Chapter 8 · Introduction

Chapter 8 — Sequences and Series

Adding up infinitely many numbers — and, surprisingly often, getting a finite answer like $\tfrac12+\tfrac14+\tfrac18+\cdots=1$.

About 2400 years ago the Greek philosopher Zeno of Elea posed a puzzle. To walk to a wall one metre away you must first cover half the distance, then half of what is left, then half of what is left again, and so on, forever. That is infinitely many tasks: $\tfrac12+\tfrac14+\tfrac18+\tfrac1{16}+\cdots$ metres. How can anyone finish infinitely many tasks in a finite time? Zeno concluded that motion is an illusion. The way out is that an infinite sum can have a finite value: the distances get smaller fast enough that their total is exactly $1$ metre, and, at constant speed, the times add up to a finite total too.

Making sense of "infinitely many additions" is the subject of this chapter. We first study sequences — endless lists $a_1,a_2,a_3,\dots$ — and what it means for them to settle down to a limit. Then we add the terms of a sequence to form an infinite series $\sum a_n$, and develop a toolkit of convergence tests that decide whether the sum is finite without having to find it. This is not just a curiosity: calculators evaluate $e^x$, $\sin x$ and $\ln x$ by adding up series; engineers, physicists and economists use them to approximate functions, solve differential equations and price long-term payments. Chapter 9 (power series) is built entirely on the tools you will learn here.

Sequences & their limits→Partial sums→Geometric & telescoping series→Divergence & integral tests→Comparison tests→Alternating series→Ratio & root tests
🎯 By the end of this chapter you can…
  • find the limit of a sequence (or show it diverges) using limit laws, L'Hôpital's rule, the squeeze theorem and the Monotone Convergence Theorem;
  • explain what it means for an infinite series to converge, in terms of its partial sums;
  • sum geometric and telescoping series exactly, and use them in applications such as repeated drug doses and repeating decimals;
  • decide whether a series converges using the divergence, integral, comparison, limit comparison, alternating series, ratio and root tests — and choose the right test quickly;
  • distinguish absolute from conditional convergence and estimate the error when a sum is cut off after finitely many terms.
Teaser — Zeno's Walk: Adding Infinitely Many Numbers

The walker on the top line hops forward by one term at a time; the lower graph shows the running total $S_n=a_1+a_2+\cdots+a_n$ (dots) above the terms $a_n$ (bars). When the totals settle on a horizontal line, the infinite sum has a finite value. Try the harmonic series last — its terms also shrink to $0$, yet the walker never stops.

1
$n$—
term $a_n$—
partial sum $S_n$—
infinite sum $S$—
gap $S-S_n$—

A first look: which test when?

Most of this chapter answers one question: does $\sum a_n$ converge? You will meet seven convergence tests and a few special series we can settle outright, each suited to a particular look of $a_n$. Here is the map you will be filling in; do not worry about the details yet — each row is explained carefully in its own section.

If the terms $a_n$ look like…try…what it tells yousection
anything — check first!Divergence testif $a_n\not\to0$, the series diverges8.3
$ar^{n-1}$ (constant ratio)Geometric seriesconverges to $\dfrac{a}{1-r}$ exactly when $|r|\lt1$8.2
$b_n-b_{n+1}$ (partial fractions)Telescoping$S_N$ collapses; take its limit8.2
$\dfrac1{n^p}$$p$-seriesconverges exactly when $p\gt1$8.3
$f(n)$ with $f$ positive, decreasing, easy to integrateIntegral test$\sum a_n$ and $\int_1^\infty f(x)\,dx$ behave alike8.3
rational or algebraic in $n$, e.g. $\dfrac{n+1}{n^3+5}$(Limit) comparisoncompare with a $p$-series or geometric series8.4
$(-1)^nb_n$ with $b_n$ decreasing to $0$Alternating series testconverges, with an easy error bound8.5
factorials $n!$, powers $c^n$, productsRatio testlook at $\left|\dfrac{a_{n+1}}{a_n}\right|$8.6
an $n$th power, e.g. $\left(\dfrac{2n}{3n+1}\right)^n$Root testlook at $\sqrt[n]{|a_n|}$8.6
⚠️ The one fact to remember from day one
If $\sum a_n$ converges then $a_n\to0$ — but not the other way round. The harmonic series $1+\tfrac12+\tfrac13+\cdots$ has terms shrinking to $0$ and still diverges (watch it in the teaser above, and see the proof in 8.2). Small terms are necessary, never sufficient.

Sections in this chapter

✅ Before you start
You will use limits at infinity constantly (Calculus I, 1.4 Limits Involving Infinity) and L'Hôpital's rule for forms such as $\tfrac{\infty}{\infty}$ and $1^\infty$ (Calculus I, 3.7 L'Hospital's Rule). The integral test in 8.3 relies on improper integrals (5.5 Improper Integrals), and telescoping sums use partial fractions (5.3 Partial Fractions). Be comfortable with sigma notation: $\sum_{k=1}^{n}k=\tfrac{n(n+1)}{2}$.