[Beginner] — one theorem about the centre, and the four results that fall out of it for free.

Before you start

What you will be able to do after this lesson:

  1. Prove the central angle theorem, and derive the angle-in-a-semicircle, same-segment and cyclic-quadrilateral results from it.
  2. Recognise which theorem a diagram is asking for, from the position of the marked angles.
  3. Prove that a perpendicular from the centre bisects a chord, and use it to find distances.
  4. Say what “subtends” means precisely enough to avoid the most common misreading of these theorems.

Study route: read 1–4, attempt the prediction in section 4, then run the lab and do the practice task.


1. Why this exists

A circular window is being fitted into a stone arch. The mason knows two points where the glass meets the frame, and needs the angle the glass makes when viewed from a third point on the rim — but the centre of the circle is behind the stonework and cannot be reached.

Circle theorems answer exactly this class of question: they relate angles at the rim to angles at the centre, and to each other, so that an inaccessible measurement can be deduced from accessible ones.

There is a second reason these are worth learning properly. Almost every one of them is a consequence of a single theorem. Learning five separate facts is a memory problem; learning one fact and four short derivations is not.

2. Terminology

TermPlain-English definitionNote
CentreThe point every point of the circle is equidistant fromUsually labelled
RadiusA segment from the centre to the circleAll radii of one circle are equal — this drives every proof
ChordA segment joining two points on the circleA diameter is the longest chord
ArcPart of the circle itselfTwo points cut the circle into a minor and a major arc
SegmentThe region between a chord and an arcNot the same as “line segment”
SectorThe region between two radii and an arcA pizza slice
SubtendTo “sit opposite” — the angle a chord opens up at a pointChord subtends at
Cyclic quadrilateralA four-sided figure with all four vertices on one circleThe condition is easy to lose sight of
Angle at the centreThe angle a chord subtends at
Angle at the circumferenceThe angle the same chord subtends at a point on the circle

Subtend deserves care. When a theorem says “the angle subtended by arc at ”, it means the angle — the angle you would see if you stood at and looked at the two ends of the chord. Which arc is named matters: the same chord subtends one angle from the major arc and a different one from the minor arc, and mixing them up is the most common error in this topic.

3. The one theorem

Central angle theorem: the angle a chord subtends at the centre is twice the angle it subtends at any point on the major arc.

              P
             / \
            /   \
           /     \
          /   O   \          angle AOB = 2 x angle APB
         /   / \   \
        /   /   \   \
       A---+-----+---B

Proof. Join to and extend it to a point on the far side of the circle.

and are both radii, so triangle is isosceles, and the base angles are equal. Call each of them :

Now apply the exterior angle theorem to triangle at . The exterior angle equals the sum of the two opposite interior angles:

The identical argument on triangle , with base angles , gives:

Adding:

Everything else in this lesson is a corollary.

Corollary 1 — angle in a semicircle is

Let be a diameter. Then , , are collinear, so . By the theorem:

Every point on the circle sees a diameter at a right angle. This is often called Thales’ theorem, and it is the oldest result here.

Corollary 2 — angles in the same segment are equal

Let and both lie on the major arc. Both angles and are half of the same central angle , so:

The angle a chord subtends does not depend on where on the arc you stand. Move around the major arc and the angle stays fixed.

Corollary 3 — opposite angles of a cyclic quadrilateral sum to

Let be cyclic. Chord subtends at and at , on opposite arcs. The two central angles at — one for each arc — together make a full turn:

Dividing by two:

Opposite angles of a cyclic quadrilateral are supplementary. The converse also holds and is often the more useful direction: if a quadrilateral’s opposite angles sum to , its four vertices lie on a circle.

4. Chords and the centre

Theorem: the perpendicular from the centre to a chord bisects the chord.

Proof. Let be the foot of the perpendicular from to chord . Triangles and have (radii), common, and right angles at . By RHS they are congruent, so .

The converse holds too: the line from the centre to a chord’s midpoint is perpendicular to it.

This gives a useful length relation. If the circle has radius , the chord has length , and the perpendicular distance from centre to chord is , then Pythagoras in triangle gives:

Two consequences follow immediately:

  • Equal chords are equidistant from the centre, and conversely. Longer chord, smaller .
  • The longest chord has , so it passes through the centre: it is a diameter, of length .

TIP

Predict before running the lab. A chord of length sits in a circle of radius . Another chord of the same circle is long. Which is closer to the centre, and by how much? Work it out before opening the answers.

Worked example — runnable

Runnable example: save as circle_theorems.py in any empty directory and run python3 circle_theorems.py. Standard library only; writes no files.

"""Every circle theorem in this lesson, checked numerically.
 
Points live on a circle of radius r centred at the origin, named by the
angle in degrees at which they sit. That makes it easy to move a point
around an arc and watch which quantities change and which do not.
"""
import math
 
EPS = 1e-9
R = 5.0
 
 
def on_circle(theta_degrees, r=R):
    t = math.radians(theta_degrees)
    return (r * math.cos(t), r * math.sin(t))
 
 
def angle_at(v, p, q):
    """Angle p-v-q in degrees."""
    u = (p[0] - v[0], p[1] - v[1])
    w = (q[0] - v[0], q[1] - v[1])
    cos = (u[0] * w[0] + u[1] * w[1]) / (math.hypot(*u) * math.hypot(*w))
    return math.degrees(math.acos(max(-1.0, min(1.0, cos))))
 
 
def dist(p, q):
    return math.hypot(q[0] - p[0], q[1] - p[1])
 
 
O = (0.0, 0.0)
 
if __name__ == "__main__":
    A = on_circle(200)      # chord AB, fixed
    B = on_circle(340)
 
    print("Central angle theorem: angle at centre = 2 x angle at circumference")
    centre_angle = angle_at(O, A, B)
    print(f"  angle AOB at the centre = {centre_angle:.6f}")
    for theta in (30, 60, 90, 120):          # four positions on the major arc
        P = on_circle(theta)
        at_P = angle_at(P, A, B)
        print(f"    P at {theta:3} deg -> angle APB = {at_P:.6f}   (2x = {2*at_P:.6f})")
        assert abs(2 * at_P - centre_angle) < EPS
 
    print()
    print("Corollary 2: the angle does not depend on where P sits on the arc")
    values = {round(angle_at(on_circle(t), A, B), 9) for t in (30, 60, 90, 120)}
    print(f"  distinct values across four positions: {len(values)}")
    assert len(values) == 1
 
    print()
    print("Corollary 1: angle in a semicircle")
    D1, D2 = on_circle(0), on_circle(180)    # a diameter
    for theta in (37, 90, 143, 250):
        P = on_circle(theta)
        a = angle_at(P, D1, D2)
        print(f"  P at {theta:3} deg -> angle in semicircle = {a:.6f}")
        assert abs(a - 90) < EPS
 
    print()
    print("Corollary 3: cyclic quadrilateral, opposite angles")
    quad = [on_circle(t) for t in (20, 100, 190, 300)]
    W, X, Y, Z = quad
    wx = angle_at(W, Z, X)
    xy = angle_at(X, W, Y)
    yz = angle_at(Y, X, Z)
    zw = angle_at(Z, Y, W)
    print(f"  angles: {wx:.4f}, {xy:.4f}, {yz:.4f}, {zw:.4f}")
    print(f"  opposite pairs sum to {wx + yz:.6f} and {xy + zw:.6f}")
    assert abs((wx + yz) - 180) < EPS
    assert abs((xy + zw) - 180) < EPS
 
    print()
    print("Chords: perpendicular from the centre bisects, and r^2 = d^2 + l^2")
    for chord_len in (6.0, 8.0, 10.0):
        half = chord_len / 2
        d = math.sqrt(R * R - half * half)   # distance from centre to chord
        print(f"  chord {chord_len:4.1f} -> half {half:3.1f}, distance from centre {d:.6f}")
        assert abs(R * R - (d * d + half * half)) < EPS
    print("  the longest chord has distance 0 from the centre: it is the diameter")
 
    print()
    print("circle_theorems: passed")

Expected output:

Central angle theorem: angle at centre = 2 x angle at circumference
  angle AOB at the centre = 140.000000
    P at  30 deg -> angle APB = 70.000000   (2x = 140.000000)
    P at  60 deg -> angle APB = 70.000000   (2x = 140.000000)
    P at  90 deg -> angle APB = 70.000000   (2x = 140.000000)
    P at 120 deg -> angle APB = 70.000000   (2x = 140.000000)

Corollary 2: the angle does not depend on where P sits on the arc
  distinct values across four positions: 1

Corollary 1: angle in a semicircle
  P at  37 deg -> angle in semicircle = 90.000000
  P at  90 deg -> angle in semicircle = 90.000000
  P at 143 deg -> angle in semicircle = 90.000000
  P at 250 deg -> angle in semicircle = 90.000000

Corollary 3: cyclic quadrilateral, opposite angles
  angles: 100.0000, 95.0000, 80.0000, 85.0000
  opposite pairs sum to 180.000000 and 180.000000

Chords: perpendicular from the centre bisects, and r^2 = d^2 + l^2
  chord  6.0 -> half 3.0, distance from centre 4.000000
  chord  8.0 -> half 4.0, distance from centre 3.000000
  chord 10.0 -> half 5.0, distance from centre 0.000000
  the longest chord has distance 0 from the centre: it is the diameter

circle_theorems: passed

The second block is the corollary made visible: four different positions on the arc, one identical angle, to the last decimal place available.

Common pitfalls and traps

  • Using the wrong arc. “Angle at the circumference” means from the major arc when the central angle is the minor one. Stand on the minor arc instead and you get minus the expected answer. If a diagram gives an angle that seems to be the supplement of the one you want, this is why.
  • Forgetting that all four vertices must be on the circle. A quadrilateral with three vertices on a circle is not cyclic and its opposite angles need not sum to anything in particular.
  • Assuming a point is the centre. A point drawn in the middle of a circle is the centre only if the diagram says so, or if two radii to it are marked equal. Every proof here uses ; without that, none of them apply.
  • Assuming a chord through the middle is a diameter. It is a diameter only if it passes through the centre exactly. “Looks like it goes through the middle” is not a reason.
  • Applying the semicircle result to a chord that is not a diameter. The comes from , which only happens for a diameter.
  • Reading the lab as proof. It confirms the theorems hold for the configurations tested, at floating-point precision. Section 3 is where the certainty comes from.

Check your understanding

  1. A chord subtends at the circumference. What does it subtend at the centre?
  2. In a cyclic quadrilateral , . Find . Can you find ?
  3. is a diameter and is on the circle. . Find .
  4. A chord of length lies in a circle of radius . How far is it from the centre?
  5. Why does the central angle theorem’s proof need on the major arc? What goes wrong otherwise?
Answers — open only after an attempt
  1. — the angle at the centre is twice the angle at the circumference.
  2. , since opposite angles of a cyclic quadrilateral are supplementary. cannot be found: the constraint links opposite pairs only, and can take any value with adjusting to keep .
  3. (angle in a semicircle). The angles of triangle sum to , so .
  4. Half the chord is , so .
  5. The proof extends to and adds . That addition is only valid when lies between and , which is what being on the major arc guarantees. With on the minor arc the two angles must be subtracted instead, and the result becomes the reflex angle at the centre — still twice , but the reflex one.

And the prediction from section 4: for the chord of length , . For the chord of length , . The longer chord is closer to the centre, by unit. Longer chords always sit nearer the centre, and the diameter sits on it.

Practice — independent task

Implement circumcircle(a, b, c): given three points, return the centre and radius of the unique circle through all three — or None if they are collinear.

  1. Find the centre as the intersection of two perpendicular bisectors, reusing the reasoning from constructions: the centre is equidistant from all three points, so it lies on the perpendicular bisector of every side.
  2. Detect the collinear case and return None rather than dividing by a near-zero determinant. Say in a comment what tolerance you chose and why.
  3. Assert that the returned centre really is equidistant from all three inputs.
  4. Then use it to test whether a quadrilateral is cyclic: build the circumcircle of three vertices and check whether the fourth lies on it.
  5. Verify your cyclic test against the theorem: for at least three cyclic quadrilaterals, confirm opposite angles sum to ; for at least two non-cyclic ones, confirm they do not.

Edge cases: three collinear points; two coincident points; a quadrilateral whose fourth vertex is close to the circle but not on it — decide what “on the circle” means numerically and defend the choice.

Done when: your cyclic test and the opposite-angle test agree on every quadrilateral you try, and you can explain why they must agree — that is, why the converse of Corollary 3 is what licenses using either one.

Tradeoffs, limits and extensions

These are theorems about a plane. On a sphere the “circle” through three points and the angle relations both change, and none of the results above survive unmodified.

The converse is often the useful direction. Corollary 3 says cyclic implies supplementary. The converse — supplementary implies cyclic — is what lets you prove four points lie on a circle, which is how these theorems are usually deployed in a harder problem. Do not assume a converse holds without checking; here it does, but the perpendicular-from-centre converse and the equal-chords converse each need their own argument.

Where this reappears. The inscribed angle theorem is the geometric core of the circle equation, and the circumcircle of the practice task is the basis of Delaunay triangulation — the standard way to turn scattered points into a mesh in graphics and terrain modelling.

Before moving on

You are done with this lesson when you can:

  • Prove the central angle theorem from the isosceles and exterior-angle results, without notes.
  • Derive all three corollaries from it rather than recalling them separately.
  • Use to move between chord length and distance from the centre.

Recap for later lookup: angle at centre angle at circumference; angle in a semicircle ; angles in the same segment are equal; opposite angles of a cyclic quadrilateral sum to ; the perpendicular from the centre bisects a chord, with .

Next: Tangents to a Circle — what happens when a chord’s two ends slide together until the line touches at a single point.

↑ Back to Geometry and Trigonometry