Calculus › Section B · Limits › Topic B.1

A preview of calculus: tangent lines and area

How steep is a curve at one exact point? How much area sits under a curved shape? Both questions lie beyond algebra, and both are answered by one new idea: the limit.

Two questions algebra cannot answer

How steep is a curve at one single point? How fast is a ball moving at one exact instant? What is the area of a region with a curved edge? Algebra answers none of these: it handles steady rates and straight-sided shapes, while the real world speeds up, slows down and bends. Calculus was invented for exactly these questions, and it rests on one new idea, the limit — the value a quantity closes in on without needing to touch it. Two classic problems preview the whole subject: the tangent line problem, which grows into differential calculus, and the area problem, which grows into integral calculus.

Average rate of change: the secant line

You already know the slope of a straight line: rise over run, \(\frac{y_2 - y_1}{x_2 - x_1}\). A line's slope is the same everywhere, which is why one number describes it. A curve's steepness changes from point to point, so one number cannot describe it — but an average can. Pick a point \(P = (a, f(a))\) on the graph and a second point \(Q = (a + h, f(a + h))\) a small step \(h\) away. The line through \(P\) and \(Q\) is called the secant line, and its slope, the difference quotient, measures the average steepness of the curve between them:

\frac{f(a + h) - f(a)}{h}

The tangent line problem

Now slide \(Q\) toward \(P\) by shrinking \(h\). Each value of \(h\) gives a new secant line, and as \(h\) approaches 0 the secants pivot toward one particular line that just touches the curve at \(P\): the tangent line. Its slope is the value the secant slopes close in on:

\lim_{h \to 0} \frac{f(a + h) - f(a)}{h}

Setting \(h = 0\) outright produces \(\frac{0}{0}\), which has no meaning, so the limit studies the journey rather than the destination. This limit of a difference quotient is the seed from which differential calculus grows.

Worked example: tangent to y = x² at (2, 4)

Find the slope of the tangent line to \(y = x^2\) at \(P = (2, 4)\). Build the difference quotient with \(a = 2\):

\frac{(2 + h)^2 - 4}{h} = \frac{4 + 4h + h^2 - 4}{h} = \frac{4h + h^2}{h}

While \(h\) is not zero it can be cancelled, leaving \(4 + h\). Now let \(h\) approach 0: the quotient approaches 4, so the tangent slope is 4. In point-slope form the tangent line is \(y - 4 = 4(x - 2)\), that is \(y = 4x - 4\). Notice the key move, used every time: simplify until \(h\) cancels, and only then let \(h\) approach 0.

Seeing the limit in a table

The same limit shows up numerically. For \(y = x^2\) at \(a = 2\), the secant slope \(4 + h\) gives: \(h = 0.1\) gives 4.1, \(h = 0.01\) gives 4.01, \(h = -0.1\) gives 3.9, and \(h = -0.01\) gives 3.99. The values squeeze in on 4 from both sides. Tables build intuition and check your algebra, but they are evidence, not proof: a function can misbehave at scales your table never samples. The algebraic route — simplify, then take the limit — is the one that carries certainty.

Velocity and other rates of change

Velocity is the same mathematics wearing different letters. If \(s(t)\) gives position at time \(t\), the average velocity from \(t = a\) to \(t = a + h\) is displacement over elapsed time, \(\frac{s(a + h) - s(a)}{h}\) — literally the secant slope on a position-time graph. The instantaneous velocity, the reading a speedometer shows, is the limit of those averages as \(h\) approaches 0. The same pattern reaches everywhere: with \(C(x)\) the cost of producing \(x\) items, the limit is the marginal cost, the cost of producing one more item. Every rate carries units: output units per input unit, like metres per second or dollars per item.

Worked example: a ball thrown upward

A ball leaves a balcony and its height is \(s(t) = 20t - 5t^2\) metres after \(t\) seconds. How fast is it moving at \(t = 1\)? With \(s(1) = 15\), the average velocity from \(t = 1\) to \(t = 1 + h\) is

\frac{20(1 + h) - 5(1 + h)^2 - 15}{h} = \frac{20 + 20h - 5 - 10h - 5h^2 - 15}{h} = 10 - 5h

As \(h\) approaches 0, the average velocity approaches 10, so at that instant the ball is rising at 10 metres per second. Over the whole interval from \(t = 1\) to \(t = 3\), the average velocity is \(\frac{s(3) - s(1)}{2} = \frac{15 - 15}{2} = 0\): the ball is back at 15 metres, so the average is zero even though it travelled up to its peak at \(t = 2\) and back down.

The area problem

The second classic problem asks for the area of the region trapped between a curve \(y = f(x)\), the x-axis and the vertical lines \(x = a\) and \(x = b\). Geometry only computes straight-sided regions, so approximate instead: slice the interval into pieces of equal width, build a rectangle on each piece whose height comes from the curve, and add the areas. The rectangles miss some curved slivers, but as the slices get thinner the missed slivers shrink. The true area is the limit of these sums as the number of rectangles grows without bound — the seed from which integral calculus grows.

Worked example: estimating an area

Estimate the area under \(y = x^2\) from \(x = 0\) to \(x = 1\) with four rectangles of width \(\frac{1}{4}\), taking each height at the right end of its slice. The right endpoints are \(\frac{1}{4}, \frac{1}{2}, \frac{3}{4}, 1\), so the heights are \(\frac{1}{16}, \frac{1}{4}, \frac{9}{16}, 1\), which sum to \(\frac{30}{16}\). Multiplying by the width:

\frac{1}{4} \cdot \frac{30}{16} = \frac{30}{64} = \frac{15}{32} \approx 0.47

Because \(x^2\) is increasing, right endpoints overshoot. With eight rectangles the estimate drops to about 0.398, closing in on the exact value \(\frac{1}{3}\), which the fundamental theorem of calculus will later confirm.

Two problems, one subject

The tangent problem measures instantaneous change; the area problem accumulates change. They look unrelated, and for centuries they were studied separately. Then, in the late 1600s, Isaac Newton and Gottfried Leibniz independently made the discovery at the heart of calculus: the two problems are inverse processes, two halves of one subject. The slope problem produces the derivative; the area problem produces the integral; and the fundamental theorem of calculus, in Section E, shows each undoing the other. What a limit builds on one side, a limit of sums builds on the other.

Common mistakes to avoid

Plugging in \(h = 0\) before simplifying gives \(\frac{0}{0}\), which is not an answer — it is a signal to simplify until \(h\) cancels. Expanding \((a + h)^2\) as \(a^2 + h^2\) drops the middle term \(2ah\); expand carefully. Do not confuse the secant slope, an average over an interval, with the tangent slope, an instant. When estimating area, do not report one rectangle sum as the answer: only the limit of the sums is exact. Another slip is forgetting to subtract \(f(a)\) in the numerator: \(\frac{f(a+h)}{h}\) answers a different question entirely.

Key takeaways

A secant slope \(\frac{f(a + h) - f(a)}{h}\) is an average rate of change, and the tangent slope at \(a\) is its limit as \(h\) approaches 0. Compute it by simplifying first and letting \(h\) shrink second. Average velocity, instantaneous velocity and marginal cost are this same difference quotient in different clothes, and every rate carries output units per input unit. Areas under curves are approximated by sums of rectangle areas and defined exactly as the limit of those sums. The tangent and area problems are inverse processes — the story the fundamental theorem will make precise.

Flashcards

Flashcard

Secant line

A line through two points on a curve

Its slope is the average rate of change between them.

Test your understanding

Question 1 of 5

For f(x) = x², what is the slope of the secant line from x = 2 to x = 2.5?

Question 2 of 5

What is the slope of the tangent line to y = x² at the point (-1, 1)?

Question 3 of 5

A ball's height is s(t) = 20t - 5t² metres. What is its instantaneous velocity at t = 1?

Question 4 of 5

Estimate the area under y = x² from x = 0 to x = 1 using two rectangles of width 1/2, with heights taken at the right endpoints.

Question 5 of 5

How are the tangent line problem and the area problem related?

Standards: 1

← Previous topicNext topic →