The evolution of discrete dynamical systems is governed by the iteration of maps. Given an initial condition $x_0$ and a map $f$, the orbit $x_0, f(x_0), f^2(x_0), \ldots$ traces out the long-time behaviour of the system. Fixed points — solutions to $f(p) = p$ — are the simplest invariant objects, and their stability determines whether nearby orbits are attracted to $p$ or driven away from it. When the derivative $f'(p)$ has modulus strictly different from one, the fixed point is *hyperbolic*, and its stability is resolved entirely by the sign of $|f'(p)| - 1$. Bifurcation theory addresses the complementary case: the *nonhyperbolic* boundary $|f'(p)| = 1$, where the linear approximation is neutral and stability depends on higher-order [derivatives](/page/Derivative). In one dimension, the constraint $f'(p) \in \mathbb{R}$ restricts the critical multiplier to exactly two values — $f'(p) = +1$ and $f'(p) = -1$ — each producing a qualitatively distinct bifurcation. The first leads to the *saddle-node* (tangent) bifurcation, in which two fixed points collide and annihilate. The second leads to the *period-doubling* (flip) bifurcation, in which a fixed point loses stability and a period-2 orbit is born. This page develops the stability theory at both critical multipliers, introduces the Schwarzian derivative as the natural higher-order invariant controlling the period-doubling case, and classifies the generic saddle-node and period-doubling bifurcations for one-parameter families of maps. [motivation] ### Linear Stability: What It Achieves The simplest stability criterion for a fixed point $p$ of a smooth map $f$ is *linear*: compute the multiplier $\lambda = f'(p)$ and check whether $|\lambda| < 1$ (stable) or $|\lambda| > 1$ (unstable). The logic is that near $p$, the map is well-approximated by its linearisation $x \mapsto p + \lambda(x - p)$, which contracts distances when $|\lambda| < 1$ and expands them when $|\lambda| > 1$. For the vast majority of fixed points — those with $|\lambda| \neq 1$ — this is the complete story, and no further analysis is needed. ### The Failure at the Critical [Boundary](/page/Boundary) At $|\lambda| = 1$, the linear approximation preserves distances: $|f(x) - p| \approx |x - p|$. The linearisation is neutral, and it cannot distinguish between attraction, repulsion, or any intermediate behaviour. Yet the actual nonlinear map $f$ need not be neutral — the higher-order terms that the linearisation discards are precisely the terms that decide the outcome. The two concrete cases are $\lambda = +1$ and $\lambda = -1$. When $\lambda = +1$, the linearisation is the identity $x \mapsto x$, so the displacement $f(x) - x$ is entirely governed by the nonlinear remainder. When $\lambda = -1$, the linearisation flips each orbit point to the opposite side of $p$, creating an oscillation $x_0, f(x_0), x_2 \approx x_0, \ldots$ whose amplitude is again controlled by the nonlinear terms. ### Two Maps, Same Multiplier, Different Stability To see that higher-order information is essential, consider two maps with the same critical multiplier $\lambda = +1$ at $p = 0$. The map $f(x) = x + x^2$ satisfies $f'(0) = 1$ and $f''(0) = 2 > 0$. Near $x = 0$, the displacement $f(x) - x = x^2 > 0$ for all $x \neq 0$, so every orbit moves to the right: points to the left of $0$ are pushed toward $0$, while points to the right are pushed away. The fixed point is *semi-stable* — attracting from one side and repelling from the other. Now consider $g(x) = x - x^3$, which also satisfies $g'(0) = 1$ but has $g''(0) = 0$ and $g'''(0) = -6 < 0$. Here $g(x) - x = -x^3$, which is positive for $x < 0$ and negative for $x > 0$. Points on both sides are pushed toward $0$, and the fixed point is *asymptotically stable*. The same multiplier produces qualitatively opposite outcomes. The resolution lies in the first nonvanishing higher-order derivative: the sign of $f''(p)$ when it is nonzero, and the sign of $f'''(p)$ when $f''(p) = 0$. Developing this idea rigorously — and extending it to the oscillatory case $\lambda = -1$ via the Schwarzian derivative — is the purpose of this page. [/motivation] ## Definition The basic objects of study are smooth self-maps of intervals and their fixed points. The regularity requirement $r \ge 3$ ensures that enough derivatives are available for the higher-order stability analysis that follows. [definition: One-Dimensional Map] Let $I \subseteq \mathbb{R}$ be an interval. A **one-dimensional map** is a [function](/page/Function) \begin{align*} f: I &\to I \end{align*} of class $C^r$ with $r \ge 3$. [/definition] A map $f$ generates a discrete dynamical system by iteration: starting from $x_0 \in I$, the orbit is the [sequence](/page/Sequence) $x_0, x_1 = f(x_0), x_2 = f(x_1), \ldots$, with $x_n = f^n(x_0)$. The simplest invariant objects for such a system are points that return to themselves after one application of $f$. [definition: Fixed Point] Let $f \in C^r(I, I)$. A point $p \in I$ is a **fixed point** of $f$ if $f(p) = p$. [/definition] The existence of a fixed point is an algebraic question — one solves $f(x) = x$. But whether a fixed point is *observable* in practice depends on stability: does a small perturbation from $p$ decay back to $p$ or grow away from it? The derivative of $f$ at $p$ provides the first-order measure of this. [definition: Multiplier] Let $p$ be a fixed point of $f \in C^r(I, I)$. The **multiplier** of $p$ is the quantity $\lambda := f'(p)$. The fixed point is **hyperbolic** if $|\lambda| \neq 1$ and **nonhyperbolic** if $|\lambda| = 1$. [/definition] The multiplier governs the local dynamics to first order: near $p$, $f(x) \approx p + \lambda(x - p)$, so the distance $|x - p|$ is scaled by $|\lambda|$ at each iterate. Whether this scaling produces convergence to $p$ is formalised by the following notion. [definition: Asymptotic Stability Of A Fixed Point] A fixed point $p$ of $f$ is **asymptotically stable** if there exists $\epsilon > 0$ such that for every $x_0 \in (p - \epsilon, p + \epsilon) \cap I$, the orbit satisfies $f^n(x_0) \to p$ as $n \to \infty$. The fixed point is **unstable** if no such neighbourhood exists — equivalently, if every neighbourhood of $p$ contains a point whose orbit eventually leaves that neighbourhood. [/definition] Between these two extremes lies a third possibility that arises at the critical multiplier $\lambda = +1$: *semi-stability*, where orbits converge from one side but diverge from the other. Semi-stable fixed points are topologically unstable (they do not satisfy the definition of asymptotic stability), but they are significant because they mark the boundary at which two fixed points collide in a saddle-node bifurcation. ## Linear Stability The first fundamental result confirms the intuition that the multiplier determines stability in the hyperbolic case. When $|\lambda| < 1$, the contraction of the linearisation persists in the full nonlinear map, and when $|\lambda| > 1$, the expansion drives orbits away. The mechanism is elementary — the [Mean Value Theorem](/theorems/186) converts a derivative bound into a geometric contraction or expansion rate — but the result is the foundation on which all subsequent bifurcation analysis rests. [quotetheorem:641] The theorem leaves the boundary case $|\lambda| = 1$ completely open. This is not a deficiency of the proof technique but a genuine indeterminacy: as the motivation examples $f(x) = x + x^2$ and $g(x) = x - x^3$ demonstrate, the same critical multiplier can produce semi-stable, stable, or unstable behaviour depending on higher-order terms. The theorem also says nothing about the *rate* of convergence in the stable case, but the proof gives a quantitative bound: the convergence is at least geometric with rate $k$ for any $k \in (|\lambda|, 1)$. [example: Stability In The Logistic Family] The logistic family \begin{align*} f_\mu: [0, 1] &\to [0, 1] \\ x &\mapsto \mu x(1 - x) \end{align*} with $\mu \in (0, 4]$ provides the simplest illustration. For $\mu > 1$, the nonzero fixed point is \begin{align*} x^* = 1 - \frac{1}{\mu}, \end{align*} obtained by solving $\mu x(1 - x) = x$ and discarding the trivial solution $x = 0$. The multiplier at $x^*$ is \begin{align*} \lambda = f_\mu'(x^*) = \mu(1 - 2x^*) = \mu\left(1 - 2 + \frac{2}{\mu}\right) = 2 - \mu. \end{align*} The stability condition $|\lambda| < 1$ gives $|2 - \mu| < 1$, which holds for $1 < \mu < 3$. At $\mu = 2$, the multiplier is $\lambda = 0$ (superstable); at $\mu = 1$, $\lambda = +1$ (saddle-node); at $\mu = 3$, $\lambda = -1$ (period-doubling). The logistic family thus exhibits both critical multipliers as $\mu$ varies, making it a natural testing ground for the bifurcation theory developed below. [/example] ## The Saddle-Node Bifurcation When the multiplier reaches $+1$, the graph of $f$ is tangent to the diagonal $y = x$ at the fixed point. Geometrically, the graph just touches the diagonal without crossing it transversally — a configuration that is generically destroyed by small perturbations of the map. The curvature of the graph relative to the diagonal, measured by $f''(p)$, determines whether the tangency creates a semi-stable fixed point (the generic case) or, when the curvature also vanishes, whether the cubic term $f'''(p)$ restores full stability or instability. The following theorem makes this analysis precise. The key mechanism is the Taylor expansion of the displacement $f(x) - x$: since $f(p) = p$ and $f'(p) = 1$, the constant and linear terms vanish, and the dynamics near $p$ are controlled by the first surviving higher-order term. [quotetheorem:642] The semi-stability in Case 1 is the hallmark of the saddle-node (tangent) bifurcation. It means that $p$ is not asymptotically stable in any neighbourhood — there is always a direction from which orbits escape — yet $p$ does attract orbits from the opposite side. This one-sided character has a direct bifurcation-theoretic interpretation: as a parameter varies and pushes the multiplier through $+1$, two fixed points (one stable, one unstable) approach each other, merge into a single semi-stable point at the critical parameter value, and then disappear entirely. The non-degeneracy condition $f''(p) \neq 0$ ensures that this collision is generic — a quadratic tangency of the graph with the diagonal. Cases 2 and 3 arise in the degenerate situation $f''(p) = 0$. While less common in one-parameter families, this case is crucial for the period-doubling analysis: when one passes to the second iterate of a map with multiplier $-1$ (see the next section), the resulting map has multiplier $+1$ and automatically satisfies $g''(p) = 0$, so the stability reduces to the sign of $g'''(p)$. [example: Saddle-Node Bifurcation In A Quadratic Family] Consider the one-parameter family \begin{align*} f_\alpha: \mathbb{R} &\to \mathbb{R} \\ x &\mapsto x + \alpha - x^2, \end{align*} so that the fixed point equation $f_\alpha(x) = x$ reduces to $\alpha - x^2 = 0$. For $\alpha > 0$, there are two fixed points $x^*_\pm = \pm\sqrt{\alpha}$. For $\alpha < 0$, no real fixed points exist. At the critical value $\alpha = 0$, a single fixed point $x^* = 0$ is born with multiplier \begin{align*} f_0'(0) = 1 - 2 \cdot 0 = 1 \end{align*} and second derivative $f_0''(0) = -2 \neq 0$. By Case 1 of the theorem, $x^* = 0$ is semi-stable at the critical parameter. For small $\alpha > 0$, the multipliers of the two fixed points are \begin{align*} f_\alpha'(\sqrt{\alpha}) &= 1 - 2\sqrt{\alpha} < 1, \\ f_\alpha'(-\sqrt{\alpha}) &= 1 + 2\sqrt{\alpha} > 1. \end{align*} The fixed point $x^*_+ = \sqrt{\alpha}$ is stable and $x^*_- = -\sqrt{\alpha}$ is unstable. As $\alpha \to 0^+$, these two points collide at the origin, their multipliers both converging to $+1$ from opposite sides. [/example] The conditions under which a one-parameter family of maps exhibits a saddle-node bifurcation can be stated precisely. The three requirements are: the multiplier must reach $+1$ at the critical parameter (criticality), the tangency must be quadratic rather than higher-order (non-degeneracy), and the parameter must actually push the graph through the diagonal (transversality). [definition: Generic Saddle-Node Bifurcation] Let $f_\alpha \in C^3(I, I)$ be a one-parameter family of maps and let $(p, \alpha_0)$ be a fixed point of $f_{\alpha_0}$. The family undergoes a **generic saddle-node bifurcation** at $(p, \alpha_0)$ if the following three conditions hold: 1. **Criticality:** $f_{\alpha_0}'(p) = 1$. 2. **Non-degeneracy:** $f_{\alpha_0}''(p) \neq 0$. 3. **Transversality:** $\dfrac{\partial f_\alpha}{\partial \alpha}\bigg|_{(p, \alpha_0)} \neq 0$. Under these conditions, two fixed points (one stable, one unstable) exist on one side of $\alpha_0$ and no fixed points exist on the other side. [/definition] The non-degeneracy condition ensures that the graph of $f$ has a genuine quadratic tangency with the diagonal at $p$, rather than a higher-order tangency that could produce more exotic behaviour. The transversality condition ensures that varying $\alpha$ actually moves the graph through the diagonal, so that the tangency is not persistent. Together, they guarantee that the local bifurcation diagram consists of a single curve of fixed points that turns around a saddle-node point in the $(x, \alpha)$ plane — a saddle-node bifurcation. ### Other Bifurcations at Multiplier One The saddle-node is the **generic** bifurcation at $\lambda = +1$ — it occurs whenever the non-degeneracy ($f''(p) \neq 0$) and transversality conditions hold. When these conditions fail, two other bifurcation types are possible. **The transcritical bifurcation** occurs when a "structural" fixed point exists for all parameter values. In biological models, the extinction state $x = 0$ is often a fixed point for every value of the growth rate parameter. When a second branch of fixed points passes through this structural fixed point, the two branches **exchange stability** but neither disappears. The normal form is $\bar{x} = x + \mu x - x^2$: for $\mu < 0$, the origin is stable and $x^* = \mu$ is unstable; for $\mu > 0$, the roles reverse. The transversality condition $\frac{\partial f}{\partial \alpha}\big|_{(p,\alpha_0)} \neq 0$ fails because the structural fixed point cannot be moved by the parameter. **The pitchfork bifurcation** occurs when the map has a symmetry $f(-x, \mu) = -f(x, \mu)$. This symmetry forces $f''(p) = 0$ at any symmetric fixed point, so the non-degeneracy condition fails. A symmetric pair of fixed points is born (supercritical) or destroyed (subcritical) at the bifurcation, while the symmetric fixed point persists. The normal form is $\bar{x} = x + \mu x - x^3$ (supercritical) or $\bar{x} = x + \mu x + x^3$ (subcritical). [remark: Persistence of Periodic Orbits] For any bifurcation at $\lambda = +1$ — saddle-node, transcritical, or pitchfork — the multiplier must pass through $+1$ for the bifurcation to occur. This is a consequence of the [implicit function theorem](/page/Implicit%20Function%20Theorem): if $f'(p) \neq 1$ (equivalently $(f(x) - x)' \neq 0$ at $x = p$), then the fixed-point equation $f(x) = x$ has a unique solution near $p$ for all nearby parameter values. Therefore, to prove that a periodic orbit **persists** for all parameter values in some range, it suffices to show that its multiplier can never reach $+1$. This argument rules out all three bifurcation types simultaneously. [/remark] ## The Period-Doubling Bifurcation and the Schwarzian Derivative When the multiplier reaches $-1$, the linearisation $x \mapsto p - (x - p)$ reflects each point across $p$, producing alternating oscillations. Unlike the tangent case, the graph of $f$ now crosses the diagonal transversally at $p$ — the fixed point persists for nearby parameter values — but the nature of the crossing changes: the fixed point switches from attracting to repelling (or vice versa), and a period-2 orbit is born. The stability analysis at $\lambda = -1$ is more subtle than the tangent case. The key insight is that the oscillating orbit $x_0, f(x_0), f^2(x_0) \approx x_0, \ldots$ has the property that the *second iterate* $g := f \circ f$ maps each point approximately back to itself. Since $g'(p) = (f'(p))^2 = 1$, the stability of $p$ under $f$ reduces to the stability of $p$ under $g$ with multiplier $+1$ — a problem already solved by the tangent bifurcation theorem. The resulting criterion is expressed in terms of the Schwarzian derivative, a classical differential expression from complex analysis that arises naturally from computing $g'''(p)$. [definition: Schwarzian Derivative] Let $f \in C^3(I, I)$ with $f'(x) \neq 0$. The **Schwarzian derivative** of $f$ at $x$ is \begin{align*} Sf(x) := \frac{f'''(x)}{f'(x)} - \frac{3}{2}\left(\frac{f''(x)}{f'(x)}\right)^2. \end{align*} [/definition] The Schwarzian derivative is a third-order differential operator that measures the "nonlinear distortion" of $f$ beyond what is captured by the first and second derivatives. It has several remarkable properties: it is invariant under Möbius transformations (linear fractional maps have $Sf \equiv 0$), and if $Sf < 0$ and $Sg < 0$, then $S(f \circ g) < 0$, so the class of maps with negative Schwarzian is closed under composition. For stability analysis, the critical property is that $Sf(p)$ encodes exactly the combination of $f''(p)$ and $f'''(p)$ that determines the sign of $g'''(p)$ when $g = f \circ f$ and $f'(p) = -1$. The following theorem formalises the reduction. Its proof — which consists of computing $g''(p)$ and $g'''(p)$ via the chain rule and identifying the result with $Sf(p)$ — reveals why the Schwarzian is the natural invariant for this problem. [quotetheorem:643] The theorem provides a complete resolution of the critical case $\lambda = -1$: compute $Sf(p)$, and its sign determines stability. The key structural observation in the proof is that $g''(p) = 0$ automatically when $f'(p) = -1$ — this is not an additional hypothesis but a consequence of the symmetry of the second iterate. This forces the stability to depend on $g'''(p)$, which is precisely $2 \, Sf(p)$. The case $Sf(p) = 0$ is not covered by the theorem and requires even higher-order analysis. In generic one-parameter families, $Sf(p) \neq 0$ at the period-doubling point, so this degeneracy is not encountered in typical applications. ### Supercritical and Subcritical Period-Doubling Bifurcations The sign of $Sf(p)$ determines not only the stability of the fixed point at the moment of bifurcation but also the *direction* of the bifurcation — whether the period-2 orbit born at the period-doubling bifurcation is stable or unstable, and on which side of the critical parameter it exists. When $Sf(p) < 0$, the bifurcation is **supercritical**: the fixed point is asymptotically stable at the critical parameter value, and as the parameter is varied past the critical value, the fixed point becomes unstable while a *stable* period-2 orbit emerges in its neighbourhood. The stable orbit inherits the role of the attractor, and the transition is smooth — physically, the system's long-time behaviour changes gradually from a steady state to a period-2 oscillation. When $Sf(p) > 0$, the bifurcation is **subcritical**: the fixed point is unstable at the critical parameter, and the period-2 orbit that exists near the bifurcation point is also *unstable*. In this case, the period-2 orbit exists on the opposite side of the critical parameter (before the bifurcation), and its disappearance at the critical value leaves no nearby attractor, often causing the orbit to jump to a distant attractor. ### Why Negative Schwarzian Matters Many families of maps arising in applications — including the logistic family, the sine family $x \mapsto \mu \sin(\pi x)$, and more generally all polynomial maps of degree $\ge 2$ — satisfy $Sf(x) < 0$ for all $x$ in the domain. A theorem of Singer establishes that maps with negative Schwarzian can have at most as many stable periodic orbits as they have critical points. For unimodal maps (single-humped, such as the logistic map), this means at most one stable orbit can exist at any parameter value. This profound constraint on the dynamics is the reason that negative Schwarzian is a standing hypothesis in much of one-dimensional dynamics. [example: Schwarzian Of The Logistic Map] For the logistic map $f_\mu(x) = \mu x(1 - x)$, the derivatives are \begin{align*} f_\mu'(x) &= \mu(1 - 2x), \\ f_\mu''(x) &= -2\mu, \\ f_\mu'''(x) &= 0. \end{align*} The Schwarzian derivative is therefore \begin{align*} Sf_\mu(x) &= \frac{0}{\mu(1-2x)} - \frac{3}{2}\left(\frac{-2\mu}{\mu(1-2x)}\right)^2 = -\frac{3}{2} \cdot \frac{4}{(1-2x)^2} = -\frac{6}{(1-2x)^2}. \end{align*} This is strictly negative for all $x \neq 1/2$ (and undefined at the critical point $x = 1/2$). In particular, at the period-doubling bifurcation point $\mu = 3$, $x^* = 2/3$: \begin{align*} Sf_3(2/3) &= -\frac{6}{(1 - 4/3)^2} = -\frac{6}{1/9} = -54 < 0. \end{align*} Since $Sf_3(2/3) < 0$, the period-doubling bifurcation at $\mu = 3$ is supercritical: the fixed point $x^* = 2/3$ is stable at $\mu = 3$, becomes unstable for $\mu > 3$, and a stable period-2 orbit is born. [/example] The conditions for a generic period-doubling bifurcation mirror those for the saddle-node, adapted to the multiplier $-1$. The non-degeneracy condition is now expressed in terms of a coefficient that is proportional to $-Sf(p)$, ensuring that the Schwarzian criterion gives a definite answer. [definition: Generic Period-Doubling Bifurcation] Let $f_\alpha \in C^3(I, I)$ be a one-parameter family of maps and let $(p, \alpha_0)$ be a fixed point of $f_{\alpha_0}$. The family undergoes a **generic period-doubling bifurcation** at $(p, \alpha_0)$ if the following three conditions hold: 1. **Criticality:** $f_{\alpha_0}'(p) = -1$. 2. **Non-degeneracy:** $\dfrac{1}{2}(f_{\alpha_0}''(p))^2 + \dfrac{1}{3}f_{\alpha_0}'''(p) \neq 0$. (This coefficient is proportional to $-Sf_{\alpha_0}(p)$.) 3. **Transversality:** $\dfrac{\partial}{\partial \alpha}\left(f_\alpha'(p(\alpha))\right)\bigg|_{\alpha = \alpha_0} \neq 0$. Under these conditions, the fixed point changes stability as $\alpha$ crosses $\alpha_0$, and a period-2 orbit is born. If the non-degeneracy coefficient is positive (equivalently, $Sf(p) < 0$), the bifurcation is supercritical and the period-2 orbit is stable. If it is negative, the bifurcation is subcritical and the period-2 orbit is unstable. [/definition] The transversality condition ensures that the multiplier $f_\alpha'(p(\alpha))$ actually crosses $-1$ as $\alpha$ varies, rather than remaining tangent to $-1$. Combined with the non-degeneracy condition, it guarantees that the bifurcation has the standard normal form $x \mapsto -(1 + \beta)x + a x^3 + O(x^4)$ in suitable coordinates, where $\beta$ is the unfolding parameter and $a$ is the non-degeneracy coefficient. ## Period-2 Orbits After a period-doubling bifurcation, the fixed point becomes unstable and a period-2 orbit appears. To analyse this orbit explicitly — find its points, determine its stability, and understand what happens as parameters are varied further — one works with the second iterate $g = f \circ f$. A period-2 orbit $\{x_1, x_2\}$ with $x_1 \neq x_2$ satisfies $f(x_1) = x_2$ and $f(x_2) = x_1$, which means both $x_1$ and $x_2$ are fixed points of $g(x) = f(f(x))$. Since the fixed points of $f$ are also fixed points of $g$, the period-2 points are found by solving $g(x) = x$ and excluding the fixed points of $f$. This amounts to factoring: \begin{align*} \frac{f(f(x)) - x}{f(x) - x} = 0. \end{align*} The roots of this quotient are exactly the period-2 points (and no fixed points of $f$). The stability of the period-2 orbit is determined by the multiplier of either of its points as a fixed point of $g$. By the chain rule: \begin{align*} g'(x_1) = f'(f(x_1)) \cdot f'(x_1) = f'(x_2) \cdot f'(x_1). \end{align*} The same expression holds at $x_2$, so both points of the orbit share the same multiplier $\Lambda = f'(x_1) \cdot f'(x_2)$, and the orbit is stable if $|\Lambda| < 1$. ### Birth of the Period-2 Orbit at a Period-Doubling Bifurcation At the exact moment of a period-doubling bifurcation, the period-2 orbit is "born" as a degenerate orbit with $x_1 = x_2 = p$ (the fixed point). Its multiplier at birth is \begin{align*} \Lambda = f'(p) \cdot f'(p) = (-1)(-1) = 1. \end{align*} Thus the newborn period-2 orbit enters the scene with the critical multiplier $+1$. As the parameter moves past the bifurcation value, the orbit opens up ($x_1 \neq x_2$), and the multiplier $\Lambda$ moves away from $+1$. In a supercritical period-doubling ($Sf(p) < 0$), $\Lambda$ moves into the interval $(-1, 1)$, making the orbit stable. The orbit remains stable until $\Lambda$ itself reaches $-1$, at which point the period-2 orbit undergoes its own period-doubling bifurcation, giving birth to a period-4 orbit. This cascade of successive period-doublings is a universal route to chaos in one-dimensional dynamics. [example: Period Two Orbit In The Logistic Map] For the logistic map $f_\mu(x) = \mu x(1-x)$, the period-2 equation $f_\mu(f_\mu(x)) = x$ can be factored. After removing the fixed-point factor $f_\mu(x) - x = \mu x(1-x) - x = x(\mu - \mu x - 1)$, the period-2 points satisfy the quadratic \begin{align*} \mu^2 x^2 - \mu(\mu + 1)x + (\mu + 1) = 0. \end{align*} The discriminant is $\mu^2(\mu+1)^2 - 4\mu^2(\mu+1) = \mu^2(\mu+1)(\mu-3)$, which is positive for $\mu > 3$. For $\mu$ slightly above $3$, the two period-2 points are \begin{align*} x_{1,2} = \frac{(\mu + 1) \pm \sqrt{(\mu+1)(\mu-3)}}{2\mu}. \end{align*} The multiplier of the orbit is \begin{align*} \Lambda = f_\mu'(x_1) \cdot f_\mu'(x_2) = \mu^2(1-2x_1)(1-2x_2). \end{align*} Using the relations $x_1 + x_2 = (\mu+1)/\mu$ and $x_1 x_2 = (\mu+1)/\mu^2$, one computes \begin{align*} (1-2x_1)(1-2x_2) &= 1 - 2(x_1 + x_2) + 4 x_1 x_2 \\ &= 1 - \frac{2(\mu+1)}{\mu} + \frac{4(\mu+1)}{\mu^2} \\ &= \frac{\mu^2 - 2\mu(\mu+1) + 4(\mu+1)}{\mu^2} \\ &= \frac{-\mu^2 + 2\mu + 4}{\mu^2}. \end{align*} Therefore $\Lambda = -\mu^2 + 2\mu + 4$. At $\mu = 3$, $\Lambda = -9 + 6 + 4 = 1$, confirming the birth multiplier. The orbit is stable when $|\Lambda| < 1$, i.e., when $-1 < -\mu^2 + 2\mu + 4 < 1$. The upper bound gives $\mu > 3$ (already ensured). The lower bound gives $\mu^2 - 2\mu - 5 < 0$, hence $\mu < 1 + \sqrt{6} \approx 3.449$. At $\mu = 1 + \sqrt{6}$, the multiplier $\Lambda$ reaches $-1$, and the period-2 orbit undergoes its own period-doubling bifurcation — the first step in the period-doubling cascade. [/example] ## References 1. J. Guckenheimer and P. Holmes, *Nonlinear Oscillations, Dynamical Systems, and Bifurcations of Vector Fields* (1983). 2. Y. A. Kuznetsov, *Elements of Applied Bifurcation Theory* (2004). 3. S. Strogatz, *Nonlinear Dynamics and Chaos* (2014). 4. D. Singer, Stable Orbits and Bifurcation of Maps of the Interval, *SIAM Journal on Applied Mathematics* **35** (1978), 260–267.