The Divergence Test

The fastest thing you can ever check about a series \sum_{n=1}^{\infty} a_n is whether its terms even bother to shrink. If they do not march to zero, the series has no hope of settling on a finite total. This is the nth-term test for divergence — a one-line sanity check that kills off hopeless series before you waste any effort on them.

\text{If } \sum_{n=1}^{\infty} a_n \text{ converges, then } \lim_{n\to\infty} a_n = 0.

Read the other way around (the contrapositive), it becomes a test you can actually apply: if the terms fail to die out, the series diverges.

\lim_{n\to\infty} a_n \ne 0 \;\;(\text{or the limit fails to exist}) \;\;\Longrightarrow\;\; \sum_{n=1}^{\infty} a_n \text{ diverges.}

Why convergence forces the terms to zero

The whole proof is a single clean cancellation. Write s_n = a_1 + a_2 + \dots + a_n for the partial sum, and suppose the series converges to a finite limit L, meaning s_n \to L.

Step 1 — recover a single term from neighbouring partial sums. Adding one more term takes s_{n-1} to s_n, so the difference is exactly the term itself:

a_n = s_n - s_{n-1}.

Step 2 — take the limit of both sides. Let n \to \infty. The left side becomes \lim_{n\to\infty} a_n; the right side is a limit of a difference:

\lim_{n\to\infty} a_n = \lim_{n\to\infty} \big( s_n - s_{n-1} \big).

Step 3 — split the limit of the difference. Both pieces converge, so the limit distributes across the subtraction:

= \lim_{n\to\infty} s_n - \lim_{n\to\infty} s_{n-1}.

Step 4 — both partial-sum limits are the same number. As n \to \infty, the index n-1 also runs off to infinity, so s_{n-1} chases the very same limit L:

= L - L.

Step 5 — collect. The two copies of L annihilate:

\lim_{n\to\infty} a_n = L - L = 0.

So convergence forces the terms to zero. Contrapose it and you have the test: terms that do not vanish guarantee a divergent series.

Let \sum_{n=1}^{\infty} a_n be a series. Then:

Heads up — the test is one-directional. Terms going to zero is necessary for convergence, but nowhere near sufficient. The arrow only runs one way:

\sum a_n \text{ converges} \;\Longrightarrow\; a_n \to 0, \qquad\text{but}\qquad a_n \to 0 \;\not\Longrightarrow\; \sum a_n \text{ converges.}

The most famous trap in all of analysis is the harmonic series,

\sum_{n=1}^{\infty} \frac{1}{n} = 1 + \frac12 + \frac13 + \frac14 + \frac15 + \cdots.

Its terms \tfrac1n \to 0 beautifully — the divergence test is stone silent on it. Yet the sum is +\infty. Oresme's medieval argument groups the terms in blocks whose lengths double, and bounds each block below by \tfrac12:

\underbrace{\tfrac13 + \tfrac14}_{>\,2\cdot\frac14 = \frac12} + \underbrace{\tfrac15 + \tfrac16 + \tfrac17 + \tfrac18}_{>\,4\cdot\frac18 = \frac12} + \cdots

Each bracket exceeds \tfrac12, and there are infinitely many brackets, so the partial sums climb past every bound — slowly, but without limit. Terms shrinking to zero is the price of admission to convergence, not a ticket to it.

Watch a series fail the test

Take terms a_n = L + \tfrac{1}{n}, which settle towards a chosen non-zero limit L. As long as L \ne 0 the terms never die out, and the partial sums s_n = \sum_{k=1}^{n} a_k climb away roughly like L\,n. Slide L to zero and the runaway slows to the gentle (still divergent!) creep of the harmonic series.