The mathematics of change — what differential equations are, how they are classified, what their solutions mean, and how slope fields reveal behavior before a formula is found
This chapter opens the study of differential equations by explaining what differential equations are, why they matter, and how they connect to the calculus you already know. Before introducing solution techniques, this chapter builds understanding. You will learn to recognize differential equations, classify them, interpret their solutions as functions, and begin to see how they describe real systems in science, engineering, biology, economics, and many other fields.
Differential equations should not feel like a new subject disconnected from calculus. They are a natural continuation of it. Calculus studies rates of change and accumulation. Differential equations use those same ideas to describe relationships between changing quantities, and then ask: given this relationship between a function and its own rate of change, what function satisfies it? That question, simple to state but rich in consequence, is what differential equations study.
Suppose you know that the rate at which a population grows is proportional to the size of the population. Larger populations grow faster; smaller populations grow slower. You want to find a function $P(t)$ that describes the population at every time $t$. You are not given $P(t)$ directly. You are given a relationship: $\dfrac{dP}{dt} = kP$, where $k$ is a positive constant. This is a differential equation. It does not say what $P$ is. It says something about how $P$ changes. This chapter prepares you to understand what such an equation means, how its solutions behave, and what questions you will learn to answer throughout this book.
Calculus gives you two fundamental tools. Differentiation finds rates of change. Integration recovers quantities from rates of change. With these tools, you can describe how things move, grow, shrink, oscillate, heat up, cool down, and respond to forces.
But calculus, as you first learned it, usually starts with a known function and asks you to differentiate or integrate it. Differential equations reverse this situation. Instead of starting with a known function, you are given a relationship that the unknown function must satisfy. The relationship involves the function and one or more of its derivatives. Your task is to find the function itself.
Consider a simple example. Suppose you know that some quantity $y$ changes at a rate equal to the quantity itself:
This equation does not tell you what $y$ is. It tells you that the rate of change of $y$ equals $y$. Any function satisfying this relationship is a solution. The function $y = e^x$ satisfies it, because its derivative equals itself. More generally, $y = Ce^x$ for any constant $C$ satisfies it, as you can verify.
This is the essential picture of what differential equations study: relationships between unknown functions and their derivatives, and the task of finding functions that satisfy those relationships.
A differential equation is an equation that involves an unknown function and one or more of its derivatives.
The dependent variable is the unknown function being solved for. The independent variable is the variable it depends on.
In $\dfrac{dP}{dt} = kP$, the dependent variable is $P$ (population) and the independent variable is $t$ (time).
| Quantity | Function Notation | Derivative | Meaning of Derivative |
|---|---|---|---|
| Population | $P(t)$ | $\dfrac{dP}{dt}$ | Rate of growth or decay |
| Temperature | $T(t)$ | $\dfrac{dT}{dt}$ | Rate of heating or cooling |
| Position | $x(t)$ | $\dfrac{dx}{dt}$ | Velocity |
| Velocity | $v(t)$ | $\dfrac{dv}{dt} = \dfrac{d^2x}{dt^2}$ | Acceleration |
| Charge | $q(t)$ | $\dfrac{dq}{dt}$ | Current |
| Concentration | $C(t)$ | $\dfrac{dC}{dt}$ | Rate of change of concentration |
In real systems, the rate of change of a quantity often depends on the current state of the system. Population grows faster when there are more individuals. A hot object cools faster when the temperature difference is larger. A drug leaves the bloodstream faster when its concentration is higher. These observations, written mathematically, become differential equations.
| Physical Principle | Differential Equation |
|---|---|
| Newton's Law of Cooling | $\dfrac{dT}{dt} = -k(T - T_{\text{amb}})$ |
| Exponential growth | $\dfrac{dN}{dt} = rN$ |
| Simple harmonic motion | $\dfrac{d^2x}{dt^2} = -\omega^2 x$ |
The central challenge of differential equations is this: you are given information about the derivative of a function, and you must recover the function. In the simplest case you already know how to do this. If $\frac{dy}{dx} = 2x$, direct integration gives $y = x^2 + C$. But when the right side involves $y$ itself, direct integration no longer suffices. The equation $\frac{dy}{dx} = y$ connects the rate of change of $y$ to the value of $y$, and a different strategy is needed.
| Equation | Structure | Solution Family |
|---|---|---|
| $\dfrac{dy}{dx} = 3x^2$ | Derivative only on right; direct integration works | $y = x^3 + C$ |
| $\dfrac{dy}{dx} = y$ | Derivative equals the function itself | $y = Ce^x$ |
| $\dfrac{dy}{dx} = xy$ | Derivative equals product of $x$ and $y$ | $y = Ce^{x^2/2}$ |
| $y'' + y = 0$ | Second derivative plus function equals zero | $y = A\sin x + B\cos x$ |
| $y'' + 2y' + y = e^x$ | Second-order with forcing term | Covered in Chapter 8 |
When you solve a differential equation, the solution is a function, not a number. Finding the solution means finding a formula that works for every point in the function's domain. This is a fundamental shift from algebra.
The first major classification separates equations involving functions of a single variable from equations involving functions of several variables.
An ordinary differential equation (ODE) involves a function of a single independent variable and its ordinary derivatives. All of Chapters 1–12 concern ODEs.
A partial differential equation (PDE) involves a function of several independent variables and its partial derivatives. PDEs are introduced as a preview in Chapter 13.
| Equation | Type | Description |
|---|---|---|
| $\dfrac{dP}{dt} = kP$ | ODE | Population growth; one variable $t$ |
| $m\dfrac{d^2x}{dt^2} + kx = 0$ | ODE | Mass-spring; one variable $t$ |
| $\dfrac{\partial u}{\partial t} = \alpha \dfrac{\partial^2 u}{\partial x^2}$ | PDE | Heat equation; variables $x$ and $t$ |
| $\dfrac{\partial^2 u}{\partial x^2} + \dfrac{\partial^2 u}{\partial y^2} = 0$ | PDE | Laplace's equation; variables $x$ and $y$ |
For ODEs, both Leibniz notation $\left(\dfrac{dy}{dx},\, \dfrac{d^2y}{dx^2}\right)$ and prime notation $(y', y'')$ are common throughout this book. In physics and engineering, dot notation $(\dot x, \ddot x)$ is also used for time derivatives. PDEs use the stylized $\partial$ to distinguish partial derivatives from ordinary ones.
The order of a differential equation is the order of the highest derivative present.
The degree is the power to which the highest-order derivative is raised, after the equation has been put in polynomial form in its derivatives.
| Equation | Order | Degree | Reason |
|---|---|---|---|
| $y' = ky$ | 1st | 1st | Highest derivative is $y'$, appears once |
| $y'' - 3y' + 2y = 0$ | 2nd | 1st | Highest derivative is $y''$, appears once |
| $\left(\dfrac{d^2y}{dx^2}\right)^2 + y = 0$ | 2nd | 2nd | $y''$ is raised to the second power |
| $\left(\dfrac{dy}{dx}\right)^3 = y$ | 1st | 3rd | $y'$ is raised to the third power |
| $y''' + xy'' - y = e^x$ | 3rd | 1st | Highest derivative is $y'''$, appears once |
Order and degree are not the same. The equation $\left(y'\right)^3 = y$ is first-order (the highest derivative is $y'$) but third-degree (that derivative is cubed). Do not confuse the exponent on the derivative with the order of the derivative.
The order tells you how many arbitrary constants appear in the general solution, and therefore how many initial or boundary conditions are needed to pin down a unique particular solution. A first-order equation needs one condition; a second-order equation needs two.
After determining order, the next major classification is linear versus nonlinear. This distinction determines which solution methods are available.
A differential equation is linear if the unknown function and all its derivatives appear to the first power only, no products of $y$ with its derivatives appear, and coefficients are functions of the independent variable only — not of $y$.
General linear first-order form: $a_1(x)\,y' + a_0(x)\,y = f(x)$
General linear second-order form: $a_2(x)\,y'' + a_1(x)\,y' + a_0(x)\,y = f(x)$
| Equation | Linear? | Reason |
|---|---|---|
| $y' + 2y = e^x$ | Yes | $y$ and $y'$ each appear to the first power |
| $y'' - 3y' + 2y = 0$ | Yes | All derivatives appear linearly; constant coefficients |
| $x^2 y'' + xy' - y = \sin x$ | Yes | Coefficients are functions of $x$, not $y$ |
| $y' = y^2$ | No | $y$ appears squared |
| $y'' + \sin(y) = 0$ | No | $\sin(y)$ is nonlinear in $y$ |
| $y\,y' = x$ | No | Product of $y$ with its derivative |
In this context, linear refers to the algebraic structure of how the unknown function and its derivatives appear in the equation — not to whether the solution curves are straight lines. A linear differential equation may have solutions that oscillate, grow exponentially, or decay. The linearity is in the equation's structure, not the shape of its solutions.
Linear equations satisfy a superposition principle: if $y_1$ and $y_2$ are both solutions to the homogeneous version, then $c_1 y_1 + c_2 y_2$ is also a solution. This powerful property allows systematic construction of general solutions and underlies all the methods in Chapters 3 through 10.
For each equation, state: (a) ODE or PDE, (b) order, (c) degree, (d) linear or nonlinear.
A solution to a differential equation is a function that, when substituted into the equation along with its required derivatives, makes the equation true for all values of the independent variable in some interval. Verifying a proposed solution always requires substitution, not just plausibility.
y = e^(2x), so y' = 2e^(2x)
y' − 2y = 2e^(2x) − 2·e^(2x) = 0 ✓
y = sin(x), y' = cos(x), y'' = −sin(x)
y'' + y = −sin(x) + sin(x) = 0 ✓
y = x², y' = 2x
y' − 2y = 2x − 2x² = 2x(1 − x)
Always verify your solutions by substitution. Even when you derive an answer using a correct method, substituting back into the original equation is the most reliable way to catch computational errors. This takes one minute and should become automatic.
The general solution of a differential equation is an expression describing the entire family of solutions, typically involving one or more arbitrary constants. A first-order equation has one arbitrary constant $C$; a second-order equation has two, often $C_1$ and $C_2$.
A particular solution is a specific member of the family, obtained when the constants are assigned specific values.
The general solution to $\dfrac{dy}{dx} = ky$ is $y = Ce^{kx}$. You can verify: $\dfrac{d}{dx}(Ce^{kx}) = Cke^{kx} = k(Ce^{kx}) = ky$ for any constant $C$. Each choice of $C$ gives a different solution curve, but every curve in the family satisfies the equation.
3 = C·e^(−0) = C·1 → C = 3
Particular solution: y = 3e^(−x)
| Type | Description | Example | Use |
|---|---|---|---|
| Initial condition | Value of solution (or derivative) specified at a single point | $y(0) = 5$ | Initial value problems; time evolution forward from a known state |
| Boundary condition | Values specified at two or more different points | $y(0) = 0,\; y(\pi) = 0$ | Boundary value problems; spatial problems with constraints at endpoints |
An initial value problem (IVP) pairs a differential equation with initial conditions. For a first-order equation, one initial condition determines a unique solution. For a second-order equation, two initial conditions are required.
y(0) = Ce^0 = C = 4 → C = 4
Particular solution: y = 4e^(−2x)
Initial conditions specify values at a single point and are associated with time-evolution problems. Boundary conditions specify values at multiple points and arise when a system has constraints at two or more locations in space or time. The theory of these two problem types differs significantly, and boundary value problems can have zero, one, or infinitely many solutions even when an IVP has exactly one.
One of the most important geometric ideas in differential equations is the slope field, also called a direction field. For a first-order equation of the form $\dfrac{dy}{dx} = f(x,y)$, the equation assigns a slope value to every point $(x,y)$ in the plane. If you draw a short line segment at each point with that slope, the collection forms the slope field.
The slope field is a field of clues. Each segment says: if a solution curve passes through here, it must be heading in this direction at this moment. You can sketch solution curves by starting at any point and following the local slopes, as if navigating a river by following its current.
Draw a coordinate plane with $x$ from $-2$ to $2$ and $y$ from $-2$ to $2$. At each grid point $(x,y)$, draw a short segment with slope equal to $y$. Along $y = 0$, all segments are horizontal. Along $y = 1$, all segments have slope $1$ (rising at 45°). Along $y = -1$, all segments have slope $-1$ (falling at 45°). Solution curves starting above the $x$-axis curve upward and away; those starting below curve downward. The line $y = 0$ is an equilibrium: it is a horizontal row of horizontal segments, and the solution $y = 0$ satisfies the equation since $0' = 0 = y$.
An equilibrium solution is a constant function $y = c$ satisfying the differential equation. In a slope field, equilibria appear as horizontal rows of horizontal segments. For $y' = g(y)$, equilibria occur where $g(c) = 0$.
A stable equilibrium attracts nearby solution curves. An unstable equilibrium repels them.
1 − y = 0 → y = 1 (equilibrium)
y > 1: y' = 1 − y < 0 (solution decreases toward 1)
y < 1: y' = 1 − y > 0 (solution increases toward 1)
The most important application of differential equations is mathematical modeling: translating a description of a real system into a differential equation, solving or analyzing the equation, and interpreting the result. A model is not complete merely because an equation has been solved. A solution must be interpreted, checked, and connected back to the real system.
1. Define — Identify the dependent variable, independent variable, and parameters. Assign units.
2. State — Describe the rate relationship in words: what makes the quantity increase or decrease, and by how much?
3. Write — Translate the rate statement into a differential equation.
4. Condition — Identify initial or boundary conditions from the problem.
5. Solve or Analyze — Use exact methods, qualitative analysis, or numerical approximation.
6. Interpret — Connect the mathematical solution back to the real system. What does the function describe? What do parameters mean?
7. Check and Refine — Verify units, test limiting behavior, and assess whether the model's assumptions are reasonable.
dP/dt = kP
Initial condition: P(0) = 1000
Differential equations appear in virtually every quantitative field. This section surveys the breadth of applications, not to teach each one in full, but to establish the scope of what you are learning to use.
| Field | Model | Equation |
|---|---|---|
| Mechanics | Free fall with air resistance | $m\dfrac{dv}{dt} = mg - kv$ |
| Mechanics | Simple harmonic oscillator | $m\dfrac{d^2x}{dt^2} + kx = 0$ |
| Circuits | RC circuit (charging) | $R\dfrac{dq}{dt} + \dfrac{q}{C} = V(t)$ |
| Circuits | RLC circuit | $L\dfrac{d^2q}{dt^2} + R\dfrac{dq}{dt} + \dfrac{q}{C} = V(t)$ |
| Biology | Logistic population growth | $\dfrac{dP}{dt} = rP\!\left(1 - \dfrac{P}{K}\right)$ |
| Epidemiology | SIR epidemic model | $\dfrac{dS}{dt} = -\beta SI$, $\;\dfrac{dI}{dt} = \beta SI - \gamma I$ |
| Chemistry | First-order reaction | $\dfrac{d[A]}{dt} = -k[A]$ |
| Finance | Continuous compounding | $\dfrac{dA}{dt} = rA$ |
| Physics | Newton's Law of Cooling | $\dfrac{dT}{dt} = -k(T - T_{\text{amb}})$ |
In every row of the table above, the differential equation is not derived from a known formula for the quantity. It is derived from a physical, biological, chemical, or economic principle about how the quantity changes. The function itself is what you need to find. That is the defining task of differential equations.
Correction: A differential equation is solved for a function, not a number. The solution to $y' = 2y$ is the family of functions $y = Ce^{2x}$.
Correction: There is a family of solutions $y = Ce^{2x}$ for any constant $C$. A unique solution requires additional information, such as an initial condition.
Correction: Direct integration only works for $y' = f(x)$ where the right side does not involve $y$. When $y$ appears on the right, different methods are needed: separation of variables, integrating factors, characteristic equations, Laplace transforms, and more.
Correction: Most differential equations arising from real applications cannot be solved exactly. Numerical methods (Chapter 11) and qualitative analysis (Chapter 12) are essential tools, not fallbacks.
Correction: $C$ represents the entire family of solutions. In applications, $C$ is determined by initial conditions and carries physical meaning: initial population, initial charge, initial temperature difference.
Correction: Slope fields are a primary tool, especially for nonlinear and autonomous equations where exact solutions may not exist. The geometric picture often reveals equilibrium, stability, and long-term behavior directly.
Correction: Order is the highest derivative that appears (here: $y'$, so first-order). Degree is the power of that derivative (here: $3$, so third-degree). This equation is first-order, third-degree.
Correction: Every method can produce computational errors. Always substitute your proposed solution and its derivatives back into the original equation. Verification takes one minute and is the most reliable check available.
This chapter has built the conceptual foundation for the rest of the book. You now have the vocabulary, classification tools, solution-checking method, and a geometric picture from slope fields. Chapter 2 introduces first-order differential equations in depth, developing slope fields and direction fields as the primary geometric tool before any symbolic method is introduced.
As you move into Chapters 2 through 5, you will draw on differentiation rules (product rule, chain rule, implicit differentiation), integration techniques (substitution, integration by parts, partial fractions), and the properties of exponential, logarithmic, and trigonometric functions. If any of these feel uncertain, Appendix A and Appendix B at the back of this textbook provide targeted review.
The question to carry into Chapter 2 is this: given a differential equation, and without solving it explicitly, what can you say about how its solutions behave? Can you identify where solutions grow, where they decay, where they level off, and which equilibria attract or repel nearby curves? Slope fields and direction fields are the tools that will answer these questions.
For each equation, state: (a) ODE or PDE, (b) order, (c) degree, (d) linear or nonlinear.
Verify by substitution that each function is a solution to the given equation. Show all steps.
Define variables, write the differential equation, and state initial conditions.
Each problem contains an error. Identify it and write the correct answer.
What is wrong? What is the correct answer?
What is wrong with the word "only one"?
Identify the error and state the correct classification.
What is the actual order, and what is the actual degree?
What should the student actually conclude?
What is missing, and why?
Find the correct equilibrium and explain how to determine it.
Chapter 1 gave you the vocabulary, classification tools, and geometric picture of differential equations. Chapter 2 deepens the geometric theory: slope fields, direction fields, autonomous equations, equilibrium solutions, stability, and the long-term behavior of solution curves. All of this geometric work comes before any symbolic solution method. By the time you reach Chapter 3 (separation of variables) and Chapter 4 (integrating factors), you will be able to look at a slope field and know immediately whether a proposed symbolic solution makes geometric sense.
The question to carry with you: given a differential equation, what can you say about how solutions behave without solving it explicitly?