Executive summary
The source introduces continuity to make 'following a branch' mathematically meaningful. A function is continuous at a point when arbitrarily small output changes can be guaranteed by sufficiently small input changes. Sums, products, quotients with non-zero denominator and compositions of continuous functions remain continuous. A complex parametric curve is a continuous function z(t) on an interval, with an orientation from its initial to final point. These ideas allow values of a multi-valued function to be selected continuously along a path, which becomes the core operation behind branched surfaces and monodromy.
What this handbook page teaches
- Use the epsilon-delta continuity definition at a point.
- Apply continuity rules for arithmetic and composition.
- Describe complex curves parametrically and interpret orientation.
- Construct curve reversals and concatenations.
- Understand continuous branch selection along a path.
Core concepts
Continuity at a point
For a real or complex function f, continuity at z0 means that for every positive tolerance ε there is a positive input tolerance δ such that |z-z0|<δ implies |f(z)-f(z0)|<ε. The modulus supplies the distance measure in the complex plane.
The source proves continuity of elementary functions and then shows that addition, subtraction, multiplication, suitable division and composition preserve continuity. This lets complicated expressions inherit continuity without repeating epsilon-delta arguments every time.
Parametric curves
A continuous curve is represented by z(t)=x(t)+iy(t), commonly with 0≤t≤1. Both coordinate functions are continuous. The point z(0) is the initial point and z(1) the final point, so the parameter gives the curve an orientation.
The same geometric trace can be traversed in the opposite direction by replacing t with 1-t. If one curve ends where another begins, they can be concatenated into a single continuous curve by rescaling the parameter on two half-intervals.
Continuous images and branch following
If f is single-valued and continuous, the image of a curve C:z(t) is the curve f(C):w(t)=f(z(t)). For a multi-valued function, one instead chooses a starting value and attempts to continue a value continuously along the curve.
When continuation is unique and avoids problematic points, the endpoint value records how the branch has evolved. Two different paths between the same endpoints can produce different branch values, which is the phenomenon later encoded by winding and monodromy.
Working method
Use the following sequence as a repeatable analysis workflow. The steps are ordered to separate definitions and admissibility checks from structural conclusions, so a result can be audited without relying on intuition or an unverified diagram.
- Write the curve as
z(t)and identify its parameter interval, start, end and orientation. - For image problems under a single-valued function, substitute
z(t)into the function and simplifyw(t). - To reverse a curve, use
z(1-t); to concatenate two curves, rescale their parameters so the first occupies the first half and the second the second half. - For a multi-valued function, select one starting branch value and demand continuous variation along the path.
- Mark branch points and non-uniqueness points that the path must avoid when uniqueness is required.
- Compare endpoint values for alternative paths to detect global multi-valued behaviour.
Following a square-root branch along two semicircles
Start at z=1 with the chosen square-root value w=1. Move to z=-1 along the upper semicircle. A continuous argument can increase from 0 to π, so the root argument increases from 0 to π/2. The continued value at the endpoint is therefore i.
If the path instead follows the lower semicircle, a continuous argument may decrease from 0 to -π. The root argument then decreases to -π/2, giving endpoint -i. The same start value and same geometric endpoint produce different continued values because the paths wind differently around the branch point.
This example makes clear why a global single-valued square root cannot be chosen continuously on the punctured plane. A cut that prevents loops around the origin can restore a single-valued branch on the cut domain.
Technical reasoning and deeper connections
The source deliberately uses some continuity facts at an intuitive level. In particular, it relies on the principle that a continuous integer-valued function on a connected interval is constant. This underpins the stability of winding number under a deformation that avoids a forbidden point.
Continuity is local, while branch switching is global. Near a non-problematic point, roots of a polynomial or branches of a radical can be followed continuously. Yet after a long closed loop, the value may return to a different branch. The later theory is largely about converting this global effect into group data.
Parametric descriptions make geometric transformations algebraic. Translation adds a constant, positive scaling multiplies by a positive real value, multiplication by a unit-modulus complex value rotates, and multiplication by an arbitrary non-zero complex number combines rotation and scale.
The distinction between the trace of a curve and its orientation is important. Winding number changes sign when orientation reverses, and loop composition corresponds to composition of branch permutations. A diagram without arrows can therefore omit essential information.
Quick-reference matrix
| Operation on curves | Parametric form | Effect |
|---|---|---|
| Translation | z(t)+z0 | Moves the curve without changing shape. |
| Scaling | a z(t) for a>0 | Scales distances from the origin. |
| Rotation | u z(t), |u|=1 | Rotates by a fixed angle. |
| Reversal | z(1-t) | Same trace, opposite orientation. |
| Concatenation | Piecewise rescaled equations | Follows one path then another. |
Common mistakes
- Treating continuity as merely 'the graph has no visible breaks' without a quantitative criterion.
- Using the quotient rule at points where the denominator vanishes.
- Confusing the curve's coordinate plane with its parameter interval.
- Reversing a curve's orientation without changing later winding signs.
- Assuming a multi-valued function has a unique continuous image along every path.
- Ignoring branch or non-uniqueness points crossed by a proposed continuation path.
Verification checklist
Before accepting a solution or proof, confirm each item below. These checks are intentionally practical: they catch the most common category errors, missing hypotheses and structure mismatches.
- Continuity claims use modulus as the complex-plane distance.
- Composite expressions satisfy denominator-domain restrictions.
- Curve start, end and orientation are explicit.
- Reversal and concatenation preserve continuity at joins.
- Multi-valued continuation begins from a specified branch value.
- Any uniqueness claim states which problematic points are excluded.
Frequently asked questions
Does a continuous path need to be differentiable?
No. The source requires continuity; differentiability is not part of the curve definition.
Why parameterise curves from 0 to 1?
It is a convenient normalisation that makes reversal and concatenation formulas simple; other intervals can be reparameterised.
Can a multi-valued function be continuous?
One speaks of continuous branches or continuous continuation of a chosen value along a path. The full value set is not a single-valued function.
Why is path orientation important?
Continuation around a loop in the opposite direction applies the inverse branch permutation and reverses the sign of argument variation.
Source scope
Several foundational continuity and deformation statements are used in the source as accepted or sketched facts. The page preserves their role and does not present them as fully developed topology.
Source basis: this page is a handbook-style synthesis of the uploaded mathematics source, principally sections 2.6, 2.7. It paraphrases the supplied material and does not reproduce source artwork or long source passages. Historical names, publisher details and personal details have been intentionally omitted.