A) = 1-1 = 0.

For a direct check, parameterize $C$ by $\mathbf{r}(t)=(1-t,t)$, $0\le t\le1$. Then $\mathbf{r}'(t)=(-1,1)$ and $\mathbf{F}(\mathbf{r}(t))=(2-2t,2t)$. Therefore,

\begin{aligned} \int_C\mathbf{F}\cdot d\mathbf{r} &=\int_0^1(2-2t,2t)\cdot(-1,1)\,dt\\ &=\int_0^1(-2+4t)\,dt\\ &=\left[-2t+2t^2\right]_0^1\\ &=(-2+2)-0=0. \end{aligned}

Reversing the endpoints gives the potential difference $\phi(A)-\phi(B)=1-1=0$, which is also the negative of the forward value. Here the sign rule holds, although the number does not visibly change because zero is its own negative.

A Common Orientation Error

A common mistake is to reverse the curve but keep the endpoint values in their original order. For the first example, the forward integral is $4-1=3$. If the curve is reversed, writing $4-1=3$ again incorrectly keeps $B$ as the finish and $A$ as the start.

Correction: Reversal swaps the starting and finishing points. The reversed integral is $\phi(A)-\phi(B)=1-4=-3$, not $\phi(B)-\phi(A)=3$.

A reliable way to avoid this error is to name the start and finish of the curve being integrated before subtracting. Always use potential at the finish minus potential at the start. Do not rely on the visual appearance of the path or on the sign of a single component of the field.

Why the Rule Is Useful

Once a potential is known, the integral along a conservative field can be found from the endpoints without parameterizing the curve. Orientation still has to be recorded correctly: the same endpoints in the opposite order give the negative result. This is useful both for efficient calculations and for checking an answer obtained by direct integration.

The rule also clarifies the difference between a path and an oriented path. A geometric route alone does not specify the sign of a line integral; the direction of traversal is part of the data. For a conservative field, reversing that direction changes the sign, while a zero integral remains zero.

Check Your Understanding

Use the direction of traversal explicitly in each question.

If $\int_C\mathbf{F}\cdot d\mathbf{r}=7$, what is $\int_{-C}\mathbf{F}\cdot d\mathbf{r}$?
A conservative field has potential values $\phi(A)=-2$ and $\phi(B)=5$. Find the integral from $A$ to $B$, then from $B$ to $A$.
In the reversal proof, what are the new integration limits after substituting $u=a+b-t$?
Why does adding the same constant to every value of a potential not change a line integral?
Can the integral be unchanged numerically when a curve is reversed? Explain when this occurs.