Eli Colman's 'box of bells' certainly counts as an unusual form of bell harp and has its own section. There are other instruments which resemble bell harps or fairy bells but may be only coincidentally similar. With more information about these instruments it might be possible to determine their relationship to the other swung zithers described on this website.

Here is an instrument which, at least superficially, looks closer to Colman's instruments than to typical fairy bells. Many thanks to Maura Barnett for the photos, measurements and other details.

It was bought in Lymington in the 1970s. Length: 64 cms, Width: 20 cms Depth 10 cms. Weight approx 2.3 Kg. Maura also owns a typical 10-string fairy bells instrument which is roughly the same length and width but weighs 1.25 Kg.  On the left-hand side there are 6 strings and 7 on the right.

As can be seen there is a metal object on the right-hand side in the playing area. Maura describes this as a metal hooked shaped device which is set into the soundboard. It rotates and can be pressed against the string in order to raise the pitch by a semitone. There isn't a corresponding device on the left-hand side.

Maura suggests this tuning: (left side) C D E F G A and right side (ascending in the opposite direction) Bb or B, depending on how the device is set, C D E F G A. 

 This instrument has no date but, hazarding a guess, it can't be from much earlier than 1900.

There are cut-outs in the sides in the playing area. Perhaps these are for gripping this quite heavy instrument when swinging and playing it. But they could be there simply as a place to place the fingers when playing with the thumbs of each hand. All the evidence that there is on the method of play for both the 18th century instruments and fairy bells is that they were played with the thumbs alone (or, in the case of the 18th century instruments, with plectrums attached to the thumbs). There wouldn't be any reason for having these cut-outs in the playing area if this instrument was played using both fingers and thumbs.

Firstly, let's compare it with a typical 10-string fairy bells (remembering that the fairy bells instrument was popular in Britain for many decades from the 1870s):

Both instruments share the positioning of the wrest planks and thus the unusual string arrangement with the string lengths getting progressively shorter, but in opposite directions (the FRAS principle). Both are rather plain, elongated rectangular boxes and are very different from the 18th century bell harps/English harps (with the exception of the problematic Kunitachi 1401). 

The obvious difference is that the 13-string instrument doesn't have a top or lid and it appears to have been designed that way. But fairy bells do have tops (or lids). And 18th century bell harps and English harps also have lids, even if they are sometimes depicted without them. Part of the bell-like sound comes not only from swinging these instruments (if they are swung) but also from the sound reverberating from the wires enclosed under a lid.  Possibly this is why the 13-string instrument has a soundboard. (Soundboards are very rare indeed on fairy bells). But still, on the 13-string instrument, although above a soundboard, the wires are played in the open, as it were. A characteristic sound of enclosed wires on fairy bells/bell harps is inevitably missing. 

Now let us compare the 13-string instrument to two of Eli Colman's instruments (also see Eli Colman).

The designs of these two Colman instruments look even more different from typical fairy bells whereas the 13-course instrument does look somewhat similar to fairy bells in the playing area: the shape of the sides and the cut-outs in the side (presumably for the fingers). But this could be entirely coincidental.

Colman is known to have made many instruments over 40 years. This 13-course instrument could be one of his even though it lacks features of the surviving ones. It could have been made by someone who knew of Colman's work or, alternatively,  someone whose work inspired Colman to start making instruments. Or the instrument's maker could be entirely independent of Colman. 

What is interesting is how this unusual 13-string instrument (presumably from the decades around 1900) relates to the popular and widely-played fairy bells. Perhaps it even has a traceable ancestry back to the 18th century. 

(The same is true of Colman's instruments which were sometimes misdescribed as fairy bells, simply because fairy bells were popular at that time).

It is still unknown what prompted Richard Cook and Co. to launch fairy bells (as a sort of novelty bells imitator) in the later 19th century. Advertisements appear from 1871 but the instrument may have existed before then. The fact that the fairy bells instrument had a somewhat distant, 18th century, predecessor seems to have played no part in its promotion, or in its popularity over many decades. 

Is this 13-string instrument simply an unusual variant of the popular fairy bells, or is it an independent creation, just as Colman's instruments seem to be independent of the fairy bells phenomenon? Both this 13-string instrument and Colman's instruments (and fairy bells, of course) follow the highly unusual FRAS principle of the scale ascending in opposite directions and only known on 18th century  bell harps/English harps. Perhaps, then, there are other lines of descent from the bell harp/English harp in the 19th century not just fairy bells.

It can be seen here:

https://collections.mfa.org/objects/51463/psaltery-bell-harp?ctx=bd18701e-137f-4bee-86aa-a815dbb586c7&idx=12

Brief summary of this section. Another instrument just like this one (below) has come to light. Both instruments have 19 strings in all, seven on the left side and eleven, plus a string on its own, on the right side.

Most fairy bells have only 8 or 10 strings.

This 19-string instrument has a strap on the left-hand side of the playing area and a sort of felt lining on the right-hand side. The right-hand thumb has to be able to reach over to pluck all twelve strings on the right-hand side. Presumably the felt has been added to cushion the right hand as it stretches over the strings. The first finger of the right hand appears to be used too. It is ideally placed to strike the outermost right-hand string which is separate from the other right-hand side strings.

This outermost string on the right side has its own wrest pin and cut-out on the top of the instrument. It fastens to a hitch pin at the back of the instrument, unlike all the other strings which are attached on the hitch pin plank. 

This outermost string is not the lowest note. Possibly, for an instrument tuned diatonically in C major, it is F#. Simple tunes like 'Blue Bells of Scotland' set in C major use an F# to cadence on the dominant. F#, a sharpened fourth, would be a useful, extra note for a diatonic instrument.

The plain rectangular box with single wire strings and cut-outs in the lid look like a very typical example of a fairy bells instrument.  But it has 19 strings whereas most surviving fairy bells have either eight or ten strings.  

The stringing follows the FRAS principle (Forward Reverse Ascending Scale: the strings get progressively shorter and the scale ascends at first from left-right towards the centre of the instrument and then from right to left, again towards the centre). There are 7 strings in a row on the left side and 11 strings in a row on the right side plus one other string by itself (non FRAS principle), also on the right side. Like fairy bells, but unlike 18th century bell harps and English harps, the hitch pins are at the playing end of the instrument.

In the late 19th century R.W. Cook and Co. produced 17-string fairy bells but these were just the 10-string model with the intervening chromatic notes on raised strings in front of the basic 10, and with an overall range of only a tenth (see later for photo).  Nevertheless there are examples of fairy bells with more than 10 strings and a range of more than a tenth: 12, 14 and 16-string instruments exist too. But this 19-string instrument is surprising.

The limited range of notes on the 8-string and 10-string models means that both hands contribute to producing one melodic line. After all, fairy bells are just meant to be fanciful bell-like sounds; the instrument was marketed by R. Cook in the 1870s as a novelty bells imitator. The possibilities for playing with  a right-hand, treble side with an accompanying left-hand bass are very limited. But with this 19-string instrument it is possible to think of 7 bass notes on the left with 12 trebles on the right and it would be possible to play a simple melodies with, admittedly, a basic accompaniment. This instrument is more like a small harp or psaltery, but nevertheless designed to look like fairy bells.

What about the string on the outermost right, with its own wrest pin beneath its own cut-out on the top? On some 18th century instruments the longest and lowest note is (surprisingly) on the outermost right side. But on this 19-string instrument this outermost right string isn't the longest. In the photos, the wire for this string doesn't appear to be of a significantly thicker gauge and it is not a wound string. This outermost string does not seem to function as the lowest in pitch so there must be some other reason for its existence.

The instrument has a strap, presumably for grasping and swinging while playing, just as on smaller instruments. But it only has one strap on the left side and there is no indication that there ever was a strap on the other side. Instead, on the right side, some felt has been glued (see below).

The playing area is in poor condition. It is very unusual because it is raised above the baseboard and so this instrument has, in effect, a small soundboard. The soundboard appears to be in two sections separated by a gap. The puzzling outermost string on the right side passes over a sort of metal rider and disappears into this gap. (The photos above are misleading because the string next to the outermost one has broken and is lying as if it were the outermost string - see below).

Fairy bells usually have no soundboard, so this too is unusual, but it's not unknown. For example:

Looking again at the  playing area of the 19-string instrument, there are many drilled holes which are probably a simple design feature. I wrote about the playing area on this instrument in my article in FoMRHI  and speculated  that something might be missing, perhaps some sort of contraption which might explain the presence of the felt on the right side. 

This photo shows again the outermost string passing over a metal rider and going into a gap. Notice that a broken string which should be on the left has fallen to the right. The string going into the gap is the outermost string.

In this closer view (above) the outermost string (with its own cut-out for the wrest pin) can be seen more clearly.  The metal rider increases the height of the string above the soundboard compared to all the other strings and this string can only be plucked from a position further forward than all the other strings. Its hitch pin is placed quite differently from the other strings too (see later).

The back of the 19-string instrument is different from other fairy bells too. Usually, as in the image below on the right, the back is made from a single piece of wood but with a reinforcing bar glued and screwed under the hitch-pin block.

The back of the 19-string  instrument, for some reason stained black, is designed in a different way. But for what purpose?

Another, very similar, 19-string instrument has surfaced. Many thanks to its owner, Maura Barnett, for the following photographs and measurements.

It is generally  in much better condition than the one above apart from its hitch pin block/area. There is a string missing on the right hand side. Length: 63cms, Width: 20 cms and Depth: 6cms. It weighs  1.25 Kg. These dimensions are very close to a typical fairy bells instrument. 

Below are the two instruments, side by side, for comparison. They are taken from different angles so the perspective is inevitably a bit distorted. Obviously they are two examples of the same design. (Both instruments have a strap on the left hand side even though the strap on one is not visible).

Rod Howell has an even more preplexing fairy bells variant. Many thanks to him for the following photographs and measurements of this instrument.

Brief summary of this section. A fairy bells instrument with 22 strings is very surprising. The 10-string fairy bells seems to have been the most common type. This 22-string instrument is larger and heavier than a 10-string instrument and it couldn't be swung and so it couldn't imitate the sound of a peal of bells.

It is a skilfully constructed instrument, obviously copying the design of fairy bells, but what it was designed to do, or how it was meant to be played, remains a mystery. It could be a one-off experiment or it could have had some didactic use in music education. Or, it could simply be a large fairy bells-type instrument, made to be played, but not played exactly like the more usual fairy bells.

It appears to have been designed to be set on a table with the instrument tilted upwards, away from the playing end. It might have been played with the thumbs as on typical fairy bells, but placed on a table it is possible that it was played in a more harp-like way.

The right-hand wrest plank has pencilled note indicating the tuning of each string. The tuning is chromatic. The left-hand wrest plank has no indication of tuning. If the left side was similarly tuned so that the instrument was chromatic throughout, the overall range would be an octave and a sixth. Perhaps the left-hand side was tuned in some other way, so the instrument was only partly chromatic.

Fairy bells instruments are diatonic instruments and can play simple traditional tunes but popular melodies in the decades around 1900 could include several accidentals. Perhaps this 22-string was designed to cope with more chromatic, popular music.

The instrument surely needs, or once had, some means of distinguishing the sharps and flats. 

Next is the instrument with the lid off and with the playing area now on the right side of the photograph:

One of the 18th century bell harps in Edinburgh is designed to be played on a table with the instrument tilted upwards, away from the player. It has a hinged stand at the other end of the instrument which lifts the instrument upwards. 

If the instrument is placed on a table, the hands don't have to grip the sides. The arms and hands are free to move around. This instrument doesn't have to be played with the thumbs, it could be played in a more harp-like way with fingers and thumbs. Or it could be played with a  plectrum held in each hand. The Edinburgh bell harp has quills associated with it. 

On the other hand this instrument has obviously been designed to look like fairy bells. Although it is too heavy to swing, perhaps the designer intended it to be played more like a fairy bells than a harp.

But there is something even more surprising. The instrument has 22 strings and this would give three octaves if tuned diatonically. But there are notes pencilled on the right (treble) side, next to each wrest pin:

On the right side the indicated notes are ascending from D to C, chromatically. There are no notes written on the left-hand side. If the left-hand side follows the same pattern then the left-hand side ascends from E flat to C#.

This  may seem somewhat improbable but if it were so, the instrument's range would be from an E flat to C, an octave and a sixth higher. Any tune in any key that fitted in that (quite small) range would be playable. Chromatic stringing suggests an instrument capable of playing more sophisticated music or, at least, capable of playing tunes in several keys. But this small range of notes overall - an octave and a sixth - is quite limiting and the possibility of two-part playing, with a melody plus some accompaniment, would be very restricted.

Perhaps the notes on the left-hand side are tuned in some other way, perhaps diatonically on the left side with the chromatic notes on the right. For example, the eleven left-hand side notes could be from a low G diatonically up an octave and a fourth up to C. That would give an overall diatonic range, in the key of C, of two octaves and a fourth (and with some chromatic notes on the right side).

Another possibility is that the left-side notes again ascend diatonically from a low G to C but with an F# rather than F natural, so that the instrument is basically in G major but with chromatic notes in the upper range. 

Yet another possibility is that the left side notes are partly chromatic. There are some constraints on all these possibilities. The instrument is clearly some sort of fairy bells type instrument, that is to say, a simple instrument, played in the decades around 1900 by amateurs, but also by professional entertainers on stage too. 

Perhaps this instrument is a basically diatonic instrument with some chromatic notes. If the left side is diatonic and the right side is chromatic, the overall range, diatonically is 18 notes, just like the 19-string instrument described above. The 19-string has one chromatic note, the 22-string has four.

Assuming for a moment that the instrument is diatonic on the left side, in the key of  C, just how useful would these chromatic notes on the right side actually be? I suggested that a single F# could be useful on the 19-string model (discussed above) for simple tunes like 'Blue Bells of Scotland' with a simple accompaniment. But, for tunes like 'Blue Bells of Scotland' having all the other chromatic notes might seem superfluous. 

On the other hand, popular songs from the early 20th century, and earlier, sometimes use accidentals and obviously cannot be played, exactly as written, on diatonic instruments. A typical music hall song such as 'Don't Dilly Dally on the Way', from 1919, has four accidentals. In the key of F the range is from C below F to D above it and the melody includes G#, F#, B natural and D#. 

With the given tuning for the right-hand side of this 22-string instrument (chromatic D-C) I couldn't find any key for a diatonic left side that would make 'Don't Dilly Dally on the Way' playable. If the left side had a diatonic scale in F, all the accidentals could be played but the song has a high D, a tone above the highest note on the instrument. In other keys only some of the accidentals are available. 

But 'Don't Dilly Dally', and tunes like it, could be played in several keys on an entirely chromatic instrument.  In that case this 22-string instrument is essentially a single-line, melodic instrument with both hands on both sides of the instrument creating one melodic line. After all, that is how melodies are played on typical 10-string fairy bells.

Perhaps, then, this curious, chromatic, fairy bells-type instrument is a response to the popular music of the decades around 1900.

Before looking at another pressing issue it's worth considering if this instrument was built as some sort of experiment. But without knowing the context of the experiment, meaningful discussion is problematic.  Similarly, it might have been built for some educational purpose: but what educational purpose?

There is another pressing, practical issue. Whether the instrument was entirely chromatic or not, if the pencilled notes are taken at face value, the right side of the instrument is definitely chromatic. Here is the hitch pin area again:

All the player sees are 22 strings. The right side strings are chromatic and the left side may also be. How can the player recognise the diatonic strings on the one hand and the sharps and flats on the other? 

Other instruments have their notes side by side along a single plane. Keyboard instruments, for example, use colour and have white for diatonic notes and black for the sharps and flats. And the sharps and flats are positioned differently, slightly apart from the white notes.  Mallet instruments (two-row xylophones, glockenspiels etc) have the diatonic notes nearer the player and the chromatic notes further away and raised above the diatonic notes.

On this instrument there is just a succession of strings side by side,  with no indication of what string is at what pitch. Perhaps something is missing that was once on this instrument.  Below is a fretless zither (sometimes referred to as a guitar zither). Like the bell harp and fairy bells it has a left side and a right side but the strings don't follow the FRAS principle. The left side has strings arranged to give a few chords for accompaniment. The right side on this kind of instrument is usually diatonic but some, like this one, are chromatic. Unlike bell harps and fairy bells, the strings on the right side ascend in pitch from  left- right.

On this fretless zither there are two octaves on the right side, that is 25 strings. The strings are simply arranged side by side as on the 22-string instrument. But the notes on the zither are indicated on a sheet under the strings. The diatonic notes are numbered 1-15. The chromatic notes are numbered in black circles. The name of each note and its position on the stave is given as well as a number.

There is a precedent for a chromatic, modified FRAS principle instrument, and that is the 24-course English harp made by John Simcock. It is described at the bottom of this page. Unfortunately the instrument is not available for inspection in the V&A and the following tuning comes from A Descriptive Catalogue of the Musical Instruments in the South Kensington Museum (Carl Engel, 1874) which is presumably referring to the same instrument. Engel says that the notes are inscribed on the instrument.

This tuning is almost fully chromatic over two octaves. Apart from the high F# on the left side, the notes in the outer positions on each side are ascending diatonically. The sharps and flats are in the middle, surrounding the low G. 

This 24-course instrument surely could not be gripped at the sides and swung when played. Apart from its weight (which is not known) the thumbs couldn't possibly reach over 11 or 12 courses with the hands fixed at the sides. And, unlike Simcock's other English harps, there are no 'lugs' on the sides.

This instrument could be quite easily played diatonically with a range of two octaves in C major, running upwards (rather oddly) from a D. Playing in other keys and playing accidentals looks quite daunting but, of course, there may have been players who did actually master this instrument.

The moral of all this could be that chromatic tuning takes bell harps/English harps/fairy bells too far away from the sound of bells.

Many thanks to Andrew Hartig for finding this curiosity. It is part of an online collection of objects from museums and collections in South Tyrol. The website describes the instrument (google trans) as a: 

[Bell harp?]: unsigned, (presumably from England, 1st half/mid 18th century)

So the instrument is possibly English, possibly a bell harp, and possibly from the first half of the 18th century.

Total length 662mm, maximum width 575 mm. The weight is not given. Materials used are wood, metal,  mother-of-pearl, and paper. The 'scale length' (if the translation is satisfactory) is given as: 'Melody strings' 218mm and 'Accompaniment strings'  700mm. (But the strings are all of different lengths, surely?). The instrument is painted on the outside and the painted decorations include King David playing  a 12-string harp and six bell ringers.

The website describes the instrument in some detail but an AI translation of the German, including in the prompt that the text is about an unusual musical instrument, is insufficient to clarify how this instrument is strung or how it could be played. 

The top area looks curiously bare, as if something is missing. On the assumption that this is some form of bell harp, the website description of the instrument interprets the two nails sticking out of the sides at the top of the instrument as indicators of where handles or lugs may have been attached. But the lugs on Simcock's English harps are solidly dovetailed into the wrest pin plank; no nails are involved. It is difficult to say what the protruding nails on this instrument indicate.  Also this zither looks quite weighty: was it even designed to be swung?

In the photo above it looks as if there is an upper row of nine courses and a lower row of five courses. A pin  is missing (and the hole filled in) on the lower row and perhaps there is a pin missing in the upper row. It is possible that the instrument originally had 14 triple courses of strings. But the website description is quite different. The pins are grouped horizontally, rather than groups of three, side by side:

Tuning Pins: * Group 1 (Upper): 26 pins in four horizontal rows (2, 9, 9, and 6 pins respectively).

○ Group 2 (Lower): Located on a cross-bolted beam; 14 pins in three rows (4, 5, and 5 pins). One hole in the first row was later filled with wood.

This is puzzling. Even more puzzling is the location of the hitch pins. It's hard to see where the strings are attached and where they pass over bridges.

The website description (in translation) is:

Interior and Stringing On the interior back, there are eight small wooden bars arranged in a slanted, asymmetrical downward pattern. These bars feature bridges topped with small steel edges.

● Bridge Distribution: The bars hold 4, 5, 5, 5, 4, 1, 1, and 3 bridges respectively.

● String Attachment: Strings are attached to pins on bars 1–3 and 5, and to hooks

driven into the soundboard below bars 4 and 6–8.

● String Path: Strings run from the first pin group to bars 1 and 3–5, and from the second pin group to bars 2 and 6–8.

● Stringing Style: Bars 1–2 are double-coursed (pairs); bar 3 has four double courses and one single string on the right edge. All remaining strings are single-coursed.

● Sound Holes: Bridge 6 is flanked by two rectangular sound holes.

This description is an automated translation of a technical description of an obscure musical instrument. I don't think that it is successful.

Obviously this instrument is some form of zither and its body is bell-shaped. There are opposing diagonal planks on each side of the soundboard and they are similar to the hitch pin planks on a bell harp. But there are five of them and it is not possible to see exactly where the strings are attached. There are also other places where some of the strings may be attached. 

With the ornate door closed there isn't much space below the lower group of wrest pins to actually pluck the strings. I don't think that it is possible to work out how this instrument was strung and played without more information. Perhaps there is a study of it somewhere.

Evidence for the bell harp/English harp in 18th century Britain is really quite limited and it is possible instruments like this Tyrolean zither were around too, but not recorded in English sources. But, on the basis of what is known about the 18th century bell harp in Britain, this instrument seems unconnected. It doesn't look anything like any of the existing instruments or illustrations; not in design, construction, nor in its decoration. 

The website suggests a date from the first half of the 18th century but that is only because it was conjectured to be an bell harp from Britain (and on the basis of inaccurate sources about the dates of bell harps). 

It is an extraordinary instrument. I don't think that is connected with the 18th century bell harp or English harp in Britain. Without that connection its dating could be later than given. Could it be an ingenious concoction, created by a local craftsman rather than a professional maker, from some time in the 19th century?