Source: Calc8 – MATH 104 Final Exam Study Guide, Penn State
Calculus rests on a precise definition of "limit," which took centuries to nail down. The (ε, δ)-definition, formalised by Weierstrass, replaced hand-wavy infinitesimals with a rigorous framework: for every tolerance ε you set on the output, there exists a distance δ on the input that keeps you within that tolerance. This study guide covers the key historical texts, the formal limit definition, continuity, and the geometric spaces (Euclidean and complex) where these ideas live.
Difficulty: Introductory to Intermediate | Prerequisites: Comfort with basic algebra and function notation. Familiarity with the intuitive idea of a limit ("what value does f(x) approach?") is helpful but not required.
This material sits right at the start of a Calculus 1 course, laying the groundwork everything else builds on. Before you can differentiate or integrate, you need to know what a limit is, and before limits make sense you need to understand continuity and the spaces in which functions operate. The (ε, δ)-definition is the centrepiece: it turns the vague notion of "getting closer" into something you can prove theorems with. If you missed the first few weeks or are rusty on what "rigour" means in calculus, start here. The historical context (Berkeley, Cauchy, Weierstrass) shows up in exams more often than students expect.
The Analyst
A 1734 book by Bishop George Berkeley criticising the use of infinitesimals in Newton's calculus as logically unrigorous. In simple terms, Berkeley said: "You can't just wave away infinitely small quantities and call it mathematics."
Cours d'Analyse
A foundational textbook by Augustin-Louis Cauchy, published in 1821. It introduced a definition of continuity using infinitesimals and an early prototype of the (ε, δ)-definition of limit. Think of it as the bridge between the old hand-wavy calculus and the modern rigorous version.
Continuity (continuous function)
A function where small changes in the input produce small changes in the output. In simple terms, you can draw its graph without lifting your pen. More precisely, a function f is continuous at a point c if the limit of f(x) as x approaches c equals f(c).
(ε, δ)-definition of limit
The formal way of saying "f(x) gets arbitrarily close to L as x approaches c." For every positive number ε (the error tolerance on the output), there exists a positive number δ (how close x must be to c) such that whenever 0 < |x - c| < δ, we have |f(x) - L| < ε. Think of ε as the acceptable error and δ as the distance you need to stay within to guarantee that error.
Weierstrass
Karl Weierstrass, a 19th-century German mathematician known as the "father of modern analysis." He formalised the concept of limit and eliminated the reliance on infinitesimals, putting calculus on a rigorous logical foundation.
Euclidean space
The standard geometric space of classical geometry: the 2D plane, 3D space, and their higher-dimensional generalisations. Every point in n-dimensional Euclidean space is determined by n coordinates. Think of it as the "default" space where most Calculus 1 problems live.
Complex plane (z-plane)
A geometric representation of complex numbers (numbers of the form a + bi). The real part maps to the x-axis, the imaginary part maps to the y-axis. Think of it as a modified Cartesian plane where the y-axis measures the imaginary component instead of a second real variable.
Berkeley's The Analyst (1734) challenged Newton's use of infinitesimals, arguing they lacked logical justification
Cauchy's Cours d'Analyse (1821) introduced early formal definitions of continuity and a prototype epsilon-delta limit
Weierstrass (19th century) completed the project: he eliminated infinitesimals entirely and formalised limits using the (ε, δ)-definition that is standard today
The progression matters: Berkeley exposed the problem, Cauchy proposed a partial fix, Weierstrass delivered the rigorous solution
The statement "lim(x→c) f(x) = L" means: for every ε > 0, there exists a δ > 0 such that if 0 < |x - c| < δ, then |f(x) - L| < ε
ε controls the output: it sets how close f(x) must be to L
δ controls the input: it sets how close x must be to c
δ depends on ε (and on the function f and the point c). The claim is that no matter how small you make ε, a suitable δ exists
The condition 0 < |x - c| is critical: x must be near c but not equal to c. The limit does not care about the value of f at c itself
Cauchy's mnemonic: ε stands for "error" (how far f(x) is from L), δ stands for "distance" (how far x is from c)
For functions of several variables, the (ε, δ)-definition generalises naturally
In two dimensions, δ becomes the radius of a disc centred at the point of interest
In three dimensions, δ becomes the radius of a sphere
In general, δ defines an open ball of radius δ around the limit point, and every point inside that ball (except the centre) must map within ε of L
Euclidean space is the standard setting for Calculus 1: the real number line (1D), the plane (2D), or 3D space. Points are specified by their coordinates
The complex plane extends the idea to complex numbers a + bi, with the real part on the x-axis and the imaginary part on the y-axis
Both spaces use the same distance notion (the Euclidean metric), which is why the (ε, δ)-framework transfers cleanly between them
Formal (ε, δ)-definition of limit
\lim_{x \to c} f(x) = L \iff \forall\, \varepsilon > 0,\; \exists\, \delta > 0 \text{ such that } 0 < |x - c| < \delta \implies |f(x) - L| < \varepsilonContinuity at a point
A function f is continuous at c if and only if:
\lim_{x \to c} f(x) = f(c)This requires three things: f(c) is defined, the limit exists, and the limit equals f(c).
Distance in Euclidean space (n dimensions)
d(\mathbf{x}, \mathbf{y}) = \sqrt{(x_1 - y_1)^2 + (x_2 - y_2)^2 + \cdots + (x_n - y_n)^2}Complex number on the z-plane
A complex number z = a + bi is plotted at the point (a, b), where a is the real part and b is the imaginary part.
The (ε, δ)-framework is not just exam material. Engineers use it when specifying tolerances: "the output voltage must stay within ε of the target, so the input signal must be kept within δ of its set point." This is exactly the limit definition applied to physical systems.
Continuity matters in control systems and signal processing. A discontinuous control function means the system jumps unpredictably, so engineers design continuous controllers precisely to avoid that. The complex plane underpins electrical engineering (AC circuit analysis uses complex impedance) and quantum mechanics.
Students often think the limit of f(x) as x→c depends on the value of f(c). It does not. The limit only cares about what happens near c, not at c. A function can have a limit at a point where it is undefined.
Students confuse ε and δ. Remember: ε is your output tolerance (how close f(x) must be to L), δ is your input constraint (how close x must be to c). ε is chosen first by someone challenging you; δ is your response.
"Infinitesimals are wrong" is an oversimplification. Berkeley's critique was about the lack of logical rigour, not that infinitesimals are inherently useless. Modern non-standard analysis (Abraham Robinson, 1960s) made infinitesimals rigorous. But for MATH 104, the (ε, δ) approach is the standard.
Students sometimes think "continuous" means "smooth" (no sharp corners). A function can be continuous everywhere but have corners, like f(x) = |x|. Continuity only requires that the limit equals the function value, not that the derivative exists.
⚠️ Be able to state the (ε, δ)-definition of limit from memory. This is a near-certain exam question in MATH 104.
⚠️ Know the three conditions for continuity at a point: f(c) is defined, the limit exists, and the limit equals f(c). Expect a question asking you to identify which condition fails for a given function.
⚠️ Historical names come up as identification questions: Berkeley wrote The Analyst, Cauchy wrote Cours d'Analyse, Weierstrass formalised the modern limit definition.
⚠️ Understand the roles of ε and δ well enough to use them in a simple proof. Even if the exam does not ask for a full epsilon-delta proof, partial credit questions often test whether you can set up the framework correctly.
⚠️ The distinction between the complex plane and Euclidean space may appear as a conceptual or matching question.
True or False: The (ε, δ)-definition of limit requires that f(c) be defined.
Answer: False. The limit concerns values of f(x) for x near c, not at c.
Fill in the blank: In the (ε, δ)-definition, ε represents the ________ and δ represents the ________.
Answer: ε represents the error (output tolerance); δ represents the distance (input constraint).
True or False: Weierstrass introduced infinitesimals into calculus.
Answer: False. Weierstrass eliminated infinitesimals and replaced them with the rigorous (ε, δ) limit framework.
Fill in the blank: Every point in three-dimensional Euclidean space is determined by ________ coordinates.
Answer: Three.
True or False: The complex plane uses the y-axis to represent the imaginary part of a complex number.
Answer: True.
Q: State the (ε, δ)-definition of the limit of f(x) as x approaches c equals L.
A: For every ε > 0, there exists a δ > 0 such that if 0 < |x - c| < δ, then |f(x) - L| < ε.
Q: What was Berkeley's main criticism in The Analyst, and why did it matter for the development of calculus?
A: Berkeley argued that Newton's use of infinitesimals was logically unrigorous. This criticism pushed mathematicians (Cauchy, then Weierstrass) to develop a formal, limit-based foundation for calculus.
Q: What three conditions must hold for a function f to be continuous at a point c?
A: (1) f(c) must be defined. (2) The limit of f(x) as x approaches c must exist. (3) The limit must equal f(c).
Q: Explain the relationship between ε and δ in the formal limit definition. Which one is chosen first, and by whom?
A: ε is chosen first, by someone "challenging" you to prove the limit. It represents the desired closeness of f(x) to L. You must then find a δ (depending on ε, f, and c) such that inputs within δ of c produce outputs within ε of L. The claim is that this works for every ε, no matter how small.
Q: How does the (ε, δ)-definition extend to functions of more than one variable?
A: The distance |x - c| is replaced by the Euclidean distance from the input point to the limit point. δ then defines an open ball (a disc in 2D, a sphere in 3D) around the limit point, and every point inside that ball (except the centre) must map within ε of L.
Q: What is the complex plane, and how does it relate to the Cartesian plane?
A: The complex plane is a modified Cartesian plane where the x-axis represents the real part of a complex number a + bi and the y-axis represents the imaginary part. A complex number z = a + bi is plotted at the point (a, b).
This material connects directly to differentiation: the derivative is defined as a limit, so without the (ε, δ)-framework you cannot rigorously define f'(x). It also connects to the Intermediate Value Theorem and the Extreme Value Theorem, both of which require continuity as a hypothesis. Later in the course, integration (Riemann sums) uses limits of sums, which again rely on the same epsilon-delta logic.
Epsilon-delta definition, ε-δ definition, formal definition of limit, limit definition calculus, continuity definition, continuous function, Weierstrass limit, Cauchy Cours d'Analyse, Berkeley The Analyst, infinitesimals critique, Euclidean space coordinates, complex plane z-plane, Argand diagram, real axis imaginary axis, MATH 104 Penn State, Calculus 1 final exam, limit proof, rigorous calculus, foundations of calculus, epsilon error delta distance