top of page

Crotal Bells

Historical Introduction


Bells are one of a very small number of artefacts that have been in virtually continuous production for over 4000 years. The earliest known examples were made in China before 2000 BC, and they were familiar everyday objects to the ancient Indians, Egyptians, Greeks and Romans. They have served a number of purposes, from ritual, magical and religious, to musical, signalling and warning. Apart from their functional role, bells have served as decorative devices throughout the ages, and continue to be popular as harness embellishments to the present day. Their longevity is reflected by the fact that the Guinness Book of Records lists the Whitechapel Bell Foundry as Britain's oldest manufacturing company, having been established in 1570, or possibly even earlier, and still producing bells today.


The earliest bells were cup-shaped and were struck externally with a separate striker, but it was not long before the attached internal clapper was invented, and the two types have co-existed ever since. The crotal bell was developed somewhat later. It differs from the preceding types in that its clapper is loose and contained within an enclosed chamber with perforations to allow transmission of the sound.


Although it is often stated that the crotal bell is much older, it is difficult to find any reliable evidence of its existence prior to the medieval period. The earliest dateable examples identified while carrying out research for the present article are some of the 9th century AD, recovered from female graves in Gotland, Sweden. They were found on chains suspended from chatelaine-type brooches, and appear to be of similar construction to English crotal bells dateable to the 13th century.


It is worth mentioning that, depending on context, sleigh bells, jingle bells, pellet bells, hawk bells and rumbler bells are all terms used to describe bells of the crotal type. Technically they are regarded as rattles, rather than true bells.

Anchor 1

The Chronology and Dating of English Crotal Bells


The earliest crotal bells found in England date to the beginning of the 13th century. They are of tin and were cast as open bells with an integral suspension loop and four ‘petals’ forming the lower body. The pellet, also of cast tin, was placed inside the open-ended bell, and the four petals were pushed inwards to meet at the centre and enclose it. Bells of this type were produced only until about the end of the 13th century. (Cf. MoL, Dress Accessories, 1668-1671; Mitchiner, Medieval & Secular Badges, 356.)

Cast copper-alloy bells based on the same principle are probably contemporaneous, or nearly so. Their integral suspension loops are frequently at the end of an extended shank, and a number of them have been found attached to harness mounts - perhaps their principal use. Bells of this type probably date from about the mid 13th to the mid 14th century. (Cf. Read, Metal Artefacts of Antiquity, 465-467.)

A variation of the previous bell is the teardrop type. Here, the bell is cast in a flat, fan-shaped form, with a loop at the apex, and four petal-shaped projections at the base. The casting is then rolled to form an elongated cone with a seam, and the four base projections are folded in to retain the pellet, as on the preceding types. Similar teardrop-shaped bells have been found on elaborate harness decorations with pendants that are dateable to the 14th century. (Cf. Read, Metal Artefacts of Antiquity, 464.)


Tin crotal bell,13th century


Copper-alloy crotal bell, 13th-14th century


'Teardrop' crotal bell, circa 14th century

Alongside the early cast crotals, copper and copper-alloy bells of sheet metal were produced. The body of these is made in two halves, formed by hammering the sheet into shaped moulds, and joined together, after inserting the iron ‘pea’, with a lead/tin solder. On the very earliest of this type, the loop was made of circular-section wire, which was inserted through a small hole in the top of the bell and its ends splayed in the manner of a modern split-pin. Slightly later, a narrow strip of sheeting was used instead of wire, and was either fitted in the same way, or formed into a ring and soldered to the top of the bell as on the example illustrated. Bells of this type have been recovered from secure contexts that span the date range circa mid-13th to mid-15th century. They are also found in a wide range of sizes, at least from 13mm to 34mm diameter, suggesting a variety of different uses. (Cf. MoL, Dress Accessories, 1644-1667; B Read, History Beneath Our Feet. p.55, No.2; Mitchiner, Medieval & Secular Badges, 350.)


Sheet-metal crotal bell, 13th-15th century

Around the end of the 13th century, a new type of white-metal (pewter and tin) crotal bell, cast in one piece, appears. The form is approximately spherical, but, as cast, the bottom half of the bell chamber is splayed. This enables the pellet to be placed inside the bell, and the splayed half to be squeezed together to retain it. It also makes support of the core within the mould relatively easy. The earliest bells of this type have several moulded parallel ribs around the circumference, both vertically and horizontally. Later ones are often plain, but some have moulded decoration of various forms. The rounded ends of the sound bow are often very close to, or interrupt, the girth rib. Bells of this type are usually quite small (typically 13mm to 17mm diameter), and many were used as dress accessories and hawking bells. The wearing of bells became fashionable in the 14th century and remained so well into the 15th century. Examples dating from the later end of this period have been found suspended from necklaces and possibly bracelets. Prior to becoming fashionable, the wearing of bells as a dress accessory was limited to jesters, acrobats, pilgrims and priests. (Cf. MoL, Dress Accessories, early: 1672-1683, late: 1689; Mitchiner, Medieval & Secular Badges, early: 351-355, late: 782-786.)


Cast one-piece white-metal crotal bell,

late 13th-15th century


Two-piece cast crotal bell,

circa 15th century

A development that occurs during the late 14th century is the casting of bells in two halves, which were then soldered at the horizontal joint line after inserting the pellet. Bells of this type were produced in both white metal (tin and pewter) and copper alloy. They are known with domed (as illustrated) and conical upper bodies, and some have moulded decoration, while others are plain. They are distinguishable from the later one-piece crotals by the mould joint lines, which run in a vertical direction, as shown, on both the upper and lower halves, and also by the absence of holes in the upper part of the body. The separate girth ribs on the two halves are also a good indication of the type. A single-point attachment of the suspension loop to the bell, via a short shank, is also a common feature. The type is not a common find, and was probably short-lived, being superseded by the one-piece cast type, probably in the late 15th or early 16th century. (Cf. MoL, Dress Accessories, 1684-1688; Mitchiner, Medieval & Secular Badges, 396.)

The one-piece cast crotal bell represents a triumph of ingenuity, the manufacturing principle of which has not changed in 400 years. Details of the process, as now implemented, are fully described below, but essentially the pellet is contained within the sand core during the moulding process, thus eliminating the need either to solder a joint, or to bend the body into shape. Bells made in this way are readily identifiable by the two ‘sound holes’ in the upper half of the body. These are, in fact, primarily to facilitate positioning of the core, rather than for transmission of the sound. The two-part moulds for bells of this type are split at the girth rib on the bell, and consequently there are no vertical mould-joint lines evident on them. The girth rib serves the useful purpose of accommodating any minor misalignment between the two halves of the mould, as well as strengthening the bell and retaining the traditional appearance of those with a soldered joint. From the 16th century, the one-piece cast crotal rendered most other types of construction obsolete. One exception was the sheet-metal type, which, has been produced ever since for hawking bells, pet bells and other uses where a small size and lightness are key considerations. (Since the 18th century, sheet-metal bells have been produced by a die-forming process, rather than the metal being hammered into a mould.)


Early one-piece cast crotal bell

16th-17th century

As the method of manufacturing the one-piece bell has changed little since the Tudor period, the determination of their chronology is dependent on differences of detail, rather than basic manufacturing concept. Close dating is often difficult, unless the bells can be associated with a maker whose period of operation is known from documentary sources. This is rarely the case prior to the late 17th century, when some makers began to put their initials on the bells. The following details are helpful in determining an approximate date.

Anchor 2

Suspension Loops

The suspension loop on the earliest one-piece crotal bells was cast as an integral solid lug and drilled afterwards as a separate operation. The fillet radii between the lug and the top of the bell are often quite generous, and there is every indication that the pattern was made in one piece.


Drilled suspension loops on one-piece cast crotal bells of circa 16th to mid-17th century date


The top of an early pattern. This type was withdrawn complete from the underside of the mould.

The sprue will have extended from the top of the lug, and will have been cropped as part of the fettling process. The accompanying illustrations show some early suspension loops of this type, and an outline drawing of the top of a typical pattern used to produce the mould. Bells with suspension loops of this type are likely to date from the 16th to the mid 17th century.

During the 17th century, an innovation in the production process eliminated the need to carry out a drilling operation. Instead, by making the pattern with a detachable ‘sprue-piece’, it was possible to create a suspension lug with a cored hole. The sand in the upper moulding box was packed around the pattern (see drawing), which was then withdrawn, as normal, from the underside.


As-cast loop on 17th to 18th century crotal bell


The top of a later pattern with detachable ‘sprue-piece’. This was withdrawn separately from the top of the mould, thus leaving a core to produce the aperture in the suspension lug.


As-cast loop on 18th to 19th century crotal bell

The detachable sprue-piece, however, was withdrawn from the top of the mould, leaving a core of sand to create the aperture. Bells with lugs produced in this way are identifiable by their uninterrupted spherical profile forming the base of the aperture, which can no longer be round. There is also little or no fillet radius where the lug joins the bell, as this would have prevented withdrawal of the sprue-piece without damaging the mould.

There were not any further fundamental changes in the process of casting crotal bells of this type, but the suspension loops tended to become proportionally larger during the 18th century, and they often have a more angular appearance. They remained this way until the traditional design was largely superseded by a new style of horse bell in the mid-19th century.

Note: It is sometimes said that the suspension loops of later crotal bells, of the type described immediately above, were separately cast. This is certainly possible, but I have found no evidence to support it on those that I have examined. It is also difficult to imagine why a manufacturer would complicate the process by producing additional moulds and adding a brazing or soldering operation. It is true, however, that some modern crotal bells are made this way in order to fit ornamental handles. It is also likely that there will be bells that have had replacement loops soldered/brazed to them to effect a repair following breakage.


I also considered the possibility of the loops being separately cast, and embedded in the mould, such that they would fuse to the body of the bell. Again, I found no evidence to support such a process, and it would present difficulties in casting, as the loop is located at the top of the mould in the position where it is desirable for the molten metal to enter.

Anchor 3


Post-medieval crotal bells may be either plain or decorated, and decoration may be applied to both the upper and lower hemispheres, or to the lower hemisphere only. Where both hemispheres are decorated, the respective designs may be of similar or differing types.

A number of different decorative devices are used, but varieties of the so-called sunburst design (sunflower would seem more appropriate) are by far the most common. This takes the form of a number of elongated ovoid petals radiating from the centre of the hemisphere. The design is found on bells throughout the post-medieval period.

The second most likely form of decoration to be found is the fish-scale pattern. This was used during the early part of the post-medieval period, but is rarely, if ever, found on bells made after the 17th century. It is often used to decorate the lower hemisphere of the bell, in combination with a sunburst design on the upper hemisphere.


16th-17th century bell with sunburst decoration on both hemispheres


16th-17th century bell with fish-scale pattern on the lower and sunburst on the upper hemisphere

There are various other forms of decoration, including crowns (Civil War period?), leaves, and human and animal faces, but the vast majority of bells are either plain or have sunburst or fish-scale decoration, or a combination of these types. As indicated, the sunburst design occurs in various varieties, and some interesting geometric patterns are to be found. It should also be noted that the founder’s mark or initials often occupy the very centre of the decorated area on the lower hemisphere. These are discussed in detail below.

Bells that are decorated only on the lower hemisphere tend to be of later date, usually late 18th to mid 19th century. Those with no decoration also usually date to this later period. There are, however, exceptions to this general rule, and plain bells of early post-medieval date are also known. All the indicators discussed should be taken into consideration when dating a bell. It should also be noted that the decoration can often have a very worn appearance, and is sometimes barely discernible. It seems unlikely that such wear occurred in use, and it is probably mostly due to the use of worn-out patterns.


16th–17th century bell with a variety of the sunburst decoration in which each of the petals contains a figure ‘9’ symbol, and there are arrowheads pointing to the centre on each side of the sound bow


Late 18th crotal bell decorated only on the lower hemisphere

Anchor 4


In the absence of analytical information, any comment on the composition of the metal from which cast crotal bells were made is inevitably speculative. The difference in appearance, however, between many earlier post-medieval bells, and the majority of later ones is such that it warrants comment, as it plays a part in their dating.


Many of the earlier bells were cast in alloys that have a distinctly white or grey appearance, and they are often casually described either as being of pewter or having a tinned surface. On closer examination, however, neither of these descriptions would appear to be correct. Pewter is a lead/tin alloy, but the corrosion products, where they occur on these ‘white-metal’ bells, are green, thus indicating the presence of copper. If the bells had a tinned surface, it would be evident when examining fracture surfaces, but this is not the case. The metal is white and consistent across the section. The most likely interpretation is that it is a high-tin copper alloy, possibly with a small addition of lead to facilitate casting. For the present purposes, the point to note is that bells cast in this white metal are unlikely to post-date the 17th century. Apart from their colour, bells of this type tend to have a thicker cross-section than their later counterparts, and are correspondingly heavier in relation to their size.


Later bells are more readily recognisable as being of copper alloy, and are invariably found with a characteristic green-brown patination. 

Anchor 5

Makers' Marks

Many crotal bells carry a maker’s mark, invariably located at the centre of the lower hemisphere, and often within a cartouche or dedicated area of the design. The mark may be a symbol, typically a bell-founder’s hammer or the initials of the maker, or in some cases both. Makers’ initials, in particular, have proved to be helpful in dating post-medieval crotal bells and establishing their chronology. Foundries that cast crotal bells also produced other types, and, for the more important ones, this included church bells. The latter were often marked with the name of the founder and the date of manufacture, and as many of the bells have remained in service over the centuries, these details are available and have been recorded for the benefit of researchers. Using this information in conjunction with that from documentary and other primary sources, it is sometimes possible to relate makers to crotal bells that bear their initials.

However, it is necessary to sound a note of caution, as simply matching a pair of initials to the name of a founder can easily result in misattribution if there is no corroborative evidence.

It is estimated that about 170 bell foundries have operated in Britain since the middle of the 13th century, varying in size from cottage industry operations to major businesses. The number in operation at any one time rose steadily from approximately five in 1250 to a maximum of nearly sixty around 1700, and then progressively declined to just two at the beginning of the 21st century. Not all bell foundries will have made crotal bells, of course, but the scope for errors of attribution will be apparent from the statistics.

Where initials occur, they may be on one side of the sound bow, or divided by it. Where they are on either side of the bow, they may be orientated at 0 degrees or 180 degrees to each other (as the RW example illustrated). In some cases they are neat and may originate from the pattern. In others they are crude, and appear to have been engraved on the bell itself.


16th-17th century crotal bell with bell-founder’s hammer mark in shield


17th century crotal bell with the initials HW (William Seller’s Foreman ?) on one side and a founder’s hammer on the other


18th century crotal bell with the initials WG (William Gwynn ?) on one side and a founder’s hammer on the other


18th century crotal bell with the initials RW (Robert Wells Foundry)

Table 1 below is an alphabetical list of initials that have been confirmed either by direct examination of crotal bells or photographs of them. It also shows the names of founders attributed to them and the related foundries and approximate dates of operation. Many other initials are mentioned in the works consulted, and a number are attributed to founders, but where it has not been possible to trace an example of the bell, they have been omitted. The list is therefore inevitably incomplete, and will be extended as more information becomes available.

Table 2 provides further information about the foundries, the bell-founders and the relationships between them. It is arranged alphabetically by foundry location. The comments in the table reinforce the point made above concerning the scope for errors of attribution.

Anchor 6

Table 1: Initials and names on crotal bells

Table 2: Foundry and founder details

Article Status

This issue dated:

15 October 2017

(Revision History)

bottom of page