Appendix 4: The Glass

Rachel Tyson

Cite this as: Tyson, R. The Glass in N. Corcos et al. Excavations in 2014 at Wade Street, Bristol - a documentary and archaeological analysis, Internet Archaeology 45.

One hundred and two fragments, including some complete vessels of glass, were submitted for specialist examination. The majority of the glass was late 17th to 18th century in date. As well as the ubiquitous wine bottles and pharmaceutical phials, there were fragments of two table vessels and a case bottle, and a few fragments of window glass. More unusual fragments included a section of possible thermometer or barometer tubing, a large dark green glass bowl, a possible linen smoother and some misshapen opaque pale blue fragments. Glass of the later 19th and early 20th centuries all came from various bottles for drink, sauces or medicine, some embossed with the name of the contents or the glassmaker. The glass is discussed below, followed by a catalogue of selected fragments. Full details of all the glass can be found in the archive.

Late 17th to 18th-century glass

Table vessels

The base of a colourless vessel forms a short pedestal, above which the broken body walls flare out at an angle (Catalogue no.1, context 1181; Figure 41). This may come from either a jelly glass or bonnet glass for serving jellies and creams, or a tankard, both dating to the mid- to late 18th century (see Bickerton 1986, 254-61, and Museum of London Archives website e.g. accession no. 34.139/164). It is not known what vessel a small vertical rim fragment of cobalt blue glass comes from (Catalogue no.2, context 1091), but tableware of blue glass was popular in the mid-18th century (e.g. jelly glasses or monteiths, Museum of London Archives website acc. nos 34.139/157-8, 160, 163, 167). A small fragment of pale greenish glass has an optic-blown rib but the form cannot be determined (Catalogue no.3, context 1185). Optic-blown ribs were common on greenish glass English table vessels such as beakers, bowls, flasks, dishes and jars in the 16th and continuing into the 17th century (Willmott 2002, 38, 47, 80, 93, 95, 96, 98). Colourless or strongly coloured glass was more commonly used for decorative vessels from the later 17th century onwards, but there are still instances of greenish glass vessels, such as a flask with optic-blown ribbing from a context of the second half of the 17th century from Number One Poultry in London (Willmott nd, G125).

Wine bottles

A total of 52 fragments of dark green wine bottles were excavated, all belonging to the hand-blown tradition of the mid-17th to the early 19th century. Compared to wine bottles from other excavations, this is a relatively fragmentary assemblage, and the glass is in poor condition. Twenty-seven fragments had features that could be dated more closely within this period (see Table 13). The earliest was an unstratified rim and neck fragment with a crude string rim applied 6-8mm below the rim edge, and a slightly tapering neck. While the shape of the body is not known, this is characteristic of a 'shaft and globe’ bottle of the second half of the 17th century, or possibly an early 'onion’ bottle of the late 17th century (Dumbrell 1983). About 13 fragments come from the rounded lower bodies of late 17th to mid-18th-century 'onion’ or 'bladder onion’ bottles (contexts 1096, 1104, 1190, 1204). Cylindrical bottles of the later 18th to early 19th century are represented by body fragments that have a slight sag on the lower body, from contexts 1094 and 1098, and a complete cylindrical base 120mm wide with a sagged lower body from an unstratified context. Another unstratified cylindrical base with a diameter of 96mm had straight sides with no visible sag, probably an example of how some glasshouses started blowing the body into a mould towards the end of the 18th century. A crude double-string rim has a likely date of c.1770-1800 (context 1146). Table 14 (see Catalogue) lists a further 25 less closely datable wine bottle fragments from this tradition, and more details can be found in the archive catalogue.

Table 13: Wine bottle fragments of dark green glass with datable features (dating based on Dumbrell 1983)
ContextNo. of fragmentsDescriptionDating
10946Cylindrical body fragments, including lower body slightly sagged at base of body, and rounded shoulders.Mid-late 18th to early 19th century
10961Curving fragment from the base edge of a curving 'onion’ wine bottle or 'bladder onion’ bottle.Late 17th to mid-18th century; most popular c. 1700-30
10981Lower cylindrical body wall, slightly sagged at base of body.Mid-late 18th to early 19th century
11042Pushed-in base edge, curving round to bulbous lower body wall.Late 17th-mid 18th century
11464Rim and neck with 2 adjoining fragments of rounded shoulders leading to missing cylindrical body. Crudely applied double string rim, neck constricted just below. RD <35mm. One further body fragment.c. 1770-1800 (Dumbrell 1983, 92)
11909*Complete rim and neck with angular string rim only just below sheared rim edge, rim very slightly everted. Short flaring neck widening out to bulbous missing body of 'onion bottle’. RD 30mm. Pushed-in base edge and lower body wall of bulbous body; 7 bulbous body fragments probably belonging to same bottle.Late 17th/early 18th century
12041Pushed-in base edge, rounded lower body walls.Late 17th-mid 18th century
100 (U/S)1Complete base from cylindrical wine bottle. Wide base diameter (c. 120mm) with high dome-shaped kick and pontil mark. Slightly sagged lower body wall.c. 1735-1820
U/S1*Rim/neck fragment. Angular string rim of uneven width 6-8mm below sheared rim edge; lower join of string rim smoothed, upper join has a small gap. Neck tapers downwards gradually, to extant height of 50mm. RD c. 30mm.Second half 17th century
U/S1Complete base. Cylindrical with straight sides, shallow pushed-in base and pontil mark. BD c. 96mm.Late 18th-early 19th century
Table 14: Other wine bottle fragments of the period from the late 17th to early 19th century
ContextNo. of fragments

Bristol had at least eight glasshouses by the end of the 17th century, out of 90 listed in England and Wales at the time, predominantly making bottles (Blakelock 2007, 1), and it is likely that the wine bottles used in Wade Street were Bristol products.

Square case bottle

A less common vessel was the base of a square 'case bottle’, excavated from context 1012 (Catalogue no.4). This is made of thick green glass, visually of a similar type to the 18th-century wine bottles above, but square examples are not common. Dumbrell describes two complete square bottles, which he dates to the early 18th century by their olive green glass and string rim (Dumbrell 1983, 144). He suggests that they may also be wine bottles, and the shape was designed for packing in wooden crates and transporting by ship. However, he observes, if this was the reason they were square, surely there would be more examples. Nevertheless, while they are not common in Britain, they are found in some quantity in the early colonies of North America, supporting this theory (Willmott 2002, 86). A limited number of other similar bottles have been excavated in Britain, for example in an assemblage of the first half of the 18th century from Smithwood Bastle in Scotland (Murdoch 2006, 32 and 40), and at Poplar High Street in east London from a 19th-century pit (Sygrave 2004, 219-20). Henkes, in his well-researched catalogue of 'Utility glass from five centuries excavated in the Low Countries 1300-1800’, includes a complete example of a probable Dutch olive-green gin bottle dated to the second half of the 18th century from Surinam, but comments 'archaeological material of these bottles only comes to light in export countries, in America and in Africa’ (Henkes 1994, 311-12, no. 64.1).

Phials/small bottles

Cylindrical phial fragments of pale greenish glass were excavated from seven different contexts (1086, 1090, 1091, 1096, 1146, 1207 and U/S), and represent a minimum of four vessels: one almost complete, as well as three bases and three rims (Catalogue nos 5-11). All have the same cylindrical form, but vary in size. All rim fragments have a short vertical neck that is wider nearer the horizontally everted rim, with a widely everted shoulder top where present. The rim diameters range from 26mm to 32mm, and the base diameters from 33mm to a large 70mm. The complete phial (Catalogue no.9) has a shoulder wide than the base, and differs from the other examples in having a rounded kick on the base, covered by a pontil mark and possibly an extra blob of glass. The remaining bottles have the kick pushed in to a pronounced point, with a small hollow tool-mark in the centre, and a circular pontil mark.

This phial form is common from the second half of the 17th century and continues through the 18th century (Willmott 2002, 90-1). While those of the second half of the 17th century may not have such pronounced horizontal rims (such as at Nonsuch Palace, Charleston 2005, 225-6, 255-7), and colourless phials become common in the second half of the 18th century, the form possibly becoming wider at the base than the shoulder at this time (Keily and Whittingham 2009), the evolution of the phial form requires more research. They are generally thought to have contained medicines and other pharmaceutical contents, a theory borne out by their presence in what appears to be the disposed contents of an apothecary's shop in Coleman Street in London of around 1745-80 (Keily and Whittingham 2009). The Coleman Street phials differ in size like those from Wade Street, and the larger bottles are thought to have been used for drinks with medicinal properties, and the smaller phials for draught medicines and drops (Keily and Whittingham 2009).


Four fragments from a large hemispherical bowl with everted rim of olive-green glass were recovered from context 1190 (Catalogue no.12). This has a similar profile to a bowl type of the first half of the 17th century described by Willmott (2002, 94, type 28.1). It is not common, with examples only from Norwich and West Bromwich Manor, as well as the glass furnace site at Haughton Green suggesting they were an English type (Willmott 2002). It is possible that green glass bowls were made at a later date, but have escaped archaeological interest. For example, 'at least three broad diameter shallow bowls with cut lips of a type previously unrecorded’ in olive-green glass were recently found among the remains of the late 18th-century chemistry stores at New College, Edinburgh (Addyman 2015, 122, colour plate 18, see 'Tubing’ below for more information about the site).


The most unusual glass artefact was a section of colourless tubing 37mm long on excavation from context 1094, 8mm in diameter with a perforation of 1.5mm (Catalogue no.13; Figure 42). It is possible that it comes from a thermometer or a barometer. The earliest surviving thermometers date to the 17th century, using alcohol in an enlarged bulb at the base, and archaeological examples of what are thought to be Venetian alcohol thermometers of the 17th or 18th century have been excavated in Langestraat, Alkmaar in the Netherlands, the scale marked on the glass with dots of blue enamel (Henkes 1994, 326, no. 66.10). A similar example attributed to mid-17th-century Florence can be seen in the National Museums of Scotland (accession no. T.1975.62) Scientist Robert Hooke described how he made 'sealed thermometers’ in Micrographia (1665): 'The stems I use for them are very thick straight, and even Pipes of Glass, with a very small perforation, and both the head and the body I have made on purpose at the Glass-house, of the same metal whereof the Pipes are made...’ and goes on to describe how he marks the divisions on the glass.

Farenheit developed the modern forerunner of the more reliable mercury thermometer with the scale using his name in the early 18th century, and three of his thermometers survive (although the glass may have been replaced), such as an example mounted on a brass backplate with the markings on the brass (Christies' Sale 6911, Lot 69, October 2012). Archaeological examples of apparent thermometer sections, including one with a bulb, were excavated at New College, Edinburgh University in 2011, among the chemical equipment of professor of chemistry Joseph Black, thought to date to the 1790s (Addyman 2015, 122; Tom Addyman pers. comm.). Curved glass tubing with a small cavity was also found in this assemblage, representing another type of chemistry equipment (Addyman 2015, colour plate 4). A section of straight colourless tubing was found in excavations in Oxford that included the remains of an 18th-century chemical laboratory; this 90mm-long section was a little larger than the Wade Street fragment, with an external diameter of 13mm and an internal bore of 5mm (Hull 2003, 20). The Museum of London has a colourless section of straight tubing, 9mm in diameter with a small bulb at one end, suggesting that this was a thermometer (Museum of London Archives website acc. no. NN23945). The lack of markings on the Wade Street tubing suggest that if it does come from a thermometer, it would have been mounted on a backplate that had the scale markings on it, rather than an independent thermometer like the earlier alcohol thermometers referred to above, or later clinical thermometers.

The barometer was also invented in the 17th century, and further developed in the 18th century, greater affordability making them more common for domestic use to predict weather changes (Hills 1999, 5). Like the thermometer, the 'stick’ barometer consisted of a glass tube, with an open cistern for mercury at the base, and was mounted on a wooden backboard. An example of a barometer and thermometer with tubing of similar diameters, mounted on the same board decorated in a Chinese style, was made in 1719 by Isaac Robelou of London (Science Museum acc. no. 1909-134). However, most 18th-century barometers appear to have glass tubing of a wider diameter than the Wade Street fragment (Hills 1999). Interestingly, there were a number of instrument makers working in Bristol in the 18th century, such as Joshua Springer, recorded as living in Clare Street in 1775 (Alison Morrison-Low pers. comm.; Goodison 1977; Sketchley's Directory 1775). The glass tubes are likely to have been made by local glassmakers, and engineer James Watt's brother John purchased eighteen barometer tubes for him in Bristol in 1761 (Hills 1999, 6).

Misshapen fragments of opaque pale blue glass

Three small fragments of glass were an opaque pale blue colour, and were misshapen in some way: a short length of glass was broken at both ends with a ridge along one side, a curving body fragment had puckered surfaces, and the third small fragment was folded and misshapen with irregular edges (Catalogue nos 14-16, contexts 1114 and 1115). Through transmitted light the glass appears brownish in places, and this dichroism/opalescence is an accidental characteristic noted in high-lime low-alkali (HLLA) glass, the type of glass used for 18th-century bottles, and noted in bottle glass-working waste from Bristol (Dungworth 2011). However, it may occur in circumstances other than glass production. While experimenting with distilling soot in 1763, experimental scientist William Lewis's glass apparatus, made from bottle glass, transformed into dichroic material, demonstrating its occurrence when glass is subject to extreme heat for a length of time (Dungworth 2011, 371, 374). Recent excavations of medieval window glass at St Mary's Abbey, Holme Cultram in Cumbria, found a number of window fragments transformed into this pale blue dichroic glass, thought to be the result of having been in a fire (Tyson forthcoming). In conclusion, the three fragments from Wade Street are insufficient evidence on their own for glass-working at this location, although there were many glasshouses in Bristol in the 18th century (see above). It is likely that they were vessel fragments affected by heat, or among debris brought from elsewhere in Bristol.


One large fragment of glass was rounded on one side, with shattered break marks on the other side (Catalogue no.17, context 1145). The surface has small hairline cracks, suggesting possible heat damage. It may be part of a linen smoother, consisting of a large lump of glass rounded on one side, made in glass from the early medieval period until the 19th century. Post-medieval examples often had a solid rod handle, giving them the appearance of an upturned mushroom, and examples have been excavated from the 16th-century glassmaking sites at Rosedale and Hutton in Yorkshire (Charleston 1972, 143-4). They have also been found in later contexts (e.g. Museum of London Archives website, accession number 5500, attributed to the 18th century).

Window Glass

Six fragments of flat pale green or almost colourless glass are likely to come from windows. Only one fragment has an indication of the original quarry shape, with lead shadows faintly visible on the weathered surface around three sides suggesting a diamond-shaped quarry with edges approximately 43mm in length (Catalogue no.18, context 1037). A fragment from context 1176 is a similar thickness (3.2mm) and colour. Windows with leaded glazing are likely to date between the 16th and 18th centuries, and diamond panes are typical of the 16th and 17th centuries. The remaining fragments are thinner at between 1.3 and 1.8mm, in pale green or colourless glass with a pale blue-green tinge (contexts 1096, 1111, 1195, 1230). At least two have indications that they were made by the cylinder method: with elongated bubbles, or being dull on one side and glossy on the other (Dungworth 2012, 10-11). The weathering and lack of bubbles on the remaining fragments prevent identification of the method. Window glass was common in domestic buildings by the end of the 17th century, and cylinder glass was less common by the early 19th century (Louw 1991). While crown glass became popular in the 18th century in sash windows, cylinder glass continued to be the more affordable choice.

Nineteenth to 20th-century glass

The glass of the 19th and 20th centuries excavated at Wade Street was smaller in quantity, and all from bottles, with no decorative vessels. Around 1821 Henry Ricketts of Bristol developed a three-piece mould for glass wine bottles (Dumbrell 1983, 115). This marked a change in the appearance of bottles, and also enabled them to be produced in larger numbers, leading to further mechanisation of the bottle-making process through the century, with the first fully automated bottle-making machine in Britain, the 'Owens’ machine, established in Manchester in 1906 (Cable 2002, 10). While the wealthy bought household goods in glass packaging in increasingly large numbers through the century, poorer families continued to make many of their own products (Licence 2015, 4-5). However, glass was recycled and re-used on a large scale; even broken glass could be sold to Swedish emery paper manufacturers (Licence 2015, 7). It is uncertain whether the relatively small number of vessels from Wade Street indicates a lack of spending power or rubbish going elsewhere.

The earliest of the new semi-mechanised bottles from Wade Street is a complete pale green octagonal-sectioned medicine bottle embossed with 'BRISTOL ROYAL INFIRMARY’ on alternate faces around the sides (U/S). The separately applied rim, and machine-moulded manufacture dates it to between the 1860s and 1890s. The title 'Royal’ was given to the Bristol Infirmary when Queen Victoria visited in 1850. A marble of pale green glass comes from a globe-stoppered bottle, a carbonated drinks bottle invented by Hiram Codd in the 1870s and in use until the 1920s (Talbot 1974, 46-54). The marbles were redistributed when children smashed the bottles to extract them (Talbot 1974, 60). A complete colourless bottle of oval cross-section was embossed 'WOODWARD CHEMIST LONDON’ and had an external screw-thread (see Five further screw-top bottles were examined as a sample from over one hundred colourless square-sectioned bottles with external screw-thread necks found in context 1027. It appears that they had been neatly stacked, presumably for re-use. They were all embossed on the base with the initials 'J L & Co’, the mark of bottlemaker John Lumb and Co. from Castleford in Yorkshire, operating from the 1870s until 1937 when they became part of United Glass Ltd (see Although the company changed the name slightly to 'John Lumb and Co Ltd’ in 1905, the suction scar on the base and the mechanised screw-top design date the bottles to between c. 1910 (or more likely 1920s) and the company's amalgamation in 1937. This bottle shape was used for table sauces and other liquid or semi-liquid foodstuffs. Fragments of another pale blue-green press-moulded bottle of octagonal section came from context 1123. A bottle stopper from context 1188 came from a late 19th or early 20th-century bottle of pre screw-top design; this type was common on sauce bottles.


While not in itself especially large, the Wade Street glass assemblage is relevant in representing a variety of activities over the post-medieval and modern periods in archaeological contexts. The late 17th and 18th-century glass includes decorative vessels indicating a comfortable living, as well as the more common wine and medicine bottles and window glass, and the green bowl rim, which although not a luxury vessel, is an unusual find. The tubing is intriguing and an important addition to a small number of similar suspected thermometer or perhaps barometer fragments, although its presence here remains a mystery in the absence of any other indications of a scientific interest. The heat-damaged fragments also raise questions. The 19th and early 20th-century glass is poorer in quantity and variety, although the embossed bottles, such as that from the Bristol Royal Infirmary, are striking. This may represent a fall in the fortunes of Wade Street, or more efficient disposal and recycling.


Internet Archaeology is an open access journal. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.

University of York legal statements