Extremes of Conventional Music Notation

Donald Byrd, Indiana University Bloomington

revised mid October 2017


  1. Notwithstanding the heading below referring to "The Game", this list has a serious purpose, namely to suggest that conventional Western music notation is far more complex and subtle than most people think. In particular, it does not have well-defined borders; it just fades away indefinitely in all directions. It's widely understood that what music notation means -- its semantics -- is ill-defined. But even what constitutes valid notation -- its syntax -- is ill-defined, and even for individual symbols regardless of context. For example, there's no limit on the shortest note duration, on the number of p's or f's in a dynamic mark, or even on the number of accidentals on a note. This list includes published music with 2048th notes, with markings of pppppppp (eight p's) and ffffffff (eight f's), and with triple sharps and triple flats. I'm confident that the existence of every one of these would surprise most musicians, including music theorists and musicologists; but no one can say that even shorter notes, larger numbers of p's or f's, or quadruple sharps or flats aren't possible or valid.
  2. Rules of The Game
    1. Only published works are considered, except for music old enough that publication may not be a meaningful concept. Also, in general, implicit notation is excluded, e.g., Renaissance notation that is equivalent to complex tuplets. Where there's a tie among several works, I've tried to emphasize those that are the most "mainstream", especially in terms of composer and publisher. But works listed are automatically biased heavily towards mainstream works, and especially those by very well-known composers, since I'm less likely to even know about details in other works. (I doubt if Bach, Beethoven, etc., were really such great pioneers of notation as this document seems to indicate!) Runners-up are often listed: this is useful to give an idea of how unusual, and therefore how significant, the extreme is -- both in practical terms and as a feature of what computer scientists would call music representation).
    2. Another question is what to do about what is sometimes called “antimusic”, “conceptual music”, and so on. This includes pieces that are obviously unplayable or inaudible, e.g., with notes on so many ledger lines they're above the hearing range even of dogs, and pieces containing zero notes or to be repeated forever. I also put jokes and parodies of music notation like the well-known "Fairies Aire and Death Waltz" in this category. I generally show such items in square brackets "[ ]", but don't take them seriously as extremes. Naturally, the borderlines are not well-defined. There are also pieces that are perfectly playable and audible but whose interest is primarily conceptual, e.g., works that consist of a single note, and works of significant musical interest but in which the extreme feature is conceptual, e.g., where the piece is to be repeated in its entirety hundreds of times or forever. These are handled on a case-by-case basis. (Examples of each of the features mentioned in this paragraph appear herein.)
    3. Jokes and parodies of music notation like the well-known "Fairies Aire and Death Waltz", with its 32,768th note (or is it a 65,536th?) and so on, form a related category. These may be amusing, but they're irrelevant for my purposes.
    4. An important borderline case of music whose interest might be primarily conceptual is the "new complexity" music. To cite just one example, Brian Ferneyhough: String Quartet no. 3 (1987; Peters ed.), II, includes a number of tuplets nested 4 levels deep, and several meters with non-power-of-2 denominators, e.g., 2/10 and 1/12. My personal inclination is strongly towards considering such work as of more conceptual than musical interest, but this is probably unfair: even if I'm right, it might still have substantial musical interest! And I've rarely (if ever) listened to this music, so I really can't say. But see below for more on Ferneyhough's use of nested tuplets.
    5. Though this list concentrates on notation, some "sound" aspects are included, e.g., extreme sounding as well as written pitches.
  3. A related collection of examples of unusual music notation appears in Byrd (1984) and a supplement, More Counterexamples in Conventional Music Notation (Byrd, 2017). That collection concentrates not on extremes but on notation that breaks the supposed rules, e.g., two clefs simultaneously active on the same staff, time-signature changes in the middle of a measure. Byrd (1994) discusses at some length a few items from the original collection. However, the dividing line between extremes and rule-breaking is often unclear. In any case, the Gallery of Interesting Music Notation webpage (Byrd, 2016) shows and comments on particularly interesting examples of both.
  4. In terms of number of contributions and ideas of various kinds, Noam Elkies and Gavin Jared Bala are in a class by themselves. Thanks also for contributions and ideas to: Hans Aberg, David Alpher, Daniel Arribas, Jacob Baltuch, Kim Bastin, Arnold Black, James Brady, Peter Burkholder, Jack Campin, Silas Cordiero, Alistair Craft, Tim Crawford, Myke Cuthbert, Samuel Dickenson, Matthew Dovey, Halvor K. Hosar, Eric Isaacson, Michael Fingerhut, Nicholas Forty, J. Gedan, Michael Good, Jay Gottlieb, Jeremy Grimshaw, Bill Guerin, Jim Halliday, Flinder Hiew, Douglas Hofstadter, Julian Hook, Sean Hunt, James Ingram, Clovis Lark, Steve Larson, David Lasocki, Joyce Lindorff, Geoffrey Liu, Tom McCanna, John McKay, David Meckler, Jose Montalvo, Scott Murphy, Nigel Nettheim, Nick Nicholas, Laurel Parsons, James Primosch, Laurent Pugin, Stan Shumway, Jeffrey Solow, Mark Starr, Russell Tinkham, Frans Wiering, Richard White, Daniel Wolf, Michael Zarky. I am not a professional musicologist, and the list could certainly be improved in many ways; contributions are invited. Asterisk ("*") indicates items I haven't seen personally.
  5. This material is based in part on work supported by the National Science Foundation under Grant No. 9909068. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
  6. Abbreviations used herein: "CMN" = Conventional Music Notation (also called Traditional Music Notation, Conventional Western Music Notation, etc.); "B&I" = Byrd & Isaacson (2016); "B & M, 1948" = Barlow & Morgenstern (1948).
  7. A small amount of this material appeared in Hewlett & Selfridge-Field (1991, 1992). Donald Byrd, Research Technologies and School of Informatics, Indiana University; e-mail, donbyrd(at)indiana.edu .


  1. Arnold, Denis, ed. (1983). The New Oxford Companion to Music. 2 vols. Oxford: Oxford University Press.
  2. Badura-Skoda, Paul & Eva (1962). Interpreting Mozart at the Keyboard. New York: St. Martin's.
  3. Barlow, Harold & Morgenstern, Sam (1948; henceforth "B & M, 1948"). A Dictionary of Musical Themes. New York: Crown Publishers.
  4. Burkholder, J. Peter, & Palisca, Claude, eds. (2006). Norton Anthology of Western Music, 5th ed. New York: W. W. Norton.
  5. Byrd, Donald (1984). Music Notation by Computer. Ph.D. dissertation, Computer Science Dept., Indiana University; available from UMI, www.umi.com .
  6. Byrd, Donald (1994). Music-Notation Software and Intelligence. Computer Music Journal 18(1), pp. 17-20.
  7. Byrd, Donald (2016). Gallery of Interesting Music Notation. Available at: http://www.informatics.indiana.edu/donbyrd/InterestingMusicNotation.html
  8. Byrd, Donald (2017). More Counterexamples in Conventional Music Notation. Available at: http://www.informatics.indiana.edu/donbyrd/MoreCMNCounterexamples.htm
  9. Byrd, Donald, & Isaacson, Eric (2016) (henceforth "B&I"). A Music Representation Requirement Specification for Academia. Based on our 2003 Computer Music Journal paper of the same title. Available at: http://www.informatics.indiana.edu/donbyrd/Papers/MusicRepReqForAcad.doc .
  10. Dumitrescu, Theodor (2007). Constructing a Canonic Pitch Spiral: The Case of Salve Radix. In Canons and Canonic Techniques, 14th-16th Centuries, ed. Kateligne Schiltz and Bonnie J. Blackburn (Leuven: Peeters), pp. 141-170.
  11. Hewlett, Walter, & Selfridge-Field, Eleanor, eds. (1991, 1992). Computing in Musicology, vols. 7 & 8. Menlo Park, California: Center for Computer-Assisted Research in the Humanities.
  12. Kirkpatrick, Ralph (1953). Domenico Scarlatti. Princeton University Press.
  13. LaRue, Jan (1988). A Catalogue of Eighteenth-Century Symphonies, Vol. I. Bloomington: Indiana University Press.
  14. Picken, Laurence E.R. (1975). Folk Musical Instruments of Turkey.
  15. Rastall, Richard (1982). The Notation of Western Music. New York: St. Martins.
  16. Read, Gardner (1969). Music Notation. 2nd ed. Boston: Crescendo.
  17. Read, Gardner (1978). Modern Rhythmic Notation. Bloomington: Indiana University Press.
  18. Risatti, Howard (1975). New Music Vocabulary. Urbana: University of Illinois Press.
  19. Sadie, Stanley, ed. (1980). The New Grove Dictionary of Music and Musicians, 1st ed. Macmillan.
  20. Sadie, Stanley, ed. (2001). The New Grove Dictionary of Music and Musicians, 2nd ed. Macmillan.
  21. Schenk, E., ed. (1997). Denkmaler der Tonkunst in Oesterreich, vol. 151. Graz: Akademische Druck- u. Verlagsanstalt.
  22. Warner, Thomas (1977). Tromlitz's Flute Treatise: A Neglected Source of Eighteenth-Century Performance Practice. In A Musical Offering: Essays in Honor of Martin Bernstein ed. by E. Clinkscale and C. Brook. Pendragon Press.


Octave notation here is in the international standard ISO system, formerly known as the ASA (Acoustical Society of America) or ANSI system. In this system, middle C (MIDI note number 60) is C4; octaves start with C, so the B just below (MIDI number 59) is B3. The lowest note of the normal 88-key modern piano is A0 (MIDI 21); the highest note is C8 (MIDI 108). (Boesendorfer Imperials, which have existed since ca. 1900, go down to F0 or even C0. And Stuart & Sons of New South Wales, Australia now make a 102-key piano with a range of C0 to F8 (contributed by Baltuch); I believe they introduced it just a few years ago.)

1. Most ledger lines {B&I 4.7}

  1. Above staff: 9 in Martino: Pianississimo (1970), for C8.
    Runners-up: 8 in Pianississimo for several A7's and B7's; 8 in Shostakovich: Piano Trio no. 2 in E minor, Op. 67, end of first movement, violin part (International ed.; contributed by Elkies); 7 in Sibelius: Violin Concerto (1905; Kalmus ed.), III, for F#7 in solo part (occurs many times); Schoenberg: Violin Concerto (1936); Davidovsky: Synchronism no. 1; etc. [Tom Johnson: Celestial Music for Imaginary Trumpets (1974) contains a chord whose top note is about 100(!) ledger lines above the staff, in treble clef.]
  2. Below staff: 6 is fairly common in piano music for B0 or A0, e.g., Brahms: Rhapsodies, Op. 79, both nos. 1 and 2 (Peters/Sauer ed.); Debussy: Pour le Piano (1901; Schirmer ed.), I.

2. Extreme pitches {B&I 4.7}

  1. Highest written and sounding: G#8 (MIDI note number 116) in Salvatore Sciarrino: Six Caprices for Violin, no. 1 (1976) (in harmonics). Runners-up: D8 in Scriabin: Piano Sonata no. 6, Op. 62 (1911, Dover & Schirmer eds.), last page (contributed by Dovey) (an editorial footnote in the Schirmer edition comments that this note "did not yet exist" on pianos; in fact, as far as I know, it didn't exist until the creation of the Stuart & Sons 102-key piano of the 21st century); also in Percy Grainger: *Let's Dance Gay in Green Meadow (Faber ed.) (contributed by Andrew Plant via Alistair Craft). Lesser runners-up include C#8 in Paganini: Violin Concerto no. 1, Op. 6 (1817), III, solo part, and Schoenberg: Violin Concerto, Op. 36 (1936), solo part (in harmonics in both pieces). [The note in Tom Johnson: Celestial Music for Imaginary Trumpets mentioned above, about 100 ledger lines above a staff in treble clef, would have a frequency of many gigahertz -- way, way above the audible range, even for dogs!]
  2. Lowest written and sounding: C0 (MIDI note 12) in Alexandre Guilmant: Organ Sonata no. 5 in c, Op. 80 (before 1909; Dover ed.), V, end (written C2 with a 32-ft. stop); in William Kraft: Encounters II for solo tuba (MCA Music, 1970) (contributed by Tinkham); and in William Bolcom: *Piano Concerto (1976) (contributed by White). This note has a frequency of 16.4 Hz, while it's generally considered that the lowest frequency that's heard as a pitch is about 20 Hz. Runners-up: E-flat-0 (MIDI 15) in *La Monte Young: Well-Tuned Piano (written for a specially-tuned Boesendorfer Imperial; apparently unpublished, like all his works, so not really in the scope of this list), frequency 18.6 Hz in Young's tuning system (contributed by Grimshaw); F0 (MIDI 17) in Bartok: Piano Sonata (1926; Boosey & Hawkes ed.), II, frequency 21.8 Hz -- this is also outside the range of all but a few pianos.

3. Extreme accidentals {B&I 4.8} and key signature {B&I 9.1}

  1. Regarding accidentals, triple sharps appear in *Fugue no. 34 of Anton Reicha: 36 Fugues for Piano (ca. 1805) (Magasin de l'imprimerie chymique ed.), m. 56, left hand (a C#x lower neighbor between two Dx's) (contributed by Nicholas); in Alkan: Etude no. 10 from Douze etudes dans tous les tons mineur, Op. 39 (1857) (contributed by Starr); and near the end of the last movement of Reger: Clarinet Sonata, Op. 49 no. 2, piano part (1904; Universal ed.) (F#x, used as a lower neighbor between two Gx's) (contributed by McCanna). The Alkan and Reger instances are each in a passage in 6 sharps; the Reicha has no key signature, making multiple sharps that much more remote in terms of tonality. Triple flats (for Bbbb) occur in Nikolai Roslavets (or Roslawez): Piano Sonata No. 1 (1914; Schott ed.), mm. 152 and 153 -- along with double flats in the same passage, their individual flats have their "stems" linked with "beams"! (contributed by Hook); they also occur in Galina Ustvolskaya: Piano Sonata No. 3, p. 30 (contributed by Cordiero).
  2. A key signature of 6 sharps and a double-sharp -- the equivalent of 8 sharps -- appears in the final pages the vocal score of *John Foulds: A World Requiem; the passage is in G-sharp major (contributed by Bala). Runners-up: Key signatures of 7 sharps and of 7 flats are not too unusual. 7 sharps is used in J.S. Bach: The Well-Tempered Clavier, Books I and II, for a prelude and fugue in C-sharp major in each (in B & M, 1948). 7 sharps is also used for C-sharp major at the end of the second movement of Schubert's last piano sonata (D.960, in B-flat) (contributed by Elkies). 7 flats is used for A-flat minor in Beethoven: Piano Sonata no. 12, Op. 26, I and III (contributed by Tinkham); also in Brahms: Horn Trio, Op.40, II (in B & M, 1948; contributed by Elkies). Note: Key signatures of up to one triple-flat + six double-flats = 15 normal flats (!) appear in Dumitrescu (2007), pp. 141-170. However, it's not clear that this qualifies as CMN. (contributed by Hook, who comments: "This is a modern realization of a 16th-century motet. The original manuscript was written in a circle. The author doing the reconstruction argues that the proper way to perform it requires adding flats at each repetition, and this is where that brings us by the third page.")

4. Extreme ottava {B&I 13.1}

  1. 15ma: Contrary to popular belief, 15ma (2 octaves transposition) is quite rare, though probably less so in the last 50 years or so than before. Still, the only instances I know of of 15ma alta in a well-known work are in the solo part of Ravel: Piano Concerto (Durand ed., 1932(), III (NB that this is incorrectly printed as "16a"). It's also used repeatedly in Messiaen's Oiseaux Exotiques (1956; Universal ed., 1959; again shown as "16"), plus piano music by Cowell (Advertisement, 1914; Associated ed.), Feldman (Durations III, no. 3, 1962), etc.
  2. 15ma bassa: The only instances I know of are for tone clusters in Cowell: The Tides of Manaunaun (probably written in 1917, though Cowell claimed 1912; Associated ed.) for piano; for the C0 in William Kraft: Encounters II for solo tuba (contributed by Tinkham); and for C0 for tuba in William Bolcom: *Piano Concerto (1976) (contributed by White).

5. Widest skip and range in a melody. (NB: of course the term "melody" isn't at all well-defined; as a result, many of these extremes are open to question.)

  1. Widest skip: 5 octaves + perfect 5th (67 semitones) in Beethoven: Piano Sonata no. 32, Op. 111 (1822), I, m. 116. (It's 5 octaves + minor 7th (70 semitones) in some editions; the old Bulow/Lebert ed. acknowledges the editors changed the higher note, justified by inference from the range of the Broadwood piano Beethoven had when he wrote the piece and by analogy with the passage with the runner-up wide skip in the same piece!)
  2. Runners-up for widest skip: 4 octaves + dim. 7th (57 semitones) in Beethoven: Piano Sonata no. 32, Op. 111, I, m. 49; 4 octaves (48 semitones) in D. Scarlatti: Sonata in c, L. 360, K. 22; 3 octaves + P 4th (41 st) in violin 1 of Beethoven: Quartet Op. 59 #3, I, ca. mm. 57-58, and in the solo part of Mozart: Piano Concerto no. 21, II (both contributed by Elkies).
  3. Widest overall range: 5 octaves + perfect 4th (65 st) in Saint-Saens: Cello Concerto no. 1, Op. 33, in a long, slow scale (pp. 53-54, Durand/Dover ed.). (This is arguably not part of a melody, but it's by far the most salient part of the texture for eight measures.) Runners-up: 4 octaves + maj 6th (57 st) in Beethoven: "Hammerklavier" Sonata, Op. 106, I -- except that 2 octaves of this is really a change of texture (via doubling). A better choice might be 4 octaves and at least a maj 3rd (52 st), perhaps a maj 6th (57 st), in Chopin: Etude in C major, Op. 10 no. 1: here the uncertainty is because it's not at all clear where the melody ends; however, none of it has a very melodic character in the normal sense.
  4. Runner-up for widest range: 3 octaves + min 2nd (37 st) in Mozart: Violin Concerto no. 5 ("Turkish"), I.

6. Most repeated notes in a melody. See above comments about the term "melody". Counts are of sounding notes, i.e., only notes with attacks, not tied notes. Only written out notes qualify, not tremolo.

  1. Without rhythmic variety (all notes the same sounding duration). 32 in Prokofieff: Toccata, Op. 11 (1912), m. 1 (first different pitch = maj 2nd up). It might be argued that this isn't a melody, but it's the beginning of the piece, unaccompanied, recurs in important places, etc. Runners-up: 15 in Khachaturian: Sabre Dance from "Gayane" (first different pitch = maj 2nd down); 14 in Beethoven: "Waldstein" Sonata, Op. 53, I, m.5 (first different pitch = maj 2nd up).
  2. With at least some rhythmic variety (one or more notes with different sounding duration). The 80 notes that comprise the entire voice part of Peter Cornelius's song "Ein Ton", Op. 3 #3 (contributed by Elkies). Runners-up: 50 in Chopin's song "Leci liscie z drzewa" ("Leaves are falling"), Op. 74 no. 17, mm. 50-67. 48 of the 50 are in one phrase (first different pitch = min 2nd up). (contributed by Nettheim). 35 in Cole Porter's "Night and Day" (contributed by Elkies). 30 in Jobim: One Note Samba (first different pitch = 4th up); here the number of repeated written notes is much larger, since many of the notes are tied. 20 in Bartok: Piano Sonata (1926), II, m.1 (first different pitch = maj 2nd down).

Rhythm and Single-Note Duration

1. Shortest notated duration

  1. For normal notes {B&I 4.5}
    1. m. 16 of Anthony Phillip Heinrich: Toccata Grande Cromatica from The Sylviad, Set 2 (ca. 1825), uses 1024th and even two 2048th(!) notes (in Byrd, 2016). However, the context shows clearly that these notes have one beam more than they should, so they should really be 512th and 1024th notes, respectively. The passage--in 2/4, marked "Grave"--also contains many 256th notes. (It's easy to argue that these durations are ridiculous: even at a tempo as slow as M.M. eighth = 40 (quarter = 20), a 1024th note would last only about 12 millisec.; cf. "Shortest performed duration" below. However, Heinrich obviously wanted a huge ritardando here, so they may not be ridiculous after all.) See J.B. Clark, The Dawning of American Keyboard Music (Greenwood Press, 1988), p. 365. (contributed by Shumway) (The first page of Ives: "Concord" Piano Sonata no. 2 (completed 1915) (Kalmus ed.), IV ("Thoreau") contains a beamed group of 64th notes ending with two 1024th notes, but it's obvious the extra beams on the "1024ths" are a mistake.)
    2. Runners-up: 256th notes appear in Vivaldi: Concerto in C for ottavino, archi e cembalo, F. IV n. 5; *Boguslav Schaeffer: 4 Movements for Piano and Orch. (mentioned in Read (1969), p. 117); the last variation in some editions (e.g. the Neue Mozart-Ausgabe) of Mozart's 12 Variations on "Je suis Lindor", K. 354 (contributed by Bala); and some editions (e.g., Barenreiter) of Beethoven: Piano Concerto no. 3, II. (In the CCARH 1987 Directory, Dydo says 256th notes occur in Telemann but gives no details. Perhaps he was thinking of the Vivaldi piece, though it's been suggested there are 256ths in the Gulliver Suite, but I haven't checked.)
    3. Interesting runners-up are 15:8 128ths in Beethoven: Piano Concerto no. 3, II, and Bartok: Rhapsody, Op. 1 (1904; Boosey & Hawkes ed.). These are very nearly as short as 256th notes.
  2. For grace notes (including appoggiaturas) {B&I 5.5}
    1. 128th grace notes in Alkan: Trois Grandes Etudes, Op. 76 (Billaudot ed.), no. 2, 3 mm. before the end (contributed by Dickenson).
    2. Runners-up: 64th grace notes are not too rare. Examples include Bartok: Rhapsody, Op. 1 and both of the Two Elegies, Op. 8B (Boosey & Hawkes eds.); Ravel: Gaspard de la Nuit (1908; Durand/Dover ed.), Ondine; Alkan: Trois Grandes Etudes, Op. 76 (Billaudot ed.), no. 2, 3 mm. before the end (contributed by Dickenson).

2. Longest notated duration, including ties

  1. In both measures and durational units (i.e., fixed units like quarters or wholes): Verdi: Otello (1887), Act I, opens with a tone cluster(!) in the organ of C2,C#2,D2, lasting 244 measures of 4/4 (= 976 quarters). (contributed by Starr)
  2. Runners-up (distant) in durational units: Bartok: The Wooden Prince (1914-17) opens with 120 measures of tied dotted halfs (= 360 quarters) of timpani roll on one pitch, but it's not clear this counts, since (despite the tie) it's really reiterated. Tied notes equivalent to 140 quarters in Beethoven: Symphony No. 9, II, basses (though peculiarities of the notation vaguely suggest subdivisions into sections the longest of which is 88 quarters). Unimpeachable runners-up are *Ravel: L'Enfant et les Sortileges, "pastorelles" section of Act I, bottom half of the alto section (84+ bars of 2/4 = 168.5 quarters) (contributed by Elkies); and Beethoven: Symphony no. 5, III, violas (43 measures of tied dotted halfs = 129 quarters).
  3. Runner-up (distant) in measures: Bach: Organ Toccata in F major begins with a pedal point of 54(?) measures (of 3/8; = 81 quarters).
  4. Notre Dame organum in CMN transcriptions has some fairly strong contenders, though nothing comparable to Otello. Cf. especially Perotin: Viderunt Omnes, "the F-major toccata of the 13th century", and Sederunt; each begins with what looks in a modern edition like 50 or 60 measures (= 150 or 180 quarters) of 6/8 of one tenor note.
  5. Longest continuous trill (considering only "real" trills in terms of both notation and sound; not timpani rolls, written-out fingered tremolos, etc.) on a single pitch: in measures (and runner-up in durational units), 10 measures of 3/4 in Chopin: Mazurka no. 53; in durational units (and runner-up in measures), 9 bars of 2/2 in Mozart: *Piano Quartet in G minor, K 478, third movement, piano part, bars 311 to 319 (contributed by Bala). Runner-up in durational units and measures: 8 bars of 2/2 in Beethoven's cadenza to the Rondo of Mozart's D-minor piano concerto, K. 466 (contributed by Elkies).
  6. Longest multi-measure rest, in measures {B&I 6.5}: the organ part of Strauss: An Alpine Symphony (1915) opens with a multibar rest labelled "tacet bis:", followed by rehearsal mark 94, then seven measures of rests with cues. The initial rest fills 695 measures. However, for a number that actually appears in the music, the largest I know of is 463 in a transcription of John Coltrane's version of the Rodgers & Hammerstein song My Favorite Things, in Coltrane Plays Standards, p. 35 (Hal Leonard ed.). Runner-up: 128 measures in Bruckner: Symphony no. 8 (1884-87), IV(?), triangle/cymbal part; this is part of a series of multibar rests and rests with cues totalling 248 measures.
  7. For grace notes (including appoggiaturas) {B&I 5.5}: Grace half notes in Mozart: Quoniam of the Great Mass in C minor, K.427, bar 85, first violins and first soprano; John Wall Callcott: "Zephyr, with thy downy wing" (both contributed by Bala). C.P.E. Bach's Versuch uber die wahre Art das Clavier zu spielen (1762?; English translation as: Essay on the True Art of Playing Keyboard Instruments) gives examples of half- and even whole-note appoggiaturas. Runners-up: There are many examples of grace quarter notes.

3. Most augmentation dots {B&I 4.6, 6.7}

  1. Quadruple dots appear in Liszt: Piano Concerto #2 (1839, rev. 1848; Kalmus ed.), Allegro deciso, mm. 327 and 331; Schumann: String Quartet no. 1, Op. 41 no. 1 (1842), III, mm. 16, 17, 33, 34, etc. (contributed by Cuthbert) (in Byrd, 2016); Verdi: Requiem (1874; Dover ed.), Rex Tremendae, mm. 356 and 358; Franck: Prelude, Chorale, and Fugue (1884; Schirmer ed.), mm. 2 and 4 (in B & M, 1948); Hindemith: Mathis der Maler Symphony (1934; Schott ed.), III, introduction; Bartok: Music for Strings, Percussion, and Celesta (1936; Boosey & Hawkes ed.), III, m. 80; and Einojuhani Rautavaara: Icons, Op. 6 (1955), I ("Jumalanäidin kuolema"), mm. 2-3 (contributed by Arribas). In every case, the dots are on a half note.
  2. Runners-up: triple dots occur in Schumann: Carnaval (1835), last movement; Chopin: Piano Sonata no. 2 in b-flat, Op. 35 (1840), I; Liszt: Mazeppa (1827; rev. 1837, 1854?), mm. 1-4 (in B & M, 1948); Sibelius: Violin Concerto, II; etc. These are all (or almost all) on notes, but a triple-dotted eighth rest appears in m. 15 of Heinrich's Toccata Grande Cromatica; triple-dotted quarter rests in mm. 2 and 3 of Rautavaara's Icons, I; and a triple-dotted half rest in Stockhausen: Zeitmasse (Universal ed., 1957).
  3. For the earliest double dots, see below.

4. Tuplet extremes. See Read (1978) for extensive discussion, with many examples.

  1. Most complex in a single level {B&I 11.1}
    1. Largest numerator: 58 (unmarked) in Chopin: Etude in C-sharp minor, Op. 25 #7 (1836; Paderewski ed.). Runner-up: 48 (marked) in Chopin: Nocturne in D-flat, Op. 27 #2 (1835; Henle ed.).
    2. Largest denominator: 32 in Sorabji: Opus Clavicembalisticum, final section (the tuplet is 40:32). Runners-up: 16 in Sorabji: Opus Clavicembalisticum, final section (25:16); 14 in Davidovsky: Inflexions, p. 13 (11:14, crossing a barline).
  2. Nested: most levels {B&I 11.13}
    1. 4 occurs several times in Brian Ferneyhough: String Quartet no. 3 (1987; Peters ed.), II: e.g., m. 6, it's 5:3 eighths, 5:3 eighths, 9:8 32nds, 3:2 32nds. However, it's doubtful if the last level--in this instance, at least--could have any audible effect, and almost certainly any effect would be far more subtle than other unintentional and (probably) intentional timing fluctuations; therefore, this notation is essentially conceptual.
    2. Runner-up: 3 in Stockhausen: Klavierstueck I (1952) (3:4, 7:8, 3).
  3. Largest "compression ratio"
    1. With small notes (suggesting grace notes or a cadenza, and therefore less impressive). 40:4 = 10:1 in Chopin: Nocturne in F#, Op. 15 no. 2 (Henle ed.), m. 51; 58:6 = over 9.5:1 in Chopin: Etude in C-sharp minor, Op. 25 #7 (1836; Paderewski ed.). The notes in the tuplet are written as 8th notes, and the group fills a measure of 3/4. The notes are all small except the first. Runner-up: 48:6 = 8:1 in Chopin: Nocturne in Db, Op. 27 no. 2 (1835; Henle ed.), all in small 8th notes, filling a measure of 6/8.
    2. Entirely in full-size notes. Haydn's *Variations in F minor (composed 1793; Henle ed.) has 20:4 = 5:1 eighth-notes in m. 186 and m. 188. While these are not small notes in the Henle Urtext edition, they are in some older editions, like the one edited by Paderewski. Runners-up: Measures 187 and 189 of the same work have 19:4 = 4.75:1 eighth-notes. (Both contributed by Bala) Also 7:2 = 3.5:1 in Chopin: Prelude in D-flat, Op. 28 no. 15; nested 3:2 and 7:4 = 21:8 = 2.625:1 in Bartok: Bluebeard's Castle (1911), p. 118.
  4. Most staves a tuplet is on (in a single system) {B&I 11.11}
    1. 3 in Varese: Ionisation (1931; Colfranc ed.), at rehearsal no. 8. But it seems likely there are many other occurrences of tuplets across 3 staves.

5. Most complex polymeter. See Read (1978).

6. Time signature extremes

  1. Simple time signatures: duration, numerator, and denominator extremes {B&I 10.1}
    1. Shortest duration. 7/128 in *Crumb: Black Angels (1970), 5th section ("Danse macabre") (7/128 time) (contributed by Hook). Runners-up: 1/16 in Stockhausen: Zeitmasse (Universal ed., 1957) (1/16 and 2/32 time). 3/32 in the Stockhausen and in Webern: String Quartet, Op. 28 (Universal ed.).
    2. Longest duration. 24 whole notes(!) in Telemann: Brobdingnagische Gigue, in Gulliver Suite (24/1 time); the shortest note value it contains is a whole. (contributed by Black)
    3. Largest numerator. 142 in Stockhausen's Klavierstuck IX (Universal, 1967): m.1 is in 142/8 time. (contributed by Hook) Runners-up: 87 in Stockhausen's Klavierstuck IX, m.2 (in 87/8 time); 47 in Nancarrow: Study no. 3a for Player Piano (publ. 1983?) (47/16 time); 43 in the same piece (43/16 time).
    4. Largest denominator. 128 in *Crumb: Black Angels, 5th section ("Danse macabre") (7/128 time) (contributed by Hook). Runners-up: 64 in the same section of the Crumb work (7/64 time) (contributed by Hook). 32 in many works, e.g., Beethoven: Piano Sonata no. 32, Op. 111, II (12/32 time); Stockhausen: Zeitmasse (2/32, 3/32); Webern: String Quartet, Op. 28 (1938) (3/32).
  2. Complex/compound/irregular meters {B&I 10.1}
    1. Most numerators in an "additive" (the term is from Read (1978)) meter: 4 in Bartok: Dance in Bulgarian Rhythm no. 5, in Mikrokosmos (1926-39; Boosey & Hawkes ed.) Book 6: specifically, (2+2+2+3)/8; 4 in Messiaen: Un Vitrail et des Oiseaux (1986): specifically, (3+2+2+2)/32.
    2. Complexity of additive meters: The Sedi Donka is a Bulgarian folk dance dance. It (or a version of it) appears in *a book by Manol Todorov, where it is written as 7/8 + 7/8 + 11/8, but it can be broken down further as (3+4)+(3+4)+(4+3+4)/8, and further still by replacing the 4's with 2+2's. In any case, the total number of numerator units is 25. (contributed by Aberg)
  3. Non-power-of-2 denominator: A time signature of 12/12 appears in the clarinet part of Franz Berwald's Quartet for Piano and Winds (1819!), II (vol. 13, p. 29 of the complete works). (The obvious way to express what seems to be intended would be simply 12/8! Perhaps the 12/12 is a mere mistake.) (contributed by Guerin) Herbert Bruen has used a denominator of 12, e.g., 5/12, I believe for what most people would write as (3+2/3)/8. Thomas Ades has written a measure of 2/6 time, consisting of a half note with a triplet 3 over it, i.e., 2/3 of a half note; he's also written 1/6 in his *Asyla, II. But this kind of thing can also be written with unconventional numerators instead of denominators: Boulez has written what is probably the same thing as (2/3)/4. Time signatures like 8/9 and 8/12 appear, e.g., in Frescobaldi, but apparently just to cancel previous proportions and not to indicate durations of 8/9 or 8/12 of anything, and this usage should not be considered CMN.

7. Slur maximums (including ties and phrase marks)

  1. Longest slur {B&I 17.5}
    1. In number of systems: Chopin, Ballade no. 2, Op. 38 (composed 1836-1839; London first edition by Wessel and Co.) has a slur extending across 9 systems, starting at the beginning. Runners-up: Alkan, Saltarelle, Op. 23 (composed 1844), p. 6, has a slur extending across 6 systems (Richault edition). (Both contributed by Bala) Berg: Wozzeck (1914-21), Act III, Scene 4, piano/vocal score has a slur across 5 systems, and several across 4 systems.
  2. Slur with most inflection points {B&I 17.17}
    1. Sorabji: Opus Clavicembalisticum (1930), IX [Interludium B] (Curwen ed., pp. 175-176) includes a slur with a total of 10(!) inflection points (in Byrd, 2016); it spans three systems, repeatedly crosses three staves (this is also the most staves in a single system for any slur I know of), and goes slightly backwards--i.e., from right to left--several times. Henri Dutilleux: Piano Sonata (1948; Durand ed.), III ("Choral et variations"), 2nd variation, also has 10 inflection points; it's slightly less impressive in that it repeatedly crosses only two staves and it never backs up. (contributed by Hook)
    2. Runners-up: 8 inflection points in *Henri Dutilleux: Piano Sonata (1948), III ("Choral et variations"), 2nd variation (contributed by Hook). Ravel: Le Tombeau de Couperin (1914-17), Minuet, has a slur with 7 inflection points, repeatedly crossing two staves (contributed by Hofstadter). Stockhausen: Zeitmasse has a slur with 4 inflection points on a single staff; Ravel: Gaspard de la Nuit, Scarbo, p. 41, has one with 3 inflection points.
    3. Tie with most inflection points. Bach: Goldberg Variations, no. 16 (Kirkpatrick/Schirmer ed.), has a tie with two inflection points(!); it's in the middle voice of three on its staves, and there's very little vertical space between the tied note and notes in the surrounding voices.
  3. Most nesting {B&I 17.4}
    1. Three levels: Schoenberg: Variations for Orchestra, Op. 31 (1928), Theme; Berg: Violin Concerto (1935), I (Universal Edition, mm. 47-48: solo violin) and II (mm. 174-75, 177: solo and orch. violins). In all cases, the innermost level is a tie; the others are presumably slur and phrase mark. Three levels is probably not too unusual.

8. Shortest performed duration

  1. Normally-played notes. Alkan: Etude No. 7 ("Symphonie, Finale") from Douze études dans tous les tons mineures, Op. 39 (1857), is marked whole note = 96, so the 16ths in mm. 357-382 last 39 milliseconds each. However, a live recording by Marc-André Hamelin is at whole note = 104 here, so that each note lasts a mere 36 ms.! (contributed by Bala) Runners-up: There are several likely candidates by Alkan. Chopin: Etude in C major, Op. 10 no. 1, in a 2005 recording by Vladimir Ashkenazy, has notes lasting about 80 ms. (16th notes at a tempo of ca. quarter = 200). NB: this is really the time between notes onsets; presumably the actual durations are slightly less.
  2. Other. Debussy: Pour le Piano (1901), I, has a glissando of 30 notes, with each note's duration less than 30 milliseconds! The glissando has a notated duration just under a half note. (Of course, glissandi for keyboard instruments, unlike those for most others, involve discrete notes.) A realistic fast tempo is something like quarter = 150; at that tempo, a half note lasts about 800 millisec., so the duration of each note is less than 27 millisec.


1. Softest {B&I 16.1}

  1. pppppppp (8 p's) in Ligeti's Etudes for Piano, 1st Book (1988-94), no. 4, m. 170 and 2nd Book, no. 9 (contributed by Hook), and in Helmut Lachenmann: Echo Andante for piano (1962), last note (contributed by Gottlieb).
  2. Runners-up: ppppppp (7 p's) in Verdi: Otello, Act II, Scene 5 (Schirmer ed., p. 187) (contributed by Forty). pppppp (6 p's) in Tchaikovsky: Symphony No. 6, Op. 74 (1893), I, m.160, bassoon; in Henze: Koenig Hirsch (1952-55), end of Act I; and in Ligeti: Etudes for Piano, 2nd Book (1988-94). ppppp (5 p's) appears in works of Tchaikovsky, Mahler, etc. Of course, it could be argued that il più piano possibile or equivalent is softer.

2. Loudest {B&I 16.1}

  1. ffffffff (8 f's) in Ligeti: Etudes for Piano, 2nd Book, nos. 13 and 14.
  2. Runners-up: fffff (5 f's) in in Tchaikovsky: The Tempest symphonic fantasia, Op. 18 (1873), at rehearsal mark 18, many instruments (contributed by Parsons); Henze: Barcarolla for Large Orchestra, near the end, many instruments; and Ives: Putnam's Camp, in Three Places in New England (Mercury ed.), last chord, piano (but this might be a misprint, since many other instruments have 4 f's). Again, of course, il più forte possibile, con tutta forza, or equivalent is arguably louder. (Tchaikovsky: 1812 Overture, a piece that's famous for its loudness, reaches only ffff (p. 81). But obviously instrumentation--i.e., fireworks and cannon in some performances--is a factor!)


    1. Vertical extremes

    1. For comparison, Stravinsky: Le Sacre du Printemps (1913) uses a maximum of 35 staves and ca. 38 notated voices.
    2. Most simultaneous notated staves {B&I 1.7}
      1. *Ligeti: Atmospheres (3rd ed., Universal) has a system of 77 staves on p. 15 (contributed by Elkies). The 1st ed. has "only" 70 staves (1961; Universal ed., p.13). The 56 strings are completely divisi in both editions.
      2. Runners-up: Penderecki: Symphony no. 1 (1975) has 71 (there are two Schott editions; it's p. 33 in one, p. 43 in another!). (He uses a large orchestra, and the 46 strings are completely divisi.) NB: the notation is typical Penderecki, and it might be argued--but not, in my opinion, very convincingly--that it doesn't belong here because it's not CMN. 61 staves in Carter: Concerto for Orchestra (1969), on a page near the end. A famous early example and more distant runner-up is a Mass by Benevoli (1628) that uses 53 staves.
    3. Most simultaneous notated voices {B&I 1.6}
      1. 79 in Ligeti: Atmospheres (1st ed., Universal, p.13).
      2. Runners-up: 71 in Penderecki: Symphony no. 1, on the page with 71 staves. Carter: Concerto for Orchestra (1969) has 69 on the same page with 61 staves.
      3. A more distant runner-up is Xenakis: Metastasis (1954-55), for 61 instruments, in which every player plays a separate part for the entire piece.
    4. Most simultaneous notated parts {B&I 1.6}. NB: "part" is not really well-defined, so neither are these numbers.
      1. At least 80 in Ligeti: Atmospheres. Runner-up: 71 in Penderecki: Symphony no. 1.
    5. Maximums for a chord (where a "chord" refers to the notes in one voice and attacked simultaneously {B&I 4.24}, and, if a stem is present, on one stem)
      1. Most notes with individual heads (i.e., non-cluster notation) : 24 in Scriabin: Piano Sonata no. 7, Op. 64, last page. Runners-up: 20 in Alkan: Quasi-Faust (2nd movement of the Grande Sonate, Op. 33, composed 1847); 19 (covering a range of 6 octaves) in Alkan: Trois Grandes Etudes, Op. 76 (Billaudot ed.), no. 2; 16 in the piano part of Ives: The Housatonic at Stockbridge, in Three Places in New England, last page.
      2. Most notes regardless of notation {B&I 4.24, 4.26): 88 for piano -- all the keys of a normal instrument -- in Joseph Schwantner: Magabunda (1983), near the beginning of the 2nd movement (cluster notation) (contributed by Primosch). Runners-up: 53 for piano in Cowell: Tiger (1928; Associated ed.) (cluster notation).
      3. Most grace notes: 4 in Paganini: Caprice in g, Op. 1 no. 10 (Dover); in Scriabin: Piano Sonata no. 6 (Dover), p. 119, and Sonata no. 7 (Dover), p. 166; and in Messiaen: Oiseaux Exotiques.
      4. Most accidentals: 16(!) in the piano part of Ives: The Housatonic at Stockbridge, in Three Places in New England, last page. Runners-up: 7 (6 on one staff) in Ives: General William Booth Enters into Heaven (Presser ed.), mm. 106 and 107. 6 (on one staff) in Ives: Putnam's Camp, in Three Places in New England, p. 46. 5 in Bartok: Piano Concerto no. 1 (Universal ed.), II, p. 95; Debussy: Preludes (Dover), La terrasse des audiences de clair de lune, p. 83; Messiaen: Quartet for the End of Time (Durand), I, mm. 4, 12, 20; and Ravel: Ravel: Gaspard de la Nuit (Dover ed. _Piano Masterpieces of Maurice Ravel_), Scarbo, p. 115.
      5. Most fingerings: 4 in Brahms: Capriccio, Op. 116, no. 7 (Breitkopf & Hartel Complete Works ed.), p. 65.
    6. Most written notes in a vertical simultaneity: (no entry yet)
    7. Most staves a stem has notes on {B&I 4.27}
      1. Crumb: Black Angels (1970) has stems extending across 4 staves, with notes on each staff.
    8. Maximums for one instrument and performer
      1. Most simultaneous notes for one instrument and performer: 57 in Cowell: Tiger, for piano (cluster notation)
      2. Most simultaneous staves for one instrument and performer: 10(!) in the piano part of Xenakis: Synophai (1969; Salabert ed.), mm. 94-99. Also 10 in Nancarrow: Study no. 27 for Player Piano (publ. 1977), but--while this is one instrument--it's no performers. Runner-up: 5 in the last pages of Sorabji: Opus Clavicembalisticum (1930), for piano.
      3. Most simultaneous notated voices for one instrument and performer: The fugato in Alkan's Quasi-Faust (2nd movement of the Grande Sonate, Op. 33, composed 1847) goes up to at least 8 voices, 4 notated in each hand (contributed by Bala). 6 in the keyboard version of Bach: 6-Part Ricercare from the Musical Offering (Dover reprint of Breitkopf & Hartel ed.).
    9. Maximums for one staff
      1. Most simultaneous notated voices {B&I 1.9}
        1. 5 voices on a staff occurs momentarily in Bach: Prelude to English Suite no.1 (1725?; Bach-Gesellschaft ed.), mm. 2-3, and in Beethoven: Piano Sonatas Op. 27 #2 ("Moonlight"; 1801), III, mm. 162-165, and no. 23, Op. 57 ("Appassionata"; 1806), I, m. 123 (Schenker eds.). But in all, it's simply a series of separately-attacked notes tied into a chord (and in the Beethoven cases, double-stemmed as both a single voice and independent voices). A clear-cut case of 4 voices on a staff occurs in the keyboard version of Bach: 6-Part Ricercare from the Musical Offering, and in Bach: "Jesu, Joy of Man's Desiring", arranged by Dame Myra Hess (Oxford ed.), mm. 29-31. 4 occurs in Beethoven: Piano Sonata no. 11, Op. 22 (1800; in 3 of 3 eds. I checked, including Schenker's), I, mm. 91-103, though it's simply a series of arpeggios with attack rhythm written out. 4 occurs in Beethoven: Piano Sonata no. 2, Op. 2 no. 2, I, p.24, m. 272, and no. 3, Op. 2 no. 3, I, mm. 221-225 (1795); Piano Sonata no. 11, Op. 22 (1800), I, mm. 91-103 (all in 3 of 3 eds. I checked, including Schenker's); and repeatedly in Brahms: Intermezzo, Op. 119 no. 1 (International ed.)--though in all it's simply a series of arpeggios with attack rhythm written out. 4 occurs repeatedly in Bach's solo violin sonatas and partitas, and in Brahms: Romance, Op. 118 no. 5, but rarely if ever are the supposed voices truly all independent.
        2. Runners-up: 3 voices on a staff is fairly common in fugues for keyboard instruments written on two staves, e.g., the c-sharp minor Fugue in Bach: The Well-Tempered Clavier, I, and Bach: Goldberg Variations, no. 16. It's not that unusual even in keyboard music that's less contrapuntal, e.g., Schubert's song Der Juengling und der Tod (Peters ed.), and perhaps on wind staves in pieces for large orchestra. It also occurs in music that's much simpler, e.g., "Vigilante Man" in A Tribute to Woody Guthrie (Ludlow Music, 1972), where the main rhythm is in normal-size notes and two variants (for different verses) in small notes, one stem up and stem down. Chopin's Nocturne, Op. 15 no. 2 (in Byrd, 2016), has beaming clearly indicating 3 voices (and suggesting 4) on the staff even though there are only 1 or 2 notes at a time.
    10. Most figures in a figured bass symbol {B&I 19.2}: 6 in "Le Cahos" from *Jean-Fery Rebel's "Les Elèments" (1737) begins with a low D and a stack consisting of: 6b 5 4 3 2 7# (contributed by Guerin).

    2. Horizontal extremes. NB: While not excluded, dramatic works are neglected here mostly because it's difficult to get useful information on lengths of major dramatic works; also, it's not clear what a "movement" is in an opera, etc.

    1. Most notes/chords
      1. notes/chords in one beamset {B&I 12.1}
        1. Unconditionally, 1440(!) in John Adams's China Gates (Associated Music Publishers, 1983); the beam extends over all 9 pages of the score, covering 48 systems. (contributed by Hook) Runner-up: 643 in Don Freund: Hard Cells (1989; MMB Music ed.), percussion; the beamset extends for more than 80 measures, across 11 systems.
        2. Broken across three systems, 132 in Liszt: Transcendental Etude no. 4 ("Mazeppa") (1827, rev. 1837; Baerenreiter Neue Ausgabe ed., 1970). This has 15 secondary beam breaks(!) and 28 segments.
        3. Broken across two systems, 70 in a cadenza apparently by Beethoven for the 1st movement of his Piano Concerto no. 2, Op. 19 (Breitkopf & Hartel ed.). Runner-up: 67 in Bartok: Rhapsody, Op. 1 (Boosey & Hawkes ed.).
        4. On one system, 59 in Liszt: Hungarian Rhapsody no. 2, and in Rachmaninoff: Piano Concerto no. 2 (Schirmer 2-piano ed.), II.
      2. grace notes/chords in one beamset {B&I 12.1}
        1. *Hugues Dufourt: Antiphysis has 28 in the solo flute. (contributed by Starr)
        2. Note that in cadenza-like passages, it may be impossible to distinguish between small normal notes and grace notes.
      3. in one measure: (no entry yet)
    2. Most appearances of a voice in one measure
      1. A voice (indicated by upstems on isolated notes) appears, disappears, and reappears five times in m. 175 of Chopin: Ballade no. 4 in f, Op. 52.
    3. Longest movement
      1. In measures (and quarter-note durations): Allan Pettersson: Symphony no. 9 (1970) is one movement of 2145 measures. Wagner: Rienzi (1840; shortened version, 1843; Schott ed., 1982) has one scene of 1384 measures; it's mostly in 2/2 or 4/4, but includes substantial passages in 2/4, 3/4, and 6/8. But this scene also contains 1042 measures from the original version, for a total of 2426, though no performance would include all (see below); nonetheless--if measures printed rather than performed is the criterion--this score holds the record. Runners-up: Mahler: *Symphony No. 8 (1906), 2nd movement, is 1572 continuous measures (contributed by Bala). Alkan: Etude no. 8 ("Concerto, Premiere Partie") from Douze etudes dans tous les tons mineur, Op. 39, is 1342 bars; it's in 3/4 (contributed by Starr). Schubert: Symphony no. 9, IV, is 1154 measures; it's in 2/4.
      2. In pages: Pettersson: Symphony no. 9 (Nordiska Musikforlager ed.) is one movement of 385 pages. Runner-up: Beethoven: Symphony no. 9, IV (Eulenberg ed.), fills 128 pages. (But this is very edition-dependent: in the Dover edition, this movement is only 88 pages.)
      3. In performance time: Satie: Vexations (unmeasured, 52 beats to be repeated 840 times, lasting a total of about 18 hours in the first performance). But this--while not unplayable--might still be disqualified as being conceptual. A clear-cut case, or at least runner-up: Pettersson: Symphony no. 9 is one movement lasting 65-70 minutes. [Chopin: Mazurka Op. 7 no. 5 is marked "Dal segno senza Fine", i.e., infinite length! (Contributed by Nettheim) A case can made that the longest work of finite duration is John Cage's Organ^2/ASLSP (As SLow aS Possible). A performance in Halberstadt, Germany, began in 2001, so it has been going on for over thirteen years; it is intended to last for 639 years! But Cage himself didn't specify the duration other than via the phrase "as slow as possible"; a group of musicologists and philosophers decided on the 639-year duration. (contributed by Brady)]
      4. In number of notes: (no entry yet, but Pettersson: Symphony no. 9 is a very good candidate.)
    4. Shortest movement
      1. In measures: no. 4 of Webern: Five Pieces for Orchestra, Op. 10 (1911-13), is 6 and 1/3 measures (in 3/4). Runner-up: no. 3 of Webern: Three Little Pieces for Cello and Piano, Op. 11, is 10 measures (of 2/4). Not surprisingly, Webern is responsible for many candidates for shortest movement, depending on the exact criteria.
      2. In pages: no. 3 of Webern: Three Little Pieces for Cello and Piano, Op. 11, is one system of three staves--perhaps 1/3 of a page.
      3. In performance time: no. 4 of Webern: Five Pieces for Orchestra, Op. 10 is 19 sec. at the tempo marked (quarter = 60 for a duration of 19 quarters).
      4. In number of notes: No. 3 of Webern: Three Little Pieces for Cello and Piano, Op. 11, contains 21 sounding notes (attacks). Runner-up: no. 4 of Webern: Five Pieces for Orchestra, Op. 10, contains 47 sounding notes. [Each of the three movements of John Cage's famous 4' 33" (1960) is entirely silent and contains zero sounding notes: clearly a piece of much conceptual interest, but no musical interest in the normal sense. *Philip Corner: One Note Once is exactly what it says; perhaps this should be considered a legitimate extreme, but I put it in the same category as 4' 33".]
      5. If the Adagio between the two fast movements of Bach: Brandenburg Concerto no. 3 is considered a movement, it wins each of the above categories: it's one measure, less than 1/10 of a page, lasts just a few seconds, and contains 20 notes. (contributed by Hiew) However, I'm inclined to consider it a mere transition between the first and third movements; besides, in performance, it's nearly always expanded with something improvisational, not played as written.
    5. Longest complete work
      1. In measures: Wagner: Rienzi (1840; revised 1843) is in five acts with 408 (overture) + 1244 + 1936 + 1487 + 605 + 937 = 6617 measures. (This seems to be the later, shortened version of the opera, in the 1982 Schott piano/vocal edition. That edition contains many additional passages without measure numbers--most marked "urspruengliche Version", i.e., original version--with an additional 134 + 1094 + 163 + 108 + 85 = 1584 measures, for a total of 8201; since many of the superseded passages have replacements, neither the original nor the revised version includes all measures. Nonetheless--if measures printed rather than performed is the criterion--the latter value applies.) Runner-up: Hans Werner Henze: Koenig Hirsch (1952-55) is in three acts with 2471 + 2225 + 990 = 5686 measures.
      2. In pages: This is very edition-dependent. The miniature score of Wagner: Die Meistersinger (in an old Schott edition, in three volumes) is 1441 pages. Runners-up: The full score of Janacek: The Excursions of Mr. Broucek ("Die Ausfluge des Herrn Broucek", 1917; publisher unknown) is 1006 pages (in two volumes). The full score of Mussorgsky: Boris Godounov (Oxford ed.) is 929 pages (this is the record for a single volume); of Die Meistersinger (Peters ed., reprinted by Dover), 817.
      3. In performance time: At least one performance of Kaikhosru Sorabji's Sequentia cyclica super "Dies irae" ex Missa pro defunctis, by a solo pianist, has lasted some 8 or 9 hours (contributed by Isaacson). Runner-up: Kaikhosru Sorabji: Opus Clavicembalisticum for piano (1930) lasts about 4 to 4-1/2 hours. (Contributed by Fingerhut)
      4. In number of sections: Bartok's Mikrokosmos contains 153 pieces, in six books. Runners-up: Ives' 114 Songs; Bartok's 44 Duos for Two Violins; Schumann's Album for the Young, Op. 68, contains 43 pieces, in two parts.
    6. Shortest duration a key signature is in effect: a single quarter-note chord in Poulenc's song "Les Chemins de L'Amour" (Eschig ed.); occurs twice, on pp. 1 and 2 (contributed by Alpher).

    3. Tempo and metronome marks

    1. Longest character string {B&I 14.1}
      1. Mussorgsky's Pictures at an Exhibition starts with: "Allegro giusto, nel modo russico, senza allegrezza, ma poco sostenuto" (69 characters). (contributed by Gedan)
      2. Runner-up: *Beethoven: Mass in C, Op. 86 has "Andante con moto assai vivace, quasi Allegretto, ma non troppo" (62 characters). (contributed by Starr)
    2. Lowest number {B&I 14.6}. For comparison, metronomes usually go down to M.M. 40.
      1. Crumb: Spiral Galaxy, no. 12 in Makrokosmos I (Peters ed., 1973), is marked "eighth = 20 = 3 sec."
      2. Runners-up: Martino: Notturno (1973) contains the marking "half note = 24"; Crumb: Black Angels (1970), Pavana lachrymae, contains "half = 30 (quarter = 60)". Martino: Impromptu for piano (1978) is marked "quarter <= 36"; Messiaen: Un Vitrail et des Oiseaux (1986) contains "eighth = 36". See also the description of the Bartok runner-up for highest number.
    3. Highest number {B&I 14.6}. For comparison, metronomes usually go up to M.M. 208. In most if not all of these examples, there's no consistent beat unit, so the metronome mark is based on a unit below the beat level, resulting in very large numbers.
      1. 16th = 906(!) in *Picken (1975, music example 17, p. 305). This is an excerpt of a dance tune for the kemence (Black Sea fiddle), from Macka in the Black Sea area of Turkey; it was transcribed from a recording. (contributed by Campin)
      2. Runners-up: Hindemith: Sonata for Solo Viola, Op. 25 no. 1 (1922), IV, is marked "quarter = 600-640". The notes for the Imai recording say that this is "completely absurd", and suggest that Hindemith wrote the marking tongue in cheek. That's a reasonable comment unless the shortest duration is a quarter and the beats are longer and irregular, which is in fact the case!
      3. Runner-up: Bartok: Dance in Bulgarian Rhythm no. 1, Mikrokosmos, Book VI, is marked "eighth = 350 (half tied to quarter tied to dotted quarter = 39)"; notice it's also a runner-up for lowest number. (The meter is (4+2+3)/8, so there's no consistent beat unit, but the numbers for all of the beat durations are somewhere in the middle.)
    4. Slowest in notated duration per unit time. Quintuplet 64th note = 75 in Stockhausen: Xi (1992) (contributed by Ingram).
    5. Shortest duration {B&I 14.4}. Quintuplet 64th note = 75 in Stockhausen: Xi. Runner-up: 64th = 288 in Crumb: Madrigal no. 1, from Madrigals, Book IV (1971).

    4. Rehearsal mark

    1. Longest {B&I 7.15}. Strauss: An Alpine Symphony has one of four characters: "114a" (all or nearly all the other rehearsal marks are the expected consecutive integers).

    5. Performance direction (in the music, not on a separate page)

    1. Longest. Berlioz: Symphonie Fantastique, Op. 14, IV, has one of 146 characters (in French): "Il faut frapper..."

    6. Part name or staff identification

    1. Longest {B&I 1.10}. Ives: Putnam's Camp, in Three Places in New England, has one of 47 characters: "Long Snare Drum (snares muffled) or small Timp." Runner-up: Haydn: Symphony no. 7 ("Midi"), II, has one of 38 characters: "Violoncello, Basso Continuo, e Fagotto".
    2. Longest abbreviated {B&I 1.11}. Webern: Five Pieces for Orchestra, Op. 10, has one of 19 characters: "Solo - Vlc. m. Dpf."

    7. Staff sizes

    1. Most simultaneous {B&I 18.3}. 3 in J. C. Bach: Concerto for Harpsichord or Piano and Strings in E-flat, Op. 7 no. 5 (Dobereiner ed., 1927). One size appears only briefly, for an ossia; this situation and number of sizes is surely not too unusual.

    8. Note modifiers (accents, articulation marks, bowings, etc., but not fingerings)

    1. Most on a single note/chord: Bartok: Ostinato, no. 146 in Mikrokosmos (1926-39; Boosey & Hawkes ed.), Book VI has several occurrences of 3 (accent, tenuto, staccato dot). Ligeti: Etudes for Piano, 2nd Book, no. 13 has many occurrences of 3 normal accents and many occurrences of 3 "hat" accents. Runners-up: 2 is not too unusual. Bartok: Allegro Barbaro (1911; Schirmer ed.) has accent and tenuto. In every other case I know of, one is a staccato dot, e.g., in Rachmaninoff: Prelude, Op. 23 no. 5; Debussy: Suite "Pour le Piano" (1901; Schirmer ed.), I; Stravinsky: Le Sacre du Printemps, Danses des adolescentes.

    9. Endings

    1. Highest ending number {B&I 20.1}. 9(!) in *"In the Heat of the Summer" in The Complete Phil Ochs Collection (Almo Publications, 1978). There's an ending marked "1.-8." for the first eight verses, and one marked "9." for the last(?) verse. (contributed by Good)

    10. Instruments to be played by one performer in a piece (excluding percussion)

    1. *Mahler: Symphony no. 5 calls for one clarinetist playing six different instruments.

Earliest Usages

For contemporary notation, Risatti (1975) and Read (1978) have numerous citations.

    1. Pitch

    1. Earliest split stem (a.k.a. "grape cluster"): Chopin: Etude Op. 10 no. 11 (1831?; Paderewski ed.). Runner-up: Chopin: Mazurka in a, KK IIb no. 4 (1840). Both cases are for augmented unisons.
    2. Earliest microtonal notation: *A piece by Nicola Vicentino (in his "L'antica musica ridotta alla moderna prattica", 1555) (contributed by McKay). Runners-up: *a four-voice work by Galeazzo Sabbatini (in Kircher's "Musurgia universalis", 1650) (contributed by McKay); *Charles Delusse: Air a la Greque (1760), which uses a + to indicate a raise of one quartertone (contributed by Wolf). In more recent times, I've seen J. Carillo (e.g., String Quartet (1895)) cited as a very early author of microtonal music.
    3. Earliest diatonic tone cluster: Heinrich Biber: Battaglia for 10 instruments (ca. 1673; in Schenk, 1997), 2nd mvmt., includes a 4-note tone cluster (in m. 9: C#,D,E,F#); several 3-note diatonic clusters; and at least one occurrence of two clusters in different scales simultaneously (D,E,F-natural, and--in a higher register--F#,G,A) (contributed by Wiering). This might be considered a 6-note cluster, but I think not, because of the change of register. (The different scales are because the passage is polytonal(!): various instruments are playing in C, D, and G.) Runners-up: Johann Kuhnau: Six Biblical Sonatas (publ. 1700), no. 1, "Der Streit zwischen David und Goliath" includes a cluster of C,D,E,G (contributed by Wolf). The earliest example with more than 4 notes is Cowell: The Tides of Manaunaun (probably 1917, not the claimed 1912) has white-key-only and black-key-only clusters of 2 octaves. (NB: as I use the term, the notes must be struck at once, and not just sound together, to count as a tone cluster. Also, I exclude acciaccaturas, examples of which can be found in many composers, including D. Scarlatti and J.S. Bach.)
    4. Earliest chromatic tone cluster: Beethoven: Piano Concerto no. 4, Op. 58 (1805-06), I, 2nd theme includes a 3-note cluster produced by a 3rd-inversion dominant 7th chord with an ornamental note filling in the major 2nd: in the exposition, adding a G# to the chord's G,A,C#,E (contributed by Elkies). However, it's very short, and it's doubtful whether it's heard as a cluster. The earliest univocal example is Verdi: Otello (1887), Act I, which opens with a very lengthy cluster in the organ of 3 notes (C2,C#2,D2). Earliest example with many notes: Cowell: The Tides of Manaunaun has almost continuous chromatic clusters of 1 octave to ca. 2-1/2 octaves.
    5. Earliest double sharp: Arnold (1983) mentions Trabaci's *Il secondo libro de ricercare (1615).
    6. Earliest 15ma: Cowell's piano piece Advertisement (1914).

    2. Duration and Rhythm

    1. Earliest tuplets (in modern notation): Emanuel Adriaensen's Pratum musicum (Antwerp, 1584) contains several triplets, one of which is indicated with a "3" below the affected notes. (Contributed by Crawford)
    2. Earliest nested tuplets
      1. Schubert: Adagio in G, D. 178 (1815!) uses triplets within triplets (contributed by Bala).
      2. Runners-up: Liszt: Transcendental Etude no.4 ("Mazeppa") (probably 1837 or earlier) uses triplets within triplets. (Caveat: This piece apparently was written in 1827, rev. in 1837, and rev. again in 1854, and I don't know in which version this feature was first used.) Chopin: *Allegro de concert, Op. 46 (finished by 1841), has sixteenth-note triplets inside eighth-note triplets (contributed by Bala). Alkan: *Etude No. 9 ("Concerto, Deuxième Partie") from Douze études dans tous les tons mineures, Op. 39 (1857), mm. 115-133, 172-175, and 186-191: they are sixteenth-note triplets within eighth-note triplets. Wagner: Tristan und Isolde (1859; Dover ed.), Act III Scene I, p. 515, uses triplets and sextuplets within a triplet.
    3. Earliest tuplets crossing a barline: In Alkan: Morte (Op. 15 No. 3, from Trois morceaux dans le genre pathetique, composed 1837), mm.223-4 have 5:4 quarter notes spanning both bars, and mm. 303-4 have 9:8 eighth notes spanning both bars. The section is in 2/4. (Caveat re the measure numbers: this edition has several editorial barlines before this, so other scores may not agree on these numberings. Subtracting 23 from them should account for that.) Runner-up: Brahms: Piano Sonata in F minor, Op. 5 (1853), V (4:3 quarter notes across the barline). (both contributed by Bala)
    4. Earliest half-note triplets (i.e., triplet with total duration of a whole note): Berlioz's Symphonie Fantastique, Op. 14 (1830), I, bassoon and cello parts in mm. 282-83 (contributed by Hook).
    5. Earliest use of 32nd notes: G. Della Casa, Il vero modo di diminuir (printed in 1584), Libro Primo, p. 27. But this is arguably white mensural notation, not CMN. (contributed by Pugin)
    6. Earliest use of 64th notes: Orlando Gibbons: Pavane Lord Salisbury (publ. in Parthenia, ca. 1612; in Davison & Apel, Historical Anthology of Music).
    7. Earliest use of polyrhythm with non-coinciding barlines after 1700: Mozart: Don Giovanni (1787), ballroom scene, has 3/4 against 2/4 against 3/8. (Relative tempi are 2/4 quarter = 3/4 quarter; 3/8 dotted quarter = quarter of the others.)
    8. Earliest feathered beams: around 1960? E.g., Roger Reynolds: Quick Are the Mouths of Earth (1965); Risatti cites *Heinz Holliger: Mobile (1964).
    9. Earliest double dots: Chambonnieres: Les Pieces de Clavecin (Paris, 1670). Double dots appear in Livre 1, p. 51, and Livre 2, pp. 2 and 49; they have their normal modern meaning. (Contributed by Crawford. The articles "Notation" in Sadie (1980) and Arnold (1983) mention this as an early example but make no claim of it being the first use, though Arnold says they were used from the 17th century onward. (Sadie (2001) appears not to mention any early examples. Read (1969) says they were first used in the late fifteenth century but gives no details, and it's a pretty far-fetched idea.)
    10. Earliest note/chord tied to a rest or to nothing at all: Schumann: Piano Sonata no. 1 in f-sharp minor, Op. 11, II (1836; Henle ed.) (contributed by Bastin). Runners-up: Ravel: Jeux d'eau (1901); Scriabin: Poem(?), Op. 31? no. 2? (1903).
    11. Earliest chord with some but not all notes tied to the next, or from the previous, chord: (1) Non-arpeggiated: Beethoven: Piano Sonata no. 7, Op. 10 no. 3, I (1797-98) has at least five; his Piano Sonata no. 28, Op. 101, I (1816) also has some. But this notation is not all that interesting, and probably rather common as well. (2) Arpeggiated (with attack rhythm specified, i.e., not just via the arpeggio symbol): Haydn: Piano Sonata in E minor, Hoboken XVI (publ. 1784; Peters-Martienssen ed.), I, has several arpeggios of 2 or 3 notes with all notes tied into a single chord following. Runners-up: Beethoven: Piano Sonata no. 5, Op. 10 no. 1 (Schenker ed.), III, m. 112 has a 10-note grace-note arpeggio across staves/hands with all but the first 4 tied into the following chords on both staves. His Piano Sonatas Op. 2 no. 3, I (1795); no. 11, Op. 22 (1800), IV; no. 14, Op. 27 #2 ("Moonlight"; 1801), III, mm. 162-165; and no. 23, Op. 57 ("Appassionata"; 1806), I, m. 123, each have arpeggios of up to 4 (5 in the "Moonlight" and "Appassionata"!) notes with all notes tied into a single chord following; in all but Op. 22, the notes are double-stemmed as both a single voice and independent voices (for all, Schenker ed.). (This notational feature is very interesting because it indicates a "blurring" of voices. The identical effect occurs much earlier, e.g., in Bach: Prelude to English Suite no.1 (1725?), mm. 2-3, but there it's notated--at least in the Bach-Gesellschaft edition!--as if the voices are really independent.)
    12. Earliest beamed group crossing a barline: Beethoven: Piano Sonata no. 7, Op. 10 no. 3 (Schenker ed.; in B & M, 1948), IV, mm. 1-2 and numerous places. Runner-up: Beethoven: Piano Sonata no. 10 in G, Op. 14 no. 1 (Schenker ed.), I, mm. 121-122.
    13. Earliest beamed group starting with a half-notehead: Chopin: Piano Sonata no. 1, Op. 4 (1828?; Schirmer-Mikuli ed.), I, p.4. But apparently this device already occurs in some guitar music published around 1810. I (contributed by Crawford)
    14. Earliest time signature uses
      1. Earliest use of different time signatures simultaneously.
        1. Different in appearance but effectively identical: J.-B. Lully: Armide (1686; Eitner/Breitkopf & Hartel ed., 1885, reprinted in Burkholder, 2006), Act II Scene 5, recitative has repeated instances of combinations like "3" for voice and 3/4 for continuo, or "2" for voice and cut time for continuo.
        2. Genuinely different: Variation 26 of Bach: Goldberg Variations (1742; Bach-Gesellschaft ed.) starts with one staff in 18/16 and the other in 3/4. (Each staff changes between the two time signatures repeatedly, at one point in the middle of a measure! But relative tempi are such that barlines always coincide.)
      2. Earliest use of multiple (alternate) time signatures on a staff: the D-major Prelude in Bach: Well-Tempered Clavier, Book II (1744; Bischoff/Kalmus ed.) has a time signature on both staves of "C" juxtaposed to 12/8.
      3. Earliest use of quintuple meter: Christopher Tye: In Nomine XXI: Trust for viol consort (mid-16th century) is entirely in 5/2 (contributed by Elkies). However, this isn't really CMN. Runners-up (and winners for unequivocal CMN): Handel's Orlando (1733), Act II, scene XI, "Ah! stigie larve...", pp. 65-66, contains a few measures in 5/8, which is also the earliest known use of that specific meter (contributed by Meckler); Chopin: Piano Sonata no. 1, Op. 4 (1828), III is in 5/4; Carl Loewe's ballad *Prinz Eugen, Op. 92 (1844).
      4. Earliest use of septuple meter:
        1. Notated as such: Alkan: Impromptu, Op. 32 No. 8 (1849!) (contributed by Bala); the meter is 7/4. Runner-up: Bartok: no. 2 of Two Elegies (for piano), Op. 8b (1909).
        2. Notated as a repeating pattern of measures of regular lengths: Reicha: 36 Fugues for Piano (1805) no. 24, is in "cut time + 3/4" throughout; it consistently alternates between the two (contributed by Murphy). Runners-up: Brahms: "Variations on a Hungarian Song" Op. 21, No. 2 (1854) is notated as 3/4 C; the theme and first eight variations consistently alternate between the two (contributed by Murphy). Brahms: Piano Trio in c, Op. 101 (1886), III is notated as 3/4 2/4, but consistently has one measure of 3/4 followed by two of 2/4.
      5. The *Fitzwilliam Virginal Book (ca. 1619) contains several pieces with metric structures tantamount to irregular meters but notated as a repeating pattern of measures of regular lengths, e.g., 11/8 = 4/4 + 3/8 (contributed by Elkies).
      6. Earliest non-integer time signature numerator: (3-1/2)/4 in Florent Schmitt: *La tragédie de Salomé (1907; Durand ed.). Runners-up: (4-1/2)/4 in Ives: *Piano Sonata no. 1 (1909; Southern Music ed.); (4-1/2)/4 in Ives: "Concord" Piano Sonata no. 2 (completed 1915) (Kalmus ed.), III ("The Alcotts"); (3-1/2)/4 in Varese: *Offrandes (1921). NB: Read (1978) lists dozens of works with fractional time signature numerators, and it's quite possible that some are earlier than those mentioned here.
    15. Earliest extended passage without barlines (after barlines came into general use, ca. 1600): "Anonyme [Louis Couperin?] [d'Anglebert?]" (so presumably mid-17th century): a four-page (23 systems) Prelude, in A minor, from "Suite III" (written entirely in whole notes, with bizarre slur-like things -- some with loops -- all over the place; this is common in unmeasured keyboard preludes in that style). Runner-up: C.P.E. Bach: Free Fantasy in D (Wotquenne 117/14), in his Versuch uber die wahre Art das Clavier zu spielen (publ. 1762), is just under a page (5 systems); the last "measure" of C.P.E. Bach: Fantasia in F (Wq 59/5) (1782) takes almost two pages (12 systems).
    16. Earliest tie: according to Rastall (1982), the first use of ties is in *Marco Antonio Cavazzoni: Recerchari, motetti, canzoni (1523).
    17. Earliest use of accelerando, ritardando, or equivalent: Slentando in Purcell's King Arthur (1691), Act III, at the end of "Cold Genius". Slentando appears in a 1710 manuscript copy; this presumably reflects Purcell's autograph. An 1813 edition is marked ritardando instead, which in any case is equivalent. (contributed by Solow) Runners-up for earliest use, and earliest use in print: calando in Mozart: Piano Sonata, K. 332 (1783), III, ending; rallentando in Beethoven's Trio, Op. 1, no. 2 (1795), Finale; Beethoven's Piano Sonata no. 4, Op. 7 (1797), IV (all contributed by Solow). Note: Rastall (1982, p. 192) says "gradual changes of tempo...were an eighteenth-century addition", and mentions terms for both accelerando and ritardano, but gives no examples.
    18. Earliest small notes: Bach: Organ Prelude in F minor, BWV 534 (Bach-Gesellschaft ed.; ca. 1708 to 1717), bottom of p. 2. Runners-up: Bach: cantatas "Sie werden euch in den Bann tun", BWV 44, V, and "Nun komm, der Heiden Heiland", BWV 62 (both 1724), II, III, and IV. All of these are in metrically-complete measures.

    3. Dynamics

    Rastall (1982) contains a fairly extensive discussion of early uses of dynamic markings. In the items below, no distinction is made between words and their abbreviations, e.g., between f and forte.
    1. Earliest use of any dynamic marking: Vincenzo Capirola's lutebook (ca. 1517) (Fig. 120 in the article "Notation" in Sadie, 2001).
    2. Earliest use of pian[o] and forte: Giovanni Gabrieli: Sonata pian e forte, from Sacrae Symphoniae (1597).
    3. Earliest use of mf: Gluck: Orfeo ed Euridice (1762), Act II Scene 1. Runners-up: Haydn: Piano Sonata no. 9 in D (1767; Peters-Martienssen ed.), II; F. X. Richter: String Quartet in B-flat, Op. 5 no. 2 (1768). NB: Rastall (1982) says C.P.E. Bach used mf, but gives no details.
    4. Earliest use of mp: Schütz: Saul, was vergolst du mich (ca. 1650) from Symphoniae sacrae III (Barenreiter; reprinted in Burkholder, 2006) (contributed by Cuthbert). Runners-up: Haydn: Quartet Op. 77 no. 2 (1799; pub. 1803), I (contributed by Solow); Liszt: Transcendental Etude no. 4 ("Mazeppa") (1827, rev. 1837). Rastall (1982) says mp appears later than mf, but gives no details. Badura-Skoda (1962) says Mozart knew of mp as well as mf, writing "pf" for the former, but gives no further details.
    5. Earliest use of fp: Johann Stamitz: Sinfonia in E-flat ("La Melodia Germanica" no. 3) (1755), I. Runner-up: Mozart: Piano Sonata in F, K.280 (1774), III.
    6. Earliest use of pp: Schütz: Saul, was vergolst du mich (ca. 1650) from Symphoniae sacrae III (Barenreiter; reprinted in Burkholder, 2006) (contributed by Cuthbert). Runners-up: Handel: The Messiah (1742), Nos. 17 ("Glory to God"), 18 ("Rejoice greatly", final version), etc. Johann Stamitz: Sinfonia in E-flat ("La Melodia Germanica" no. 3) (1755), I; Haydn: Symphony no. 7 ("Midi") (1761), I. NB: Rastall (1982) says pianissimo "appeared early in the 17th century".
    7. Earliest use of ff: J.-J. Rousseau: Le Devin Du Village (1752), Scene 1, air "J'ai perdu tout mon bonheur". Runners-up: Johann Stamitz: Sinfonia in E-flat ("La Melodia Germanica" no. 3) (1755), I; Haydn: Symphony no. 7 ("Midi") (1761), I.
    8. Earliest use of ppp: Beethoven: Piano Concerto no. 5 in Eb ("Emperor"), Op. 73 (1809), I. Runner-up: Schubert: Erlkoenig, D. 328 (1815). NB: Warner (1977) says J. G. Tromlitz's Flute Treatise of 1791 mentions ppp.
    9. Earliest use of fff: Haydn: Sieben letzten Worte (1787; Artaria), string quartet version and *orch. version (contributed by Hosar). Runners-up: Beethoven: Leonore Overture no. 3, Op. 72a (1806; Eulenberg ed.), m. 610; Beethoven: Piano Concerto no. 5 in Eb ("Emperor"), Op. 73 (1809), I. NB: Warner (1977) says J. G. Tromlitz's Flute Treatise of 1791 mentions fff.
    10. Earliest use of pppp: Berlioz: Symphonie Fantastique, Op. 14 (1830; Breitkopf & Hartel ed.), V, m. 345, double bass. Runners-up: Berlioz's Roméo et Juliette (1839; Breitkopf & Hartel ed.), no. 4, "La reine Mab, reine des songes", in the celli four bars before rehearsal no. 54 (contributed by Bala); Wolf: In der fruehe (1888; Peters ed.).
    11. Earliest use of ffff: Unknown. Tchaikovsky: 1812 Overture (1880) uses it, but his The Tempest, Op. 18 (1873) already used fffff (5 f's).
    12. Earliest use of forte possibile: Dussek: Piano Sonata in E-flat, Op. 44 (publ. 1800), I. An interesting runner-up is Chopin: Etudes, Op. 25 (1832-36; Paderewski ed.), nos. 10 in b and 12 in c, but there were almost certainly other uses between 1800 and this.
    13. Earliest use of hairpins: According to Rastall (1982), hairpins are first used in the violin sonatas of Piani (1712). Distant runners-up: Haydn: Piano Sonatas no. 9 in D (1767), I; no. 2 in e (1778), I; and no. 5 in C (1780), I (all Peters-Martienssen ed.); Mozart: Piano Sonata in D, K. 576 (1789), II (Presser/Broder ed.). But surely there were many uses between 1712 and 1767!

    4. Other

    1. Earliest metronome marks: Beethoven published metronome marks for all of his (then) eight symphonies in 1817.
    2. Earliest chord with explicit "arpeggiate downward" indication: Chambonnieres' (d. ca. 1672) table of ornaments; d'Anglebert's table of ornaments (1689). (Both contributed by Zarky; both use a notation completely different from the modern form.) Runners-up: Bartok: Fourteen Bagatelles (1908; Schirmer ed.), no. 10 (indicated with the normal arpeggio sign, but after the chord instead of before); Krenek: Piano Piece, Op. 39 no. 5 (1926) (this is the earliest with the now-standard arpeggio sign before the chord with an arrow on the bottom); Bartok: Piano Sonata (1926; Boosey & Hawkes ed.), I. But it's very hard to believe there were no uses between 1689 and 1908!
    3. Instrument-specific
      1. Bowed strings: Earliest explicit indication of what string something should be played on: Haydn: *Cello Concerto in D (1783), I (2nd theme). Runner-up: Beethoven: *Trio, Op. 1 No. 3, III (cello part). (contributed by Solow)
      2. Bowed strings: Earliest bowing marks: Arnold (1983) mentions Corelli: *Folia, Op. 5 no. 12 (1700).
      3. Bowed strings: Earliest glissando in natural harmonics: Rimsky-Korsakov: *Suite from the opera Christmas Eve (1895), in the section "Demonic Carol" ("Besovskaya Kolyadka"). Runner-up: Ravel: Rapsodie espagnole (1908; Durand ed.), p. 40. (both contributed by Burkholder)
      4. Keyboard: Earliest explicit glissando: D. Scarlatti: Keyboard Sonatas K. 379 (1754 or earlier) & *468 (glissandi in both marked "con dedo solo") (contributed by Lindorff); cf. Kirkpatrick (1953), p. 188. Runners-up: Padre Antonio Soler used glissandi in some of his keyboard sonatas (contributed by Lindorff). An interesting runner-up, perhaps the earliest specifically for piano, is Clementi: Piano School (1801), Waltz 9 (contributed by Solow).
      5. Piano: Earliest explicit use of una corda pedal: Beethoven: Piano Concerto no. 4, II.
      6. Piano: Earliest clear indication of "Ped. 3" (sostenuto pedal): *Percy Grainger: "Four Irish Dances (1907). Runners-up: *Percy Grainger: In a Nutshell (1916); K.S. Sorabji: Piano Sonata No. 2 (1920); *Busoni: "Mit Anwendung des III. Pedals" in the 9th book of the Klavierubung (1923); *Ruth Crawford-Seeger: Prelude No. 1 (1924). (all contributed by Gottlieb)
      7. Single-manual keyboard: Earliest use of more than two staves: In open score, Samuel Scheidt's *Tabulatura Nova (1624) (contributed by Zarky). In conventional keyboard notation, G. J. Vogler, Variations sur l'air de Marlborough (1791) (Fig. 125 in the article "Notation" in Sadie, 2001).
      8. Voice: Earliest use of sprechgesang (sprechstimme): *Humperdinck: Königskinder (1897) (cited in the article "Notation" in Arnold, 1983).