Berço manuelino is a breech-loading small cannon, loaded with an interchangeable chamber, shooting bullets at a distance of about 400 meters. The gun was mainly used on sailing ships, mounted on a swivelling stand and installed on the fore- and aftercastles. Berço manuelino from the wreck of Santiago dates most probably from the reign of King João III (1521–1557). This type of cannon was developed during the reign of the king Manuel I O Venturoso (1495–1521)—hence the name.

The breech-loading swivel guns appeared in Europe in the 15th century. They had two distinguishing features: they were mounted on a swivel (forquet) allowing easy rotation of the gun and used a breech loading system. These guns were loaded using movable chambers, previously loaded with a gunpowder charge. The loaded chamber was inserted into the recess in the rear part of the cannon, blocked with a wedge, and then fired. Because charging was done in advance and separately, the breech-loading swivel guns were exceptionally quick-firing for their time. Their disadvantage was a leaky connection between the chamber and the barrel, causing the loss of part of the powder gases. However, this was compensated by the high rate of fire, as a large number of charged chambers could be prepared in advance. The Portuguese name for such guns is ‘berço’, which literally means “a cradle”, but in this case it is a gun lock—a recess into which the loaded chamber is inserted (Fig. 1).

Being in collections of the KwaZulu Natal Museum in Pietermaritzburg, South Africa, the berço manuelino Durr 876 comes from the Portuguese carrack Santiago, which was wrecked on the reefs of the Bassas da India atoll in the Mozambique Channel on the 19^th August 1585, while on a voyage from Lisbon to Cochin on the route known as Carreira da ĺndia. Santiago was a very big carrack (port. nau). Described as “the biggest and best vessel sailing on the Indies route” (Duffy, 1955: 56), but such pompous descriptions apply to many ships of the time. Nevertheless, it was Santiago that was chosen in 1585 by the admiral of the Indian fleet for his own sea voyage. So it must have been a large and impressive ship. Santiago’s displacement was not less than 900 tons. A ship of this size must have been about 50 meters long (LOA), with a keel length of 33 meters and a maximum width of 16.5 meters.

In December 1977, the Swiss treasure hunter Ernst Klaar easily located the site of the Santiago disaster. During repeated illegal explorations of the site, Klaar acquired numerous artefacts from Santiago. Some of them were purchased by the Natal Museum in Pietermaritzburg, South Africa, several others by the Museu de Marinha in Lisbon.

The museum collections (KwaZulu Natal Museum in South Africa, Museu de Marinha in Portugal) include 16 cannons from Santiago. These are:

Colonel Nuno Valdez dos Santos believes that in total Santiago was armed with 24 cannons of various sizes (Santos, 1986: 46).

Most muzzle-loading cannons in South Africa have been identified, registered, measured and catalogued in a registry known as “Durr1 Record of Muzzle Loading Cannon in South Africa” (de Vries & Hall, 2001: 23-24). The register was created by Gerry de Vries and Jon Hall and currently records 1,056 cannons.2

Durr 876 ex Santiago is a swivel gun berço of 72 mm bore. The gun was breech-loaded with an iron chamber, which was not found on the wreck; probably corroded away completely. The assumption that the chamber was made of iron is confirmed by another artefact. At the archaeological site where the Durr 876 gun comes from, one severely corroded powder chamber was found. It was made of iron and thanks to conservation efforts, it was preserved. It is highly probable that the missing chamber from Durr 876 was also made of iron. The cannon is cast in bronze (gunmetal). Noteworthy in this piece is the asymmetric nature of the receiver, the recess in which the powder chamber was located. The thickness of the walls on the left side is 45 mm, and on the right side it is 58 mm (Van Oordt & De Vries, 2006: 24). There are no signs of serious corrosion or erosion and it is likely that the weapon was manufactured, albeit unintentionally, in this asymmetric way. The receiver near the bore has a hollow cup shape on the right side. It can be assumed that at some point the bullet fell out of the bore and over time, affected by the sea water, it eroded into the shape of a cup. The cannon was probably loaded when it sank. There are remnants of a wadding still in the bore. The berço on the top of the chase has heavily faded markings, made in raised relief: the coat of arms of Portugal, an armillary sphere and a single “C”, most likely the founder’s mark. None of these markings are clear and all have been heavily obliterated by erosion.

The Durr 876 cannon shows three markings foremost on the upper chase: Portuguese coat of arms, armillary sphere and supposed manufacturer’s mark, all made in raised relief. The markings are in raised relief on a raised shield.

The Portuguese royal coat of arms is located on the upper chase, closest to its muzzle. The drawing is heavily eroded, only fragments of the outer frame of the coat of arms are clearly legible. The original form of the coat of arms is debatable. The existing interpretation (Van Oordt & De Vries, 2006: 37) suggests that the lower part of the coat of arms has a triangular shape.

However, a closer examination of the coat of arms suggests otherwise. In the author’s opinion, the base of the motif is more likely flat or very gently curved rather than pointed as they illustrate. It appears to have been the same motif with a flattish base on another falconet in the collection of the Kwa Zulu Natal Museum (Durr 881). Unfortunately, the motif is not well enough preserved to be sure. However, the muzzle-loading cannon (Durr 884) does have a motif that is similar to, but not exactly the same, as the Van Oordt and De Vries interpretation (Fig. 2).

The drawing also suggests, that there are only five 'quinas' shields on the coat of arms, and there are no castles on the bordure. From 1557, the coat of arms should include seven castles. There seems to be no basis for such an interpretation. Not having any castles on the bordure would be a serious mistake for the founder. Although the gun is badly damaged, it appears that the outline of a bordure is visible on the shield (Fig. 2, left)

The author has attempted to reconstruct the original appearance of the coat of arms drawing. This proposal appears to be the most likely (Fig. 3).

The armillary sphere is located as a central mark on the upper chase. The ecliptic band runs NW-SE. The drawing is heavily eroded, only the outer frame of the sphere is clear (Fig. 4).

Farthest from the muzzle, as the third mark, is a slanted ‘C’. This is interpreted as a mark of the foundry or founder. This marking has yet to be satisfactorily associated with any known founder. As a possible hypothesis, it should be assumed that these may be foundry marks (not the founder's) or trademarks of the company ordering the cannon.

The Durr 876 gun (as well as all other guns from the wreck of Santiago) has no visible proof marks. It is not known if there was a standard method for proofing firearms in the early 16th century, it is assumed that satisfactory results were obtained by the trial and error method.

A cannon coming out of the foundry may differ slightly from the design due to manufacturing methods, allowable tolerances, shrinkage of castings, and aesthetics of the finished product.

If an internal barrel diameter of 72 mm is assumed, due to the required difference between the diameter of the bore and the diameter of the projectile3 of 1/16 of the diameter of the projectile, the diameter of the projectile (a ball) would be 67.5 mm. If we accept that this 72 mm diameter may have grown 1.5 mm due to wear and corrosion, then we will get the probable designed barrel diameter of 66 mm. The unit of measure used in Portugal in the early 16th century was p’e (foot) = 329,8 mm. A diameter of 66 mm is equal to ¹/₅ p’e. This is approximate equivalent to 2 pounders, according to the British calibre system. The actual weight of the ball was probably 0,45 kg (stone) (de Sousa, 2021: Quadro 4, Van Oordt & De Vries, 2006: 31), a lead one 0,67 kg (Marzia, 2014: XX).

The formula of the bronze alloy has been modified over the centuries. In the mid-16th century, an alloy of 100 parts copper, 10 parts latten and 8 parts tin was used to obtain a ‘gunmetal’. Latten was an alloy of copper and zinc (Blackmore, 1976: 407). Impurities of the 16th century alloys, unintentional or perhaps also intentional—added experimentally—incorporation of zinc, lead, antimony and other elements, and residual and imperfect mixing and casting methods lead to estimates of the strength of the 16th century bronze at 18 to 20 MPa. By comparison, modern true bronze or bronze is 90% copper and 10% tin, resulting in a strong, resilient alloy with a yield strength of 145 MPa. (Van Oordt & De Vries, 2006: 30)

Serpentine is a very finely ground gunpowder. In Portugal, the use of corned gunpowder is reported from the late 14th century, used primarily in handguns, like the serpentines, from where the serpentine gunpowder name arose. The artillery gunpowder would have a bigger grain, it is already in use in the late 16th century. Its ingredients: saltpeter, sulfur and charcoal, were first ground separately, then mixed and ground again. The final product had the consistency of fine dust. The formula of gunpowder in the middle of the 16th century was 66% saltpeter, 17% sulfur and 17% charcoal (which would give a maximum pressure of 50 to 55 MPa) (McConnell, 1988: 273). Peter Whitehorne (1573) reports that gunpowder from the early 16th century contained about 50% saltpeter in the formula (Ffoulkes, 1937: 85-90). The propellant charge therefore burned slower and had less power than modern gunpowder. The fact that the compressed charge of this powder did not have significant spaces between the grains through which the flame could advance rapidly further reduced the efficiency of the weapon. The process of the formation of granules of gunpowder,4 was not yet widely used in the first half of the 16th century. The pressure obtained from modern gunpowder is 90 to 100 MPa, and the pressure of serpentine from the beginning of the 16th century is estimated at about 25 MPa (Gillet, Lefebvre and Galloy, 2004: 22).

Berço manuelino Durr 876 was salvaged from the wreck of the Santiago, which sank in the Mozambique Channel in 1585. The date of the disaster allows us to unambiguously determine the year after which the cannon was definitely not manufactured (terminus ante quem).

The assumption that the cannons were thrown just before the ship’s maiden voyage does not seem very likely. The gun could have been in active service for 20, 30, or even much longer. For example: the age range of cannons salvaged from the wreck of Santissimo Sacramento is 63 years; the oldest of the cannons is dated to 1590, and the newest to 1653. (Guilmartin, 2002: 62-69). Guns in service listed for HMS Britannia in 1736 included both cannons dated 1579 and 1705, a span of 114 years! Years of service don't really matter for bronze guns. The decrease in their condition resulted primarily from the number of shots fired, which negatively affects the calibre. Their gun carriages wore out more often than bronze cannons. Often the carriages were replaced with new ones two or even three times during the life of the gun. Cannons often outlived their ships by being transferred to the newer ones. At the beginning of the fifteenth century, Portugal, building and strengthening its empire, suffered from a constant shortage of artillery. The old guns were kept in service as long as possible. The date of the sinking of the ship from which the guns were recovered is only the latest possible date of manufacture, but it may well be a gun several decades old at the time of the disaster. In no way can it be assumed that the guns aboard each wreck were new or relatively new when the ship sank.

A closer dating of the Durr 876 cannon is possible due to the presence of an armillary sphere on the upper chase. Dom Manuel I (reigned from 1495 to 1521) was the first Portuguese king to use the depiction of the armillary sphere as his royal standard. Thus, the cannon on which the armillary sphere is depicted was not made in the period preceding the reign of this king. However, the armillary sphere continued to be placed on cannons for many decades after the king's death. The armillary sphere was commonly used on Portuguese guns during the 16th century, most probably in the reigns of João III (1521–57) and Sebastião (1557–78) (Auret & Maggs, 1982: 6). Thus, its presence is only an indication, narrowing down the time period in which the cannon could have been manufactured.

Determining the exact date of manufacture of the Durr 876 cannon is therefore not possible.