Infinite Series (Appendix)

Every finite sum you have ever computed obeys the same quiet rule: add enough numbers and eventually you stop, and there is a definite answer. This appendix asks a stranger question — what happens when you never stop? Add $1 + \dfrac12 + \dfrac14 + \dfrac18 + \cdots$ forever, with infinitely many terms and no last one, and the running total does not run away to infinity. It creeps closer and closer to exactly $2$ , and never overshoots. That single observation — that an infinite sum of shrinking numbers can settle down to a finite, exact value — is the idea this chapter builds an entire toolkit around. Chapter 9 introduced sequences and series and taught you to sum a finite geometric progression. This appendix asks what survives when the progression never ends, and the answer turns out to be one of the most productive ideas in all of mathematics. It reappears in three disguises here. First, the Binomial Theorem you met for a positive integer index is generalised to any index at all — negative, fractional, irrational — at the cost of trading a finite sum for an infinite one that only makes sense when $|x| < 1$ . Second, the infinite geometric series itself is given a precise sum-to-infinity formula, $S = \dfrac{a}{1-r}$ , resting entirely on what happens to $r^n$ as $n$ grows without bound. Third, two named infinite series are built from scratch: the exponential series, which defines Euler's famous number $e$ and is squeezed rigorously between $2$ and $3$ , and the logarithmic series, which expresses $\log_e(1+x)$ as an infinite sum. None of this is required for your board examination, but it is the seed of calculus, real analysis, and the convergence tests you will meet in higher mathematics — read it as your first honest look at infinity behaving itself.

Key takeaways

An infinite series is the indicated sum $a_1 + a_2 + a_3 + \cdots$ of an infinite sequence, written compactly with sigma notation as $\sum_{k=1}^{\infty} a_k$ .
The Binomial Theorem for any index: $(1+x)^m = 1 + mx + \dfrac{m(m-1)}{1\cdot2}x^2 + \dfrac{m(m-1)(m-2)}{1\cdot2\cdot3}x^3 + \cdots$ , valid whenever $|x| < 1$ , for any real number $m$ — integer, negative, or fractional.
Unlike the positive-integer binomial theorem, this binomial series never terminates when $m$ is negative or fractional — it has infinitely many terms, and the restriction $|x| < 1$ is not optional; violate it and the equality breaks down.
For $(a+b)^m$ with $a \neq 0$ , factor out $a^m$ and apply the theorem to $\left(1 + \tfrac{b}{a}\right)^m$ ; this is valid whenever $|b| < |a|$ .
An infinite geometric series $a + ar + ar^2 + \cdots$ has a finite sum to infinity $S = \dfrac{a}{1-r}$ if and only if $|r| < 1$ , because the finite-sum formula $S_n = \dfrac{a(1-r^n)}{1-r}$ has $r^n \to 0$ exactly when $|r| < 1$ .
The exponential series $1 + \dfrac{1}{1!} + \dfrac{1}{2!} + \dfrac{1}{3!} + \cdots$ defines Euler's number $e$ , and a comparison with a geometric series proves rigorously that $2 < e < 3$ ; the general exponential series is $e^x = 1 + \dfrac{x}{1!} + \dfrac{x^2}{2!} + \dfrac{x^3}{3!} + \cdots$ .
The logarithmic series states $\log_e(1+x) = x - \dfrac{x^2}{2} + \dfrac{x^3}{3} - \cdots$ for $|x| < 1$ , and remarkably the expansion continues to hold at $x = 1$ , giving $\log_e 2 = 1 - \dfrac12 + \dfrac13 - \dfrac14 + \cdots$ .
All three series share one structural feature: each is an infinite sum that only equals a finite number under a stated restriction on the variable — dropping that restriction produces nonsense, as the chapter demonstrates with an explicit false substitution.
This chapter is classified as enrichment: it sits outside the core board syllabus, but it is the direct ancestor of Taylor series, power series, and the convergence tests of higher-secondary and undergraduate calculus.
Key idea: an infinite series converges to a finite sum only under a precise condition on its terms — for a geometric series it is the condition |r| < 1, and the same idea of a restricted infinite sum builds the binomial series for any index, defines Euler's number e via the exponential series, and produces the logarithmic series for log_e(1+x).

The core insight

An infinite series is a genuinely infinite sum that can still equal a finite, exact number, but only under a stated condition — $|r| < 1$ for the geometric series ( $S = a/(1-r)$ ), $|x| < 1$ for the binomial series $(1+x)^m$ and the logarithmic series $\log_e(1+x)$ — and the exponential series $e^x = \sum x^n/n!$ converges for every $x$ and, at $x=1$ , defines and bounds Euler's number $2 < e < 3$ .

What this chapter covers

The appendix opens (A.1.1) by recalling the definition of an infinite series as the indicated, never-ending sum of an infinite sequence, introducing sigma notation as shorthand. Section A.1.2 generalises the Binomial Theorem to any index $m$ — not just a positive integer — stating (without proof, as NCERT does, since the proof needs tools beyond this chapter) the infinite binomial series for $(1+x)^m$ and the crucial restriction $|x| < 1$ ; the section works through the general form for $(a+b)^m$ , lists the four standard special cases used constantly in practice, and solves a worked expansion. Section A.1.3 revisits the geometric progression from Chapter 9 and asks what happens to the finite-sum formula as the number of terms grows without bound, arriving at the sum-to-infinity formula $S = a/(1-r)$ and applying it to a worked example. Section A.1.4 builds the exponential series from first principles, uses a comparison argument to trap Euler's number $e$ rigorously between $2$ and $3$ , states the general series for $e^x$ , and applies it to find a coefficient in an expansion and to estimate $e^2$ to one decimal place. Section A.1.5 closes with the logarithmic series for $\log_e(1+x)$ , its special value at $x=1$ giving $\log_e 2$ , and a worked example that uses the additive property of logarithms to expand $\log_e(1+px+qx^2)$ for a quadratic with roots $\alpha, \beta$ . This article follows that order exactly, with two figures and three fully worked derivations along the way.

The core ideas

Infinite series and sigma notation

Recall from Chapter 9 that a sequence $a_1, a_2, a_3, \ldots, a_n, \ldots$ with infinitely many terms is an infinite sequence, and its indicated sum

a_1 + a_2 + a_3 + \cdots + a_n + \cdots = \sum_{k=1}^{\infty} a_k \qquad (1)

where: the symbol $\sum_{k=1}^{\infty}$ is read "the sum from $k$ equals $1$ to infinity," and $a_k$ is the general ( $k$ th) term of the sequence, is called an infinite series. The dots after the last written term are doing real work — they signal that the pattern continues forever, with no final term to add. Whether such a sum equals a finite number, and if so which number, is the entire subject of this appendix, taken up through three specific families of series: the binomial series, the geometric series, and the exponential and logarithmic series.

The Binomial Theorem for any index

In Chapter 8 you proved $(1+x)^n = {}^{n}C_0 + {}^{n}C_1 x + \cdots + {}^{n}C_n x^n$ for $n$ a non-negative integer, using the counting symbol $^{n}C_r$ . That symbol stops making sense the instant $n$ is replaced by a negative integer or a fraction — you cannot "choose $2$ objects out of $-3$ ." Yet the pattern of coefficients can be rescued by rewriting $^{n}C_r$ in its product form and simply allowing $n$ (renamed $m$ here, to stress that it need no longer be a whole number) to be any real number. This gives, without proof, the general Binomial Theorem for any index:

(1+x)^m = 1 + mx + \frac{m(m-1)}{1\cdot 2}x^2 + \frac{m(m-1)(m-2)}{1\cdot 2\cdot 3}x^3 + \cdots, \qquad |x| < 1. \qquad (2)

where: $m$ is any real number at all, and the restriction $|x| < 1$ — that is, $-1 < x < 1$ — is essential whenever $m$ is a negative integer or a fraction. NCERT gives a sharp warning about what goes wrong if this condition is ignored: take $x = -2$ and $m = -2$ in equation (2), formally plugging in without checking $|x|<1$ . The left side gives $(1-2)^{-2} = (-1)^{-2} = 1$ . The right side gives $1 + (-2)(-2) + \dfrac{(-2)(-3)}{1\cdot2}(-2)^2 + \cdots = 1 + 4 + 12 + \cdots$ , a sum of ever-larger positive terms that clearly cannot equal $1$ . The formula is simply not valid outside $|x| < 1$ , and using it anyway produces contradictions like this one. The second structural difference from the Chapter 8 theorem is length: when $m$ is a positive integer, the series in equation (2) terminates naturally, because the factor $(m - m) = 0$ eventually appears in the numerator and every subsequent term vanishes — you recover the familiar finite expansion. But when $m$ is negative or fractional, no factor in the product $m(m-1)(m-2)\cdots$ is ever exactly zero, so the series genuinely never ends; it has infinitely many terms, every one of them potentially nonzero. For a general binomial $(a+b)^m$ with $a \neq 0$ , factor out $a^m$ :

(a+b)^m = a^m\left(1+\frac{b}{a}\right)^m = a^m\left[1 + m\frac{b}{a} + \frac{m(m-1)}{1\cdot2}\left(\frac{b}{a}\right)^2 + \cdots\right] = a^m + m\,a^{m-1}b + \frac{m(m-1)}{1\cdot2}a^{m-2}b^2 + \cdots, \qquad (3)

where: this expansion is valid precisely when $\left|\dfrac{b}{a}\right| < 1$ , equivalently $|b| < |a|$ . From this, four special cases are used constantly and are worth memorising outright: $(1+x)^{-1} = 1 - x + x^2 - x^3 + \cdots$ ; $(1-x)^{-1} = 1 + x + x^2 + x^3 + \cdots$ ; $(1+x)^{-2} = 1 - 2x + 3x^2 - 4x^3 + \cdots$ ; and $(1-x)^{-2} = 1 + 2x + 3x^2 + 4x^3 + \cdots$ — all valid for $|x| < 1$ .

Infinite geometric series and the sum to infinity

From Chapter 9, a geometric progression (G.P.) in standard form $a, ar, ar^2, \ldots, ar^{n-1}$ has a known finite-sum formula, $S_n = \dfrac{a(1-r^n)}{1-r}$ . The natural next question — the one that gives this appendix its name — is what happens to that sum as you keep adding terms forever, letting $n \to \infty$ . NCERT answers it with a concrete example: the G.P. $1, \tfrac23, \tfrac49, \ldots$ has $a = 1$ , $r = \tfrac23$ , and a short table of $\left(\tfrac23\right)^n$ for $n = 1, 5, 10, 20$ shows the values $0.6667$ , $0.1316\ldots$ , $0.01734\ldots$ , $0.000300\ldots$ — shrinking toward zero. The conclusion generalises: whenever $|r| < 1$ , the term $r^n \to 0$ as $n \to \infty$ , so the finite-sum formula's second piece vanishes and $S_n \to \dfrac{a}{1-r}$ . This limiting value is the sum to infinity, denoted $S$ :

S = \frac{a}{1-r}, \qquad |r| < 1. \qquad (4)

Two worked identities from the chapter illustrate the formula immediately: $1 + \tfrac12 + \tfrac14 + \tfrac18 + \cdots = \dfrac{1}{1 - \tfrac12} = 2$ , and $1 - \tfrac12 + \tfrac14 - \tfrac18 + \cdots = \dfrac{1}{1-\left(-\tfrac12\right)} = \dfrac{1}{\tfrac32} = \dfrac23$ .

Figure

A graph of the partial sums of the geometric series with first term one and common ratio two thirds plotted against the number of terms added, with the points climbing quickly at first and then levelling off just under a dashed horizontal line marking the infinite sum of three

Example 2 in the chapter applies the formula directly: for the G.P. $-\dfrac54, \dfrac{5}{16}, -\dfrac{5}{64}, \ldots$ , we read off $a = -\dfrac54$ and $r = -\dfrac14$ , and since $|r| < 1$ the sum to infinity exists: $S = \dfrac{-5/4}{1-\left(-\tfrac14\right)} = \dfrac{-5/4}{5/4} = -1$ .

The exponential series and Euler's number e

Leonhard Euler (1707–1783) introduced a number, now written $e$ , that is to calculus what $\pi$ is to the geometry of circles. It is defined as the sum to infinity of a specific series:

e = 1 + \frac{1}{1!} + \frac{1}{2!} + \frac{1}{3!} + \frac{1}{4!} + \cdots \qquad (5)

where: every term is positive, so the sum — if it exists at all — must exceed every partial sum, in particular exceed $1 + 1 + \tfrac12 = 2.5$ , and so $e > 2$ . The far more interesting question, worked out in full below as a derivation, is why the series does not run off to infinity: NCERT proves, by comparing it term by term against a geometric series with ratio $\tfrac12$ , that $e$ is also less than $3$ . So $2 < e < 3$ , and more refined estimates (which you will meet again in calculus) pin it down to $e \approx 2.71828$ . The series generalises immediately to any real $x$ :

e^x = 1 + \frac{x}{1!} + \frac{x^2}{2!} + \frac{x^3}{3!} + \cdots + \frac{x^n}{n!} + \cdots \qquad (6)

with no restriction on $x$ at all — unlike the binomial and logarithmic series, this one converges for every real number. Two chapter examples put it to work. First, replacing $x$ by $2x+3$ in equation (6) and picking out the coefficient of $x^2$ across the whole (infinite) series turns, after using $n! = n(n-1)(n-2)!$ to peel off a factor, into exactly $2e^3$ times a repackaged copy of the exponential series for $e^3$ — the coefficient of $x^2$ in $e^{2x+3}$ is $2e^3$ . Second, putting $x=2$ directly into equation (6) and bounding the tail of the series the same way as for $e$ itself shows $7.355 \le e^2 < 7.4$ , so $e^2$ rounded to one decimal place is $7.4$ .

The logarithmic series

The last named series in the chapter expresses the natural logarithm of $1+x$ as an infinite sum, stated (again without proof) as a theorem:

\log_e(1+x) = x - \frac{x^2}{2} + \frac{x^3}{3} - \frac{x^4}{4} + \cdots, \qquad |x| < 1. \qquad (7)

where: the alternating signs are the signature of this series — every even-indexed term is subtracted, every odd-indexed term added. A notable feature NCERT flags explicitly is that the expansion continues to hold even at the boundary value $x = 1$ (a subtlety beyond this chapter's scope to justify), giving the striking identity

\log_e 2 = 1 - \frac12 + \frac13 - \frac14 + \cdots. \qquad (8)

The chapter's Example 5 shows the series' algebraic power: if $\alpha, \beta$ are the roots of $x^2 - px + q = 0$ (so $\alpha+\beta = p$ and $\alpha\beta = q$ ), then applying equation (7) separately to $\log_e(1+\alpha x)$ and $\log_e(1+\beta x)$ and adding uses the identity $\log_e A + \log_e B = \log_e(AB)$ to prove $\log_e(1+px+qx^2) = (\alpha+\beta)x - \dfrac{\alpha^2+\beta^2}{2}x^2 + \dfrac{\alpha^3+\beta^3}{3}x^3 - \cdots$ , valid whenever both $|\alpha x| < 1$ and $|\beta x| < 1$ .

Figure

A number line from negative two to positive two with the open interval from negative one to one shaded to mark where the binomial series and the logarithmic series converge, and points just outside that interval marked with a cross to show the formulas fail there

Derivations

Derivation ASum to Infinity of a Geometric Series

This derivation makes precise the claim behind equation (4): why the infinite geometric series $a + ar + ar^2 + \cdots$ collapses to the exact finite value $\dfrac{a}{1-r}$ whenever $|r| < 1$ , and why that same reasoning fails completely the moment $|r| \geq 1$ . The whole argument rests on nothing more exotic than the already-known finite-sum formula and a careful look at what the power $r^n$ does as $n$ grows without any bound — no new machinery is needed, only the willingness to let $n$ run forever and watch a single term shrink to nothing.

Step 1. Start from the finite geometric series sum, proved in Chapter 9 by the standard subtraction trick (write $S_n = a + ar + \cdots + ar^{n-1}$ , form $rS_n$ , and subtract to telescope every middle term away): $S_n = \dfrac{a(1-r^n)}{1-r}$ , valid for any $r \neq 1$ and any positive integer $n$ .

Step 2. Split this single fraction into two separate pieces by distributing the numerator: $S_n = \dfrac{a}{1-r} - \dfrac{a\,r^n}{1-r}$ . The first piece, $\dfrac{a}{1-r}$ , does not depend on $n$ at all — it is a fixed number, the same for every partial sum. All of the dependence on how many terms have been added is concentrated entirely in the second piece, through the factor $r^n$ .

Step 3. Examine what $r^n$ does as $n$ increases, restricting attention to $|r| < 1$ . NCERT's own table for $r = \tfrac23$ makes this concrete: $r^1 = 0.6667$ , $r^5 \approx 0.1317$ , $r^{10} \approx 0.01734$ , $r^{20} \approx 0.0003$ . Each time $n$ grows, $r^n$ shrinks further, because multiplying a number already smaller than $1$ in size by another number smaller than $1$ in size can only make it smaller still. The pattern is general: for any fixed $r$ with $|r| < 1$ , repeated multiplication by $r$ drives the power toward zero, however slowly. This is written $r^n \to 0$ as $n \to \infty$ .

Step 4. Substitute this limiting behaviour back into the split formula from Step 2. As $n \to \infty$ , the second term $\dfrac{a\,r^n}{1-r} \to \dfrac{a \cdot 0}{1-r} = 0$ , since $r^n \to 0$ and the denominator $1-r$ is a fixed nonzero number (fixed because $r \neq 1$ is guaranteed by $|r|<1$ ). So the whole expression $S_n \to \dfrac{a}{1-r} - 0 = \dfrac{a}{1-r}$ . This limiting value is defined as the sum to infinity, $S = \dfrac{a}{1-r}$ , exactly equation (4).

Step 5. Confirm the necessity of $|r| < 1$ with two quick counterexamples. If $r = 2$ (so $|r| > 1$ ), then $r^n = 2^n$ grows without bound rather than shrinking, so $S_n$ diverges to infinity and no finite sum exists. If $r = -1$ (so $|r| = 1$ ), the series $a - a + a - a + \cdots$ has partial sums that oscillate forever between $a$ and $0$ , never settling on a single value — again there is no sum to infinity. The condition $|r| < 1$ is therefore not a technicality; it is exactly the boundary between a series that settles down and one that does not.

Go deeper · Derivation B · 2 minSqueezing Euler's Number e Between 2 and 3

Derivation BSqueezing Euler's Number e Between 2 and 3

This derivation reconstructs NCERT's proof that the exponential series in equation (5) actually adds up to a genuine finite number trapped strictly between $2$ and $3$ , rather than growing without bound the way, say, $1 + 1 + 1 + \cdots$ would. The technique — bounding an unfamiliar series above by comparing it term by term with a familiar geometric series whose sum to infinity is already known from the previous derivation — is a template you will reuse constantly once you study convergence tests formally in later mathematics.

Step 1. Write the series to be bounded: $e = 1 + \dfrac{1}{1!} + \dfrac{1}{2!} + \dfrac{1}{3!} + \dfrac{1}{4!} + \cdots$ . Since every single term $\dfrac{1}{n!}$ is positive, the partial sums only ever increase, and already $1 + \dfrac{1}{1!} + \dfrac{1}{2!} = 1 + 1 + 0.5 = 2.5 > 2$ , so $e > 2$ is immediate and needs no further argument.

Step 2. Establish the key comparison inequality: for every integer $n > 2$ , $\dfrac{1}{n!} < \dfrac{1}{2^{\,n-1}}$ . Check the pattern directly, exactly as NCERT tabulates it: for $n=3$ , $\dfrac{1}{3!} = \dfrac16$ against $\dfrac{1}{2^2} = \dfrac14$ , and indeed $\dfrac16 < \dfrac14$ ; for $n=4$ , $\dfrac{1}{4!} = \dfrac{1}{24}$ against $\dfrac{1}{2^3} = \dfrac18$ , and $\dfrac{1}{24} < \dfrac18$ ; for $n=5$ , $\dfrac{1}{5!} = \dfrac{1}{120}$ against $\dfrac{1}{2^4} = \dfrac{1}{16}$ , and $\dfrac{1}{120} < \dfrac{1}{16}$ . The reason this pattern continues forever is that $n!$ contains the factors $3, 4, 5, \ldots, n$ once each beyond $2!$ , and from $n=3$ onward every one of those factors is at least $3$ , so $n!$ eventually outpaces $2^{n-1}$ , which only ever multiplies by $2$ at each step.

Step 3. Sum the inequality from Step 2 over all $n$ from $3$ to infinity, term by term: $\dfrac{1}{3!} + \dfrac{1}{4!} + \dfrac{1}{5!} + \cdots < \dfrac{1}{2^2} + \dfrac{1}{2^3} + \dfrac{1}{2^4} + \cdots$ . The right-hand side is itself an infinite geometric series with first term $\dfrac{1}{4}$ and common ratio $\dfrac12$ , so by the sum-to-infinity formula from the previous derivation, $\dfrac{1}{2^2} + \dfrac{1}{2^3} + \cdots = \dfrac{1/4}{1 - 1/2} = \dfrac{1}{2}$ .

Step 4. Add the two fixed opening terms $1 + \dfrac{1}{1!} + \dfrac{1}{2!} = 2.5$ to both sides of the inequality from Step 3. The left side becomes exactly $e$ (the whole series from equation (5), reassembled). The right side becomes $2.5 + \dfrac12 = 3$ . So $e < 2.5 + \dfrac12 = 3$ .

Step 5. Combine Step 1 and Step 4: $e > 2$ and $e < 3$ , so $2 < e < 3$ . This is a genuine proof, not an estimate — it guarantees $e$ is a specific finite number strictly between two integers, obtained purely by comparing the unfamiliar factorial series against a geometric series whose behaviour was already fully understood from the first derivation.

Go deeper · Derivation C · 2 minThe General Term of the Binomial Series for Any Index

Derivation CThe General Term of the Binomial Series for Any Index

This derivation shows exactly how the compact statement $(a+b)^m = a^m + m\,a^{m-1}b + \dfrac{m(m-1)}{1\cdot2}a^{m-2}b^2 + \cdots$ from equation (3) continues, term after term, so that any single term of the infinite expansion can be written down directly without listing every term before it — the analogue, for a general index, of the general-term formula $T_{r+1} = {}^{n}C_r\,a^{n-r}b^r$ you already know for a positive integer index. The goal is to identify the exact pattern of numerator and denominator in the $(r+1)$ th term and confirm it against the known special cases.

Step 1. Begin from the stated theorem for $(1+x)^m$ in equation (2): $(1+x)^m = 1 + mx + \dfrac{m(m-1)}{1\cdot2}x^2 + \dfrac{m(m-1)(m-2)}{1\cdot2\cdot3}x^3 + \cdots$ , valid for $|x|<1$ . Look at the coefficients in order: the coefficient of $x^0$ is $1$ (an "empty product," by convention); the coefficient of $x^1$ is $m$ ; the coefficient of $x^2$ is $\dfrac{m(m-1)}{1\cdot2}$ ; the coefficient of $x^3$ is $\dfrac{m(m-1)(m-2)}{1\cdot2\cdot3}$ .

Step 2. Extract the pattern governing the coefficient of $x^r$ for a general non-negative integer $r$ . In the numerator, the coefficient is a product of exactly $r$ consecutive factors starting at $m$ and decreasing by $1$ each time: $m(m-1)(m-2)\cdots(m-r+1)$ — note the last factor is $(m-r+1)$ , not $(m-r)$ , so that there are precisely $r$ factors total (check: for $r=2$ the factors are $m$ and $(m-1)$ , matching Step 1; for $r=3$ they are $m$ , $(m-1)$ , $(m-2)$ , again matching). In the denominator, the factors are simply $1\cdot2\cdot3\cdots r = r!$ , the ordinary factorial. So the coefficient of $x^r$ is $\dfrac{m(m-1)(m-2)\cdots(m-r+1)}{r!}$ , and the $(r+1)$ th term of $(1+x)^m$ is $T_{r+1} = \dfrac{m(m-1)\cdots(m-r+1)}{r!}\,x^r$ .

Step 3. Extend this to the general binomial $(a+b)^m$ using the factoring from equation (3): $(a+b)^m = a^m\left(1+\tfrac{b}{a}\right)^m$ . Substitute $x = \tfrac{b}{a}$ into the term formula from Step 2, then multiply the whole thing by the outer factor $a^m$ : the $(r+1)$ th term of $\left(1+\tfrac{b}{a}\right)^m$ is $\dfrac{m(m-1)\cdots(m-r+1)}{r!}\left(\dfrac{b}{a}\right)^r$ , so the $(r+1)$ th term of $(a+b)^m$ itself is $a^m \cdot \dfrac{m(m-1)\cdots(m-r+1)}{r!}\left(\dfrac{b}{a}\right)^r = \dfrac{m(m-1)(m-2)\cdots(m-r+1)}{1\cdot2\cdot3\cdots r}\,a^{m-r}b^r$ , since $a^m \cdot a^{-r} = a^{m-r}$ .

Step 4. Verify against a memorised special case: take $a=1, b=x, m=-1$ , i.e. $(1+x)^{-1}$ . The $(r+1)$ th term is $\dfrac{(-1)(-2)\cdots(-r)}{r!}x^r = \dfrac{(-1)^r\,r!}{r!}x^r = (-1)^r x^r$ , since $r$ consecutive integers starting at $-1$ multiply out to $(-1)^r r!$ . So the series is $1-x+x^2-x^3+\cdots$ , matching the first special case after equation (3).

Step 5. Check one more case: $m=-2$ , giving $(1+x)^{-2}$ . The $(r+1)$ th term is $\dfrac{(-2)(-3)\cdots(-(r+1))}{r!}x^r = \dfrac{(-1)^r(r+1)!}{r!}x^r = (-1)^r(r+1)x^r$ . This gives $1-2x+3x^2-4x^3+\cdots$ , matching the third special case exactly and confirming the general-term formula from Step 3.

Worked examples

Board level - expanding a binomial with a negative fractional index

Expand $\left(1 - \dfrac{x}{2}\right)^{-1}$ , where $|x| < 2$ .

Step 1. Match the given expression to the general binomial series with $m = -1$ and, writing $\left(1-\tfrac{x}{2}\right)^{-1}$ in the form $(1+X)^{-1}$ with $X = -\tfrac{x}{2}$ , apply the special case $(1+X)^{-1} = 1 - X + X^2 - X^3 + \cdots$ , valid for $|X| < 1$ , i.e. $\left|\tfrac{x}{2}\right| < 1$ , i.e. $|x| < 2$ — exactly the condition stated in the problem.

Step 2. Substitute $X = -\tfrac{x}{2}$ term by term: $1 - \left(-\dfrac{x}{2}\right) + \left(-\dfrac{x}{2}\right)^2 - \left(-\dfrac{x}{2}\right)^3 + \cdots = 1 + \dfrac{x}{2} + \dfrac{x^2}{4} + \dfrac{x^3}{8} + \cdots$ , where each squared or higher even power of the negative $X$ turns positive and each odd power stays negative, but the outer alternating minus sign in the formula flips every sign back, leaving every term positive.

Step 3. Cross-check the first three terms directly using the general-term formula from the third derivation, with $m=-1$ , $a=1$ , $b = -\tfrac{x}{2}$ : $T_{r+1} = (-1)^r\left(-\dfrac{x}{2}\right)^r = \left(\dfrac{x}{2}\right)^r$ for every $r$ , confirming a series of entirely positive terms $1, \tfrac{x}{2}, \tfrac{x^2}{4}, \tfrac{x^3}{8}, \ldots$ , matching Step 2 term for term.

Step 4. Write the final expansion using NCERT's own presentation, stopping at the cubic term as the chapter does. Answer: $\left(1-\dfrac{x}{2}\right)^{-1} = 1 + \dfrac{x}{2} + \dfrac{x^2}{4} + \dfrac{x^3}{8} + \cdots$ , valid for $|x| < 2$ .

Board level - sum to infinity of a geometric progression with a negative ratio

Find the sum to infinity of the G.P. $-\dfrac54, \dfrac{5}{16}, -\dfrac{5}{64}, \ldots$

Step 1. Identify the first term and common ratio: $a = -\dfrac54$ , and $r = \dfrac{\text{second term}}{\text{first term}} = \dfrac{5/16}{-5/4} = \dfrac{5}{16}\times\left(-\dfrac45\right) = -\dfrac14$ . Confirm the pattern continues: the ratio of the third term to the second should also be $-\tfrac14$ , and indeed $\dfrac{-5/64}{5/16} = -\dfrac{5}{64}\times\dfrac{16}{5} = -\dfrac14$ , matching.

Step 2. Check the convergence condition before applying the formula: $|r| = \left|-\dfrac14\right| = \dfrac14 < 1$ , so the sum to infinity in equation (4) applies and a finite answer exists.

Step 3. Substitute directly into $S = \dfrac{a}{1-r}$ : $S = \dfrac{-5/4}{1-\left(-\tfrac14\right)} = \dfrac{-5/4}{1+\tfrac14} = \dfrac{-5/4}{5/4}$ .

Step 4. Simplify the resulting fraction: $\dfrac{-5/4}{5/4} = -1$ , since the two quarters cancel and only the sign and the equal numerators remain. Answer: the sum to infinity is $S = -1$ .

Higher-order thinking (HOTS) - coefficient of x squared in the expansion of e to the power 2x plus 3

Find the coefficient of $x^2$ in the expansion of $e^{2x+3}$ as a series in powers of $x$ .

Step 1. The cleanest route is to split the exponent using the index law $e^{2x+3} = e^3 \cdot e^{2x}$ , since $e^3$ is a fixed constant that can be pulled entirely outside the series, leaving only $e^{2x}$ to expand using equation (6).

Step 2. Expand $e^{2x}$ using equation (6) with $x$ replaced by $2x$ : $e^{2x} = 1 + \dfrac{2x}{1!} + \dfrac{(2x)^2}{2!} + \dfrac{(2x)^3}{3!} + \cdots = 1 + 2x + \dfrac{4x^2}{2} + \cdots = 1 + 2x + 2x^2 + \cdots$ . The term containing $x^2$ here is exactly $\dfrac{(2x)^2}{2!} = \dfrac{4x^2}{2} = 2x^2$ .

Step 3. Multiply back by the constant factor $e^3$ pulled out in Step 1: $e^{2x+3} = e^3\left(1 + 2x + 2x^2 + \cdots\right) = e^3 + 2e^3 x + 2e^3 x^2 + \cdots$ . Reading off the coefficient of $x^2$ directly from this product gives $2e^3$ .

Step 4. Cross-check against NCERT's own (longer) method of substituting $(2x+3)$ directly into the series for $e^x$ and isolating the coefficient of $x^2$ inside the general term $\dfrac{(3+2x)^n}{n!}$ , summed over all $n \geq 2$ — that method also collapses, after using $n! = n(n-1)(n-2)!$ , to the same sum $2\left(1 + \dfrac{3}{1!} + \dfrac{3^2}{2!} + \cdots\right) = 2e^3$ , confirming the shortcut in Steps 1–3 is exact, not an approximation. Answer: the coefficient of $x^2$ in the expansion of $e^{2x+3}$ is $2e^3$ . Timing note: budget about two minutes. The factoring shortcut $e^{2x+3} = e^3 \cdot e^{2x}$ is dramatically faster than substituting $(2x+3)$ directly into the series and hunting for the coefficient of $x^2$ inside an infinite sum over $n$ — recognise instantly that any exponent of the form (constant + variable term) should be split by the index law before expanding, never expanded as a single messy binomial-inside-exponential unless the question specifically forces it.

Common traps

Forgetting the convergence condition entirely. The formulas $(1+x)^m = \sum \ldots$ , $S = \dfrac{a}{1-r}$ , and $\log_e(1+x) = \sum \ldots$ are conditional statements, not universal identities — each one is only true under its stated restriction ( $|x|<1$ , $|r|<1$ , $|x|<1$ respectively). Quoting the formula without checking the condition is the single most common error in this chapter.
Confusing the binomial series with the binomial theorem. For a positive integer index the expansion terminates after $n+1$ terms; for any other index it never terminates, and writing " $\ldots$ " followed by silence, without stating the validity condition, is an incomplete answer.
Getting the general term's factor count wrong. In $\dfrac{m(m-1)\cdots(m-r+1)}{r!}$ , the last factor is $(m-r+1)$ , giving exactly $r$ factors in the numerator — writing $(m-r)$ instead by analogy with $^nC_r$ silently shifts the whole coefficient by one factor and gives a wrong answer.
Mixing up $a$ and $r$ or misreading which term is "first." In sum-to-infinity problems the ratio must be computed as (second term) divided by (first term), consistently, and a sign error in $r$ (as in the second worked example above) is easy to miss if the calculation is rushed.
Assuming $e$ is "just" an irrational number to be looked up. The chapter's entire point in Section A.1.4 is that $2 < e < 3$ is provable from the series definition by comparison with a geometric series — treating this as a fact to memorise rather than a result to derive misses the chapter's real content.

Common trapThe single biggest misconception: an infinite series formula is only valid where its convergence condition holds

Every headline formula in this chapter — the binomial series, the sum to infinity, the logarithmic series — is stated together with a strict condition on the variable ( $|x|<1$ or $|r|<1$ ). Students routinely treat these formulas as unconditional algebraic identities and plug in any value they like. The chapter's own counterexample makes the danger vivid: substituting $x=-2$ , $m=-2$ into the binomial series (ignoring $|x|<1$ ) produces the false statement $1 = 1+4+12+\cdots$ . Before using any formula from this appendix, the very first check must be whether the stated condition on the variable is satisfied — if it is not, the formula simply does not apply, and no algebraic manipulation can rescue it.

You should be able to…

State the general Binomial Theorem $(1+x)^m = 1 + mx + \dfrac{m(m-1)}{1\cdot2}x^2+\cdots$ for any real index $m$ , and quote the validity condition $|x|<1$ without being asked.
Expand $(a+b)^m$ for a negative or fractional index by factoring out $a^m$ and applying the theorem to $\left(1+\tfrac{b}{a}\right)^m$ , correctly identifying the condition $|b|<|a|$ .
Recall and apply, without re-derivation, the four standard special-case expansions of $(1+x)^{-1}$ , $(1-x)^{-1}$ , $(1+x)^{-2}$ , and $(1-x)^{-2}$ .
Derive the sum-to-infinity formula $S=\dfrac{a}{1-r}$ from the finite geometric sum formula by reasoning about the limiting behaviour of $r^n$ , and state clearly why the condition $|r|<1$ is essential.
Compute the sum to infinity of any given infinite G.P., after first checking that $|r|<1$ .
Explain, using the comparison-with-a-geometric-series argument, why $2 < e < 3$ , and state the general exponential series $e^x = \sum \dfrac{x^n}{n!}$ .
Use the exponential series to extract a specific coefficient from an expansion such as $e^{2x+3}$ , ideally via the index-law shortcut $e^{a+b}=e^a e^b$ .
State the logarithmic series $\log_e(1+x) = x-\dfrac{x^2}{2}+\dfrac{x^3}{3}-\cdots$ for $|x|<1$ , and recall the special identity $\log_e 2 = 1-\tfrac12+\tfrac13-\cdots$ obtained at $x=1$ .
Recognise, in any problem from this chapter, that quoting a series formula is incomplete without also checking or stating its convergence condition.

Ask the tutor: Want to see how this appendix's convergence conditions turn into the formal limit tests you will meet in Class 12 calculus and beyond?

Why does the binomial series need |x| less than 1 when the original binomial theorem for a positive integer index has no such restriction at all

The two situations differ in length. When the index is a positive integer $n$ , the expansion of $(1+x)^n$ always has exactly $n+1$ terms — it terminates naturally because the factor $(n-n)=0$ appears in every coefficient beyond the $n$ th term, killing them all. There is nothing to "converge," so no restriction on $x$ is needed. When the index $m$ is negative or fractional, no factor in $m(m-1)(m-2)\cdots$ is ever exactly zero, so the series has infinitely many nonzero terms, and an infinite sum only settles to a finite value if its terms shrink fast enough — guaranteed precisely when $|x| < 1$ , as the chapter's own $x=-2, m=-2$ counterexample shows.

Is Euler's number e the same e that appears in e to the x expansion and in compound interest problems

Yes, it is exactly the same constant, defined here as $e = 1 + \dfrac{1}{1!}+\dfrac{1}{2!}+\dfrac{1}{3!}+\cdots \approx 2.71828$ . The general exponential series $e^x = 1+\dfrac{x}{1!}+\dfrac{x^2}{2!}+\cdots$ reduces to this definition exactly when $x=1$ . This same number also arises as the limit of $\left(1+\dfrac1n\right)^n$ as $n\to\infty$ in compound-interest and growth problems, and later, in calculus, as the unique base for which the exponential function is its own derivative. All these routes converge on the same irrational number; this appendix gives you the series definition and proves the bound $2<e<3$ rigorously from it.

Since this whole chapter is beyond the core syllabus, is it safe to skip it entirely before the board exam

For a board examination focused strictly on the main chapters, yes — this appendix is explicitly enrichment material and questions from it are not typically part of the standard board paper. However, it is genuinely worth reading rather than skipping outright, because it is the direct conceptual bridge to calculus and higher mathematics: the idea of a series only "working" under a convergence condition, the technique of bounding one series by comparing it to another, and the number $e$ itself all reappear immediately in Class 12 calculus and in any competitive exam that tests mathematical reasoning beyond rote formula recall. Treat it as a preview, not a burden.

How is the sum to infinity formula S equals a over 1 minus r different from just adding up a lot of terms on a calculator

Adding many terms on a calculator only ever gives you an approximation — however many terms you add, you have technically stopped short of infinity, and you cannot be certain how close your running total is to the true limit without more theory. The formula $S = \dfrac{a}{1-r}$ , by contrast, is an exact value, derived rigorously by showing that the "leftover" piece of the finite-sum formula, $\dfrac{ar^n}{1-r}$ , genuinely shrinks to exactly zero as $n\to\infty$ whenever $|r|<1$ — not approximately zero, but the limiting value zero. This is the crucial difference between numerical approximation and a proven, closed-form limit, and it is why the chapter insists on the derivation rather than just quoting the final formula.