A rectangle is a two-dimensional, closed, four-sided figure with four right angles.^[1] It is a type of quadrilateral in which opposite sides are equal in length and parallel to each other, making it a special case of a parallelogram.^[2] Key properties of a rectangle include the fact that its diagonals are congruent in length and bisect each other at their midpoints.^[3] Additionally, all angles measure exactly 90 degrees, and the figure exhibits bilateral symmetry along the lines connecting midpoints of opposite sides.^[4][5] These characteristics distinguish rectangles from other quadrilaterals, such as rhombuses or trapezoids, though a square represents a rectangle where all four sides are of equal length.^[6] In practical applications, rectangles are fundamental in geometry for calculating measurements, with the area given by the formula area = length × width and the perimeter by perimeter = 2(length + width).^[7][8] These formulas underpin computations in fields like architecture, engineering, and computer graphics, where rectangles model structures, screens, and layouts.^[9]

Definition and Characterizations

Formal Definition

A rectangle is defined as a quadrilateral in the Euclidean plane with four right angles, each measuring 90 degrees.^[10] This configuration ensures that the figure is a closed planar shape bounded by four line segments.^[11] This definition is equivalent to that of a quadrilateral possessing four equal interior angles and pairs of opposite sides of equal length, as the right angles imply congruence among all angles and parallelism of opposite sides.^[10][12] The term "rectangle" derives from the Latin rectangulus, combining rectus (meaning "right" or "straight") and angulus (meaning "angle"), thus signifying a "right-angled" figure; it entered English usage in the 1570s via French.^[13] In ancient geometry, Euclid implicitly referenced rectangles in his Elements (circa 300 BCE) through the concept of a "rectangular parallelogram," described as a parallelogram contained by two straight lines forming a right angle.^[14][15] Visually, a rectangle consists of two pairs of parallel sides, with all four angles congruent at 90 degrees, distinguishing it as a specific type of parallelogram.^[10]

Equivalent Characterizations

A rectangle can be characterized as an equiangular quadrilateral, meaning all four interior angles measure 90 degrees. This equivalence holds because in Euclidean geometry, a quadrilateral with four equal angles must have each angle as a right angle, given that the sum of interior angles in any quadrilateral is 360 degrees.^[16] Equivalently, a rectangle is a parallelogram that contains at least one right angle. In a parallelogram, opposite angles are equal and consecutive angles are supplementary; thus, if one angle is 90 degrees, all angles must be 90 degrees. Another equivalent condition is that a parallelogram has congruent diagonals. To see this, consider a parallelogram ABCD with position vectors: let vector $\overrightarrow{AB} = \mathbf{u}$ and $\overrightarrow{AD} = \mathbf{v}$. The diagonals are $\mathbf{u} + \mathbf{v}$ and $\mathbf{u} - \mathbf{v}$. Setting their magnitudes equal gives $|\mathbf{u} + \mathbf{v}|^2 = |\mathbf{u} - \mathbf{v}|^2$, which simplifies to $2\mathbf{u} \cdot \mathbf{v} = -2\mathbf{u} \cdot \mathbf{v}$, implying $\mathbf{u} \cdot \mathbf{v} = 0$. Thus, adjacent sides are perpendicular, confirming it is a rectangle.^[17] In vector terms, a rectangle is defined by two adjacent sides represented by vectors $\mathbf{u}$ and $\mathbf{v}$ such that opposite sides are equal and parallel ($\overrightarrow{DC} = \mathbf{u}$, $\overrightarrow{BC} = \mathbf{v}$), and adjacent sides are perpendicular ($\mathbf{u} \cdot \mathbf{v} = 0$). This ensures the figure has the required right angles and parallel sides.^[18] In coordinate geometry, an axis-aligned rectangle has vertices at $(x_1, y_1)$, $(x_2, y_1)$, $(x_2, y_2)$, and $(x_1, y_2)$, where $x_1 < x_2$ and $y_1 < y_2$. The horizontal and vertical sides are parallel to the axes, and adjacent sides are perpendicular by construction.^[19]

Classification

Within Quadrilaterals

A rectangle is a special type of parallelogram, characterized by having opposite sides that are both parallel and equal in length, with the additional property of all interior angles measuring 90 degrees.^[20] This distinguishes it from a general parallelogram, where angles may not be right angles.^[21] Within the category of rectangles, subtypes exist based on side lengths: a square is a rectangle where all four sides are equal, while an oblong, also known as a non-square rectangle, has unequal adjacent sides but retains the four right angles.^[22] The square represents the most symmetric subtype, overlapping with other quadrilateral forms.^[21] Rectangles differ from other quadrilaterals such as trapezoids, which feature exactly one pair of parallel sides, and kites, which have two pairs of adjacent sides that are equal in length.^[20] Unlike these, rectangles require both pairs of opposite sides to be parallel. In the taxonomy of quadrilaterals, rectangles form a subset of parallelograms, which in turn are a subset of all quadrilaterals; rhombuses, with all sides equal, also subset parallelograms, and the square lies at the intersection of rectangles and rhombuses.^[20] This hierarchical structure illustrates how rectangles occupy a specific position among convex quadrilaterals with parallel opposite sides.

In Broader Geometric Hierarchies

In non-Euclidean geometries, the rectangle's classification extends beyond the Euclidean plane, revealing how its defining properties—such as right angles and opposite equal sides—adapt or fail under different metrics and axioms. In hyperbolic geometry, quadrilaterals with four right angles are impossible because the sum of interior angles in any hyperbolic quadrilateral is less than 360 degrees, a consequence of the constant negative curvature.^[23] However, analogous figures known as hyperbolic rectangles can be constructed as equiangular quadrilaterals with four equal angles each less than 90 degrees and opposite sides of equal length; in these, the diagonals are unequal, contrasting the equal diagonals of Euclidean rectangles.^[24] In taxicab geometry, also known as Manhattan geometry and governed by the L1 norm where distance is the sum of absolute differences in coordinates, rectangles are defined as four-sided figures with four right angles (measured 90 degrees in the Euclidean sense but interpreted via the taxicab metric) and opposite sides congruent. These "rectangles" exhibit distinct properties from their Euclidean counterparts, such as diagonals whose taxicab length equals the sum of adjacent side lengths rather than the Euclidean hypotenuse, leading to altered notions of symmetry and congruence criteria like SASAS for triangles.^[25][26] Beyond plane geometries, the rectangle fits into abstract hierarchies as a special case of higher-dimensional polytopes. Specifically, a rectangle is a two-dimensional orthotope, the generalization of a rectangle to n dimensions defined as the Cartesian product of n closed intervals along mutually orthogonal axes; for example, in three dimensions, this yields a rectangular prism (or cuboid) with all right angles and pairwise perpendicular faces.^[27] Rectangles also belong to the family of zonotopes, centrally symmetric polytopes formed as the Minkowski sum of line segments; a rectangle arises from the sum of two perpendicular line segments of appropriate lengths, placing it within broader polyhedral classifications including parallelograms and higher-dimensional analogs like zonoids.^[28] These classifications highlight the rectangle's role in unifying geometric structures across metrics and dimensions, with extensions to tessellations in non-Euclidean spaces noted in specialized tilings.^[23]

Core Properties

Sides and Angles

A rectangle is defined by its four interior angles, each measuring exactly 90 degrees, resulting in a total sum of 360 degrees, consistent with the interior angle sum of any quadrilateral.^[6] These right angles ensure that the figure's corners form perfect perpendicular intersections, distinguishing the rectangle from other parallelograms where angles may vary.^[29] The sides of a rectangle consist of two pairs of opposite sides that are equal in length, typically denoted as length $ l $ and width $ w $, where adjacent sides may differ unless the rectangle is a square.^[30] This equality of opposite sides, combined with the property that opposite sides are parallel to each other, positions the rectangle as a special type of parallelogram.^[29] The parallelism arises from the right angles, which align the sides such that no side intersects another except at the vertices.^[6] Regarding the possibility of an inscribed circle tangent to all four sides, a rectangle admits such an incircle only if it is a square, as non-square rectangles fail the condition that the sums of the lengths of opposite sides must be equal—a requirement established by Pitot's theorem for tangential quadrilaterals.^[31] In a general rectangle with $ l \neq w $, the sums are $ 2l $ and $ 2w $, which are unequal, preventing the existence of an incircle.^[32] This property highlights the rectangle's geometric constraints, briefly noting its duality with the rhombus in quadrilateral classifications.^[30]

Diagonals and Symmetry

In a rectangle, the two diagonals are congruent in length and bisect each other at their midpoint.^[30][33] This bisection occurs because a rectangle is a special type of parallelogram, where the diagonals always intersect at their midpoints, and the congruence follows from the equal right angles at the vertices.^[30][33] The rectangle exhibits reflection symmetry across two lines: the horizontal axis passing through the midpoints of the top and bottom sides, and the vertical axis passing through the midpoints of the left and right sides.^[34] These reflection symmetries, combined with the identity transformation and a 180° rotation about the center, form the full symmetry group of a non-square rectangle, known as the Klein four-group or dihedral group $D_2$.^[35] Each non-identity element of this group has order 2, reflecting the pairwise commuting nature of the rotations and reflections. In the special case of a square, the symmetry group expands to the dihedral group $D_4$ of order 8, incorporating additional 90° and 270° rotations as well as reflections over the diagonals.^[35] Applying Varignon's theorem to a rectangle, the quadrilateral formed by connecting the midpoints of its sides is a rhombus. This result stems from the theorem's general statement that such midpoints always form a parallelogram, with the additional property of equal side lengths arising specifically when the original quadrilateral has congruent diagonals, as in the rectangle. The rhombus's sides are parallel to the diagonals of the original rectangle and half their lengths, highlighting the rectangle's inherent symmetry in midpoint constructions.

Rectangle-Rhombus Duality

In geometry, rectangles and rhombi exhibit a fundamental duality, where the rectangle emphasizes equality among its angles—all four being right angles—while allowing for unequal adjacent sides, and the rhombus prioritizes equality among all four sides while permitting unequal adjacent angles.^[36] This complementary relationship trades uniformity in one attribute (angles versus sides) for flexibility in the other, reflecting a broader duality in Euclidean plane geometry between angular and linear measures.^[36] The square serves as the fixed point of this duality, satisfying both conditions simultaneously: all angles are right angles, and all sides are equal.^[36] In this configuration, the trade-off resolves, yielding a shape with maximal symmetry that aligns with both the rectangle's angular precision and the rhombus's side congruence. This intersection underscores the square's unique position within the family of parallelograms. This duality manifests in optimization problems, such as the isoperimetric inequality for quadrilaterals, where for a fixed area, the square achieves the minimal perimeter among both rectangles and rhombi.^[37][38] Specifically, the inequality $P^2 \geq 16A$ holds for any quadrilateral with perimeter $P$ and area $A$, with equality only for the square, highlighting how deviating from squareness in either direction—increases angular variance in rhombi or side disparity in rectangles—results in larger perimeters for the same area.^[37] This table illustrates the dual properties, where each feature of the rectangle corresponds inversely to that of the rhombus.^[36]

Formulas and Calculations

Area and Perimeter

The area of a rectangle is calculated by multiplying its length by its width. For a rectangle with length $ l $ and width $ w $, the area $ A $ is given by the formula

$$A = l \times w.$$

This formula arises from the general principle of area for parallelograms, where the area equals the base length times the corresponding height; in a rectangle, the right angles ensure that the height perpendicular to the base equals the width $ w $.^[39] The concept of this area formula traces back to ancient Egyptian mathematics, as evidenced in the Rhind Mathematical Papyrus, a document from around 1650 BCE that includes practical problems computing rectangular areas via multiplication of dimensions.^[40] The perimeter $ P $ of a rectangle, which measures the total length of its boundary, is the sum of all four sides; since opposite sides are equal, it simplifies to twice the sum of the length and width:

$$P = 2(l + w).$$

This follows directly from adding the two lengths and two widths.^[41] There is no single fixed formula for the length of a rectangle, as length is one of its two dimensions (conventionally the longer side). However, the length can be calculated from other known quantities using rearrangements of the standard formulas:

• Given area $ A $ and width $ w $: $ l = \frac{A}{w} $
• Given perimeter $ P $ and width $ w $: $ l = \frac{P}{2} - w $

In practical applications, such as measuring a room with dimensions 5 meters by 4 meters, the area would be 20 square meters (useful for flooring estimates), while the perimeter would be 18 meters (for baseboard length).^[42] Units for area are typically square units (e.g., square meters), and for perimeter, linear units (e.g., meters), ensuring consistency in measurements.^[43] Rectangles illustrate the isoperimetric inequality, which states that for a given perimeter, the circle encloses the maximum possible area; among rectangles of fixed perimeter, a square achieves the largest area, but all rectangles are less efficient than a circle of the same perimeter.^[44][45]

Diagonal Length and Coordinates

The length of a diagonal in a rectangle with side lengths $ l $ (length) and $ w $ (width) is given by the formula $ d = \sqrt{l^2 + w^2} $. This formula arises directly from the Pythagorean theorem, as each diagonal forms the hypotenuse of a right triangle with legs of lengths $ l $ and $ w $.^[46] Rearranging this formula, if the diagonal $ d $ and width $ w $ are known, the length $ l $ (conventionally the longer side) can be calculated as $ l = \sqrt{d^2 - w^2} $, provided $ d > w $ (which always holds for a non-degenerate rectangle). To prove this using the distance formula in the coordinate plane, place the rectangle with one vertex at the origin $(0,0)$ and adjacent vertices at $(l,0)$ and $(0,w)$. The opposite vertex is then at $(l,w)$. The distance between $(0,0)$ and $(l,w)$ is $\sqrt{(l-0)^2 + (w-0)^2} = \sqrt{l^2 + w^2}$, confirming the diagonal length; the same applies to the other diagonal by symmetry.^[47] In the Cartesian coordinate plane, an axis-aligned rectangle can be defined by four vertices: $(x_1, y_1)$, $(x_2, y_1)$, $(x_2, y_2)$, and $(x_1, y_2)$, where $ x_1 < x_2 $ and $ y_1 < y_2 $. The side lengths are then $ l = x_2 - x_1 $ and $ w = y_2 - y_1 $. For a rotated rectangle, apply a rotation transformation matrix to these coordinates. The 2D rotation matrix by an angle $\theta$ counterclockwise about the origin is

$$\begin{pmatrix} \cos \theta & -\sin \theta \\ \sin \theta & \cos \theta \end{pmatrix},$$

which transforms each vertex $(x, y)$ to $(\cos \theta \cdot x - \sin \theta \cdot y, \sin \theta \cdot x + \cos \theta \cdot y)$.^[48][49] For a tilted (non-axis-aligned) rectangle, represent it using two perpendicular side vectors $\mathbf{u}$ and $\mathbf{v}$ with lengths $ l = |\mathbf{u}| $ and $ w = |\mathbf{v}| $, where $\mathbf{u} \cdot \mathbf{v} = 0$. The vertices are then the origin, $\mathbf{u}$, $\mathbf{v}$, and $\mathbf{u} + \mathbf{v}$, providing an oblique coordinate basis for the rectangle. The diagonal vectors are $\mathbf{u} + \mathbf{v}$ and $\mathbf{u} - \mathbf{v}$, with lengths $\sqrt{l^2 + w^2}$ by the Pythagorean theorem in vector form.^[50][51]

Key Theorems

Pythagorean Applications

The Pythagorean theorem finds a direct application in rectangles through the relationship between the sides and the diagonal. In a rectangle with adjacent sides of lengths $a$ and $b$, each diagonal forms the hypotenuse of a right triangle with legs $a$ and $b$, yielding the equation $d = \sqrt{a^2 + b^2}$.^[10] This relation holds because the right angle at each vertex ensures the triangle's legs align with the rectangle's sides.^[52] The theorem also provides a straightforward proof that the diagonals of a rectangle are equal in length. Dividing the rectangle along one diagonal creates two congruent right triangles, each with legs $a$ and $b$; applying the Pythagorean theorem to both yields the same hypotenuse length $d = \sqrt{a^2 + b^2}$ for each diagonal.^[10] This congruence arises from the rectangle's parallel sides and right angles, ensuring the triangles share identical side lengths.^[53] An important extension involves rectangles inscribed in a circle. A rectangle can always be inscribed in a circle, with the diagonals serving as diameters because each subtends a right angle at the opposite vertices, consistent with the theorem that an angle inscribed in a semicircle is a right angle.^[10] The circle's radius is then half the diagonal length, $R = \frac{1}{2} \sqrt{a^2 + b^2}$, confirming the diagonals' equality as a necessary condition for such inscription in the Euclidean plane. Historically, the Pythagorean theorem, attributed to Pythagoras in the 6th century BCE, underpins many Euclidean proofs involving rectangles, including those in Euclid's Elements (ca. 300 BCE), where Book I Proposition 47 establishes the theorem's core relation used to derive rectangle properties like diagonal equality.^[54] Euclid's geometric constructions, such as rearranging areas in right triangles, extend naturally to rectangular figures, influencing classical treatments of symmetry and inscription.^[55]

Advanced Theorems

The British flag theorem is a distinctive property of rectangles, stating that for any point $P$ in the Euclidean plane and a rectangle $ABCD$ with $A$ and $C$ as one pair of opposite vertices and $B$ and $D$ as the other pair, the sum of the squares of the distances from $P$ to $A$ and $C$ equals the sum of the squares from $P$ to $B$ and $D$:

$$PA^2 + PC^2 = PB^2 + PD^2.$$

This equality holds regardless of whether $P$ lies inside, outside, or on the boundary of the rectangle, highlighting the theorem's robustness and the rectangle's orthogonal symmetry.^{[56] [57]} The theorem can be proved using coordinate geometry by placing the rectangle's center at the origin with sides aligned to the axes; the distances then simplify via the expansion of squared terms, where cross terms cancel due to orthogonality, yielding a constant sum independent of $P$'s position.^[56] Van Aubel's theorem provides another advanced perspective on rectangles within the broader context of quadrilateral constructions. In its general form, for any quadrilateral, if squares are erected outwardly (or inwardly) on each side, the line segments joining the centers of opposite squares are of equal length and perpendicular to each other.^[58] When applied to a rectangle, this configuration exploits the rectangle's right angles and equal diagonals: the centers of the squares on opposite sides align such that the joining segments maintain the theorem's equal-length and perpendicular properties, but the rectangle's symmetry further ensures these segments are parallel to the rectangle's diagonals or axes, simplifying the resulting varignon-like figure into a square.^[59] Extensions of Van Aubel's theorem to rectangles erected on the sides of a quadrilateral yield dual generalizations, where similar rectangles (rather than squares) are constructed outwardly on each side. For a base rectangle, this adaptation preserves the perpendicularity and equality of the segments joining opposite centers, with the aspect ratios of the erected rectangles influencing the orientation but not the core relation; specifically, if the erected rectangles share the same aspect ratio, the joining lines remain perpendicular and equal, mirroring the original theorem's outcome. In vector terms, this can be interpreted by decomposing the rectangle's sides into orthogonal basis vectors $\vec{u}$ and $\vec{v}$, where the centers of the erected figures are offset by scaled rotations of these vectors (e.g., 90-degree rotations for squares), leading to vector sums that confirm the perpendicularity via zero dot product.^[60] In vector geometry, rectangles admit a natural decomposition as the parallelogram spanned by two orthogonal vectors $\vec{a}$ and $\vec{b}$ with $\vec{a} \cdot \vec{b} = 0$, enabling theorems on diagonal equality and symmetry to emerge from norm properties: the diagonals are $\vec{a} + \vec{b}$ and $\vec{a} - \vec{b}$, satisfying $|\vec{a} + \vec{b}|^2 = |\vec{a} - \vec{b}|^2 = \|\vec{a}\|^2 + \|\vec{b}\|^2$ due to the orthogonality condition.^[61] This representation underpins applications in coordinate systems, where the British flag theorem follows directly from the invariance of squared distances under orthogonal projections.^[56]

Variant Forms

Crossed Rectangles

A crossed rectangle is a self-intersecting quadrilateral formed by two opposite sides of a rectangle together with its two diagonals, resulting in a bowtie shape where the diagonals intersect at their midpoint, the center of the original rectangle.^[62] This configuration distinguishes it from a simple rectangle by introducing self-intersection, with the "sides" of the crossed figure consisting of the selected rectangle sides and the crossing diagonals.^[63] Unlike a standard rectangle, a crossed rectangle does not have all interior angles measuring 90 degrees; instead, it features two acute angles and two reflex angles, with the sum of its interior angles totaling 360 degrees, consistent with properties of quadrilaterals. It possesses two perpendicular axes of symmetry passing through the intersection point of the diagonals and exhibits point symmetry at that center.^[62] As a special case of an antiparallelogram, its opposite sides are equal in length, but the parallel pairs from the original rectangle become the non-parallel elements in the crossed form.^[64] The area of a crossed rectangle is calculated via triangle subtraction, where the figure is divided into two triangular regions formed by the intersecting diagonals, and the areas are combined accounting for any signed contributions or overlaps in the self-intersecting path, often yielding the algebraic difference or net enclosed region relative to the original rectangle's configuration.^[62] In complex geometry, crossed rectangles are classified as star quadrilaterals, representing non-simple polygons with density greater than 1 at the intersection point, analogous to other self-intersecting forms like the pentagram but for four sides.^[64] Historically, crossed rectangles appear in discussions of pseudopolygons, which encompass self-intersecting figures beyond convex polygons, though formal study emerged in the context of quadrilateral variants in the 19th century. In modern computational geometry, they are relevant for algorithms handling self-intersecting polygons, such as decomposition into simple components for rendering, clipping, and intersection detection in computer graphics and CAD systems.^[65]

Other Non-Standard Rectangles

Curvilinear rectangles arise in the context of orthogonal curvilinear coordinate systems on surfaces, where they represent regions bounded by four coordinate curves that intersect at right angles, though the curves themselves are generally curved rather than straight. In such systems, the coordinate lines form an orthogonal grid, ensuring 90-degree angles at intersections, but the "sides" follow the curvature of the coordinate curves. For example, in polar coordinates $(r, \theta)$, a curvilinear rectangle is defined by constant ranges of $r$ (along radial rays, which are straight) and $\theta$ (along circular arcs, which are curved), creating a sector-like region with orthogonal boundaries. This concept is fundamental in differential geometry for parameterizing surfaces and computing integrals over curved domains.^[66] In spherical geometry, a rectangle is a quadrilateral on the sphere bounded by four geodesic arcs (great circle segments) that meet at right angles, analogous to Euclidean rectangles but adapted to the sphere's positive curvature. These geodesics serve as the "straight" sides, with interior angles typically exceeding 90 degrees due to the geometry, and the figure encloses a region of positive area proportional to the excess angle sum over 2π. Such spherical rectangles are used in cartography and global positioning to model bounded regions on curved surfaces, like latitude-longitude grids where meridians and parallels approximate geodesics. Properties include uniqueness up to isometry based on the angle between pairing great circles, with the modulus approaching zero for small angles.^[67] Degenerate cases of rectangles occur as limiting configurations where one or more dimensions approach zero, reducing the figure to a line segment or a point while preserving certain topological or metric properties in the limit. For instance, a rectangle with fixed perimeter but height tending to zero collapses to a line segment of length half the perimeter, retaining the boundary length but losing interior area and the four-sided structure. Similarly, further degeneration to a point arises when both dimensions vanish, representing a zero-area limit with no extent. These cases are important in geometric computations and topology to handle boundary conditions without singularities.^[68] In computer graphics, rounded rectangles are modified rectangles featuring filleted corners, where each corner is replaced by a circular arc of specified radius to soften sharp edges. Mathematically, this shape is the convex hull of a rectangle with four equal-radius circles centered at its corners, blending straight segments with quarter-circles at the vertices. The corner radius determines the fillet size, adjustable independently for aesthetics or functionality, and is commonly implemented in vector graphics tools for UI elements like buttons. Rounded rectangles with top-rounded corners are commonly used in UI elements like bottom sheets, modal sheets, cards, or panels that slide up from the bottom of the screen; this creates a clean, modern look often seen in mobile apps and web interfaces. This form enhances visual appeal by evoking approachability while maintaining rectangular proportions for alignment.^[69][70][71]

Geometric Applications

Tessellations

Rectangles are among the simplest polygons capable of tessellating the Euclidean plane, forming a complete covering without gaps or overlaps through repeated translations along their sides. This periodic tiling creates a lattice structure where copies of the rectangle are placed adjacent to one another, aligning edges perfectly due to the parallel opposite sides and right angles inherent to the shape.^[72][73] Variations on this basic arrangement include staggered layouts, where rows of rectangles are offset by a fraction of their width, producing patterns like the running bond used in masonry. Rotated configurations, such as 90-degree alternations, are feasible primarily for squares but can approximate similar effects with non-square rectangles when combined with translations, though they may introduce minor asymmetries depending on the aspect ratio. The aspect ratio—defined as the ratio of length to width—impacts the tiling's structural and visual efficiency; for instance, more elongated rectangles (higher aspect ratios) in staggered patterns enhance shear resistance in load-bearing applications, while near-square ratios promote uniformity and ease of alignment.^[74] In architecture, rectangular tessellations manifest prominently in bricklaying patterns, such as the common bond or Flemish bond, where bricks (rectangular prisms in three dimensions, but planar rectangles in facade views) interlock to form durable walls that distribute loads effectively across surfaces. These patterns leverage the rectangle's geometric stability to cover large areas efficiently, minimizing material waste and ensuring aesthetic continuity in structures from ancient Roman walls to modern facades.^[75][76] Beyond Euclidean geometry, regular rectangular tilings appear in the hyperbolic plane as uniform tilings denoted by the Schläfli symbol {4, n} for n ≥ 7, where quadrilateral tiles meet n at each vertex, forming infinite patterns with angles less than 90 degrees that cover the curved space without gaps. These hyperbolic analogues differ from Euclidean rectangles by having acute interior angles to accommodate the geometry's negative curvature, enabling denser vertex figures than possible in flat space. Squared variants of rectangles, which involve dissections into smaller squares, represent finite applications of tessellation principles but are explored in greater detail elsewhere.^[77]

Tiled and Squared Rectangles

A tiled rectangle is a rectangle dissected into a finite number of smaller rectangles that cover it exactly without overlaps or gaps. These dissections, also known as rectangular tilings, are always possible for any rectangle and form the basis for more specialized geometric constructions. They are classified as simple if no four rectangles meet at an interior point, ensuring a more structured arrangement without unnecessary junctions. Such tilings have applications in optimization problems, including efficient packing of rectangular objects and computational geometry algorithms for layout design.^[78] A squared rectangle represents a particular type of tiled rectangle where all the component tiles are squares rather than general rectangles. In this case, the dissection uses squares that may or may not have equal side lengths; if all squares have distinct sizes, the squared rectangle is termed perfect, whereas if any sizes repeat, it is imperfect. A perfect squared rectangle is further specified as simple if it contains no proper subset that itself forms a squared rectangle. The order of a squared rectangle is defined as the number of squares in the dissection. Imperfect squared rectangles are easier to construct and were among the earliest examples; for instance, early imperfect squared rectangles were constructed by Michio Abe in 1930, who generated over 600 such tilings. In contrast, perfect examples require all unique sizes, making them more challenging but mathematically elegant. The historical development of squared rectangles traces back to the 1930s, when R. L. Brooks, C. A. B. Smith, A. H. Stone, and W. T. Tutte at the University of Cambridge pioneered systematic constructions using graph-theoretic methods and analogies to electrical networks. Their work produced the first simple perfect squared rectangle of order 9—a 33×32 rectangle tiled with squares of side lengths 18, 15, 14, 10, 9, 8, 7, 4, and 1—demonstrating that perfect tilings were feasible for non-square outer shapes. This remains the lowest-order simple perfect squared rectangle known. Building on this, A. J. W. Duijvestijn advanced the field in 1978 by computationally discovering the lowest-order simple perfect squared square (a special case of squared rectangle where the outer shape is square) of order 21, using exhaustive enumeration on early computers.^[79] Construction algorithms for tiled and squared rectangles often rely on recursive subdivision or graph-based approaches. For general tiled rectangles, guillotine algorithms perform successive straight cuts parallel to the sides, partitioning the space into sub-rectangles until the desired number of tiles is achieved; this method guarantees a valid dissection and is efficient for computational implementations. Squared rectangles, particularly perfect ones, are constructed using "c-nets" (cycle networks), as formalized by Tutte in 1947, where vertices represent junctions, edges correspond to square sides, and electrical flow analogies ensure side length compatibility via Kirchhoff's laws. Duijvestijn's enumeration for higher orders employed backtracking over possible square placements, verifying uniqueness and simplicity. These methods prioritize avoiding repeated sizes in perfect cases and have been extended to generate examples up to order 30 or more, though exhaustive catalogs remain computationally intensive. Representative imperfect examples include a 4×3 rectangle tiled with one 3×3 square and three 1×1 squares, illustrating repeated sizes for lower complexity.^[80]

Notation and Representation

Mathematical Symbols

In classical Euclidean geometry, rectangles were primarily represented through diagrams where the vertices were labeled with sequential letters, such as A, B, C, and D, to denote the quadrilateral ABCD. This labeling convention allowed for precise references to sides and angles in proofs, as seen in Euclid's Elements, where rectangles are described as figures "contained by" specific line segments without a dedicated symbol beyond the point labels.^[81] In modern mathematical notation, a rectangle may be denoted by a boldface letter R to represent it as a geometric object or vector in abstract contexts, such as when treating it as a subset of a vector space. Vertices are still commonly labeled ABCD for clarity in diagrams and proofs. Standard variables for its dimensions include $ l $ for length, $ w $ for width, and $ d $ for the diagonal, facilitating calculations like the Pythagorean relation $ d = \sqrt{l^2 + w^2} $. These conventions appear consistently in geometry textbooks and emphasize the rectangle's right-angled structure. For typesetting in LaTeX, there is no standard dedicated command for a rectangle symbol in core packages, but more complex or labeled diagrams are often created using the TikZ package for custom drawings.^[82]

Unicode Characters

In the Unicode Standard, rectangles are represented by specific characters primarily in the Geometric Shapes block (U+25A0–U+25FF), which includes symbols for various geometric forms suitable for text-based diagrams and illustrations.^[83] Key examples include U+25AC BLACK RECTANGLE (a solid black rectangle used for filled shapes); U+25AD WHITE RECTANGLE (an outlined or empty rectangle for unfilled representations); U+25AE BLACK VERTICAL RECTANGLE (a tall, narrow solid form often employed as a histogram marker); and U+25AF WHITE VERTICAL RECTANGLE (its outlined counterpart).^[83] These characters enable precise depiction of rectangular elements in digital text, supporting applications from simple annotations to complex geometric layouts. For approximating solid or shaded rectangles, the Block Elements block (U+2580–U+259F) provides versatile options, with U+2588 FULL BLOCK (█) being particularly useful as a dense, solid square that can be repeated horizontally and vertically to form larger rectangular areas in monospaced fonts.^[84] This character, equivalent in appearance to a black square, facilitates the creation of custom-sized rectangles without relying on dedicated shape symbols, though it requires careful alignment to maintain proportions. Emoji offer informal proxies for rectangles, often leveraging box-like icons for visual representation in modern interfaces. The PACKAGE emoji (📦, U+1F4E6) depicts a rectangular parcel, serving as a common stand-in for boxed or rectangular objects, while the RED SQUARE (🟥, U+1F7E5) from the Geometric Shapes Extended block acts as a colored rectangular proxy in contexts requiring emphasis or variation.^[85] These are not strict geometric rectangles but provide accessible, platform-rendered alternatives for non-technical use. Despite their utility, Unicode rectangle characters face limitations in text rendering, as support varies across fonts, devices, and applications; for instance, older systems or basic ASCII environments may display them as placeholders or fail to align them properly in proportional fonts. In such cases, alternatives like ASCII art using box-drawing characters from the Box Drawing block (U+2500–U+257F)—such as U+2502 VERTICAL LINE (│), U+2500 BOX DRAWINGS LIGHT HORIZONTAL (─), and U+2514 BOX DRAWINGS LIGHT UP AND RIGHT (└)—allow construction of rectangular frames through combinations, ensuring compatibility in legacy text displays. This method, while more labor-intensive, preserves rectangular intent without advanced encoding.