An array is a systematic arrangement of similar objects, usually in rows and columns.^[1] The term is used across various fields. In mathematics and computing, it refers to ordered collections of data, such as matrices or data structures for efficient storage and access. In physical sciences and engineering, arrays describe configurations like antenna or telescope arrays for signal processing. Biological applications include DNA microarrays for gene analysis and protein arrays for diagnostics. Other uses appear in music (e.g., sound arrays) and military contexts (e.g., historical formations).

Mathematics and Computing

Mathematical arrays

In mathematics, an array is defined as a systematic arrangement of numbers, symbols, or expressions organized in rows and columns, forming a rectangular structure that facilitates organized data representation and computation.^[1] This concept is often used interchangeably with the term "matrix" in elementary contexts, where a matrix specifically denotes a two-dimensional array equipped with algebraic operations, though arrays can extend to higher dimensions as ordered lists of lists with uniform lengths at each level.^[1] One-dimensional arrays, such as row vectors (arranged horizontally) or column vectors (arranged vertically), represent linear sequences, while multidimensional arrays generalize this to tensors in advanced settings.^[2] The historical development of mathematical arrays traces back to early tabulations in the 18th and 19th centuries, where they served as tools for organizing complex calculations. Leonhard Euler, in his work around 1782, explored square arrays of symbols known as Graeco-Latin squares, which are orthogonal arrangements ensuring unique pairings in rows and columns, laying groundwork for combinatorial designs. Carl Friedrich Gauss advanced their application in 1809 through his "Theoria Motus Corporum Coelestium," where he employed array-like structures to solve systems of linear equations via least squares methods, treating observations as rectangular tabulations for astronomical data reduction.^[3] These early uses evolved into the formal matrix theory formalized by Arthur Cayley in the 1850s, emphasizing arrays as foundational for linear algebra.^[3] Key properties of mathematical arrays include indexing, which assigns positions to elements—denoted as $ A_{ij} $ for the element in the $ i $-th row and $ j $-th column in a two-dimensional array—and support for various operations when dimensions align. For addition, if two arrays $ A $ and $ B $ have the same dimensions, their sum $ C = A + B $ is defined component-wise such that

$$C_{ij} = A_{ij} + B_{ij}$$

for all $ i, j $.^[2] Scalar multiplication scales each element by a constant $ k $, yielding $ (kA){ij} = k \cdot A{ij} $, while transposition swaps rows and columns, producing $ A^T $ where $ (A^T){ij} = A{ji} $.^[2] These operations preserve the array structure and enable manipulations like solving linear systems, where a coefficient array $ A $ (e.g., a 2×2 matrix $\begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}$) multiplies a column vector $ \mathbf{x} $ to equal another vector $ \mathbf{b} $, as in $ A\mathbf{x} = \mathbf{b} $.^[2] Examples illustrate these concepts: a row vector like $ [1, 2, 3] $ arrays scalars horizontally for sequence representation, a column vector $ \begin{pmatrix} 1 \ 2 \ 3 \end{pmatrix} $ does so vertically for vector spaces, and a simple 2D array such as $\begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}$ models coefficients in linear equations, solvable via methods like Gaussian elimination.^[2] In statistics, arrays underpin contingency tables, which are two-dimensional count arrays displaying frequencies of categorical variables' joint occurrences, enabling analyses like chi-squared tests for independence. Data matrices, as rectangular arrays of observations and variables, further support multivariate statistical techniques, such as principal component analysis.^[4]

Arrays in computer science

In computer science, an array is a fundamental data structure that stores a fixed-size collection of elements of the same data type in a contiguous block of memory, allowing efficient access via indices that typically start from 0.^[5][6] This structure enables direct indexing to retrieve or modify elements, making it suitable for scenarios where the number of elements is known in advance and random access is frequent.^[7] Arrays come in various types to accommodate different needs. Static arrays have a fixed size determined at compile time, with memory allocated once and unresizable during execution, while dynamic arrays allow resizing at runtime through mechanisms like heap allocation, though this may involve reallocation and copying for efficiency.^[8] One-dimensional arrays represent linear sequences, such as a list of numbers, whereas multidimensional arrays simulate grids or matrices, like a 2D array for image pixels accessed as array[i][j].^[9] Jagged arrays, a variant of multidimensional arrays, consist of arrays of varying lengths within a single dimension, enabling irregular structures without wasting space in rectangular formats.^[10] Key operations on arrays include initialization, which sets all elements to a default value; access and modification, both achieving O(1) time complexity due to direct index calculation; insertion and deletion, which require shifting elements and thus take O(n) time in the worst case; and searching, which is O(n) for linear scans but O(log n) for binary search on sorted arrays.^[11][12] Traversal, a common operation, can be implemented via a simple loop, as shown in the following pseudocode:

for i from 0 to length-1:
    process array[i]

This iterates through all elements sequentially in O(n time.^[13] Memory management for arrays relies on contiguous allocation, where elements occupy sequential addresses to facilitate cache-friendly access and constant-time indexing via offset calculations. Many programming languages incorporate bounds checking to verify indices before access, preventing buffer overflows that could lead to security vulnerabilities or crashes, though this adds overhead in performance-critical code.^[14] The concept of arrays originated in the 1950s with the development of FORTRAN, the first high-level programming language designed for scientific computing, where arrays enabled efficient numerical processing on early computers.^[15] Over time, arrays evolved to support parallelism in modern languages, such as through coarrays in Fortran standards, allowing distributed memory access across processors for high-performance computing applications.^[16] Static arrays face limitations due to their fixed size, which can lead to inefficiency or failure if the required capacity changes unpredictably, prompting alternatives like linked lists that offer dynamic sizing at the cost of slower access times.^[17][18]

Physical Sciences and Engineering

Antenna arrays

An antenna array is a configuration of multiple antennas arranged to function collectively as a single radiating or receiving system, enhancing directivity, radiation patterns, and beamforming capabilities in electromagnetic applications. By exploiting interference effects among the elements, arrays achieve narrower beams and higher gain than individual antennas, enabling precise control over signal direction and strength.^[19] Key types of antenna arrays include linear arrays, with elements aligned along a straight line for one-dimensional beam steering; planar arrays, featuring elements in a two-dimensional grid for broader coverage and shaping; and circular arrays, arranged in a ring for omnidirectional or azimuthal scanning. Phased arrays, applicable across these geometries, facilitate electronic beam steering by varying the phase and amplitude of excitation signals to each element, eliminating the need for physical repositioning.^[20][21] The underlying principle governing array performance is the array factor, which describes the far-field radiation pattern resulting from element interactions. For a uniform linear array of $N$ isotropic elements spaced by distance $d$, the array factor is expressed as

$$AF(\theta) = \sum_{m=0}^{N-1} e^{j(k d \sin\theta \, m + \phi_m)},$$

where $k = 2\pi / \lambda$ is the wave number, $\theta$ is the observation angle relative to the array axis, and $\phi_m$ is the progressive phase shift for the $m$-th element. This formulation captures how phase differences and spacing influence constructive interference in desired directions and destructive interference elsewhere, determining beamwidth and sidelobe levels. Historical development traces to the early 20th century, when Guglielmo Marconi pioneered directional transmission using multiple antennas for transatlantic radio signals in 1906, marking an initial step toward array concepts for improved range and selectivity. Subsequent advancements led to modern adaptive arrays, which integrate digital signal processing to dynamically adjust weights for interference cancellation, enhancing robustness in multipath and jammed environments.^[22][23] Antenna arrays find critical applications in radar systems for target detection and velocity estimation through Doppler processing; in 5G wireless communications via massive MIMO setups, where large arrays support spatial multiplexing for higher throughput; and in radio astronomy, enabling synthesis imaging with high angular resolution. These uses leverage arrays' ability to form directive beams that concentrate energy efficiently.^[24][25][26] Arrays offer advantages such as significantly increased gain—scaling with the number of elements—and superior resolution for distinguishing closely spaced signals, far surpassing single-element performance. However, challenges arise from mutual coupling, where electromagnetic interactions between closely spaced elements alter impedance and distort the intended pattern, potentially reducing efficiency and beam accuracy.^[27]

Telescope arrays

Telescope arrays in astronomy consist of networks of individual telescopes that function collectively as a single, much larger instrument through the technique of interferometry, enabling the achievement of high angular resolution beyond the capabilities of any single telescope. This approach synthesizes signals from multiple telescopes to simulate a virtual aperture with a diameter equal to the separation between the farthest telescopes, or baseline, thus resolving fine details in celestial objects. The primary types of telescope arrays are radio telescope arrays and optical/infrared arrays. Radio arrays, such as the Karl G. Jansky Very Large Array (VLA) in New Mexico, which has been operational since 1980 and features 27 movable antennas spanning up to 36 kilometers, are designed to observe emissions at radio wavelengths for mapping extended sources like galaxies and supernova remnants. In contrast, optical and infrared arrays, exemplified by the CHARA array on Mount Wilson in California with six 1-meter telescopes providing baselines up to 330 meters, target visible and near-infrared light to image stellar surfaces and binary systems. The fundamental principle of telescope arrays relies on the baseline length to determine angular resolution, approximated by the formula $\theta \approx \frac{\lambda}{B}$, where $\theta$ is the resolution angle in radians, $\lambda$ is the observing wavelength, and $B$ is the maximum baseline between telescopes. Data from each telescope is combined through correlation processes, where the interference patterns of incoming wavefronts are analyzed to reconstruct high-fidelity images, often requiring complex algorithms to account for phase differences. Applications of telescope arrays span a wide range of astronomical investigations, including detailed mapping of radio sources such as pulsar distributions and the structure of active galactic nuclei. A landmark achievement came from the Event Horizon Telescope (EHT), a global array of radio telescopes that in 2019 produced the first image of the supermassive black hole in the galaxy M87, revealing its shadow against surrounding plasma emissions at a resolution of 20 microarcseconds. Subsequent observations confirmed the persistent nature of the M87* black hole shadow in January 2024 using 2017 and 2018 data. In September 2025, new EHT images from multi-year observations (2017-2021) revealed unexpected polarization flips in the magnetic fields around M87*, indicating a dynamic environment near the event horizon. Additionally, an October 2025 study demonstrated how EHT black hole images can serve as ultra-sensitive detectors for dark matter annihilation signals. The EHT also imaged Sagittarius A* at the Milky Way's center, with results published in 2022 from 2017 data.^[28][29][30] Historically, the VLA's completion in 1981 marked a pivotal milestone, providing unprecedented sensitivity and resolution for radio astronomy and influencing subsequent designs. Advancements in very long baseline interferometry (VLBI) have enabled global-scale arrays like the EHT, with continued improvements after 2020 enhancing capabilities, incorporating telescopes across continents for Earth-sized baselines exceeding 10,000 kilometers, as demonstrated in the 2022 imaging of Sagittarius A* at the Milky Way's center. Key challenges in operating telescope arrays include precise synchronization of signals over vast distances, often requiring atomic clocks and high-speed data recording to maintain coherence in VLBI setups. Optical arrays face additional hurdles from atmospheric turbulence, which distorts wavefronts and necessitates adaptive optics or closure-phase techniques to preserve image quality.

Biological Sciences

DNA microarrays

DNA microarrays, also known as DNA chips or gene chips, are small solid supports, typically glass slides or silicon chips, onto which thousands to millions of microscopic DNA probes are arranged in a grid pattern to enable the simultaneous analysis of gene expression, genetic variations, and other genomic features through hybridization with target nucleic acid sequences.^[31][32] The technology's historical development began in the late 1980s with the introduction of very large scale immobilized polymer synthesis (VLSIPS) by researchers at Affymax, leading to the first microarray for peptide synthesis published in 1991.^[33] Affymetrix, established as a spin-off from Affymax starting in 1990, developed key methods for light-directed in situ synthesis of DNA probes, leading to patents such as USPTO 5,744,305 issued in 1998 (filed in 1989), launching the GeneChip system in 1994 and receiving a $31.5 million Advanced Technology Program grant from the U.S. Department of Commerce.^[33] By the early 2000s, DNA microarrays achieved widespread adoption in genomics research, particularly in cancer studies where they facilitated gene expression profiling to identify tumor subtypes and biomarkers, with over 130 peer-reviewed studies published before 1999 and federal funding from the NIH accelerating their integration into academic and clinical workflows.^[33][34] Two primary types of DNA microarrays exist: cDNA microarrays, which use longer DNA fragments (typically 500–2000 base pairs) derived from PCR-amplified cDNA and spotted onto slides via robotic printing, and oligonucleotide microarrays, which employ shorter synthetic probes (25–60 base pairs) either spotted or synthesized in situ using photolithography (e.g., Affymetrix GeneChips) or inkjet printing (e.g., Agilent arrays), offering greater specificity for distinguishing similar sequences.^[35] The fabrication process involves attaching DNA probes to the chip surface, followed by hybridization where fluorescently labeled target DNA or RNA from a sample binds to complementary probes; unbound targets are washed away, and a laser scanner detects fluorescence intensities to quantify binding.^[31][35] Data analysis typically compares signal intensities between experimental and control samples—often using ratio-based metrics—to identify differentially expressed genes or variants, with software normalizing for background noise and technical variability.^[35] Applications of DNA microarrays include gene expression profiling to measure mRNA levels across thousands of genes simultaneously, genotyping to detect single nucleotide polymorphisms (SNPs) with call rates exceeding 99.5% for over 1 million markers, and identifying mutations such as those in BRCA1/BRCA2 for cancer risk or HIV-1 drug resistance.^[35] They played a key role in the Human Genome Project (completed in 2003) by enabling large-scale mutation detection and population genomics studies, contributing to the mapping and sequencing of the human genome.^[31][36] Despite their impact, DNA microarrays have limitations, including cross-hybridization where related sequences bind non-specifically to probes, leading to false positives in complex genomes, and their static design, which only detects predefined sequences and misses novel or low-abundance transcripts.^[35][37] Post-2010, integration with next-generation sequencing has addressed some shortcomings by providing higher resolution for dynamic genomic analysis, though microarrays remain cost-effective for targeted applications. As of 2025, the DNA microarray market continues to expand, projected to reach USD 6.85 billion, driven by applications in personalized medicine and integration with NGS technologies.^[35][38]

Protein and tissue arrays

Protein and tissue arrays represent high-throughput platforms in proteomics and pathology, featuring grids of immobilized proteins or tissue samples that facilitate simultaneous assays for protein expression, interactions, and modifications across numerous targets. These arrays enable the miniaturization of traditional assays, allowing researchers to analyze hundreds to thousands of samples on a single slide or chip, thereby accelerating biomarker discovery and validation in biological research. Unlike single-plex methods, they support multiplexed detection, reducing reagent use and experimental time while preserving limited biological materials.^[39] Protein arrays, often termed protein microarrays, are categorized into analytical, functional, and reverse-phase types based on their fabrication and purpose. Analytical protein arrays capture native proteins from complex mixtures, such as serum or cell lysates, using immobilized capture agents like antibodies to quantify specific analytes. Functional protein arrays, in contrast, display purified recombinant proteins or peptides to study interactions, such as enzyme-substrate binding or ligand-receptor affinities. Reverse-phase protein arrays (RPPAs) involve printing diluted cell or tissue lysates onto surfaces and probing them with antibodies to profile activation states in signaling pathways, particularly useful for detecting low-abundance phosphorylated proteins. Applications include high-throughput antibody screening, where antigen arrays identify monoclonal antibodies for therapeutics, and kinase activity assays that map dynamic signaling cascades in disease models. For instance, RPPAs have been instrumental in dissecting cancer signaling pathways by quantifying pathway nodes like PI3K/AKT and MAPK across tumor samples.^[40][41] Recent advancements, such as Illumina's Protein Prep launched in 2025 measuring 9500 unique human protein targets, highlight evolving applications in proteomics.^[42] Tissue microarrays (TMAs) extend this technology to histopathology by coring multiple paraffin-embedded tissue specimens—typically 0.6 to 2 mm in diameter—and embedding them into a single recipient block for sectioning and parallel analysis. The foundational method emerged in 1987 with Wan et al.'s syringe-based sampling for multitissue blocks, but the high-density TMA format was pioneered by Kononen et al. in 1998, permitting the interrogation of up to 1,000 cores per array for molecular profiling. Standardization in the 2000s, including automated punching devices and digital imaging, has made TMAs a staple in clinical research. Detection relies on immunohistochemistry with primary and secondary antibodies, often visualized via chromogenic or fluorescent signals, or increasingly by mass spectrometry for multiplexed protein quantification; image analysis software then enables semi-automated scoring of staining intensity and distribution.^[43][44][45] These arrays drive applications in drug discovery, where functional protein arrays screen compound libraries for binding affinities, and in biomarker identification, notably in oncology trials post-2015, where TMAs have validated predictive markers like HER2 expression in breast cancer cohorts from large-scale studies. For example, TMAs constructed from archival tumor tissues have supported pharmacogenomic analyses in soft tissue sarcoma trials, correlating protein markers with treatment outcomes. Relative to Western blots, which analyze one sample per gel with limited multiplexing, protein and tissue arrays provide superior throughput—processing 500+ samples concurrently—along with reduced sample requirements and integrated spatial context in TMAs; however, challenges include intra-array variability from tissue heterogeneity and the necessity for orthogonal validation to confirm array-derived signals. Protein arrays can integrate with DNA microarrays to link genomic alterations to proteomic outcomes, offering a holistic view of disease mechanisms.^[46][47][48]

Other Uses

Arrays in music

In advanced music theory, particularly serialism and combinatorial composition, an "array" refers to a structured, ordered arrangement of musical elements such as pitches, durations, or dynamics, often represented mathematically. This usage appears in modern techniques where composers manipulate these structures for variation and unity, as in the spatial organization of parameters in electroacoustic music.^[49] Historically, arrays appear in modern music theory through serialism, where Arnold Schoenberg's twelve-tone technique treats the tone row as a fixed ordering of all twelve chromatic pitches, arranged to eliminate tonal centrality. Developed in the early 20th century, this method involves generating derivative forms—such as the inversion array, which mirrors the row's intervals around a central axis, or the retrograde array, which reverses the sequence—to create permutations for thematic development. Schoenberg's approach, detailed in his theoretical writings, influenced composers seeking emancipation from traditional harmony, with the tone row serving as a foundational structure for entire works. The term "array" is particularly used in extensions by composers like Milton Babbitt, who developed all-partition arrays for complex combinatorial serialism.^[50][51][49] In performance and synthesis contexts, the term is sometimes used analogously, such as describing the coordinated layers in choral polyphony or computational arrays of oscillators in electronic music to generate timbres. In electronic music, synthesizer design may employ arrays of multiple waveform generators in sequence or parallel for additive synthesis, layering harmonics.^[52][53] Central to composition, permutations and transformations of musical arrays facilitate variation without repetition, as composers reorder or modify elements to evolve motifs across sections. In serialism, for instance, applying operations to the prime row yields a matrix of interrelated forms, ensuring combinatorial unity. Karlheinz Stockhausen's 1950s electronic works exemplify this, with pieces like Studie II (1954) using serialized arrays to govern parameters such as frequency, amplitude, and duration, creating pointillistic structures from precise algorithmic arrangements. This integration of array-based serialization marked a pivotal advancement in electroacoustic composition.^[54][55]

Arrays in military and history

Historically, the term "array" has referred to the ordered arrangement or formation of troops, ships, or weapons in rows, columns, or geometric patterns to optimize combat effectiveness or ceremonial display.^[56] These formations have evolved from rigid ancient structures to adaptive modern configurations, emphasizing coordinated positioning to leverage terrain, firepower, and mutual protection.^[57] One of the earliest prominent examples is the Roman testudo formation, dating to the 1st century BCE, where legionaries arranged in a compact rectangular formation held shields vertically on the front and sides while raising them horizontally overhead to form a protective "turtle shell" against arrow fire and projectiles.^[58] This tactic proved effective during sieges, as described by ancient historians like Plutarch, but revealed vulnerabilities to mobile cavalry attacks, such as at the Battle of Carrhae in 53 BCE.^[59] In the early 19th century, Napoleonic line infantry formations represented a shift toward linear deployments for maximizing musket volleys, with battalions forming extended lines to deliver devastating firepower while minimizing exposure on flanks.^[60] French forces often transitioned from marching columns to these lines for assaults, though British defenders exploited the formation's rigidity at battles like Vimeiro in 1808 and Waterloo in 1815.^[60] In modern warfare, naval task force formations emerged during World War II, where U.S. carrier groups adopted circular defensive arrangements to shield central aircraft carriers with concentric rings of battleships, cruisers, and destroyers.^[61] These arrangements, as seen in Task Force 58 operations, concentrated anti-aircraft fire and allowed rapid maneuvers to evade threats like kamikaze attacks.^[62] As of 2025, developments since 2020 have introduced drone swarms—coordinated groups of unmanned aerial vehicles operating in autonomous formations analogous to arrays—for reconnaissance, electronic warfare, and strikes. These have been demonstrated in the Ukraine conflict, with swarms of up to hundreds of drones, and by China's People's Liberation Army in exercises integrating AI for real-time adaptation and saturation attacks that overwhelm defenses.^[63][64] Tactical principles underlying military arrays focus on maximizing collective firepower while minimizing vulnerabilities through mutual support and terrain exploitation.^[65] Formations evolved from static squares, used in the 18th and early 19th centuries to repel cavalry charges, to more flexible arrangements that incorporate skirmishers and dispersed units for maneuverability in open or urban environments.^[66] This progression reflects advancements in weapons technology, from muskets to precision-guided munitions, allowing arrays to balance density for impact with agility to avoid concentrated enemy fire.^[67] Beyond combat, arrays hold cultural significance in ceremonies, such as military parades where troops form precise ranks to symbolize discipline and national unity. Historical examples include the Roman triumph of 223 BCE, where victorious legions marched in ordered formations displaying spoils, and the 1865 Grand Review in Washington, D.C., featuring 145,000 Union soldiers in linear arrangements to celebrate the Civil War's end. These displays, often incorporating marching band arrangements akin to musical arrays, reinforce historical narratives through reenactments of ancient battles.