86

3.0 Results of the Archaeological Investigation

3.1 Methodology

3.1.1 Program of Archaeological Works The main phase of archaeological excavation commenced in October 2008. Initially, the Casey & Lowe team worked in conjunction with the site’s demolition program and then in advance of the main bulk excavation works for the basement. The basement area is located in the southeastern corner of the development site, which occupies a larger area bounded to the north by Bathurst Street, the east and south by Harbour and Liverpool Streets respectively, with Tumbalong Park and the Darling Harbour Precinct forming the western limit (Figure 3.1.1). In response to the development program, the archaeological excavation began in the south and progressed northwards as the site was cleared. Concurrent with the basement works, the site of the new Children’s’ Theatre, located northeast of the basement, was excavated. Facilitating basement perimeter piling, a 12m wide corridor along the eastern edge and an area in the southwest corner remained unexcavated until the piling rigs had completed their work. The main archaeological program culminated on this eastern basement boundary 31 March 2009. Following a short break, work in the southwestern part of the basement archaeological excavation was completed by mid April 2009.

A second phase of archaeological investigation linked with civil works for the public domain began in February 2010. The public domain was located within the northwest part of the development site. Three separate excavations within the public domain were completed by August 2010. The archaeological excavation, necessitated by development impacts, consisted of in total eight months of on-site work, and took place over two years, from October 2008 to August 2010.

Figure 3.1.1: An aerial view of the site (red outline) during the basement excavation, taken at the very end of 2008 (or early 2009) during the investigation of Area 6. Archaeological work began in October 2008 in the south and continued north as the site was cleared. Google Maps Satellite, C&L additions. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 87

3.1.2 Archaeological Areas The development site was divided into nine distinct areas based on historic property boundaries (Figure 3.1.2). The basement was located within Areas 5 to 9 and the public domain excavation work affected parts of Areas 4, 6 and 8. The following is a list of the main associations with each area in the 19th century: • Area 1 - Wearne’s Wharf • Area 2 - William Orr Engineering • Area 3 - Anchor Flour Mill, Sawmill, Sugarworks • Area 4 - PN Russell & Co Engineering Works • Area 5 - PN Russell & Co Carriage Works and Boiler House • Area 6 - Miller & Harrison Timber Yards and Sawmill • Area 7 - Captain Brooks’ Slaughter House, Soap and Candle Factory • Area 8 - Workers’ Housing • Area 9 - Barker’s Mill

Figure 3.1.2: The site (red outline) was divided into Areas based on mid 19th-century property boundaries (blue lines). The basement (dashed purple), children’s theatre (dashed green) and public domain (dashed yellow) excavations were located within Areas 4 to 9. Trigonometric Survey 1865, City of Sydney Archives, C&L additions.

A Note on the Area Divisions The Area divisions are based on boundaries established by the 1860s, following a period of intensive reclamation, wharf construction and development. Prior to this, much of the site was below the high water mark. In the early 1800s the land between Bathurst and Liverpool Streets was divided between three property holdings located for the most part outside of the development site to the east. By the late 1820s there were two properties; to the north the land was owned by ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 88

Thomas Barker, and to the south by Captain Richard Brooks. The site however, was analysed and excavated and recorded in accordance with the later historical property divisions. These Area divisions also formed the framework for the technical excavation reports and detailed site plans, artefact catalogue and specialist reports. In order to best illustrate development of the site, in particular up to the 1840s, this synthesis of results does not strictly adhere to the Area division framework.

3.1.3 Archaeological Phasing Historical research determined the phases of development and in general, these were consistent within each Area. The following describe the ten main phases that apply to all Areas:  Phase 1: Natural Landscape  Phase 2: Aboriginal Occupation  Phase 3: 1788 to 1820s Early Foreshore Activity  Phase 4: 1820s to 1830s Barker’s Mill Pond and Jetty  Phase 5: Late 1830s to 1840s Reclamation  Phase 6: 1840s to 1860s Residential and Industrial Development  Phase 7: 1860s to 1900 Residential, Industrial and Commercial Development  Phase 8: 1900s to 1920 Resumption and Railways  Phase 9: 1920s to 1980s Railways and Commercial Development  Phase 10: 1980s Demolition and Redevelopment

The detailed description of the archaeological excavation (trench reports and site plans) are presented within this chronological framework. This section (Section 3) is a synthesis of the excavation results and analysis, and presents the findings strictly within this framework as it also considers them archaeology thematically, such as ‘Reclamation and Land Consolidation 1825 to 1870s’ and ‘Workers’ Housing 1840 to 1900’.

3.1.4 Excavation Methodology The programming of the archaeological excavation was linked with the demolition and perimeter piling components of the construction works. The excavation was divided into areas based on the mid 19th-century historical streetscape and property boundaries. Each Area was excavated as it became available after demolition and removal of overburden. A 12m bund for the piling rig was left around the southern, eastern and northern basement perimeter. As the piling rig completed work and the capping beam constructed, this bund was reduced and included in the archaeological excavation. Work began in the southern area, Area 7, and progressed north to Area 6, then Area 8 and Area 5. Area 9 on the eastern side of the site was the last area to be investigated.

The bulk removal of overburden to around RL 2m was completed by 20 ton excavators. A 7-ton machine then undertook more detailed bulk excavation work, removing the remaining modern fill and exposing the archaeological levels. This machine work was undertaken with archaeological supervision and direction. The small machine was used throughout the excavation to assist the archaeologists in the removal of the large dumps of fills that existed between archaeological phases, sub-surface modern disturbances, and to excavate large test trenches through reclamation fills. This work was always supervised by an experienced archaeologist.

The archaeological excavation methodology was a combination of ‘open area’ excavation and test trenching. The nature of the site as a landscape, formed by reclamation from the harbour and undergoing several phases of industrial, and also residential developments, made it necessary to employ both methods of archaeological excavation. In general, the occupation/industrial phases were excavated using open-area excavation methodologies, and the reclamation phases and natural landscape were investigated using a test-trenching methodology. Open-area excavation

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 89 aims to expose, investigate and record archaeological features, fills and deposits in their entirety phase by phase. Test-trenching provides a detailed sample of archaeological features and fills, or phases and events, such as the reclamation events in the Darling Quarter site.

The archaeological remains were excavated and recorded within a site grid. The site grid was established from a baseline located in the southern area, Area 7. The baseline was roughly east- west aligned. The origin point was given an arbitrary value of 100m North 200m East. Grid points were set at intervals of 20m east and west, and 10m north and south. The grid was established and maintained by the site surveyor. Several datum points were also established throughout the excavation and all levels (RLs) are calculated to Australian Height Datum (AHD).

The 20m by 10m grid system facilitated the planning of the site at a scale of 1:50 using A3 sized sheets of permatrace. A series of ‘top-plans’ were produced for the site, 24 in total. The top-plans recorded multi-phase and multi-context archaeological remains at a ‘start level’, and a series of overlays were produced as features and fills were excavated. Test-trenches were located on the main top-plans and were also recorded in section. Section drawings were produced at a scale of 1:10 or 1:20.

The physical excavation and recording of the archaeological remains was undertaken by professional archaeologists. A small number of students (Honours year) were also employed to assist the archaeologists during the excavation and to gain field experience. Archaeological excavation requires a great deal of manual labour. The tools used to excavate were mattocks, shovels, spades, hoes and of course, hand-held trowels. A small mechanical excavator was utilised as much as possible to expedite the excavation of fill layers and large test-trenches.

Archaeological structures, features and deposits were, for the most part, assigned a context number and were recorded on a context sheet that detailed general and specific context characteristics such as colour, soil matrix, stratigraphic and physical location, dimensions, building methods and materials, artefact quantity and type, and preliminary phase et cetera. Not all individual extensive fills or modern disturbances were given a context number. In some cases, one number was assigned to a group of fills or a general level. This practice was discretionary and was generally employed where bulk removal of fill layers between archaeological phases was carried out by machine, or where a group of deposits with ill-defined boundaries existed, but were of the same general event, such as demolition material within a room displaying slightly different properties, but clearly relating to the same general event.

Along with the drawn and written record, the archaeological remains were also photographed. Most of the photography was digital, though main photographs were also taken using Black & White and Colour Slide film for archive purposes. For consistency, most of the photography was undertaken by the Site Director, however due to the scale of the site, this was not always possible. A professional photographer was also present to take final Area and major structure/feature photographs. In general, most investigated features were photographed. General area shots were also taken to provide context to the features. Videography was also carried out during the excavation.

Artefacts were collected according to context number and processed on site. There was not a 100 per cent collection of artefacts from either machine, or hand excavated fills and deposits, as this is unnecessary and unrealistic for historic sites of this type. For the most part, diagnostic items and a representative range of artefact category and type were kept. Items discarded were noted on the context sheet or a discard sheet. The processing of the artefacts on site included washing, drying, sorting by category and type, labelling and boxing.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 90

A special methodology for the excavation of underfloor occupation deposits was employed. The system of excavation of underfloor spaces is linked with the artefact database designed by Casey & Lowe. In rooms where underfloor deposits were present, a 500mm by 500mm grid was set up and excavated by context and in spits of 50mm. The material was 100 per cent wet sieved to guarantee the collection of the small artefacts and ecofacts expected in this deposit type (such as pins, buttons, beads, seeds, small mammal and fish bones). This system also allows for spatial and comparative analysis of the artefacts to be undertaken.

The sampling strategy for the excavation included the collection of environmental, timber, industrial waste material and building materials samples. Natural deposits were sampled to provide information on the soil matrix and properties, as well for as pollen analysis. Significant occupation deposits, historical accumulation layers and fills were also sampled, for both pollen and soil analysis. These samples provide additional environmental information, such as landscape and vegetation, to the archaeological results. A number of large timber-built structures were present on site. The timber elements of the structures were sampled to identify the wood species. Building materials that were sampled included bricks and mortar from structural remains, construction and demolition deposits. Other samples collected from the site include roofing slate, 19th-century ceramic service pipes, slag and other industrial waste material.

3.1.5 Excavation Team A large team of archaeologists were involved in the excavation at the Darling Quarter site. The excavation was co-directed by Dr Mary Casey and Abi Cryerhall, and they were assisted by the trench supervisors: Amanda Dusting, Mike Hincks and Nick Harrop. The excavation team consisted of: Co-Excavation Director: Dr Mary Casey Co-Excavation Director: Abi Cryerhall Trench Supervisors: Amanda Dusting, Mike Hincks and Nick Harrop Site Planners: Franz Reidel and Shane Willis Artefacts Supervisor: Rowan Ward Archive Supervisor: Jill Miskella Sieving Supervisor: Sue Hearne Archaeologists: Glen Suey, Ronan Mc Eleney, Jenny Winnett, Rhian Jones, Robert Maxwell, Beau Spry, Dave Marcus, Lyndon Patterson, Kylie McDonald, Stirling Smith, Darran Jordan, Tom Eley, Balazs Hansel, James McGuinness, Sarah Peisley, Katie Molyneaux, Conn Herriott, Anna Biggs, Bernadette McCall, Tessa Boer-Mah, Jenna Weston, Ngaire Richards and Garth Henderson. Student Archaeologists: Peter Howard, Richard Burchfield, Ryan Desic, Annika Korsgaard , Vanessa Alexander and Daniel Blick. Comber Consultants undertook the Aboriginal testing and excavation component of the project. Jillian Comber directed and Tracy Appel supervised the Aboriginal archaeological program. Representatives from the Metropolitan Local Aboriginal Land Council also participated. Russell Workman, professional photographer, and Dr Wayne Johnson, archaeologist for SHFA, were also regular visitors to the excavation. Bovis Lend Lease facilitated the excavation works. Mechanical assistance was provided by Cardinal in the early stages and by Theos Bros for the majority of the excavation. Joseph Consentino from C.M.S Surveyors was the surveyor.

3.1.6 Excavation Limitations As Darling Quarter was a former industrial site, there were several excavation limitations due to environment, health and safety (EH&S) concerns. Known areas of contamination were not investigated archaeologically, such as the large area of Benzene contamination on the eastern side of Area 7 (Figure 3.1.3). Unexpected EH&S concerns, such as large diesel storage tanks,

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 91 hydrocarbon-rich fill layers and asbestos piping were encountered during the excavation. These required remediation and removal by appropriately trained and certified people, and within the parameters of the developer’s EH&S policy. In some case, this had some impact on the surrounding archaeological remains.

Figure 3.1.3: The pink area highlights roughly the location of PASS. In the southeast corner the green area approximates the location of high levels of benzene contamination. Aecom with C&L additions.

Several areas of Potential Acid Sulphate Soils (PASS) were also identified before the excavation commenced (Figure 3.1.3). As this material has potential to cause environmental damage, these areas could not be excavated during the archaeological program. PASS occurs in areas where organic-rich soils are trapped in an anaerobic environment, and once exposed to the air, can release sulphuric acid and turn highly acidic. This type of material is common in wetland and foreshore areas. Unfortunately the presence of PASS in Area 6 and Area 9 hindered the archaeological investigation of two of the early structures, Barker’s jetty and mill pond dating from the 1820s. The archaeological investigation and recording of these two features was incomplete, though as much as possible was done within these limitations.

3.1.7 Report Methodology for Section 3 This section (Section 3) presents a synthesis of results from the archaeological excavation. It provides a comprehensive description and discussion of the main archaeological features and findings. Information from the detailed trench reports, field reports and matrices (Vol 2, Section 7; Vol 4, Section 11), detailed site plans (Vol 4, Section 10), historical research and cartographic sources (Vol 1, Section 2), interpretive plans and graphics (Vol 4, Section 9), artefact specialists reports and environmental samples analysis (Vol 3, Section 8), and the artefact catalogue (Vol 6) have been utilised to write this section.1

1 The final miscellaneous report (Vol 3, Section 8.2) was not available for most of the analysis and writing of this section. Importantly, a draft version with analysis of the artefacts from the Area 8 underfloors and cesspits was available for the writing of Section 3.7 – the synthesis of results from the Area 8 workers’ housing. Further details and analysis of the small finds that are mentioned and included in the main report (Section 3) can be found in Section 8.2. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 92

It must also be noted that there may be some conflicts between the analysis and interpretations presented in this section and the descriptions and phasing presented in the various trench reports, matrices, plans and specialist reports. Every effort has been made during the final report writing stage to address these, and footnotes have been included in the trench reports to indicate where a re-interpretation has occurred. Not all archaeological features detailed in the trench reports, or findings from the artefact and environmental samples have been presented in this section.

The excavation and recording of the site was undertaken within defined areas (Section 3.1.2) and within an overall framework of historical phases (Section 3.1.3). These area divisions and overall site phases determined the structure and framework for all the technical reports, specialist analysis, data and plan production. However, the synthesis of results (Section 3) has not been presented within these parameters. The site has been considered as a whole, especially up to the 1850s, and the archaeology of its transformation, development and use has been presented more thematically.

The research questions formulated for the archaeological investigation prior to the excavation have also been considered in the structure and presentation of the archaeology in this section (see Vol 1, Section 1.4). The research questions; concerning residential housing and material culture, industrial and landscape archaeology, have been addressed throughout Section 3. A separate section for the response and discussion of these is also presented in Vol 1, Section 6 of this report.

It also must be noted that this section focuses on the archaeology of the site up to about 1880, with some comment on activity up to resumption in 1900. A history of the site’s development up to the resumptions and through the 20th century can be found in the historical background of this report (Section 2).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 93

3.2 The Natural Environment This section discusses the natural environment of the site and the surrounding area. Major environmental changes occurred quite swiftly after the establishment of the British colony at Sydney Cove in 1788. Land was cleared of trees, local resources harvested, buildings were constructed and streets laid-out. However, there was little detectable change in the Darling Quarter site’s natural environment until the 1820s when development along the foreshore began. This section discusses the evidence and considers the natural environment of the site up to the 1820s (beginning of Phase 3). Archaeological evidence of the natural environment was recorded in all Areas.

3.2.1 Introduction In recreating the pre-1788 landscape, underlying geology and soils can suggest the plant communities and fauna that once populated Sydney Harbour. Crucial works such as Benson & Howell’s Taken for Granted2 and more recently the comprehensive study of the ecology of Pyrmont by Broadbent3 have informed our view of Sydney’s and Darling Harbour’s natural landscape. Historical records such as early accounts, paintings, etchings and surveys can also add to the description of the topography and natural environment at the advent of British settlement. Of course these early sources are not always accurate, but combined with modern scientific analysis, a model for a broad view of Darling Harbour prior to British settlement can be created.

Archaeological evidence of the prehistoric physical landscape and topography, soils, pollens and organic matter, such as wood and shell, can test and further refine this model. The Darling Quarter excavation uncovered the physical remains of the original eastern foreshore, including the sandstone shoreline and intertidal sand flat, long since buried below reclamation fills and modern development. Environmental sampling allowed for analysis of the fossil pollens and soil properties. Samples of wood and marine shell from contexts predating the first development in the 1820s also provided further evidence and insight into the natural environment, of both the site and the broader Darling Harbour area.

Combining the geological, historical and archaeological evidences, the natural foreshore of the site can be reconstructed. This is key to understanding why, and how this site was initially developed to include a substantial jetty and mill pond servicing Barker’s flour mill in the 1820s. The nature of the foreshore also facilitated successful large-scale reclamation in the 1830s and 1840s that created both valuable harbour-fronting sites for industry and land for housing development.

Using Darling Quarter as a case study, the reconstruction of the natural environment is also useful in demonstrating the environmental impacts of 19th-century development and industrialisation were. Environmental change is also considered part of the reason for the pattern of continuous redevelopment of the harbour waterfront, in particular the maritime infrastructure servicing the industries established on Darling Harbour. Increasing levels of pollution and shoaling around the initial wharfage and jetties led to a decrease in water depths and suitability for shipping. This was corrected by either wharf and jetty extension or periodic dredging. Subsidence and erosion of the made-land led to almost continuous levelling and ground consolidation throughout the 19th century.

2 Benson & Howell 1990. 3 Broadbent 2010. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 94

3.2.2 Physical Geography and Plant Communities

3.2.2.1 Geology Sydney Harbour is within a broader Sydney region geological landscape that began taking shape 300 million years ago when it lay at the mouth of a swampy river basin (Figure 3.2.1). During the Triassic Period (230 million years ago) successive layers of sediment from river-eroded mountains filled the Sydney Basin. The sandy sediments were cemented into sandstone and the finer silts and clays into mudstone and shale. The lowest levels of sediments became what we know today as the Narrabeen Group of sandstone and shales, the middle layers as Hawkesbury Sandstone, and the upper layers as the Wianamatta Shale. As the continental plates separated rivers cut deep gorges through the sandstone throughout the Tertiary Period. In the Cumberland Plain, the predecessor of the Napean-Hawkesbury River deposited silts, sand, clays and gravels.

More recently in geological time during the Quanternary Period, alluvium continued to be deposited in the flood plains of all the main rivers and streams. Sea levels dropped some 120m from present day levels about 20,000 years ago during the last of the Pleicstocene ice ages. When levels rose again, to present day levels, some 6,000 years ago, river valleys from Broken Bay to Port Hacking, including Sydney Harbour were inundated.4 This geological landscape of Darling Harbour has been illustrated by Broadbent (Figure 3.2.2). Hawesbury Sandstone dominates either side of the harbour, with Quanternary river delta deposits at the head and swathes of Ashfield Shale either side.

Figure 3.2.1: Sydney Harbour is located on the eastern edge of the roughly disc-shaped Sydney Basin, characterised by Hawkesbury Sandstone and Wianamatta Shale. Benson & Howell 1990: 8.

4 Summarised from Benson & Howell 1990: 7-8. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 95

Figure 3.2.2: Interpretation of the geological landscape by Broadbent showing the site (red circle). The natural environment of the immediate area would have been influenced by the underlying river delta deposits and the Hawkesbury Sandstone of the eastern foreshore. Broadbent 2010: 43.

3.2.2.2 Soils Despite the limited range of rock types, the Sydney region has quite a variety of soils.5 However there are four soil landscapes within the Darling Harbour area:6  Gymea – on Hawkesbury Sandstone  Lucas Heights – on plateau surfaces of the Mittagong Formation  Deep Creek fluvial deposits  Blacktown – on Wianamatta Group Shales.

Bordering the site is a Gymea soil landscape that is characterised by a topography of undulating low hills with slopes of between 10-25 per cent and isolated rock outcrops; soil depths would have varied depending upon relief and were mainly derived from shale deposits with additional siliceous clays and leached sands along the drainage lines. Of note is the fact that these soils are prone to sheet erosion when disturbed.7 Further to the east in the Cockle Bay Creek catchment area is a Lucas Heights soil landscape that is common on landforms of gently undulating slopes, crests, ridges and plateau surfaces. This soil landscape consists of silty sands in the upper layers and sandy clays at the base of the soil layers and is also prone to erosion.8 At the head of the harbour is a ‘Deep Creek’ soil landscape that is described as a level to gently undulating alluvial floodplain draining the Hawkesbury Sandstone.9 In Broadbent’s interpretation of the soil landscapes, sedimentation and deposits of alluvium at the head of the harbour that formed mudflats have not been included (Figure 3.2.3).

5 Benson & Howell 1990: 9. 6 Broadbent 2010: 44. 7 Broadbent 2010: 45, cf. Chapman and Murphy 1989. 8 Casey & Lowe 2011: 65. 9 Broadbent 2010: 45, cf. Chapman and Murphy 1989: 74. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 96

Figure 3.2.3: Based on Broadbent’s interpretation of the soil landscapes of Pyrmont, the Darling Quarter site is indicated by the red circle. This interpretation does not include natural and once on-going deposition of Quanternary alluvium at the head of the harbour. Broadbent 2010: 44.

3.2.2.3 Vegetation The differing soil landscapes in the vicinity of the site would have supported several types of plant communities (Figure 3.2.4). Vegetation on the eastern side of the harbour across to Sydney Cove would have consisted of ‘Sandstone Ridgetop Woodland’ and heath.10 On the ridges and upper slopes, low dry sclerophyll open-woodland would have dominated, with species commonly present including Red Bloodwood Eucalyptus gummifera, Scribbly Gum Eucalyptus haemastoma, Brown Stringybark Eucalyptus capitellata and Old Man Banksia Banksia serrata. More sheltered slopes would have commonly supported Black Ash Eucalyptus sieberi, Sydney Peppermint Eucalyptus piperita and Sydney Red Gum Angophora costata. The understorey would have consisted of shrubs from the families Ericaceae, Myrtaceae, Fabaceae and Proteaceae.11 Scrub such as She-oak Casuarina distyla would also have been characteristic.12

To the east of this, the Wianmatta shale soil landscape would have supported tall open-forest with species such as Blackbutt Eucalyptus pilularis and Angophora costata, with a shrubby understorey. Also near the site and associated with this soil landscape, was the eastern edge of the Turpentine- Ironbark forest.13 The estuarine wetlands at the head of the harbour would have been characterised by mangroves and saltmarsh.14 Species present would have been Grey Mangroves Avicannia marina, various saltmarsh species, rushes including the Common Reed Phragmites australis, and low open-forest including Swamp Oak Casuarina glauca.15 Surrounding the creeks

10 Broadbent 2010: 142. 11 Broadbent 2010: 45, cf. Chapman and Murphy 1989. 12 Broadbent 2010: 142. 13 Benson & Howell 1990: 44 14 Benson & Howell 1990: 27. 15 Broadbent 2010: 142. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 97 that flowed through present day Haymarket into Darling Harbour, coastal swamp-forest species such as Swamp Mahogany Eucalyptus robusta and Casuarina glauca were likely present.

Figure 3.2.4: Vegetation types and plant communities around the Darling Harbour area according to Broadbent. The site is indicated with the red circle. Broadbent 2010: 142.

Figure 3.2.5: Vegetation around Sydney and Darling Harbour according to Benson & Howell differs slightly and identifies the mudflats along the eastern foreshore, including the Darling Quarter site circled. Benson & Howell 1990: 42. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 98

There are slight variations in the two views presented by Broadbent and Benson & Howell of the vegetation types in the immediate vicinity of the site. Broadbent interprets the vegetation of the site and its immediate surroundings as being ‘Sandstone Ridgetop Woodland’ with species such as Red Bloodwood, Scribbly Gum, heath-leafed Banksia and She-Oak. Benson & Howell’s version differs slight as included are estuarine wetland or mudflat and swamp forest species; Swamp Mahogany and Swamp Oak (Figure 3.2.5). Added to the woodland species of the sandstone slopes to the east of the site, Benson & Howell include Blackbutt, Sydney Peppermint and Sydney Red Gum as likely species present.

3.2.2.4 Darling Harbour Shoreline Darling Harbour is one of the several sandstone-fringed bays within Sydney Harbour. The natural shoreline has been altered radically by extensive reclamation, wharf construction and quarrying throughout the 19th century. It once would have consisted of sandstone ledges, cliffs and rocks, with extensive sandy intertidal areas near the head and perhaps metered with small sandy beaches towards the mouth. An interpretation of Darling Harbour’s shoreline has been created by Broadbent for the Pyrmont ecology study (Figure 3.2.6)

Parramatta and Lane Cove Rivers, along with many creeks, flow into Sydney Harbour. Several creeks exited at the head of Darling Harbour, the main one being Cockle Bay Creek that once existed through modern-day Haymarket. The deposition of soil materials from eroded land surfaces into the bay resulted in an on-going natural process of sedimentation.16 This had created the large intertidal sandy mudflat at the head of Darling Harbour. The build up and consolidation of such fluvial deposits was ensured by the sheltered waters of the bay. The broad tidal zone at the head led to shallow waters, with an initially narrow deeper channel transecting the harbour. Further north towards the mouth and the rockier shoreline, the water deepened to a maximum of c.20m.17

Figure 3.2.6: Reconstruction of the Darling Harbour shoreline indicating the extent of silty, sandy or clayey shore heading north from the head before changing to a rocky or cliffy shoreline. The study area is indicated with the red circle. Broadbent 2010: 57.

16 McLoughlan 2000: 184. 17 McLoughlan 2000: 185, cf. Roy 1981: 75. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 99

3.2.3 Historical Accounts18 Darling Harbour was initially named ‘Long Bay’ by the First Fleeters in 1788, reflecting its physical appearance.19 It soon became known as Cockle Bay, until 1826 when it was officially changed to Darling Harbour. Cockle Bay was no doubt a reference to the abundant natural resource of cockles living in the mud and sand flats around the head of the harbour. These resources were exploited by the Aboriginal people as a food source prior to British settlement. The shell middens and natural shell beds within the intertidal zone were exploited by the early colonists as a source of lime for mortar. In 1818, the public were warned not to remove shells from the middens at Cockle Bay as they were reserved for the government’s use.20

Historical documents are an important source of information on the natural environment of Sydney Harbour. Much of the early evidence consists of observations made by lay people and officers, rather than expert scientists or botanists.21 Several early maps and illustrations of the colony also feature or include depictions of the natural environment. It is important to note however, that many of these were created as romanticized and stylized impressions, rather than as accurate representations of the landscape.22 Furthermore, much of this early evidence does not specifically refer to Cockle Bay, but instead provides an idea of the area surrounding the early settlement of Sydney. For example, in 1789 Captain Watkin Tench observed Port Jackson as being covered with trees growing up to the edge of the water, and later goes on to describe: The general face of the country is certainly pleasing, being diversified with gentle ascents, and little winding vallies, covered for the most part with large spreading trees, which afford a succession of leaves in all seasons. In those places where trees are scarce, a variety of flowering shrubs abound, most of them entirely new to an European, and surpassing in beauty, fragrance and number, all I ever saw in an uncultivated state.23

A scene similar to the one described by Tench is represented in the painting of Port Jackson in 1804 (Figure 3.2.7). The hills, valleys, trees and shrubs described by Tench are spread across the landscape and give a general impression of the natural environment at the time of British settlement. Similarly, in Figure 3.2.8 of Ultimo House c.1813 below, a tall eucalypt forest is visible in the background,24 and in an 1812/14 engraving (Figure 3.2.9), it can be seen that while much of the vegetation had been cleared by this time, a line of tall rainforest trees remain.25

Another indication of the natural environment of Cockle Bay is provided by Governor Phillip, who states that ‘the necks of land that form the coves are mostly covered with timber, yet so rocky that it is not easy to comprehend how the trees could have found sufficient nourishment to bring them to so considerable a magnitude…’.26 Once again, this description is not specific to a particular location, but provides a general impression of the environment. The view across Cockle Bay in Figure 3.2.10 demonstrates the rocky nature of the land, depicting the rocky projection south of Pyrmont Bay.27 Similarly in Figure 3.2.9, Millers Point is depicted as a rocky outcrop in the left middle-ground.28

18 Section 3.2.3 has been written by Sandra Kuiters, Archaeologist, Casey & Lowe. 19 Johnson & Parris 2008: 11 20 Johnson & Parris 2008: 16. 21 Benson & Howell 1995: 29; Aplin 1988: 19. 22 Aplin 1988: 19. 23 Tench 1789: 38, 64 in Broadbent 2010: 125. 24 Broadbent 2010: 19. 25 Broadbent 2010: 26. 26 Phillip 1790: 74 in Broadbent 2010: 125. 27 Broadbent 2010: 21. 28 Broadbent 2010: 26. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 100

Figure 3.2.7: William Westall, Port Jackson 1804 (ML V1/1804/1). Image from McCormick, p. 92.

Figure 3.2.8: West, A., 1813/14, Views in New South Wales, Plate 14: Ultimo House (ML PX*D 65)(digital order no. a1474022) SLNSW.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 101

Figure 3.2.9: West, Absalom, 1812/14, Views in New South Wales, Plate 8: A native camp near Cockle Bay, Sydney, A. West (ML PX*D65; digital order no. a1474013). SLNSW.

Figure 3.2.10: Detail from Taylor, James, 1820, ‘Sydney looking south from Flagstaff Hill, ca 1820” (ML 69), SLNSW.

One example of a historical account specific to the natural environment of the Cockle Bay area was made by picnickers on Macarthur’s estate in Cockle Bay in 1806. The article states that the guests admired the ‘picturesque beauties’ of the ‘romantic scene’ from underneath the shelter of a fig tree, and that nearby they found a ‘pure and uncontaminated spring’.29 Though not greatly detailed, this account compliments the broader view of the written and visual sources. Broadbent combines the evidence of early maps and illustrations with written historical accounts and contemporary studies of the physical geography of the area to conclude that mangroves were likely to have grown along Cockle Bay’s the western foreshore, south of Pyrmont Bay, and across

29 Sydney Gazette 21 December 1806, p 2. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 102 the mouth of Cockle Creek.30 The reeds within Cockle Bay and low-lying mangroves in the southern inlet are faintly visible in Figure 3.2.10.31 He also suggests the mudflats of Pyrmont Bay may have also been covered in mangroves, however he has found no historical evidence for this.32

3.2.4 Cartographic Analysis Historic surveys and plans also provide useful information regarding natural environment in the early years of settlement. Analysis of these sources can also assist in tracing development and identify other possible changes and alterations to the physical landscape. The cartographic sources can be divided into land-based surveys and nautical charts; both offering different environmental information about the Darling Harbour area. It must be noted that often these surveys are inaccurate and present stylised information.

A nautical chart from 1788 depicts the shoreline of Darling Harbour and the depths of the waters (Figure 3.2.11). At the head the water is recorded as being very shallow, at 1 to 2 feet (300mm to 600mm). The depths increase further north. This very early survey may not have been accurate in the detail, but does indicate that the head of Darling Harbour was extremely shallow and implies a broad intertidal zone. Meehan’s 1807 survey confirms this as he annotates the head of the harbour as being ‘all a sand shoal at low water’ along the eastern side, and ‘sand and mud shoal dry at low water’ on the west (Figure 3.2.12).

The land surveys from 1788 and the 1790s detail the embryonic settlement at Sydney Cove and it is not until Lesueur’s ‘Plan of the Town of Sydney’ dated to 1802 that the head and eastern foreshore of Darling Harbour is represented, alluding to its topography and surrounding development (Figure 3.2.13). While this plan is not an accurate survey it does portray the area around the site as having cliffs or gradients along the shore and hints at the intertidal zone with shading. Creeks meander through the head of the harbour. Inland there is some vegetation depicted. A version of this plan published in 1873 annotates the land as ‘open forest land or bush’.33 It is likely though that the land in this area was either substantially deforested or in the process of becoming so by the turn of the 19th century. Development at this time is still concentrated around Sydney Cove with a second settlement area, ‘the Brickfield’s’, spanning Cockle Bay Creek to the east of the harbour head.

The 1822 plan of Sydney records the growing town and the eastern shoreline of Darling Harbour more accurately (Figure 3.2.14). The town’s grid pattern of streets is presented as a two- dimensional survey. This plan is not an accurate survey but does depict the extent of urban development and provide some information regarding the extent of the intertidal range and harbour depths near the Darling Quarter site. Some topographical details are still depicted, such as hills east of the Brickfields and the cliffy shoreline around Farm and Palmer’s Coves. On Darling Harbour’s eastern shore only the steeper gradients at the end of Market Street and the cliffs at the northern end near present-day Barangaroo are represented. The creeks at the head of the harbour are still depicted with numerous tributaries. This plan also indicates that land south of the town limits still had some vegetation with stylised trees or scrub scattered around the Ultimo Estate and to the east in modern-day Surry Hills area. This plan also includes information about Darling Harbour, providing soundings in fathoms and illustrating substantial intertidal areas. The head of the harbour is annotated as being ‘dry at low tide’. The Darling Quarter site is depicted as including both shore and intertidal area with harbour water of 1½ to 2¼ fathoms at low tide.

30 Broadbent 2010: 129. 31 Broadbent 2010: 21. 32 Broadbent 2010: 129. 33 Map of Sydney showing in contrast the town of 1802 with the city of 1873: supplement to the Sydney Mail 1873 MAP RM 1274 (copy 1). ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 103

Figure 3.2.11: Detail from 1788 nautical chart of Port Jackson showing the extent of the shallow area at the head of Cockle Bay/Darling Harbour. The approximate location of the site is circled in red, though it must be noted that this is not an accurate survey. Raper, G., 1788, Chart of Port Jackson New South Wales: as survey’d by Capt.n John Hunter, ML M2 811.15/1788/1.

Figure 3.2.12: Detail from Meehan’s 1807 plan of Sydney showing the head of Cockle Bay/Darling Harbour annotated as sands shoals. The site location is indicated by the red circle. NLA, http://nla.gov.au/nla.map-f105b.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 104

Figure 3.2.13: Topographical detail of the land between Sydney Cove and Darling Harbour, such as the shoreline gradient, are depicted on Lesueur’s 1802 plan of Sydney. The creeks entering the head of the harbour through the ‘Brickfields’ and vegetation are also represented on this early plan. http://nla.gov.au/nla.map-f307.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 105

Figure 3.2.14: Plan of the town and suburbs of Sydney 1822 with the approximate location of Darling Quarter circled in red. The inset shows more detail of the site and the intertidal area and harbour soundings. This plan does not provide and accurate survey but does indicate the extent of urban development and an idea of the intertidal range and water depths near the site. NLA .

William Harper produced a much more accurate survey of Sydney in 1822 and a relatively detailed depiction of the shoreline within the site boundary (Figure 3.2.15). It would appear at this stage that the shore within the site had not been developed to provide wharfage , though to the north and south such developments had begun. The site appears to contain two rocky projections, one to the north and one in the middle, with more regular or uniform shoreline. The intertidal zone is also depicted by shading, darker and more extensive in the southern half of the site.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 106

Figure 3.2.15: Detail from William Harper’s survey dating to 1822 showing the undeveloped shoreline through the Darling Quarter site (red outline). The east-west positioning of the site boundary is inaccurate and is based on the sketch of the planned junctions of Sussex, Bathurst and Liverpool Streets. SRNSW SZ435 with C&L additions.

3.2.5 Archaeological Evidence The excavation provided an opportunity to expose sections of the original shoreline, associated intertidal zone, and to collect environmental samples for analysis. Archaeological remains from Areas 5, 6, 7 and 9 contributed information about the natural landscape and environment (Figure 3.2.16, Table 3.1). Area 6 provided the greatest opportunity to investigate the nature of the original shore. In both Area 6 and 7, the intertidal zone was exposed in plan; though extensive excavation or test-trenching was not possible due to the presence of PASS (see Section 3.1.6). Analysis of the soil properties from a core sample taken in Area 6 confirmed wetland characteristics in this area prior to development.34 Pollen samples were taken from both Area 5 and Area 9, providing information about vegetation in the surrounding area and the subsequent impact of British settlement and development.35 In Area 7, comparison of marine shells from a thick ancient shell deposit below the harbour sands and samples taken from foreshore deposits dated to the early years of British settlement also demonstrated impacts on the natural environment.36 In Area 5, the remains of an Aboriginal shell midden was located on the shore just above the mean high water

34 Lawrie 2011: 5, Vol 3, Section 8.7. 35 Macphail 2010: 6-7, Vol 3, Section 8.6. 36 Carter 2010: ii, Vol 3, Section 8.5. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 107 mark (MHWM). This feature was excavated and recorded by Comber Consultants, and a separate report has been prepared (see Section 8.9, Vol 3).37 Post-excavation analysis of soil samples confirmed this feature was prehistoric.38 Tree root remains from the midden site were identified by Roy Lawrie as Swamp oak (Casaurina glauca), a species typical of swamp forest environments.39

Figure 3.2.16: The site boundary (red outline), basement (purple outline) and archaeological areas (blue outline and numbers) with Harper’s 1822 plan. This overlay suggests that the shoreline crossed through Areas 9, 5 and 8 only. This overlay gives an approximate location only as these early surveys are inaccurate, especially in east-west positioning. SRNSW SZ435, C&L additions.

Above HWM Intertidal Soil Pollen Natural Shell Deposits Pre-1788 Wood Shoreline Zone Sample Sample (context #) Remains Area 5   Site 1 #216, #212   Area 6   Site 2 #37, #65 8384 (post-1788) Area 7  #11, #17 8082 (pre-1788). 7919,8058/59, 8073/74, 8078/79 (all likely post-1788) Area 9  #203, #211 Table 3.1: This table summarises the excavation data from the different areas and whether evidence for the shoreline or intertidal zone was recorded.

37 Comber Consultants 2012, Aboriginal Archaeological Excavation Report Vol 3, Section 8.9. 38 Lawrie 2011: 1-4, Vol 3, Section 8.7. 39 The presence of Swamp oak in the Aboriginal midden site confirms Benson & Howell’s reconstruction of the vegetation in the immediate vicinity of the site as being rocky shore fringed by swamp forest. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 108

3.2.5.1 Sea Levels and the Tidal Range Understanding the tidal range of Darling Harbour is important for discussing the archaeological evidence of the natural environment and subsequent impacts due to development of maritime infrastructure and reclamation. It is generally accepted that at the time of British arrival in 1788, sea levels around Sydney Harbour had remained relatively constant for 6000 years, although some data has suggested a positive oscillation to around 1.4m above present mean sea level as recent as 2000 years BP.40 Today, the harbour’s sea levels are measured relative to zero on the Fort Denison Tide Gauge (ZFDTG) and the mean high water mark (MHWM) is 1.48m above ZFDTG. The mean low water mark (MLWM) is at 0.415m above ZFDTG. The general tidal range for Darling Harbour therefore is from low tide at 0.415m to high tide at 1.48m, with fluctuations above and below these values.41 These values do not equate to the survey levels taken on site and presented throughout this report. Height values are relative to the Australian Height Datum (AHD) and the MHWM is equivalent to 0.555m AHD. Therefore the AHD calibrated average tidal range is from -0.51m at low tide to 0.555m at high tide, with the usual small fluctuations above and below. The intertidal zone is the land or area between these two height values. The height values relative to AHD are also referenced in this report as a reduced level or RL.

Fort Denison Tide Gauge Australian Height Datum (FDTG) (AHD) MHWM Mean High Water Mark 1.48m above Zero 0.555m

MLWM Mean Low Water Mark 0.415m above Zero -0.51m

3.2.5.2 The Shoreline The review of the historic and cartographic sources, and confirmed by geotechnical investigations prior to excavation, placed a large portion of the site as being below the high water mark. Overlays of site and historic plans indicated that the original shoreline would cross through the site in an southeast to northwest trajectory, through Areas 9, 5 and 8 (Figure 3.2.16). During the excavation, parts of the rocky shoreline were exposed in Areas 5 and 6, and located in test pits in Areas 8 and 9 (Figure 3.2.17). Its presence in Area 6 confirmed that Harper’s 1822 can be considered a reasonably accurate and reliable survey.

40 Sloss et al 2007. 41 The lowest astronomical tide is equal to zero on the Fort Denison Tide Gauge, or 0.415m less than MLWM. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 109

Figure 3.2.17: Graphic showing the location of the test trenches and open area excavations that revealed natural environment features and deposits such as the shoreline and sand flat. The basement is outlined in purple, the grey dashed lines are limits of excavation areas, and the black lines are sections of the original shoreline. The site is overlaid on the 1822 Harper plan. Cryerhall.

Area 6 Evidence In Area 6, water and weather-worn sandstone (contexts 8448 and 8474) formed an undulating and gradual or gently sloping foreshore (Figure 3.2.17, Figure 3.2.18). The height dropped about 1m over a 20m area of exposed bedrock; from 1.3m in the east to 0.25m further west (Vol 4 Plan 10.8). The bedrock met grey harbour-deposited sands at around 0.05m. Low southeast to northwest orientated ledges or steps were present further east and away from the shore. Erosion due to water pooling was most pronounced on the highest parts of the exposed sandstone at the shore edge, with circular and interconnected concave depressions of up to 250mm depth (Figure 3.2.19). Shallow and parallel undulations suggested the movements of tides that scarred the lowest parts of the sandstone with unevenly spaced horizontal bands (Figure 3.2.20). A test trench (TT27) through the sands against the bedrock showed that the same pattern of weathering on the sandstone beneath the sandy harbour deposits (Figure 3.2.19).

Further inland from Area 6, bedrock in Area 9 was exposed both in plan and located at the base of several test trenches (TTs 68, 72 and 74; Vol 4: Plan 10.38). Along the western limit of the area the sandstone was relatively flat, with a surface between 1.25 and 1.45m. To the south and west were low ledges that dropped sharply to about 0.7m. Further east again the north-south aligned series of three test trenches over 15m confirmed the bedrock to again be flattish in nature, with a height range of 0.95m to 1.1m. The most southerly of the test pits (TT68) demonstrated the edge of the ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 110 shoreline had become closer to being east-west aligned at this point. This detail was first depicted in the 1822 survey (Figure 3.2.16). Though not fully exposed during the excavation it would seem that this section of shore consisted of a sandstone ledge abutted by deep deposits of harbour sands. In TT68 the top of these sands was at 0.88m and were still present below 0m in TT59 to the southwest.42

0.25m

1.3m

Figure 3.2.18: View to the southwest across Area 6 of the weathered sandstone shore above the MHWM. The bedrock gently sloped from 1.3m to 0.25m where it met coarse grey water-deposited harbour sands (arrows). Scale 1m.

42 As demonstrated by the core soil sample taken by Roy Lawrie during the excavation. See Lawrie 2011, Vol 3, Section 8.7. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 111

Figure 3.2.19: Erosion from water pooling was evident along the shoreline through Area 6. A small test trench (TT27) was excavated through the sand against this rock and revealed similar erosion below. Scale 1m.

Figure 3.2.20: View of the shoreline in Area 6 looking northeast and showing erosion patterns that resulted in smooth surfaces with horizontal banding (yellow arrow). In the background low ledges in the bedrock can be seen (red arrow).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 112

Area 5 Evidence Through Area 5, the shoreline continued in its northwest trajectory and the foreshore as undulating and gently sloping (Figure 3.2.21). The height of the bedrock ranged from 1m at the shore edge to 2m over an area of 10m east-west by 20m north-south (Figure 3.2.22). At the boundary with the harbour sand deposits the weathered bedrock contained some criss-crossing fissures filled with organic-rich harbour sands (Figure 3.2.23). The remains of an Aboriginal shell midden was located over this section of foreshore (Section 3.3). Similar to Area 6, erosion due to water pooling was also evident on the surface. Several test trenches (TTs 52-54) east of the harbour front located the flattish bedrock at around 1.6m (Vol 4: Plan 10.2).

Area 8 Evidence In Area 8, below the remains of the 1840s housing development, bedrock was found in several locations (Vol 4: Plans 10.29, 10.30). In general it was below the MHWM at between RL 0.3m and 0.5m, as was the case at the bases of cesspit 8717, 8716 and 8662 and within TT48. In the northwest, within a 3m by 1m test trench (TT45), it was also located between RL -0.05m and 0.26m, sloping from northeast to southwest. Several water-eroded horizontal channels or ledges were present at the junction with the harbour sands at RL -0.1m (Figure 3.2.24). The absence of harbour sands above a large area of flat rock but below the MHWM suggests a slightly more dynamic movement of water in this part of the foreshore than to the south, in Areas 5, 6 and 7, where water-deposited sands formed sand flats over and against the rocky shore around the MHWM.

Figure 3.2.21: View of the eastern boundary of the site with Area 5 on the left. The shoreline is indicated by the yellow line. View to the southeast across Areas 5, 9, the eastern part of 6 and the very eastern corner of 7 (mid right of photo).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 113

Figure 3.2.22: Extract from the detailed site plan of Area 5 showing the fissured and pock-marked sandstone that formed the shoreline in this location. This was the location of the Aboriginal shell midden. Extract from Vol 4: Plan 10.2.

Figure 3.2.23: Details of the rocky shore in Area 5. On the left the arrow points to the organic-rich sand filled fissures and low ledges. On the right the arrow indicates the high tide water abutting this area. Left photo is taken looking southeast and on the right the photo is taken looking northwest.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 114

Figure 3.2.24: TT45 in the northwest of Area 8 exposed gradually sloping bedrock below the high water mark with eroded narrow channels and ledges at the junction with the harbour sands (arrow). The photo is taken looking north. Scale 1m.

3.2.5.3 Original Shoreline Reconstruction Overall the impression of the original shore through the site was of an exposed low gradually sloping platform of water-eroded and weathered sandstone bedrock. Above the high water mark, this surface was occasionally interrupted by rockier or more jagged elements, low ledges and areas of steeper declines towards the harbour edge. Pockets of water-deposited and wind-blown coarse grey sands had collected in crevices and below ledges. Below high tide level, where it was not covered by harbour sands, tidal action had smoothed the bedrock surface and created narrow channels and indented horizontal bands. Detailed and comprehensive investigations below the mean sea level, or roughly RL 0m were not undertaken, though it may be possible to produce a contour reconstruction of the sandstone bed from geotechnical borehole results. Below the harbour sands and through Area 8, the sandstone dropped-off forming a substantial and deeper ledge (Figure 3.2.25). Using information from the exposed sections of the sandstone shore and the height of rock in various test trenches across the site, and the 1823 survey, the original shoreline and its location crossing through the site could be reconstructed (Figure 3.2.26, Vol 4: Plan 9.1). This reconstruction is useful in demonstrating what the original foreshore looked like and how these natural attributes facilitated later reclamation and development of maritime infrastructure.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 115

Figure 3.2.25: During the basement bulk excavation the sandstone platform (dashed line and arrow) that existed just above and below the mean sea level was exposed across the north of Area 8 and Area 5. View to northeast.

Figure 3.2.26: Reconstruction of the original shoreline above high tide (orange) and rocky intertidal platforms (yellow) based on the excavation data. Overlay with the 1822 survey, the modern street plan, site and basement outlines. Extract from Cryerhall, Plan 9.1 (Vol 4). ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 116

3.2.5.4 The Intertidal Zone The surface of coarse grey sands that formed the expansive foreshore through Area 6 and 7 were part of an ever-evolving intertidal environment. Sedimentation is a natural process and accumulations of sands, silts and clays in the harbour during the Holocene have been calculated at 0.8mm per year using an average of 7m of deposits.43 Homogenous layers of sand recorded in test pits below the foreshore surface in both areas are likely to represent this process and perhaps hundreds, if not thousands of years of sedimentation. The first years of the British colony witnessed dramatic landscape transformation with extensive land clearance and subsequent erosion. This led to increased sedimentation in the harbour and there are numerous references from the early 1800s, to massive silting and shoaling around Sydney Cove, Darling Harbour and Woolloomooloo, and the subsequent negative effects on water supply and access to deep water.44 Overlaying the homogenous sand bodies, the thin layers of banded sands and silts with frequent organic detritus that presented as the surface of the foreshore in Areas 6 and 7, are unlikely to be pre-1788 in date. These well-sorted sandy layers represent the foreshore deposits prior to the extensive reclamation works that began in the area some 30 years or so after British settlement.

Figure 3.2.27: Graphic showing the basement outline (purple) with the key areas of excavation within the foreshore, including test trenches and open-areas. Cryerhall.

43 McLoughlin 2000: 186, cf. Roy 1981: 65. 44 McLoughlin 2000: 187 cites many references from the early 1800s to silting and various associated problems with the water supply and deep water access. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 117

Area 7 Evidence One deep test trench (TT11) was excavated in Trench A, Area 7 to sample and record the harbour- floor stratigraphy below the water table (Figure 3.2.27, Figure 3.2.28). Excavation limitations due to potential acid sulphate soils (PASS) restricted extensive testing of the harbour deposits (see Section 3.1.6). TT11 was located some distance from the shoreline and was excavated from the MLWM of RL -0.5m (Vol 4: Plans 10.18, 10.24). At a depth of RL -1.5m, or about 1m below the MLWM, the top of geological clay layers was encountered (Figure 3.2.29, Vol 4: Plan 10.24). These clays consisted of dense pink and white mottled clay with ironstone inclusions followed by a 200mm thick layer of similarly dense bright orange clay also with ironstone and degrading sandstone fragments (contexts 8084 and 8083). These layers were once terrestrial B-horizon clays prior to the last sea level rise during the Holocene.

Once submerged, the clays were sealed by accumulations of alluvial sands and silts. Over time, these become homogenised and devoid of visible organic material, except marine shell. In TT11 overlaying the clay layers was a 300mm thick deposit of coarse grey sand with a high frequency of degrading shell and shell fragments (context 8082, Figure 3.2.29). This layer was an ancient layer, and had accumulated perhaps several thousand years ago. The analysis of the environmental samples from this deposit are discussed below. The shell-rich layer was sealed by 700mm of homogenous mid-grey coarse sand (context 8081) with very few inclusions of shell or other material. This final layer of sand had been accumulating prior to 1788. The dense frequency of shell in the lower deposit is the result of the shell fragments filtering through the sands over time and settling in a layer above the impervious clay.

Ancient Shell Bed Analysis of the shell samples taken from the shell-rich layer 8082 confirmed that this deposit, and another similar shell-rich context, 8074 from TT9, were pre-1788 in date and the result of natural harbour depositional processes (Figure 3.2.30).45 The dominant species by weight was Mud oyster (Figure 3.2.32), a species observed in large numbers, along with Sydney cockle, in the estuaries around Sydney Harbour by Europeans in the late 18th century.46 However as this is a large species, the MNI quantities are more appropriate in determining the species range for the natural environment (Figure 3.2.32, Table 3.2). While only nine MNI of Mud oyster were identified, much larger numbers of Rock oyster, Club mud whelk, but particularly Sydney cockle (63 MNI) were recorded. The low numbers of oyster comparatively to cockle and ‘other’ could be explained by the distance of this sample from the rocky shoreline. Oyster habitats are the rocky shoreline while cockles, and the majority of the ‘other’ species range, dwell in intertidal and mudflat areas. The size range of Sydney cockle demonstrates a large proportion of valves measuring between 2-4cm and 4- 6cm.47 The large MNI of ‘other’ types comprises a total of 20 different shell types (see Carter 2010: Vol 3, Section 8.5). The small-sized (<2cm) species Caltholotia fragum and Nassarius burchardi, in addition to juvenile Cerithidae shells make up the majority of this assemblage. These species are common in the intertidal mud and sand flats of estuaries. It is the presence of such large quantities of small-sized shell types, and the large quantities of juvenile Sydney cockle shell, that confirm that contexts 8082 and 8074 are natural deposits and not the result of cultural selection for consumption (Figure 3.2.31).48

45 Carter 2010: 17, Vol 3, Section 8.5. 46 Carter 2010: 18 cf. Bradley 1969, Beaglehole 1955. 47 Carter 2010: 14, Vol 3, Section 8.5. 48 Carter 2010: 15-17, Vol 3, Section 8.5. See also Carter 2010: 22 - discusses the dominance of cockle shell (92%) from the Aboriginal shell midden in Area 5 to illustrate a cultural selection rather than a natural deposition. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 118

Figure 3.2.28: Plan of archaeological deposits and test trenches relating to the intertidal zone within Trench A, Area 7. Extract from Plan 10.18 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 119

Figure 3.2.29: South-facing section of TT11 excavated to RL -1.7m showing geological clay layers sealed by dense shell and sand layers that formed the harbour foreshore prior to 1788. The upper reaches of the final sand layer would have formed the foreshore prior to late 1830s reclamation. Above the sand in this photo is a sandstone and clay reclamation fill. Scale 1m.

Shell type Total weight (g) % weight Total MNI %MNI Mud oyster 1596.7 11.5 9 4.3 Sydney cockle 1018.9 21.8 63 9.8 Rock oyster 537.6 21.1 27 3.5 Club mud whelk 986.2 34.2 23 1.4 Other types 531.9 11.4 520 81.0 Totals 4671.3 100 642 100 Table 3.2: Shell types from the combined analysis of natural harbour contexts 8082 and 8074. Carter 2010: 15.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 120

Figure 3.2.30: Cockle and other shell fragments in context 8074, Area 7, below post-1788 sand deposits in TT9. Scale 500mm.

Figure 3.2.31: Unbiased sample from context 8082 displaying the range of shell types. In the centre on the bottom line are juvenile to adult Sydney cockle shell. The other small species are also common to estuarine mud and sand flat habitats. Top row: example of shell debris (#4100). Second row (left to right): Austrocochlea porcata (#4097), Velacumantus australis (#4091), Polinices sorditus (#4102), Bembicium auratum (#4096). Third row: Corbula tunicata (#4094), Saccostrea glomerata (#4095), unidentified bivalve (#4098), Tapes dorsatus (#4092), mussel (#4093). Fourth row: Pyrazus ebeninus (#4089), Nassirius burchardi (#4101), Bedeva hanleyi (#4088), Velacumantus australis (#4090). Bottom row: Calthalotia fragum (#4087), Anadara trapezia (#4086), family Cerithiidae (#8082). Scale 10cm. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 121

Figure 3.2.32: Mud oyster shell from context 8074. All are longer than 10cm. Scale 10cm.

Analysis of microflora from a sample (pollen sample #11) of context 8082 confirm the vegetation model for the foreshore of Darling Harbour prior to 1788. This sample was moderately sparse in microflora, not unusual given the context’s location below MLWM. Most common within the sample was casuarina, followed by low numbers of mangrove and eucalypt.49 The sample also contained an unidentified ‘cup’ micro-alga that had previously been found in estuarine sediments on the southwest coast of Western Australia along with trace numbers of banksia, broom spurge, native hops and broom heath, common to rocky shorelines above the MHWM.50

Two cockle shell samples from the shell bed (context 8082) were submitted for radiocarbon age determination.51 One of the samples yielded a result of 5012 ± 29 BP and the other 3119 ± 28 BP.52 These results confirm that shell within the deposit was ancient, suggesting that the layer had accumulated several thousand years ago. The 2000 or so year age difference of the two samples indicates that shell had been accumulating and forming this shell bed for some time.

Harbour Sand Deposits Pre and Post-1788 Above the final homogenous sand layer 8081 were many thinner, localised bands and deposits of sands and silts (Figure 3.2.33, Table 3.3). These formed the slightly undulating foreshore surface around the low tide mark roughly between RL -0.25. and -0.7m, just prior to the commencement of extensive reclamation in the late 1830s (Vol 4: Plans 10.18, 10.19, 10.23). Several context numbers were assigned to these deposits and they were all interpreted as the result of harbour depositions after British settlement. Mostly they consisted of coarse mid to dark grey sands with varying amounts of organic inclusions such as wood fragments, charcoal and shell fragments (Figure 3.2.34). Localised patches of dark grey to black fine silty clay (contexts 7920 and 8069) were present across Trench A (Figure 3.2.35). This type of material forms a mudflat environment when it accumulates and stabilises in greater depths. The patchiness of mudflat material in Trench A is due to the distance from the shoreline, head of the harbour, and the location below the MLWM. This area was therefore subject to more dynamic harbour water movement that did not allow for the accumulation of mudflat silts.

49 Macphail 2010: 29, Vol 3, Section 8.6. 50 Macphail 2010: 29, Vol 3, Section 8.6. 51 Analysis by Radiocarbon Dating Laboratory, University of Waikato. See also Vol 3, Section 8.10. 52 For a more detailed discussion see Vol 3, Section 8.10. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 122

TT1 TT6 TT7 TT8 TT9 TT10 TTA 7920 dark 8053 coarse 8059 coarse 8069 dark 7919 dark 8078 coarse 8058 coarse grey/black grey sand grey sand, grey/black grey coarse grey sand, grey sand, silty fine with shell grit infrequent silty fine clay sand, infrequent infrequent clay shell infrequent shell shell shell 7919 dark 8079 coarse 8073 coarse 8074? 8088 coarse 8053 light grey coarse grey sand, grey sand, grey sands grey sand sand, infrequent infrequent with high shell with shell infrequent shell shell frequency grit shell Table 3.3: List of intertidal sand deposits from hand-excavated test trenches in Trench A, Area 7.

Figure 3.2.33: Grey sand littered with wood and shell fragments, and patches of silty clay formed the foreshore surface at low tide in Trench A, Area 7. The foreshore was located almost 3m below modern ground level. View to southwest. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 123

Figure 3.2.34: Below the MLWM the harbour deposits consisted of coarse grey sands and shell-rich layers. These deposits had been accumulating prior to 1788 and continued to do so until reclamation in the late 1830s. Photo of TT9 (background) and TT10 within Trench A, Area 7. View to east. Scale 500mm.

Figure 3.2.35: Section of TT8 in Trench A, Area 7, showing the fine silty clay build-up (8069) at the MLWM just prior to 1840s reclamation. This material seals the homogenous grey sand that had accumulated, perhaps over centuries. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 124

The analysis of the shell samples from these deposits (contexts 7919, 8058, 8059, 8073, 8078, 8079) suggest a post-1788 date. The deposits overall had a different species composition to the pre- historic deposits, most notably in the dominance of Rock oyster (Figure 3.2.39, Figure 3.2.40). The increase in Rock oyster numbers and decrease in Mud oyster is one of the examples of environmental change as a result of British settlement.53 Sydney cockle was still present in considerable quantities, however the overall trend demonstrated by the post-1788 harbour deposits indicate a decline in the local Sydney cockle population.54 Pollen analysis from context 8053 also confirms a likely post-1788 date to the upper layers of sand. Along with microflora from natives such as casuarina, eucalypt, mangrove, broom spurge and raspwort was small number of cereal pollens.55 The cereal pollens were likely to have originated from either Dickson’s mill (1815) or Barker’s (mid 1820s). Also identified in this sample was Cloacasporites sydneyensis.56 This evidence of human sewerage is a clear indicator of a post-1788 date. It should be noted that it is possible, given the context’s location in the intertidal zone, that the cereal pollen and the human waste indicator had washed in from a later phase but prior to reclamation in the 1830s and 1840s.

Figure 3.2.36: This graph of the species distribution by weight of shell samples from the natural harbour shell deposits (8082 and 8074) shows the slight dominance of Mud oyster with the other estuarine species well-represented. Carter 2010: 14

Figure 3.2.37: This graph demonstrates the dominance of Rock oyster over lesser numbers of other species in the intertidal deposits (7919, 8058, 8059, 8073, 8078, 8079) dated between 1788 and extensive reclamation work in the 1840s. Carter 2010: 13.

53 Carter 2010, Vol 3, Section 8.5. 54 Carter 2010: 17, Vol 3, Section 8.5. 55 Macphail 2010: 28, Vol 3, Section 8.6. 56 Macphail 2010: 29, Vol 3, Section 8.5. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 125

Three cockle shell samples from harbour sands that formed the post-1788 surface material of the foreshore in Area 7 were submitted for radiocarbon age determination. These contexts were 7920, 8053 and 8078. The samples provided a wide range in ages; the shell from context 7920 yielded a date of 4192 ± 29 BP; the shell from 8053 yielded a date of 2710 ±26 BP; and context 8078 had a date of 549 ± 25 BP. The date range of the shells further demonstrates that the natural foreshore environment was dynamic, and the sand bodies and other deposits that formed its surface, contained a mix of ancient shell deposits with newer ones, and included native and exotic, or post- 1788, pollens.

Radiocarbon Age Determination of Area 7 Cockle Shell Analysis of shell from the naturally deposited sand bodies and shell bed in Area 7 provided information regarding the natural environment and changes that occurred following British settlement. To further illustrate this, several shell samples were submitted for carbon-14 dating to the Radiocarbon Dating Laboratory, University of Waikato. Two shell samples from the shell bed 8082 were submitted to confirm that this deposit was an ancient shell bed. Three samples from the upper levels of harbour sands (contexts 8053, 8078 and 7920) were also submitted. The contexts were part of the foreshore surface prior to reclamation in the 1830s. The shell ages from these contexts were intended to illustrate the nature of foreshore environments, and whether it was possible to distinguish ancient deposits from post-1788 ones.

Methodology 57 The species of shell dated from Darling Quarter was Sydney Cockle, Anadara trapezia. This species was selected as it was both a suspension feeder, which was recommended for carbon-14 dating,58 and also the second-most abundant shell species by weight in both natural harbour deposits and intertidal deposits in Area 7. Atomic Mass Spectroscopy (AMS) carbon-14 dating was selected in preference to radiometric carbon-14 dating methods, as the sample sizes required for AMS dating allowed for individual shells to be dated, rather than a combined sample.

The shell samples for dating were selected from those shells which had already been catalogued as part of the post-excavation analysis of Darling Quarter. For contexts 7920, 8053 and 8078, this shell had been hand-collected from the deposits during excavation, while the catalogued shell from context 8082 came from an unbiased sample. As part of the cataloguing process, these shells had been hand-washed in water, dried, and then stored in polyethylene snap-lock bags together with a paper shipping tag. Before dispatching, the samples were weighed and placed in new polyethylene snap-lock bags. They were double-bagged to separate the paper tags from the shell.

All samples were dated through the University of Waikato, Radiocarbon Laboratory, New Zealand. The shells were dispatched to the laboratory on 31 March 2012. The results were supplied to Casey & Lowe on 5 June 2012. These have been summarised in Table 3.4 and are discussed in detail in Vol 4, Section 8.10 of this report.

Results The two shell samples from the shell bed 8082 dated between 1060-820 calBC and 3510-3325 calBC (95.4% probability). These dates confirm the interpretation of this layer from the other archaeological data, that the shell bed was ancient, and that it contained shell several thousands of years old. The two individuals selected from the context were around 2000 years apart in age. This suggests that the shell bed had been forming for thousands of years. Perhaps the shell bed was still in the process of forming at the time of British settlement, as individual shells were being filtered, with the aid of tidal movements, through the sand body above into the shell layer.

57 The methodology section has been written by Nick Pitt. Vol 3, Section 8.10. 58 Waikato Radiocarbon Dating Laboratory n.d. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 126

There was also a wide age range from the shell samples submitted from the foreshore surface material; contexts 7920, 8053 and 8078. Stratigraphic relationships, pollen analysis and shell species analysis indicated that these contexts were likely deposited after 1788 and prior to reclamation in the 1830s.

The shell sample from context 8053 was dated to between 595-355 calBC (95.4% probability). This context was sand that accumulated on the western side of a paling fence, erected along the low tide level in the early 1800s. The pollen within the sand consisted of mostly natives, but also some exotics, in particular cereal pollens. One fragment of blue-edged pearlware, dated c.1780-c.1860, was also found within context 8053. The shell sample from context 7920 was dated to 2450-2200 calBC (95.4% probability). Context 7920 was a silt-rich deposit that was located above sand 8078, and was located around the low tide mark. It was sealed by 1830s reclamation fill. Within context 7920 were a small number of historic artefacts. This context was most definitely deposited in the early decades of the 1800s and yet contained ancient shell. Context 8078 was located below 7920 and was a sand deposit that was, like 8053, interpreted as being deposited in the early 1800s. The shell sample from context 8078 was dated to 1675-1895 calAD (95.2% probability). This was the youngest shell date obtained from the samples, with a age range that includes a post-1788 date. This may be evidence that not all the cockle shell found in the foreshore harbour sands were from ancient deposits, and that some do represent molluscs that were alive in the harbour in the last few hundred years.

The age of the shell samples from these contexts indicated that the contexts contained a mix of ancient shell and pollens, and also a post-1788 shell species range, exotic cereal pollens and historic artefacts. This demonstrates that the surface of the foreshore was in a constant process of formation, as the natural tidal influence moved, deposited, and mixed sands, shell and other materials.

13 14 Waikato Sample Context Cat # Sample δ C D C (‰) F14C (%)60 Result Calibrated Calibrated lab # # weight (‰)59 (BP)61 date (68.2% date (95.4% (g) prob.) 62 prob.) Wk- 500- 1-8053 8053 3148 3 1.3±0.2 -286.4±2.3 71.4±0.2 2710±26 595-355 calBC 33746 385 calBC Wk- 2410- 2450- 2-7920 7920 3162 4 1.3±0.2 -406.6±2.2 59.3±0.2 4192±29 33747 2265 calBC 2200 calBC 1675- Wk- 1700- 3-8078 8078 3042 3 1.0±0.2 -66.0±2.9 93.4±0.3 549±25 1895 calAD 33748 1815 calAD (95.2% prob) Wk- 1000- 1060- 4-8082 8082 4086 15 0.3±0.2 -321.8±2.4 67.8±0.2 3119±28 33749 880 calBC 820 calBC Wk- 3465- 3510- 5-8082 8082 4086 4 1.0±0.2 -464.1±2.0 53.6±0.2 5012±29 33750 3355 calBC 3325 calBC Table 3.4: Summary of AMS radiocarbon dating of shell from Area 7.

59 Note from University of Waikato, Radiocarbon Dating Laboratory reports, “The isotopic fractionation, δ13C, is expressed as ‰ wrt [with regards to] PDB [Peedee Formation].” Vol 3, Section 8.10. 60 Note from University of Waikato, Radiocarbon Dating Laboratory reports, “F14C is also known as Percentage Modern Carbon (pMC).” Vol 3, Section 8.10. 61 Note from University of Waikato, Radiocarbon Dating Laboratory reports, “Result is Conventional Age or Percent Modern Carbon (pMC) following Stuiver and Polach, 1977, Radiocarbon 19, 355-363. This is based on the Libby half-life of 5568 yr with correction for isotopic fractionation applied.” Vol 3, Section 8.10. 62 Calibrated dates from University of Waikato, Radiocarbon Dating Laboratory reports, using OxCal v4.1.7, (Bronk Ramsay 2010), using data from Reiner et al (2009), ∆R=4.25. Vol 3, Section 8.10. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 127

Area 6 Evidence Similar stratigraphy below the MLWM was present in Area 6, to the north and northwest of Trench A, Area 7. Detailed archaeological recording below the water table was not undertaken in this area, however monitoring the remediation of an area of diesel contamination provided an opportunity to identify and photograph the harbour deposits below the reclamation fills (Figure 3.2.38, Vol 4, Plan 10.7). Forming part of the harbour floor was dense orange clay, similar to clay 8083 recorded in TT11, Area 7 (see above). Overlaying this was an approximately 200mm thick deposit of shell and fragmented shell mixed with some grey sand, similar to deposit 8082 in TT11. During the mechanical excavation through this layer it appeared to be dominated by juvenile and adult cockle shell. Overlaying the shell layer was a 500 to 800mm deep homogenous deposit of coarse grey sand, similar to context 8081 in TT11. The upper levels of sand appeared more banded, with relatively thin layers of sand interdispersed with clay deposits and dark grey to black silty clay lenses. These sand and silty clay bands also appeared to be littered with woody debris. The upper level of sands containing wood fragments and silt certainly accumulated post-1788. An accurate height reading for the level of the foreshore in this part of Area 6 was not obtained but can be estimated at around RL -0.2m.

Figure 3.2.38: Interpretation of the naturally deposited material in Area 6 (west). The foreshore was at roughly RL -0.2m prior to Barker’s jetty development works in the late 1820s. Below this level were successive layers of silty clay, sands and clay, some deposited naturally after 1788. Photo taken looking west from close to the western limit of the basement excavation in Area 6. 5m Scale held by Amanda.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 128

A large portion of the foreshore was exposed in plan in the northeastern part of Area 6 (Figure 3.2.37, Figure 3.2.40). Like Trench A, Area 7, a similar difficulty in confidently identifying pre-1788 harbour deposits was also encountered. Major development of this part of the foreshore began in the 1820s with the construction of a substantial jetty involving reclamation. Therefore, the sands and silts recorded in this area formed the natural foreshore environment prior to this date. Several small hand-dug test trenches through the upper layers of harbour sand revealed similar homogenous medium-grained, mid-grey sands to those recorded in Area 7 (Figure 3.2.41, Figure 3.2.42). The top of these sand deposits (context 8447 in TT26 and TT27, context 8378 in TT20 and context 8461 in TT22) were located just above the MLWM at between RL -0.35m and -0.2m, displaying a gradual slope westwards to the harbour zone. It is likely that these contexts accumulated prior to 1788.

The upper level harbour deposits were more likely post-1788 in date and were a mix of slightly undulating sands, sticky silty clay patches and woody detritus.63 Abutting and partial overlaying the sandstone shoreline were deposits of mid-grey sands with woody debris and charcoal inclusions and pockets of sticky silty clay (contexts 8481 and 8443). These sands were located just above the MHWM at RL 0.6m at the highest point overlying the bedrock and gradually sloped westward into the intertidal zone (Figure 3.2.43). At the western limit of excavation, similar organic-rich sands, of various shades of grey, were located at around RL -0.6m, the MLWM (contexts 8428, 8484, 8376). The sandy foreshore therefore sloped gradually westwards into the harbour from the rocky shoreline at just above high water at RL 0.6m to -0.6m over a distance of 15m. As this range is essentially the full extent of the foreshore at low tide, this part of Area 6 is a good representation of the sand flats abutting the rocky shore prior to development. There was some evidence for organic-rich silt or mud build up, but nothing that would constitute a ‘mudflat’. This part of the shoreline jutted out into the harbour and would appear to have been exposed to more dynamic tidal movements. This has also been discussed for the foreshore further northwest from this point where sands did not accumulate in the intertidal zone (see Section 3.2.5.2 Area 8 Evidence).

Two samples from the intertidal zone in Area 6 were analysed for pollen remains. Context 8445 consisted of dark grey to black humic silt naturally deposited around a post-1788 timber ‘boat- ramp’ structure. The sample included casuarina, eucalypt, raspwort and mangrove, all representative of a natural environment. The small percentage of cereal pollen suggests an early 1800s date for the deposit.64 The similarly dated context 8483 also contained the same range of native species types, along with a slightly higher percentage of cereal pollen. Exotic weeds, such as dandelion and wire-weed, were also present in the sample, indicating the spread of non-native weeds on what was vacant land at the time.65

63 Cryerhall, Hincks, Harrop 2010: 21-24 details the complicated sequence of deposition of these post-1788 harbour deposits. Vol 2, Section 7.2, Area 6 Trench Report. 64 Macphail 2010: 21-22, Vol 3, Section 8.6. 65 Macphail 2010: 21, Vol 3, Section 8.6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 129

Figure 3.2.39: The foreshore in Area 6 abutted the rocky shoreline (yellow dashed line). Sand had accumulated above the high water mark but mostly the junction between the sand and the sandstone was at about 0.4m, just below the MHWM. Photo taken looking east towards the city showing the foreshore in relation to the modern street level.

Figure 3.2.40: The full extent of the foreshore was exposed within Area 6 excavation limits. This photo is taken looking west from the shoreline and the arrow points to the low tidal waters resting above grey sands littered liberally with organic detritus.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 130

Figure 3.2.41: Detail of TT27 showing homogenous mid grey sands 8447 below later organic build-up and a timber boat ramp. Scale 1m.

Figure 3.2.42: Detail of if TT20 at the western limit of Area 6 showing homogenous grey sands 8378 below 1820s jetty infill (bands of yellow and grey sands with tip lines). The water indicates the low tide level. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 131

Figure 3.2.43: Harbour sands 8481(yellow arrows) overlying the rocky shoreline 8474 in the eastern part of Area 6. These sands accumulated above the high water mark at around 0.6m and sloped gradually westwards towards the harbour zone. View to the southeast.

Area 5 Evidence Soil and pollen samples were taken from the sediments within and below the Aboriginal midden site but above the sandstone bedrock. A detailed discussion of the results of the analyses is presented in a separate report (Aboriginal Archaeological Excavation Report in Vol 3, Section 8.9). Three layers of pre-British settlement soil over sandstone bedrock were identified and described by Lawrie as yellow and grey mottled sandy clay C-horizon followed by pre-midden black loamy sand A-horizon, then followed by pre-British settlement topsoil consisting of black sand with brown streaks and many shell inclusions.66 The two soil samples that were analysed confirmed the pre- 1788 date for the midden site. This was indicated by the low nutrient content, suggesting that the midden site had not received deposits of animal manure or human waste common to Colonial- period soils.67 The lack of contamination from industrial activity was also noted in the lack of heavy metal concentrations in the samples.68

The microflora from the pre-midden soil horizon is wholly dominated by casuarina and is likely to represent the Casuarina glauca-Eucalyptus robusta swamp forest that was characteristic of the Darling Harbour foreshore environment before British settlement.69 Microflora in the midden soil layer is again dominated by casuarina, and also fungal spores and microbial nodules. Macphail suggests that casuarina colonised the midden after its abandonment and that the high fungal spore

66 Lawrie 2011: 1, Vol 3, Section 8.7. 67 Lawrie 2011: 2, Vol 3, Section 8.7. 68 Lawrie 2011: 3, Vol 3, Section 8.7. 69 Macphail 2010: 17, Vol 3, Section 8.6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 132 content reflects nutrient-rich waste percolating into the sand from the shell midden and therefore contemporary with the midden.70

Area 9 Evidence Only a small portion of the southwest corner of Area 9 was within the basement. The excavation demonstrated that the shoreline was in almost an east-west alignment through this area (Figure 3.2.26). This was abutted by coarse grey sands that formed the surface of the intertidal zone. Open-area excavation to expose the foreshore was not undertaken in Area 9. Instead test trenching (TT69 and TT59) and analysis of auger-cored soil and pollen samples provided information about the natural environment in this location (Figure 3.2.17).

TT69 was excavated in an east-west alignment and the top of the sandy intertidal zone located between RL 0m in the east and -0.3m a further 15m west (Vol 4: Plan 10.38, 10.40). Only the top of the gradually sloping mid grey coarse sand body was exposed (context 9272). The sand had infrequent inclusions of dark grey to black silt pockets and a small hand-dug test pit into the deposit confirmed that is was homogenous and devoid of organic inclusions. This material formed the foreshore surface in this area prior to extensive landscape modification associated with the 1820s jetty construction. The sand appeared to be subjected to the same tidal dynamics as Area 7 and Area 6 as there was little build-up of silts and organic matter prior to development.

Figure 3.2.44: TT69 was excavated in an east-west alignment and was 15m in length. The pre-development foreshore surface was represented by context 9272 and was located around the mean sea level (arrow). The material above this in the photo consisted of redeposited harbour sands, likely the result of a natural deposition process that formed a sand bank. View to southwest. Scale 1m.

70 Macphail 2010: 16-17, Vol 3, Section 8.6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 133

TT59 was located to the northeast of TT69 and a little closer to the shoreline (Vol 4: Plan 10.38; Figure 3.2.17). An auger core soil sample was taken from the base of a smaller test pit within TT59 (Table 3.5). At approximately RL 0.3m the top of natural deposits were located below 1820s fills (Figure 3.2.45). The various clay and sand layers gave some indication of how the natural soil profile was formed in this location.71 The lowest layer in the sample around the MLWM consisted of dark grey waterlogged clayey sand with bands of light grey sandy clay, located from RL -0.35m to -0.55m+. Over this was 350mm of bluish grey waterlogged light to medium dense and impermeable clay. These layers formed in the late Holocene and were permanently waterlogged. Lawrie suggests that the thin clay and sand layers at the low tide level formed as a result of individual rainfall or flood events, and probably in shallow non-tidal water.72 The bluish clay above he interprets as forming in a quiet and protected area, most likely in a shallow fresh water lagoon.73 The fresh water environment interpretation is based on the very low chloride concentration in the sample, that suggests the soil was waterlogged with fresh not salt water.74 Nutrient levels in both these layers were very low, typical of bushland soils that have never been cleared or cultivated.75

Figure 3.2.45: An augured soil sample was taken from natural deposits identified in TT59 in Area 9. The core was taken (yellow circle) from the top of natural at around 0.3m to a depth of 900mm below, or RL -0.6m. Scale 2m.

71 Lawrie 2011: 6, Vol 3, Section 8.7. 72 Lawrie 2011: 6, Vol 3, Section 8.7. 73 Lawrie 2011: 6, Vol 3, Section 8.7. 74 Lawrie 2011: 6, Vol 3, Section 8.7. 75 Lawrie 2011: 6, Vol 3, Section 8.7. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 134

Above the bluish clay were thinner layers of sands and clays that formed more recently due to the increased present of organic matter. At RL 0.15m was 150mm of dark grey clayey sand with plant debris. This was followed by greyish brown light clay with plant debris inclusions. At the top of the sample and forming the youngest layer of natural soil was 100mm of dark grey light clay with many fibrous roots resting on a layer of brown harbour sand. The layers of sand and clay with plant debris were formed as a result of rain and flood events. Lawrie suggests that this material signifies a quiet backswamp environment.76 The upper layers containing plant roots indicate the absence of tidal action, even though the elevation is below the MHWM. Some natural or man-made barrier along the shoreline may have been protecting this layer from tidal erosion.77 In the mid 1820s, a substantial jetty, involving reclamation, was built in this location. The construction process would have taken some time, and would have allowed for naturally deposited material to continue to accumulate, and plants to propagate in a tide and wave-free lagoon-type environment.

Lawrie’s interpretations of the environment from this soil sample as a one-time freshwater lagoon or backswamp in Area 9 is interesting as it differs from the intertidal foreshore environments of Areas 6 and 7. While detailed soil analysis was not undertaken for these other areas, the foreshore presented as being a sand flat subject to dynamic tidal action as there was little build-up of silts or inclusions of organic materials such as plant roots and debris. The foreshore in Area 9 was closer to, and may have been sheltered by the sandstone shoreline to the north. The projecting part of the shoreline surveyed on the 1823 plan and confirmed archaeologically in Area 6 may have influenced the harbour tidal movements and depositions to the southeast in Area 9, causing perhaps sandbanks that sheltered the area. Freshwater influences are likely due to creeks falling from the eastern sandstone ridge line and exiting into the harbour at this point. Though none are depicted in this location on early surveys, creeks or water channels are present on sketches and plans from the later 1820s.78 The construction of a mill pond to collect freshwater in the later 1820s in this location also strongly suggests the presence of at least one creek (discussed below).

A sample from a sandy natural deposit above the MHWM in Area 9 (context 9514) was analysed for fossil pollen content. Context 9514 was identified in TT70 and TT72 and was located between 0.8m and 1.1m, therefore above high tide levels. It consisted of blackish grey humic sand and rested on the sandstone bedrock. The microflora was dominated by casuarina, eucalypt and trilete fern spores, also present were trace numbers of mangrove and the exotic weed, dandelion.79 Macphail suggests that the presence of mangrove and dinoflagellates (marine plankton, also common in freshwater habitats) demonstrate that context 9514 was influenced by periodic saltwater inundations. The sample demonstrated that the natural environment, as expected, was dominated by Casuarina glauca-Eucalyptus robusta swamp forest and the sample was taken in a location that was close to the mouth of a creek that was vegetated by ferns as indicated by the fern spores.80 The date of the deposit is post-1788 due to the presence of the exotic weed but most likely prior to 1815 and the establishment of the first flour mill by Dickson just to the south of the site as there were no cereal pollens present.

76 Lawrie 2011: 6, Vol 3, Section 8.7. 77 Lawrie 2011: 6, Vol 3, Section 8.7. 78 See Section 3.2.5.5 for discussion of the evidence for creeks. 79 Macphail 2010: 38, Vol 3, Section 8.6. 80 Macphail 2010: 38, Vol 3, Section 8.6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 135

Depth cm Profile features Comment Height datum RL 1.02m at top of drain, 40cm High tide level approx. RL 0.3m, above a layer of mixed fill 25 cm thick; abrupt equivalent to 72cm below top of boundary to drain

0 (RL 0.3m) Dark grey light clay with many fibrous roots; rests Lower part of former topsoil, on a layer of brown beach sand 5cm thick; high with remnant of beach sand tide level at approx depth of 7cm underneath

10 Grey brown light clay containing preserved plant Lagoon or wetland floor; debris sampled 0-15cm (including topsoil from above but omitting beach sand layer)

15 Dark grey sandy clay to clayey sand with plant Sampled 15-30cm debris

30 Bluish grey waterlogged light medium clay, dense Sampled 30-45cm and impermeable; abrupt boundary to Potential acid sulphate layer

65 Dark grey clayey sand with bands of light grey Sampled 65-90cm sandy clay; waterlogged throughout potential acid sulphate layer

90 (RL -0.6m) Maximum depth of inspection Table 3.5: Description of soil layers from the auger core through natural deposits in TT59, Area 9. Lawrie 2011: 5.

3.2.5.5 Evidence for a Creek and a Sandbank The early surveys include the major creeks, and several smaller ones, entering the head of Darling Harbour. No other creeks are depicted on either side of the harbour. Broadbent suggests that the small bays and coves along the western side of the Pyrmont peninsula would have been fed by small creeks and uses his reconstructed topography to extrapolite the courses of nine likely watercourses.81 Though the harbour’s original eastern shoreline was not as indented by as many coves and bays as the Pyrmont side, within the Darling Quarter site a small bay or cove was created by the change in the shoreline’s trajectory towards the northwest through Areas 9, 6, 5 and 8. This area is just to the southwest of the base of a steeply graded decline (Bathurst Street) from the ridge (George Street). Based on this topography, a watercourse of some sort was very likely to have existed, either a permanent small creek or one that flowed during heavy rain, and entered the harbour in this location.

When this land was advertised for sale by John Smith in 1822, the sale notice said that the land included ‘a constant supply of excellent fresh water’.82 Later development in this area also lends support to there once being a natural creek mouth here. In the mid 1820s, Thomas Barker bought Cooper & Levey’s steam-powered flour mill. He invested in a development program that included the construction of a mill pond to supply freshwater for the mill’s steam engine. This mill pond was located in Areas 9 and 7, within the small cove or bay area and its construction included extensive earthworks and a timber retaining wall on the harbour-side. It was presumably located here in order to retain water flowing into the harbour from a watercourse. A sketch of Barker’s mill dated to 1828-1830 depicts what appears to be a watercourse entering the mill pond from the northeast

81 Broadbent 2010: 59. 82 Sydney Gazette 21 June 1822, p 2(2). ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 136

(Figure 3.2.46). Several mid 19th-century plans and sketches record that the mill pond was fed from the east by one or two watercourses. The 1833 survey shows a watercourse crossing below Kent and Sussex Street, and though presumably it entered Barker’s land, it was not recorded as flowing through his property (Figure 3.2.47). A detailed sketch of Barker’s mill yard dating to 1847 shows a channelised ‘drain’ in the same location as depicted on the 1828-1830 sketch, and another entering the pond from the east (Figure 3.2.48). Barker was collecting and storing freshwater from enhanced watercourses or drainage channels into the mill pond. However, it is not known whether these drains had evolved from natural creeks that existed prior to 1788. Changes in natural drainage patterns may have occurred after settlement due to increased erosion following land clearance and development, creating new ‘natural’ watercourses that were later enhanced, redirected and channelized as part of stormwater management in the general area.

Figure 3.2.46: Sketch of Barker’s mill dating from 1828-1830 showing the location of the mill pond (green arrow) and a watercourse entering it from the northeast (purple arrow). A Hallen: Surveyor’s Field Book 347 [c.1828-1830]. Reel 2628 (SRNSW).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 137

Figure 3.2.47: Two sections of the 1833 survey showing the watercourse crossing below Kent and Sussex Street just to the northeast of Barker’s mill and the site. Extract from City Section Survey Plans, 1833 (with 1880s additions in red) Section 10 and 11, City of Sydney Archives.

Figure 3.2.48: Sketch of Barker’s mill dating to 1847 showing two ‘drains’ entering the mill pond from the east and northeast. Sydney University Archives, 1847. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 138

The environmental samples also indicate that there was influences from a natural source of freshwater in this part of the site. Lawrie interprets the soil found below the mill pond in Area 9 between approximately RL -0.35m and 0m as being formed during the late Holocene in a quiet, protected location behind a beach, and was most likely part of a shallow freshwater lagoon capable of trapping muddy sediments.83 The very low chloride concentration in this soil layer adds to Lawrie’s interpretation that the soil layer was waterlogged with freshwater not saline. A creek flowing into this part of the harbour would have been the source for this freshwater.

To the west and southwest in Area 6 and 7, there was little convincing evidence for a natural barrier, such as a beach, sandbank or levee, that would have helped form and protect such a freshwater lagoon. Perhaps the churned sand and silt deposits above the harbour floor sands revealed in TT69 in Area 6 can be interpreted as evidence of a naturally formed sandbank in the intertidal sand flat (see Figure 3.2.27 for location of TT69). The north-facing section of this test trench illustrated that the churned sands and silt (contexts 9270 and 9271) sloped from west to east, back toward the shoreline (Figure 3.2.49). Perhaps this represented a cross-section through a naturally formed low sandbank that acted to retain a lagoon of freshwater as suggested by the soil analysis, and protect it from saltwater inundation. This possible feature and creek location has been interpreted in the graphic below (Figure 3.2.48).

Figure 3.2.49: North-facing section of TT69 with a yellow dashed line to indicate the top of the possible naturally formed bank material located in the intertidal sand flat west of the likely location of the creek mouth. View to southwest. Scale 1m.

83 Lawrie 2011: 6, Vol 3, Section 8.7. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 139

In Area 7 shells from a naturally accumulated layer (8082) below the MLWM demonstrated the species range expected from an estuarine habitat; a tidal environment where fresh, and salt water mix. The context contained 20 different shell type (listed in Appendix 9 of Carter’s 2010 report; Vol 3, Section 8.5). The small sized (<2cm) species Caltholotia fragum and Nassarius burchardi, in addition to juvenile Cerithidae shells make up the majority of this assemblage. These species are common in the intertidal mud and sand flats of estuaries.84 Trilete fern spores from a sample of naturally deposited sand above the MHWM in Area 9 suggest that a creek existed nearby as the ferns are common to creek banks.

3.2.5.6 Intertidal Zone Reconstruction The early historic plans do not present an accurate survey of the intertidal zone along Darling Harbour’s eastern shore, instead it is illustrated with shading (for example Harper’s 1822 plan, Figure 3.2.15). The archaeological evidence from Darling Quarter can be used to reconstruct fairly accurately the nature and extent of the intertidal zone just prior to, and in the early years of British settlement (Figure 3.2.50). The Darling Quarter site is located not far to the north of the head of Darling Harbour where extensive intertidal sand and mud flats existed. Within the site the intertidal zone was extensive, consisting of a gradually sloping surface comprised of thick deposits of medium and coarse mid to dark grey sands capped with thinner bands of sands, silt pockets and woody detritus. Closer to the low water mark the rock surface displayed gentle undulations, a result of water action and deposition.

Under the homogenous sand deposits and the low water mark was a layer densely packed with whole and fragmented marine shell. This distinctive shell bed contained a whole range of shell species, with notably high proportions of juvenile to adult cockles. Also included were many other smaller species common to intertidal sand flat environments. The abundance of cockle shells in the natural environment was noted in Darling Harbour’s early name Cockle Bay. Smaller areas of flattish weathered sandstone bedrock with no sand deposits formed part of the intertidal zone to the north of the site.

In the southern part of the site the foreshore stretched westward from the shore for at least 50m (Figure 3.2.50, Vol 4: Plan 9.2). This area was likely part of a small cove or bay formed and protected to the north by the change in the shoreline’s trajectory from north to northeast. Further north the sand flats decreased significantly in size to perhaps just 10m to 20m off the rocky platform below the high water mark in Area 8. This is likely due to changes in harbour water dynamics and deposition patterns as the shore jutted out in a northwestly trajectory and the bedrock shelved steeply below the low water mark.

A creek (or creeks) flowed into this sheltered cove and the environment around it was influenced by freshwater. The archaeology suggests that a lagoon of freshwater had formed here and that it was protected from tidal saltwater inundation by a sandbank (Figure 3.2.51, Vol 4: Plan 9.3). These natural features were later exploited by the flour millers that developed the land in the 1820s. The creek-filled lagoon was enhanced to make the mill pond that supplied the mill’s steam engine with fresh water.

The shell bed was ancient and had started to form several thousand years ago. The sand layers forming the surface of the harbour floor above the shell bed were more influenced by dynamic tidal movements. These sands contained shell that ranged from several thousand to just a several hundred years in age. They also contained the range and ratio of shell species that the harbour supported following British settlement just over 200 years ago. Pollens from both native and introduced or exotic species were trapped with the sand layers, or perhaps filtered through them.

84 Carter 2010: 15, Vol 3, Section 8.5. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 140

The intertidal zone was ever evolving and changing. The formation of the harbour floor was a naturally occurring process, which on the Darling Quarter site showed some evidence of the environmental changes that occurred following British settlement but prior to its development in the 1820s.

There was no clear evidence in this phase for the reported extensive shoaling and increased sand and silt depositions resulting from land clearance and development in the early years of settlement. For the most part, the sand layers that formed the surface of the intertidal zone represented normal deposition processes of an intertidal harbour environment.

Figure 3.2.50: Reconstruction of the intertidal zone and sand flat within the Darling Quarter site using Harper’s 1822 plan as a background and also showing the wider context and location of modern streets and buildings. It consisted of an expansive sand flat (grey shading) and areas of flattish sandstone platforms (mustard shading). Extract from Cryerhall, Plan 9.2 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 141

Figure 3.2.51: Interpretation of the foreshore environment prior to development in the 1820s using Harper’s 1822 plan as the background. Possibly two creeks flowed into the harbour. Sheltered within a small cove and possibly protected by a sandbank, a lagoon of freshwater formed. Extract from Cryerhall, Plan 9.3 (Vol 4).

3.2.6 Discussion of Results A model for the natural environment of the site has been created from analysis of the geology and soil types, descriptions from early sources, illustrations and surveys. The Darling Quarter site is located on the eastern foreshore of Darling Harbour. The site included a bay or cove with intertidal sand or mudflat abutting a rocky sandstone shore. Offshore, the harbour waters were shallow, recorded on an early chart as being just three or so feet at low tide. To the south, the head of the harbour consisted of extensive mudflats that supported mangrove and saltmarsh plant communities. Benson & Howell suggest the mudflats, and therefore the mangroves, extended northwards past the Darling Quarter site. To the immediate east, the site was fringed with swamp forest species such as Swamp Mahogany and Swamp Oak. Further east, the Hawkesbury sandstone ridge supported woodland including Red Bloodwood, Scribbly Gum, Banksia and She Oak. On early historic plans, a creek or watercourse, with perhaps a tributary, existed into the harbour within the site. The estuarine sand or mudflat environment supported a number of shell species such as Sydney cockle, an abundant resource at the time of British settlement.

The archaeological evidence further refined this model. Open area excavation in Area 6, 5 and 9 confirmed that the rocky shoreline changed trajectory as the early surveys indicate, thereby

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 142 creating a sheltered bay or cove to the south. The intertidal area was mostly formed with coarse grey sands with very little evidence for substantial build-up of silts or other mudflat material. Fossil pollens from the sand flat material and other natural deposits indicate that the immediate area was dominated by swamp forest species, with a minor mangrove component. Also confirming this the wood samples from the Aboriginal midden site were identified as Casuarina glauca, or Swamp Oak. The evidence from the Darling Quarter site discounts the existence of mudflat in this location. Instead, extensive sand flats stretched from the shore for up to 50m at low tide. This means that mangroves and saltmarsh species of mudflat habitats at the head of Darling Harbour appear not to be dominant further north.

The mouth of predicted creek would have been located just within the site boundaries, and the archaeological evidence does support its existence. Analysis of soil samples from Area 9, within the cove or bay area and close to the shore where the creek is likely to have existed, confirm freshwater influences during the late Holocene. In fact, this area may have once been a shallow freshwater lagoon, protected behind a sandbank or other such naturally formed feature. This freshwater lagoon may not have been long-lasting or permanent, and it is not known with certainty that it existed in the natural environment in the early decades of British settlement. The fern spores identified within a natural deposit in Area 9 originated from a damp habitat nearby, such as a creek bed. Such evidence suggests the creek existed prior to British settlement, though none were recorded on early surveys. Lending further support to this interpretation, this area was later developed as a mill pond to retain water from two watercourses.

Samples from an ancient shell bed below the water table contained a wide range of estuarine species. Within the sample was a high proportion of Sydney cockle, common to intertidal sand flat habitats. Shell from the Aboriginal midden consisted of mostly cockle. The dominance of cockle within the natural environment certainly supports the historic accounts and gives substance to the name Cockle Bay. Also identified in the ancient shell bed were mud and rock oyster, and club mud whelk. Rock oysters as the name suggests attach to rocks in the mid to low tide range, and in the pre-1788 environment were attached to the rocky shore. Mud oysters were also noted by Europeans in the late 18th century to be in abundance in the estuaries of Sydney Harbour. It would seem from the shell samples that they were not abundant in the natural environment of the Darling Quarter site. This may explain why mud oyster was not found within the Aboriginal midden material, rather than their absence being explained by a cultural preference for or easier harvesting of cockle over mud oyster in this locality.

The radiocarbon age determination of cockle shell samples from the natural deposits yielded a wide range in age. Shell from the ancient shell bed were several thousands of years old, and the two dates obtained were almost couple of thousand of years apart. Above the shell bed were deep deposits of grey sands, the surface of which contained organic detritus, such as twigs and other plant-related litter, typical of an intertidal sand flat. Shell and pollen within these layers displayed a range of pre-1788 and post-1788 characteristics. The archaeological evidence, in particular the environment samples, from the site illustrates the natural processes of harbour deposition in an intertidal sand flat environment. Heavy organic material, such as shell, filters overtime through the sand and becomes trapped above the clay and sandstone floor forming a shell bed over thousands of years. During this process further sands and silts are deposited by harbour tidal movements and from the rivers and creeks that flow into the harbour. This action facilitates the movement of material through the sands, and also mixes the heavier content, such as the shell.

Refining the model for the natural environment of the Darling Quarter site is important in understanding why and how this location developed in the 19th century. The natural attributes of the site and its extensive intertidal sand flat both necessitated and facilitated the establishment a substantial jetty that included an area of reclaimed land for Barker’s mill in the 1820s. The likely

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 143 existence of a natural lagoon fed with freshwater from a creek became an obvious location for the mill pond to retain water for the mill’s steam engine. The success of the extensive reclamation projects undertaken by Barker and Brooks to the south in the 1830s and 1840s was partly due to the existence of the sand flats.

3.3 Aboriginal Occupation

3.3.1 Introduction Prior to 1788, the site was part of the Eora territory.85 The Eora people comprised several clans that shared a common language and a saltwater economy.86 They inhabited a varied landscape that stretched from the Georges River in the south to Pittwater in the north, and inland along the drowned river valley mouth to Parramatta.87 The Cadi clan occupied the southern side of Port Jackson, extending from South Head to Long Cove, and incorporating the study area. The antiquity of the Cadi people remains unknown, but archaeological evidence has confirmed a continued presence of Australian Aboriginal stone technologies and exploitation of the resources around the harbour over several thousand years.88

Figure 3.3.1: This engraving by Philip Slaeger of a John Eyre drawing indicates continued Aboriginal occupation in the area into the 19th century. A Native Camp Near Cockle Bay, New South Wales, 1813. From Absalom West, Views in New South Wales, 1813–1814 [and] historical account of the colony of New South Wales, 1820-1821.

85 The term Eora has typically been applied to describe a group of clans for which there may have been no collective word prior to European settlement. The area south of Port Jackson may have been populated by speakers of a coastal Darug dialect, or some other dialect. This is discussed in Attenbrow 2002: 30-36. 86 http://www.sl.nsw.gov.au/discover_collections/history_nation/indigenous/eora/ 87 Willsteed, Smith & Bourke 2006. 88 Willsteed, Smith & Bourke 2006. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 144

The earliest radiometric dates for Aboriginal occupation in the Sydney region stretch beyond 40,000 BP, but the area may have been occupied prior to this.89 Dates become more frequent from about 14,000 BP, with numerous dates from a variety of sites throughout the area in the last 6000 years. Multiple Aboriginal sites have been identified near the Darling Quarter site, including stone tool scatters, shell middens, and engravings, all within 3km of the Darling Quarter site.90 The KENS site, one kilometre to the north of Darling Quarter and in a similar proximity to Darling Harbour yielded close to 1000 artefacts and dated from 2,800 BP to British arrival.91

3.3.2 Archaeological Evidence Monitoring and testing for archaeological evidence of Aboriginal occupation was undertaken during the archaeological program (Figure 3.3.2, Figure 3.3.3).92 A number of 1m by 1m test trenches were excavated in Area 7 on the northern boundary with Area 9. These test trenches were located well within the intertidal zone and as expected did not contain any Aboriginal artefacts or other cultural material. Test trenches through the historic reclamation fill to the bedrock or natural sands in Area 8 were monitored and no Aboriginal archaeological remains were uncovered. In Area 5 however, monitoring and test trenching revealed the remains of a shell midden on the rocky foreshore. This was excavated by Comber Consultants and a detailed report on the results and analysis is included as Section 8.9 in Vol 3 in this report (Vol 4: Plan 10.2).

Figure 3.3.2: Graphic showing the location of the Aboriginal test trenches in Areas 8 and 7 and the shell midden salvage excavation in Area 5, within the basement area (purple outline). Harper’s 1822 plan has been used as a background to indicate where the original shoreline was. Cryerhall.

89 Attenbrow 2002: 18-21. 90 Comber Consultants 2012, Vol 3, Section 8.9. 91 Steele 2006. 92 Work undertaken by Comber Consultants. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 145

3.3.2.1 Aboriginal Shell Midden In Area 5, a total of 37 1m by 1m test trenches were excavated in the location of the shell deposit (Figure 3.3.3, Vol 4: Plan 10.2).93 The shell deposit was spread over a rocky outcrop that was above the high water mark and mixed with harbour sand material in the intertidal zone (Figure 3.3.4). Of the 145kg of shell material collected for analysis from the midden, Sydney cockle (Anadara trapezia) was the dominant species at 90.57 per cent of the MNI. Other species included Sydney Mud Whelk and Sydney Rock Oyster. There was a notable absence of juvenile shells, indicating that the shells were culturally selected for consumption. The soil analysis also supports this interpretation as charcoal and ash from cooking fires were present in the soil.94

Ten artefacts were found in association with the shell deposit.95 These consisted of four flaked pieces of chert, three chert flakes, a silcrete flake and a flaked piece of quartz. There are no known sources of chert in the immediate area, and the closest source is Plumpton Ridge about 40km to the northwest. This suggests that the Cadi traded with the people from this region (Bool-bain-ora) to obtain suitable raw materials for stone tools.96 The artefacts may have been used to open the shellfish or gut fish.97

The location of the shell midden material on a rock outcrop above the high water mark and within the sands in the intertidal zone, indicated that this midden had most likely been re-deposited by tidal action.98 The inclusion within the midden material of rounded pebbles, some water-worn shells and marine grit further supports this interpretation.

Figure 3.3.3: Aboriginal shell midden during excavation. Comber Consultants. Scale 1m.

93 Comber Consultants 2012: 14, Vol 3, Section 8.9. 94 Comber Consultants 2012: 32, Vol 3, Section 8.9. 95 Comber Consultants 2012: 21, Vol 3, Section 8.9. 96 Comber Consultants 2012: 34, Vol 3, Section 8.9. 97 Comber Consultants 2012: 32, Vol 3, Section 8.9. 98 Comber Consultants 2012: 32, Vol 3, Section 8.9. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 146

Figure 3.3.4: Detail of the shell deposit that formed the remains of the Aboriginal shell midden in Area 5. Scale 1m. Comber Consultants.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 147

3.4 Early Foreshore usage 1788 to mid 1820s

3.4.1 Introduction In the early years of British settlement, the new colony initially centred around Sydney Cove. Its harbour frontage was for the most part reserved for Government use; the western side for the government dockyard and commissariat stores, and the eastern side was set aside for the grounds of Government House. Private merchants established businesses at the mouth of the Tank Stream and in the remaining areas on the western side, such as Robert Campbell who leased land for his private wharf and warehouses.

Initially there was little development along the foreshore of Darling Harbour and by 1804 there were just 18 houses recorded in the area.99 Though not very well documented, early activities around Darling Harbour included lime production, market gardening and salt panning.100 Utilising the shell resource to process for lime is perhaps the activity most commonly associated with Darling Harbour in the early years of settlement. As Sydney’s maritime and commercial businesses expanded, the need for harbour access increased. Not only was Sydney Cove’s harbour frontage largely reserved for government use, increased sedimentation at the mouth of the Tank Stream affected its viability as a port for private enterprise. Darling Harbour, or Cockle Bay as it was known until 1827, was the obvious area for expansion. In the first decades of the 19th century, merchants and manufacturers were leasing large waterfront allotments and developing wharf facilities along the eastern side of Cockle Bay.

In 1810, Governor Macquarie gave orders to construct a new wharf at Cockle Bay to provide facilities for the unloading of produce brought in from the Hawkesbury.101 A new market was established up the hill, now the location of the Queen Victoria Building. Market Street was laid out to connecting the wharf to the market. Moving the markets and associated wharf to this location created a new focal point in the town’s development, away from Sydney Cove. Sussex Street was also laid out in 1810, running parallel with Kent Street, between it and Cockle Bay, from the barracks to the burial ground (now Town Hall). This extended the town grid to include the eastern side of Cockle Bay. Initial development and use of Darling Harbour was confined to Millers Point, where John Leighton operated three windmills.102 Shipbuilding, a long standing industry associated with Darling Harbour, also began around this time.

The southern part of the harbour was not initially attractive to early developers and manufacturers as essentially it was beyond the town limits and it consisted of swamp and shallow harbour waters. A number of creeks drained into the head of the harbour, and these sources of freshwater were essential for some industries. In 1813, John Dickson arrived in the colony with goods and machinery to establish a steam-powered mill. He was granted 15 acres of land at the southern end of Cockle Bay that included access to the freshwater creek, necessary to run the steam engine. By 1815 the flour mill was in operation, making it the first use of steam power in Australia.103

The Darling Quarter site is located between the southern extent of the laid-out town grid and Dickson’s mill. The first detailed survey carried out by William Harper in the early 1820s depicts the extent of development around Cockle Bay (Figure 3.4.1). Allotments with houses line the parallel streets between Cockle Bay and George Street as far south as Market Street. Further south development was patchy. The alignments of streets, such as Sussex and Druitt, are indicated on

99 Johnson & Parris 2008: 16. 100 Johnson & Parris 2008: 16. 101 Johnson & Parris 2008: 20. 102 Johnson & Parris 2008: 16. 103 Johnson & Parris 2008: 23. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 148 plan, but are not formalised with allotments and houses at this stage. In the immediate vicinity of the site there is little recorded development and it consisted of large allotments between the alignment of Sussex Street and the foreshore.

Figure 3.4.1: Harper’s survey from 1822 showing the extent of development along Cockle Bay and in the immediate vicinity of the site (red circle). The immediate surroundings are mostly undeveloped at this time. SRNSW SZ435, C&L additions.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 149

The archaeological evidence from the site confirms this lack of intensive use and development in this part of Cockle Bay during the first 30 years or so of the colony (Figure 3.4.2). Above the high water mark in Area 9, a localised layer of crushed shell overlying the sandstone shore may be evidence for lime production. Within the foreshore were the remains of two timber structures. In Area 6, the remains of a roughly constructed small boat ramp were located within naturally deposited harbour sands. To the south in Area 7, were the remains of timber fencing that appeared to form a western property boundary that extended into the intertidal zone to the low tide mark. Also recorded across the intertidal zone were deposits of harbour sands and debris that built up after 1788 but prior to major development of the area from the mid 1820s. Pollen samples record the changes in the surrounding environment, as native plants were cleared and new plant species introduced. Cereal pollen, presumably initially originating from Dickson’s mill to the south or from the windmills to the north, appear in large numbers in the samples. Shell analysis also confirmed environmental change shortly after British settlement, as species such as the mud oyster become less common and rock oyster dominant.

Figure 3.4.2: Graphic showing the location of the main trenches and features associated with Phase 3 Early Foreshore Activity 1788 to 1820s within the basement outline (purple). The thin orange lines represent fencelines and boundaries in the intertidal area of Brooks’ land. Cryerhall.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 150

3.4.2 Early land grants Parts of three large land grants were within the site boundary (Figure 3.4.3). To the south Captain Richard Brooks was granted land at Cockle Bay by Governor Macquarie in 1814. He established a slaughterhouse and curing yard to provide provisions for ships, government stores and to the public. Harper’s 1822 survey indicates a building, presumably the slaughterhouse, on the harbour edge, just outside the southeast corner of the site boundary (Figure 3.4.3). It appears that the land to the north was not granted until in the 1820s. Immediately to the north of Brooks’ land, Cooper & Levey had established a steam mill by 1825, though a formal land grant was not issued to Daniel Cooper until 1 June 1827, at about the time the property was sold to Thomas Barker, and became known as ‘Barker’s Mill’. North of Cooper & Levey’s mill site was a large parcel of land that was initially applied for by David Ramsay.104 The two acre plot did not appear to have been developed in anyway prior to its subsequent sale to Thomas Barker in 1829.

Figure 3.4.3: Early land grants associated with the site were large parcels of land between Sussex Street and the foreshore; Ramsay to the north, then Cooper & Levey and in the south Richard Brooks. South of the site was John Dickson’s mill. As the background is Harper’s 1822 plan. Cryerhall.

104 Casey & Lowe 2008b: 22. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 151

3.4.2.1 David Ramsay’s Allotment105 This block of land was originally applied for by David Ramsay in 1829 (Figure 3.4.4),106 but was referred to earlier as belonging to ‘Raine & Ramsay’, in an 1825 advertisement for the lot purchased by Cooper & Levey,107 in Daniel Cooper’s 1827 grant108 and on surveyor’s sketch of Thomas Barker’s property from 1828 (Figure 3.4.5).109 According to the 1829 surveyor’s sketch, the block was 2 acres, 1 rood and 3 perches in area.110 Raine & Ramsay was the business partnership between Thomas Raine and David Ramsay, founded in 1822 and involved in a range of trading activities, including importing the steam engine in 1824.111 This engine probably is the same one as at the Darling Mills in Parramatta, set up by John Raine, Thomas’ brother.112 The partnership of Raine & Ramsay broke up in 1828,113 and their block of land at Cockle Bay was sold to Thomas Barker in the first half of 1829, through a complicated series of transactions, involving the release of the site from Thomas Raine to David Ramsay (1 January 1829), the lease and release of the land from Ramsay to JT Goodsir (4 and 5 April 1829), and the lease and release of the land from Goodsir to Thomas Barker (13 and 14 April 1829) as well as a mortgage of £1200 plus interest from Goodsir to Barker.114 There appears to be little evidence for the development of the site prior to its sale to Barker.

In 1831, Thomas Barker received a grant which consolidated all his holdings adjoining his mill at Cockle Bay, namely his lease on the southwest corner of Sussex and Bathurst Streets, the grant containing the steam engine and mill, and the allotment originally belonging to Raine & Ramsay. Like Brooks’ grant of the same date (see below), this new grant was part of a scheme to regularise town grants, leases and permissive occupancies.115 It replaced the earlier grants and lease held by Thomas Barker. The new grant gave the total area of the site as 6 acres, 1 rood and 32 perches (2.61 ha),116 while the combined areas of the older allotments of Raine & Ramsay, Barker and Cooper & Levey was 5 acres, 3 roods and 1 perch (2.33 ha). This made the new grant 2 roods and 31 perches, or 0.69 acres (2807.5 m2), larger than the older allotment surveys. This discrepancy in area may be evidence of the early land reclamation in the late 1820s, causing the site to grow in area by the time of the 1831 grant. Alternately, it may due to the original survey underestimating the area of the site, although the scale of the required underestimate (10.7%) is probably too large to be attributed to error in measurement in normal circumstances.

105 Written by Nick Pitt, Archaeologist, Casey & Lowe. 106 SRNSW Surveyor General’s Sketch Books Vol. 1 Fol. 43, AO Reel 2778, #240. Dates on plan, ‘19th Febry 29’ and ‘25th Febry’. 107 Sydney Gazette 28 April 1825, p 1(4). 108 SRNSW NRS 13836; Item 7/482, Reel 2704, grant dated 1 June 1827. 109 SRNSW NRS 13836; Item 7/482, Reel 2704; SRNSW Surveyor General’s Sketch Books Vol. 1 Fol. 27, AO Reel 2778, X750A. 110 SRNSW Surveyor General’s Sketch Books Vol. 1 Fol. 43, AO Reel 2778, #240. 111 Maude 1967; McMartin 1967; Sydney Gazette 1 April 1824, p 2(2). 112 The Australian 28 July 1825, p 4(4); Sheehan 1967. 113 McMartin 1967. 114 Barker papers A 5398/3, No. 73, Schedule of deeds 14 March 1851 (Mitchell Library). 115 Sydney Gazette 11 June 1829, p 1(1), 31 May 1831, p 1(2). 116 Sydney Gazette 31 May 1831, p 1(2); Sydney Gazette 12 April 1832, p 2(2); Attested copy of 1831 land grant, dated 19 October 1831, in Barker papers A 5398/3, No. 29, Abstract of title of Thomas Barker Esq to Land in Bathurst and other Streets Sydney (1847). (Mitchell Library); cf Register of Land Grants and Leases, SRNSW NRS 13836, Item 7-473, Reel 2700, which does not mention the earlier, cancelled grants. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 152

Figure 3.4.4: ‘An allotment in Bathurst Street applied for by David Ramsay’, dated 19 and 25 February 1829, J Thompson, Assistant Surveyor, SRNSW Surveyor General’s Sketch Books Vol. 1 Fol. 27, AO Reel 2778, X750A, SRNSW.

Figure 3.4.5: Sketch of Thomas Barker’s property on Sussex Street. Surveyor Thomas Florance. Dated 26 August 1828. SRNSW Surveyor General’s Sketch Books Vol. 1 Fol. 27, AO Reel 2778, X750A.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 153

3.4.2.2 Cooper & Levey’s Allotment117 The earliest reference to this allotment appears be an advertisement for sale by James Smith in 1822: TO be SOLD, a most valuable PIECE of LAND, situate in Cockle Bay, near Captain Brook's Slaughter-house, containing about three acres, enclosed by a strong fence, and bounded on one side by the water. Its situation is truly valuable for maritime purposes, and commands a scope of water frontage, for wharf and other improvements, not to be equalled in the whole of Cockle Bay. There is also a constant supply of excellent fresh water. Added to these invaluable advantages, there is now erected a Stone Building, suitable for the residence of a family. Further particulars may be known by application to James Smith, Sydney.118

As John Dickson’s steam engine was situated to the south of Brook’s slaughter house, this advertisement would appear to refer to land to the north of Brook’s property. The allotment later developed by Cooper & Levey bears similarities to that described in this advertisement, including a supply of water (later used to supply the steam engine), and an area of 3 acres and 2 perches.119 The stone building mentioned in the advertisement is also visible in this allotment on the block in Harper’s 1822 plan (Figure 3.4.6), again providing evidence that it was already being developed by the early 1820s. The same building again is shown in an 1828 surveyor’s sketch, near Sussex Street, and is outside the study area (Figure 3.4.5).120

Figure 3.4.6: Detail of Harper’s 1822 plan. Red arrow shows the location of a building, likely the stone building referred to in James Smith’s June 1822 advertisement. Note, this building lies outside the current study area. SRNSW SZ 435.

117 Written by Nick Pitt, Archaeologist, Casey & Lowe. 118 Sydney Gazette 21 June 1822, p 2(2). 119 SRNSW NRS 13836; Item 7/482, Reel 2704. 120 SRNSW Surveyor General’s Sketch Books Vol. 1 Fol. 27, AO Reel 2778, X750A. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 154

James Smith was a builder and surveyor in Sydney and this property was possibly being sold as a speculative development. He had previously advertised at least two other properties bordering Cockle Bay for lease or sale.121 It is not known if Smith managed to sell this allotment in 1822, but the block was advertised for auction again in April 1825. At the time it was advertised as: THAT valuable PIECE of LAND, situate in Cockle-bay, adjoining Captain Brooks's Premises, on the South side ; Messrs. Raine and Ramsay's on the North; Sussex-street on the East ; and the Water on the West; with an unfinished Building, and sundry Building Materials there on. Conditions will be made known at the Time of Sale.122

It is likely that around this time, the property was purchased by Daniel Cooper and Solomon Levey, who constructed a steam mill on the site between July and December 1825.123 Cooper & Levey, in a formal business partnership operated this mill until mid-1827, when they sold it to Thomas Barker. Cooper & Levey and their steam mill is discussed in greater detail in the next section.

3.4.2.3 Richard Brooks’ Allotment124 Richard Brooks was a sea captain who made repeated trips to New South Wales, between 1802 and 1806, and then for trading trips in 1808, 1810 and 1811. He arrived as a free settler with his wife and children in March 1814. By that time, he had already developed business interests in the colony, possessing a large herd of cattle and the promise of an allotment in Sydney to replace land previously promised to him at Farm Cove.125 His land at Cockle Bay may have been granted to him in fulfilment of this promise of land.126

On his land at Cockle Bay, Brooks built ‘a slaughter house and other buildings for the slaughter and curing of meat’127 while his main residence was situated on Pitt Street.128 In 1816, the land was being used at one stage to hold 100 cattle before auction, while the slaughter house had been constructed by 1822.129 During 1823, Brooks repeatedly advertised several properties for sale, including his Cockle Bay slaughter house,130 probably as a result of his move to Denham Court, near Liverpool.131 In 1823 this property was advertised as ’consisting of a large substantial built slaughter-house, with two extra rooms, and a paddock well watered’.132 Richard Brooks was unsuccessful in selling the property at that time, and in 1827 advertised salt beef for sale at his stores ’joining DICKSON’S Steam Engine, in Cockle Bay’.133 Harper’s 1822 plan shows a single building on Brooks’ property, along its southern boundary (Figure 3.4.3). A sketch plan in Hallen’s Surveyor’s Field Book (c 1828-1830) (Figure 3.4.20) shows another three, smaller buildings on the allotment, as well as the building previously shown on Harper’s plan.134

121 Sydney Gazette 12 December 1818, p 2(2) [note James Smith was the proprietor of 68 George St at that time, see for example Sydney Gazette 10 January 1818, p 3(1).]; Sydney Gazette 26 August 1820, p 2(1). 122 Sydney Gazette 28 April 1825, p 1(4). 123 The Australian 7 July 1825, p 4(4), 29 December 1825, p 3(2); Sydney Gazette 7 July 1825, p 2(3), 19 December 1825, p 1(4). 124 Written by Nick Pitt, Archaeologist, Casey & Lowe. 125 Vivienne 1966. 126 Vivienne 1966. 127 Casey & Lowe 2008b: 51, citing Bk 41 No. 67 (L&PI). 128 eg Sydney Gazette 8 February 1812, p 1(3), 3 October 1812, p 1(3), 2 June 1814, p 2(3), 6 September 1817, p 2(2), 11 January 1822, p 2(2). 129 Sydney Gazette 27 April 1816, p 1(3), 21 June 1822, p 2(2). 130 Sydney Gazette 16 January 1823, p 4(2), 20 November 1823, p 4(3). 131 Vivienne 1966, Sydney Gazette 27 September 1822, p 2(2). 132 Sydney Gazette 16 January 1823, p 4(2). 133 Sydney Gazette 17 October 1827, p 1(4). 134 SRNSW Surveyors’ Field Books No. 347, A Hallen, 2/5195, Reel 2628. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 155

In 1831 Brooks received a formal grant of his land, which at the time was described as: Richard Brooks, ac. 1, 1 r. 7 p. One acre, one rood, seven perches; bounded on the east by Sussex- street, bearing north fifteen degrees west two hundred and seventy-three links ; on the north by number 2 allotment, bearing west three degrees thirty minutes south six hundred and seventy links; on the west by the Waters of Darling Harbour; and on the south by Liverpool street, bearing east one degree thirty mi- nutes north three hundred and ninety-five links ; being allotment number 1 of section 10.—Quit rent, £5 3s. 6d. per annum, commencing 1st July, 1823.135

The issuing of this grant does not mean that Richard Brooks did not previously have title over the land, but rather it was part of a process of formalising the allocation of land, now referred to by section and allotment number.136

In July 1833, Brooks leased this property to the architect John Verge, for £100 a year for a period of 21 years.137 Brooks died on 16 October 1833,138 and the Cockle Bay slaughter house and land were left in trust for his widow,139 who died in 1835.140 The block was then further subdivided and developed before Richard Brook’s estate was finally put to auction in February 1855.141

3.4.3 Cooper & Levey’s land – Archaeological Evidence To the north of Brooks’ property was a large parcel of land acquired by the firm Cooper & Levey in 1825. Part of this land parcel and its associated foreshore was within the basement excavation area – Areas 5 and 6, and part of Area 9 (Vol 4: Plans 10.2, 10.8, 10.38). The allotment stretched from Sussex Street to the harbour frontage, and was abutted by an extensive intertidal sand flat. The northern and southern boundaries are recorded on Harper’s 1822 survey (Figure 3.2.15, Figure 3.4.3) but there were no buildings at this time. The basement excavation area included the shoreline, and a portion of the land to the east (Area 9) and the harbour foreshore to the west and southwest (Areas 5, 6 and part 9).

Prior to Cooper & Levey’s development of the land from 1825 there was archaeological evidence for activity within the lot (Figure 3.4.2). On the landward side of the shore, evidence for some form of shell processing was revealed within test trenches (locations in Area 9 illustrated in Vol 4: Plan 10.38). Resting on a rocky intertidal platform and extending westward towards the harbour zone, was a timber log construction that may have functioned as an informal or temporary boat ramp (in Area 6 located on Plan 10.8 in Vol 4). Surrounding this were several accumulations of sand, organic and woody debris, formed by natural harbour deposition processes prior to extensive landscape modifications in the mid to late 1820s. Extensive areas of woody detritus overlaying the harbour sands in the upper reaches of the foreshore were also recorded. Pollen samples from this area echo the tale of widespread environmental change during the early years of the colony.

3.4.3.1 Shell Processing Evidence Located in several hand-excavated test trenches (TTs 70, 72, 74 and located on Plan 10.38) across Area 9 was a dense layer of shell-rich fill (context 9513). It was located above natural sandy deposits and sandstone bedrock between RL 1.07m and 1.2m (Figure 3.4.7, Figure 3.4.8). The slightly damp material was dominated by crushed shell and grit, with the occasional whole cockle

135 Sydney Gazette 31 May 1831, p 1(2); SRNSW NRS 13836; Item 7-473, Reel 2700, grant dated 19 October 1831. 136 Sydney Gazette 11 June 1829, p 1(1), 31 May 1831, p 1(2). 137 Casey & Lowe 2008: 51, citing G28 dated 5 July 1833 (L&PI). 138 Vivienne 1966. 139 Casey & Lowe 2008b: 51, citing Bk 41 No. 67 (L&PI). 140 Vivienne 1966. 141 Casey & Lowe 2008b: 51. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 156 shell visible, within a light grey clay and coarse blight brown sand matrix. Also visible were occasional inclusion of charcoal and a slag-like material with imbedded shell fragments. Many of the shell grit fragments were burnt, though the whole or mostly whole cockle shell did not display evidence for burning. The shell content gave this layer an overall whitish cream colour. This context differed from the Aboriginal shell-midden material located to the northwest, as it was predominantly composed of finely crushed shell and grit. Its damp qualities enabled it to almost form a paste when rubbed between the fingers. Located within the immediate area was an extensive shell bed, and the remains of an Aboriginal sell midden. This suggests that large quantities of cockle and oyster shell was available and accessible from this location. There are many documentary references to shell lime production on the shores of Cockle Bay. The crushed and burnt shell material has therefore been interpreted as evidence for such shell-processing and lime production.

Figure 3.4.7: This photograph illustrates the layer of crushed shell (context 9513) at the base of TT72 being just above natural (blackish sand and bedrock) and sealed by sandy fill. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 157

Figure 3.4.8: The crushed shell layer was exposed during machining and it was revealed to be fairly extensive but petered-out towards the south (photo background) View to south. Trowel in foreground for scale.

Lime Production from Shell142 Lime was used in the 18th and 19th centuries for a variety of purposes, including construction and agriculture.143 It was produced by strongly heating (to between 812°C -1100°C) limestone, marble, chalk, or shell, all natural forms of calcium carbonate, CaCO3, in a process generally known as ‘burning lime’. The lime, also referred to as quicklime,144 produced by lime-burning was predominantly calcium oxide, CaO, along with varying amounts of other impurities. This lime was then ‘slaked’ by adding water, converting it to calcium hydroxide, Ca(OH)2, before it was used in lime mortar or plaster. Slaked lime, also known as hydrated lime, then gradually reacts with carbon dioxide in the air to harden and form calcium carbonate.145

There is no limestone available around Sydney, meaning that initially lime had to be obtained by burning shells.146 Both oysters and cockles are named as being burnt for lime around Sydney and

142 This section has been written by Nick Pitt, Casey & Lowe. 143 Hebert 1836: 91; Ure 1847: 780. 144 “lime, n.1, 3a” OED Online 2011; “quicklime, n.” OED Online 2011. 145 Hebert 1836: 89-91; Pearson 1990: 28; Harrington 2000: 3-4. 146 Phillip 1789 [2003]:125; Worgan 1978 [2003]: 7 (letter June 12-June 18 1788); Tench 1789 [1998]: 132; Collins 1798 [2003]: 433, 555; Wentworth 1819 [2003]: 55-56; Nagle [c 1829]: 93-94; Mundie 1852 [2003]: 369. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 158

Newcastle.147 The exact circumstances of shell collection and lime burning appear to have varied over the first decades of settlement. Tench’s early account mentions female convicts gathering shell, while another early witness c.1788, Jacob Nagle, remembered that a lighter (or small boat) was used to gather shells for lime and other building material.148 Later, it seems that lime was burnt at Botany Bay,149 Broken Bay150 and Newcastle151 and brought to Sydney Cove by boat. However, there is also reasonable evidence for lime burning at Cockle Bay in the early 19th century. There appears to have been a privately operated lime kiln on land claimed by the Government near the market wharf until around 1830.152 Another lime kiln was constructed at Cockle Bay near Millers Point by one Arthur Martin around 1811 and operated until at least 1833.153 Outside of Cockle Bay, but still on Sydney Harbour, another lime kiln was depicted on Bennelong Point between 1822 and 1845.154 Pearson suggests that early 19th-century Sydney developed a system where shell was brought from around Sydney Harbour and the Parramatta River to centrally located kilns.155

The simplest method for lime production used in colonial New South Wales was heap and pit burning. This involved layering shell and fuel, such as dry wood, in either heaps or in shallow pits.156 This is the method described by WC Wentworth as being used at Newcastle.157 However, these methods did not require permanent kilns, as such. Intermittent kilns, so called because they burn lime in discrete batches rather than continuous production, were the other dominant means of producing lime in the 19th century.158 The most common early forms were rectangular ‘D’ kilns, bell kilns, inverted bell kilns and cylindrical kilns.159 Intermittent kilns could be operated either as flare kilns, where the fuel was laid on top of the shell (or limestone), or as mixed-feed kilns, where the fuel was mixed with the shell (or limestone), such as by layering. Out of these two methods, flare kilns potentially produced a more pure product, but required a more experienced operator, and appear to have been uncommon in the larger Australian context.160 The appearance of charcoal and slag-like material intermingled with shell in the shell-rich (context 9513) suggests that it may be the product of either heap burning, or burning in a mixed-feed, intermittent kiln.

The end product, lime, was used in the production of mortars and plaster. It was also used in agriculture, where lime was added to soil as fertiliser. According to the 1847 edition of Ure’s Dictionary of Arts etc ‘the most extensive employment of quicklime is in agriculture’.161 Also it is clear that the lime used for agriculture in this period was considered the same as that used for construction. For instance, a report of the Australian Agricultural Company, described its site at Port Stephens as follows, ‘in the immediate neighbourhood adjoining the shore, are beds of oyster-

147 Worgan 1978 [2003]: 7 (letter June 12-June 18 1788); Wentworth 1819 [2003]: 56. 148 Tench 1789 [1998]: 132; Nagle [c 1829]: 93-94. 149 Sydney Gazette 23 October 1803, p 2(2). 150 Sydney Gazette 25 December 1803, p 4(1); 24 June 1804, p 4(1); 25 November 1804, p 4(2); 6 April 1806, p 2(3); 2 April 1809, p 2(3). 151 Wentworth 1819 [2003]: 55-56; Harris 1847 [2003]: 23. 152 The Australian, 20 January 1827, p 2(3); Sydney Monitor, 24 March 1830, p 2(3); For Cockle Bay more generally, see The Monitor 12 November 1827, p 7(2). 153 Sydney Gazette, 27 March 1819, p 2(3); 25 October 1832, p 3(4); 27 October 1832, p 3(2); Sydney Herald 25 October 1832, p 3(1); 1833-1880 map section 92. 154 Pearson 1990: 29. 155 Pearson 1990: 28. 156 Pearson 1990: 28; Harrington 2000: 7-8. 157 Wentworth 1819 [2003]: 56. 158 Pearson 1990: 31. 159 These are described at length by Pearson 1990. 160 Harrington 2000: 8-9. 161 Ure 1847: 780. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 159 shells, convertible into the finest lime, both for building and agriculture’.162 Another use for lime was as a pesticide, being recommended against a wide variety of insects and moss.163 Other uses for lime (calcium oxide) in the early 19th century included soap manufacturing, tanning, glass manufacturing, glue making, tanning, sugar refining, and cleaning freshly plucked feathers.164 Lime was also occasionally used to hasten the decomposition of animal and human corpses,165 although public sentiment in Great Britain appears to have been against the practice, particularly among the ‘middling and lower classes’.166

3.4.3.2 Timber Boat Ramp Located on the western edge of the intertidal sandstone platform (context 8448) within Area 6 was the remains of a timber structure, most likely a boat ramp, or slipway, (context 8377) (Vol 4: Plan 10.8). The ramp was orientated northeast to southwest and extended from the sandstone 8.5m into the sand flat (Figure 3.4.9, Figure 3.4.10). This orientation was a response to the natural topography of the shoreline in this location. It displayed a gradual slope towards the harbour zone from RL 0.25m on the bedrock to -0.1m in the sands; meaning that at high tide the end of the ramp would be about 550mm (almost 2 feet) underwater. It was constructed from roughly cut ironbark and blackbutt timbers laid directly onto the rock and sands with no evidence of nails or lashing.167 The timbers were askew in places, leaving large and uneven gaps. Several timbers had been displaced and were scattered over the sand flat. Remnants of a possible consolidation fill (context 8442) consisting of sandstone fragments, white coarse-grained sand, and some clay was found beneath timbers and over the harbour sands.

The construction of the ramp appeared to have been undertaken in three sections; a shoreward, mid-section and seaward parts. The shoreward section was characterised by split logs laid directly onto the sandstone, parallel to the edge of the sandstone foreshore (Figure 3.4.11). The timbers were roughly triangular in cross-section and although they had been neatly sawn at each end, they had only been roughly split longitudinally. Many of the timbers still exhibited the rough, splintered surface from this process, perhaps indicating a short usage of this structure. Bark still adhered to the exterior and un-worked surface of many of the timbers. In this section, the ramp was between 2m and 2.5m in width, usually reflecting the lengths of individual timbers, but occasionally utilising timbers of different lengths to achieve it. Individual elements averaged 150mm in thickness.

The mid-section was more complex in its construction method, using split logs laid perpendicular to the bedrock edge as ‘sleepers’ for the overlying timbers (Figure 3.4.12). The overlying timbers were shorter than those of the shoreward section, averaging 1.25m in length and were laid end-to-end across the sleepers to maintain the 2.5m width of the ramp. The overlying timbers were laid in the same parallel fashion as those of the shoreward section. They were cut and split in the same way, bearing the same characteristics. The seaward part of the ramp was constructed with similar timbers of around 1.9m length. In this section, the timbers were laid perpendicular to the shoreline, in the same fashion as the lower sleepers of the mid-section (Figure 3.4.13). The timbers

162 Report of the Australian Agricultural Company quoted in Quarterly Review 1828: 12. 163 Loudon 1824 lists many examples including quicklime used to control insects by dusting leaves, p 437; used against turnip fly (as recommended by the Sydney Gazette), p 629; used against caterpillars (by shaking the caterpillars into quicklime spread out on the ground or washing the bark of a tree with a mixture of quicklime and water) p 735, used to control ‘aphis’ as part of a mixture, p 827; for reprinted advice see Sydney Gazette, 12 March 1803, p 2(3) (against moss), 14 October 1804, p 3(1) (against turnip fly), 25 August 1805, p 2(2) (against the wheat caterpillar), 17 March 1810, p 2(1) (against apple-eating insects). 164 Hebert 1836: 91; Ure 1847: 780. 165 Harrington 2000: 4. 166 [House of Commons] 1842 –see especially summarised evidence in the index, pp 249-250. 167 Identification by Dr Jugo Ilic, Vol 3, Section 8.8. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 160 were two-deep and closely packed together. The width of the ramp in this section was considerably narrower, at only 1.25m.

The timber boat ramp appeared to be in a state of disrepair when it was covered by fills during the construction of the jetty in the mid 1820s (see next section). This may have been due to a poor construction method or damage caused by the jetty’s construction. The ramp was somewhat of an informal structure as it was constructed using roughly worked or unworked timbers that were clearly not securely bound or indeed fastened into the sand flat. Among the timber used was blackbutt and grey ironbark. Both these hardwood species would have been available locally, were relatively easy to work and suitable for use in structures such as the boat ramp.

It was a small ramp and likely was used for small vessels, such as rowboats, to land and launch from the shore. The ramp pre-dates the jetty, built in the mid 1820s by Cooper & Levey. The boat ramp may have been constructed prior to Cooper & Levey acquiring the land, and therefore belong to a more informal or opportunistic use of this part of the harbour foreshore, such as for fishing or shell gathering for lime production. It is also possible that the boat ramp was constructed at the beginning of Cooper & Levey’s occupancy and utilised during the early construction phase of the mill buildings and jetty.

Figure 3.4.9: Detail of the timber boat ramp (also called ‘slipway’) extracted from Plan 10.8 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 161

Figure 3.4.10: Annotated plan of the north part of Area 6 showing the timber ramp (highlighted in brown) in context, resting on the bedrock and intertidal harbour sands. The red arrow in the inset points to the location of this area within the basement. Annotated extract from Plan 10.8 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 162

Figure 3.4.11: The shoreward end of the timber ramp. The sandstone bedrock (context 8448) foreshore can be seen in the background. View to the east.

Figure 3.4.12: The mid-section of the timber ramp context 8377. View to the north. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 163

Figure 3.4.13: The seaward end of the timber ramp 8377. View to the east. Scale 1m.

Figure 3.4.14: View of the timber ramp extending into the harbour zone showing the exposed bedrock that formed the shoreline (left foreground) and the foreshore sands (left and right midground). View to southwest. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 164

3.4.3.3 Harbour Deposits The timber ramp was constructed on both bedrock and the sandy foreshore (Figure 3.4.10). The sand that formed the intertidal zone prior to the ramp’s construction was likely to have been a mix of pre and post-1788 depositions. These have been previously described as part of the natural environment in Section 3.2.5.4. After the ramp was constructed, coarse grey sands, organic-rich material and woody debris accumulated around and partially sealed the structure. Typical of the type of post-construction accumulation was context 8390 located on the northern side of the ramp (Figure 3.4.15). This deposit was dominated by decayed timber fragments. Very small chips and splinters were thoroughly mixed with pale grey, medium-grained sands with a small silt component. The deposit was evenly spread and elements were equally distributed throughout. A similar wood splinter-rich sandy deposit (context 8483) was also located in TT30 further to the north (Figure 3.4.16).

Deposited in the same locality above 8390, was an unevenly spread mix of pale to white, fine- grained sands containing occasional patches of darker material. These sands (context 8389) were deepest at the seaward end of the ramp and very thin above the sandstone bedrock. There were no inclusions and the deposit was well-sorted. Further to the north of the timber ramp this deposit was represented by contexts 8482 and 8458 in TT30. Above 8389, a second splinter-rich deposit (context 8384) had washed over the sands and timber ramp (Figure 3.4.17). The deposit was grey- black in colour, well consolidated and dominated by splinters of wood and other small timber fragments. The organic component was well mixed within a fine-grained sand matrix. This deposit also contained small pieces of rope, other non-diagnostic fibrous material and a small amount of shell fragments. Further to the north in TT30, this deposit was represented by context 8459. Further west and into the harbour zone, the sand deposits post-dating the ramp were much ‘cleaner’ and did not include such an amount of wood or other organic material. Representative of this type of accumulation was context 8376 (Figure 3.4.17). The lack of debris within this sand was probably due to its location further into the intertidal zone, and therefore influenced by greater tidal dynamics that washed away the lighter pieces of wood and other debris, rather than allowing it to accumulate.

8377

8428

8390

Figure 3.4.15: Section through the organic sandy deposit 8390 that abutted the timber ramp (context 8377), showing the paler pre-1788 natural sands 8428 beneath. View to the south. Scale 1m. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 165

Figure 3.4.16: East-facing section of TT30 located northwest of the timber ramp shows a mix of wood splinters and sand (8483). This was typical of the type early 19th-century harbour deposition.

8376

8377

8419

8384

Figure 3.4.17: On the right the organic and woody sand deposit 8384 that was deposited after the ramp was constructed. Further to the west and into the harbour zone the harbour deposits, such as 8376, were more sandy and contained very little organic debris. View to west. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 166

Elsewhere in Area 6, several other types of similarly phased deposits were recorded. To the southeast of the timber ramp there was an extensive deposit of grey sand (context 8446) mostly recorded in section (Figure 3.4.18, Figure 3.4.19). This context consisted of mottled light to mid grey sands with small nodules of red and yellow clays, pockets of dark, sandy silt organic-rich material. While it was recorded in section, it was possible to see that this material had once formed a low bank within the intertidal zone just to the south of the sandstone platform (context 8448). The top of this bank material was around 0.5m (high tide level) and it petered out into the intertidal zone to the west at around 0m (mean tide level). This bank material may have been part of a bank that extended further to the southeast and formed a barrier that trapped freshwater from a creek to create a lagoon-type feature. This evidence for this has been discussed and presented in Section 3.2.5.5 and Figure 3.2.51. Bank material 8446 has been included in the interpretative illustration for Phase 3 activity (Figure 3.4.32, Vol 4: Plan 9.4).

Jetty fill

8444

8446

8448

Figure 3.4.18: Sand deposited by water movement (context 8444) overlying the low bank of sand (context 8446) with a clear horizon created by the harbour tidal action above. This was later sealed by jetty infill in Phase 4 (post-1825). View of Section 25 to the east with bedrock (context 8448) to north. Scale 1m.

Figure 3.4.19: Section drawing showing the sandstone bedrock 8448 and low bank of sand 8446 with swirls of sands 8444 between. West-facing section. Extract from Plan 10.13 (Vol 4). ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 167

Between this low sandy bank and the bedrock, was a deposit that clearly illustrated a number of individual depositional events. Context 8444 displayed a multitude of sand, silt and clay bands. It consisted of coarse-grained white sands lensed with mid-to dark grey clays and silts (Figure 3.4.18, Figure 3.4.19). The lensing betrayed the movement of water, with swirls of light and dark material. The top of both the low sand bank and the swirling sand, clay and silt bands had been flattened by the tidal action and presented as a clear horizon prior to the later infilling as part of the jetty construction.

Within the higher reaches of the intertidal zone, and located around the shoreward part of the timber ramp and further southeast following the shoreline, were two deposits with a notably high organic component. Contexts 8445 and 8449 were fairly extensive and presented as patchy deposits of brown to black fine-grained silt with high organic content (see location in Figure 3.4.10). This material was deposited during or after the lifespan of the timber boat ramp. The organic content of the silty material may be the result of rubbish or other pollutants being dumped into the harbour during the early years of the colony.168 The material was only present in the higher reaches of the intertidal zone and close to the shoreline.

3.4.3.4 Pollen Samples Two pollen samples from harbour deposits post-dating the timber ramp construction were analysed. An early 1800s date for these deposits was confirmed by the high percentage of native plant species and low percentage of exotic species present.169 Context 8445, the patchy silt with apparent high organic content, included casuarina (64%), eucalypt (8%), raspwort and mangrove. There was also a small percentage of cereal pollen (6%) in the sample. The presence of such a high percentage of casuarina relative to cereal and other weed exotics, and the presence of mangrove pollen, suggest that the western and/or eastern foreshore of Darling Harbour still contained areas of native vegetation, in particular swamp forest landscapes. The cereal pollen may have originated from Dickson’s Mill located just to the south of this site from 1815. Therefore, the suggested date for this deposit is the early 1800s and possibly post-1815, when Dickson’s flour mill become operational. While this context appeared highly organic, the sample contained no human sewerage indicators. The source of the organic material may just have been decaying plant matter.

Context 8483 from TT30 north of the timber ramp also contained the same range of native species types, along with a slightly higher percentage of cereal pollen. Exotic weeds, such as dandelion and wire-weed, were also present in the sample, indicating the spread of non-native weeds on what was vacant land at the time.170 Along with having a relatively high diversity of pollen types from native shrubs typical to rocky foreshore or creek banks, this sample also contained significant numbers of samphire pollen. This may be representative of a saltmarsh environment or an introduced herbaceous species. There was also a diverse range of fern spores, most typical of moist substrates such as damp sandstone or creek banks.171

168 Witnesses to the NSW Legislative Assembly inquiry into the silting of the harbour in 1866 frequently referred to the long-standing practices of dumping rubbish and the exiting of sewers into the harbour as a major contributing factor. McLoughlin 2000, 31: 2, 183-208. 169 Macphail 2010: 21-22, Vol 3, Section 8.6. 170 Macphail 2010: 21, Vol 3, Section 8.6. 171 Macphail 2010, Vol 3, Section 8.6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 168

3.4.4 Brooks’ Land – Archaeological Evidence During this phase the foreshore associated with Brooks’ land was within Area 7. Brooks’ land stretched from the western side of Sussex Street to the harbour. The early surveys depict a building just outside the excavation area. Presumably this was the slaughterhouse that Brooks established on his property after 1814. Prior to reclamation in the 1840s, an extensive intertidal sand flat extended westward from the shore, part of which was in the excavation area (Area 7). Harper’s 1822 plan indicates that Brooks’ northern and southern property boundaries extended into the foreshore. A sketch from Hallen’s field book 1828-1830 also records that these property boundaries included a portion of the intertidal area (Figure 3.4.20). Remains of three sets of timber fencing dating to the early 19th century were found in Area 7 (Figure 3.4.2, Figure 3.4.21).

3.4.4.1 Timber Fencing Trench A within Area 7, was a large open area that was excavated to the top of the intertidal sand flat (Figure 3.4.21, Vol 4: Plan 10.18). Within this trench were the remains of an east-west orientated post-and-rail fence (context 7997) and a north-south paling fence (context 8057). Further to the north in Area 7, a large machine-excavated test trench (TT19) revealed two phase of fencing on the boundary with Area 6, the earlier of the two (context 8363) is interpreted as being the pre-1825 northern property boundary fence of Brooks’ land. The locations of Trench A and TT19 within Area 7 are illustrated in Figure 3.4.2.

Figure 3.4.20: The early survey by Harper 1822 (left) and field book sketch by Hallen 1828 (right) depict lines within the foreshore to the west of Brooks’ land. These fencelines represent the extension of the property boundaries and a perception of ownership into the intertidal area. Plans are orientated north.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 169

Figure 3.4.21: Annotated plan of Trench A with the post-and-rail fence 7997 and paling fence 8057 highlighted. The location of the trench within the basement excavation area is indicated by the red arrow in the insert box. Annotated extract from Plan 10.18 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 170

Figure 3.4.22: Remains of the post-and-rail fence 7997 within Trench A, Area 7. This photo shows the grey harbour sands of the intertidal zone, and the extent of subsequent reclamation and land fill. Also in the photo is the north-south aligned paling fence 8057. View to west. Scale 1m.

Figure 3.4.23: Detail of the post-and-rail fence 7997. View to southwest. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 171

Figure 3.4.24: Detail of the mortise and tenon joint in fence 7997. Scale 300mm.

Paling Fence 8057 Crossing the post-and-rail fence in a north-south alignment were the remains of a paling fence (Figure 3.4.25, Vol 4: Plans 10.18, 10.22). The remains consisted of one collapsed post and the base of at least 50 fence palings driven into and/or later surrounded by harbour sands. The fence remains were found at around RL -0.5m (low tide level) and survived to a height of RL -0.4m where they had been previously cut down, likely during the first phase of reclamation in the late 1830s (Figure 3.4.26). Rock oysters were also found adhering to both the west and east sides of the palings (Figure 3.4.27). In a number of the timber palings small holes had been made by marine borers, such as teredo worm. The rock oyster and evidence for marine borers both indicate that this fence existed within the foreshore for some time while it was still influenced by tidal movements.

Figure 3.4.25: Remains of the paling fence 8057 within the harbour sands. View to southeast. Scale 1m. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 172

Figure 3.4.26: Section drawing of the western face of the paling fence remains (8057) showing several rock oysters adhering to the timbers within the intertidal range. Also evident was damage to the timber from marine borers such as teredo worm. Extract from Plan 10.22 (Vol 4).

Figure 3.4.27: Detail of the rock oyster adhering to the western (outer) face of the timber paling (8057). Scale 250mm.

Boundary Fence 8363 To the north of Trench A a large north-south test trench (TT19) was mechanically excavated to establish the northern property boundary of Brooks’ land (Figure 3.4.2, Vol 4: Plans 10.19, 10.25). Within TT19 was evidence for two phases of boundary fence (Figure 3.4.28). The earlier of the two fences, context 8363, was found at RL 0.255m (within the tidal range) and did not survive well. The remains consisted of a series of fragmentary timber elements including three truncated vertical

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 173 timber palings with a maximum surviving height of 250mm. The fence was in an east-west alignment, though slightly off to the northwest compared with a later 19th-century boundary fence. Boundary fence 8363 is the first phase of three phases of boundary fence identified within TT19 (Figure 3.4.29). The projected alignment of the fenceline is illustrated in Figure 3.4.2 by the orange lines.

8320

8363

Figure 3.4.28: Detail of the west-facing section of TT19 showing the very scant remains of the early boundary fence 8363 and the better preserved Phase 5 boundary fence 8320 (on right). Scale 1m.

Figure 3.4.29: Extract from the Area 7 Phase 5 & 6 plan showing TT19 with the three phases of boundary fences. The remains of the earliest, 8363, and its predicted alignment is highlighted in red. The trench was around 1.5m in width and the plan is orientated north. Extract from Plan 10.19 (Vol 4). ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 174

Discussion The scant fence remains within TT19 can clearly be interpreted as an early northern property boundary of Brooks’ land due to the location and more substantial remains of a further two phases of boundary fencing. The early fence was probably constructed between 1814 and the mid 1820s. Harper’s survey in 1822 and Hallen’s field book sketch 1828-1830 both indicate that this boundary did indeed extend into the intertidal zone (Figure 3.4.20, Figure 3.4.30). The location of the post- and-rail fence does not correspond to any boundaries or fences recorded on the early surveys. It is however on the same alignment of the later post-reclamation property subdivision in the 1850s. This fence indicates that the subdivision of Brooks’ property may have occurred prior to reclamation.

The paling fence was in a north-south alignment, and though not in the correct east-west position it may correspond to a similarly aligned line depicted on Hallen’s 1828 sketch. The surrounding sand that this fence was driven into was located at about RL -0.5m, the mean low tide mark. The paling fence may have delineated the initial extent of Brooks’ property claim into the foreshore, at the low tide mark. As it was constructed using palings, rather than a more economical post-and-rail fence, it may also have had a function in preventing debris and rubbish being washed into Brooks’ foreshore portion.

Brooks’ estate was described in the Sydney Gazette as having a slaughterhouse, store, paddocks and facilities for curing meats.172 It is possible that timber fencing within the intertidal zone may also have been the result of the way Brooks managed and organised his land, creating enclosed spaces to corral animals waiting for slaughter.

Figure 3.4.30: Detail of the locations and predicted alignments of the fencelines (red lines) in Area 7 with Harper’s 1822 plan also depicting property boundary fences. Brooks’ slaughter house is the illustrated building on the adjacent land to the east. The plan is orientated north. Cryerhall.

172 Sydney Gazette 21 June 1822, p 2b, 28 January 1826 p 4c 717, October 1827, p 1d. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 175

3.4.4.2 Harbour Deposits The top of the intertidal zone was exposed in plan within Trench A (Figure 3.4.21, Vol 4: Plan 10.18). It consisted of several distinct deposits of coarse grey sand, shell, patches of organic-rich silt and woody debris. The sand flat within Trench A was located between 0m and -0.7m, essentially below the mean water mark, and at the western extent of the trench, below the mean low tide level. The sand deposits were recorded in plan and within several hand-excavated test trenches (TT1 and TTs 6 to 9). These have been described as part of the natural environment (Section 3.2.5.4). Test trenches excavated on the harbour side of the paling fence (TTs 6 and 7) revealed accumulations of sand and shell against the fence, demonstrating the continuing process of harbour deposition after it was constructed. Dense silty deposits (contexts 7920, 8069 and 8086) continued to accumulate around the paling and post-and-rail fence. Extensive deposits of this type of material were not encountered in the open areas of foreshore, suggesting that the material accumulated due to the presence of the fences.

Shell samples from these sand deposits display a change in the dominant species type from the pre- 1788 natural deposits. In the seven combined contexts described as post-1788 harbour sand deposits, Rock oyster comprised of almost 60 per cent of the total weight of the shell assemblage. As the next most common shell types, Sydney cockle and Club mud whelk comprise near proportionally equal components of the assemblage, followed by much smaller proportions of Mud oyster and Other shell types (Figure 3.4.31). Rock oyster in the pre-1788 natural harbour deposits comprised of just 11.5 per cent of the total weight. There was also notably higher weight percentage values for cockle (the dominant species in the pre-1788 harbour deposits), whelk, mud oyster and other estuarine species. Although Sydney cockle is present in considerable quantities in the intertidal wash deposits, the overall trend demonstrated by the post-1788 deposits indicate a decline in the availability of local Sydney cockle populations.173

4000

59.7% 3500

3000

2500

2000

Shell weight (gm) 1500 19.3% 16.4% 1000

500 2.4% 2.2%

0 Rock oyster Sydney cockle Club mud whelk Mud oyster Other types

Figure 3.4.31: Proportion of shell types (weight) in combined natural intertidal-wash deposits from Area 7. Carter 2010.

173 Shell information and statistics are from Carter 2010: 12-18, Vol 3, Section 8.5. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 176

One sample from the post-1788 harbour deposits was submitted for pollen analysis. Context 8053 contained predominantly casaurina pollen (72%) followed by other natives, eucalypt, mangrove, broom spurge and raspwort. Significant numbers of mangrove (Avicennia marina) pollen, an unidentified 'cup' micro-alga previously found in estuarine sediments on the southwest coast of Western Australia (Macphail 2008), foraminiferal trochospiral liners and fragments of a crustacean egg confirm that the sample is sediment deposited below the high water mark.174 Also within the sample were cereal pollens, and these are likely to have originated from Dickson’s mill to the south of the site, indicating a post-1815 date for the deposit. Also identified in this sample was Cloacasporites sydneyensis, making this sample the earliest record for the human sewerage indicator.175

The radiocarbon dates from shell in the three harbour deposits identified as post-1788 offered a wide age range. Shell samples from harbour sands 8053 and 8078, and silty deposit 7920 produced a date range of between 4192 to 549 BP. This wide age range indicates that the harbour foreshore environment was quite dynamic, and resulted in the sand body that formed the foreshore surface included a mix of ancient shell deposits and newer ones, and also native and exotic post-1788 pollens, and the human sewerage indicator Cloacasporites sydneyensis. The radiocarbon dating results of the shell samples have been discussed in more detail in Section 3.2.5.4 and in Vol 3, Section 8.10.

3.4.5 Discussion of the Early Foreshore Use Evidence Transport, industry and trade were intrinsically linked to the waterways in the 19th century, and properties with potential for harbour access held high value. Sydney’s harbours and rivers were vital for its early development, growth and success. The early grant holders associated with the site would have recognised the potential in these properties, for their own business ventures or as real estate and development speculations. Prior to the realisation of this potential, the earliest colonial activity on the site involved a somewhat opportunistic use of the foreshore and its resources. The intertidal zone was seen as useable land, and with some of it being more formally claimed and enclosed with fencing, creating field divisions more reminiscent of a rural setting than that of an urban fringe. As Darling Harbour’s former name, Cockle Bay, suggests, and confirmed in the archaeological record, there was an abundant shell resource, not only in Aboriginal shell midden remains but also extensive naturally formed shell beds within the foreshore. This resource was exploited by the early settlers in the production of lime, a product used in construction and agricultural industries. A reconstruction of the site’s original foreshore and early 19th-century activity is presented in Figure.

Captain Richard Brooks was the first land-holder to develop a business on land associated with the site. He established a slaughterhouse just to the east of the site boundary. To the west of his waterfront, and within the site (Area 7), was an extensive intertidal sand flat. Brooks identified the potential of this foreshore as usable land, and staked his claim to it by demarcating and enclosing the space with timber fences. The northern boundary with Cooper & Levey’s land was extended out into the sand flat with a timber pailing fence, and presumably a similar structure delineated the southern boundary of his foreshore-claim from that of John Dickson’s land. The land-claim extended as far west as the low tide mark, where a pailing fence was erected to further illustrate this natural boundary with the harbour. This pailing fence may have also served to prevent sands, flotsam and detritus from the harbour being deposited, by the tides, within the enclosed land. Within these boundaries, a substantial east-west post-and-rail fence divided this large parcel of foreshore land in two. Though it may be possible that other ‘field’ divisions also existed, it

174 Macphail 2010: 29, Vol 3, Section 8.6. 175 Macphail 2010: 29, Vol 3, Section 8.6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 177 appeared that Brooks’ foreshore was bisected by this fence to form two east-west orientated ‘fields’. These were most likely to have been used as holding areas for the livestock waiting to be processed at his slaughterhouse.

The early use of the foreshore to the north was quite different. The roughly built-timber boat ramp constructed from the smooth and flat intertidal rock platform into the harbour zone marked a safe entry point and provided more stable access to the water for small vessels at high tide. Located just to the east above the high water mark was evidence for shell processing for lime production. Shell from the harbour is known to have been burnt to produce lime in the early decades of the colony. The shell resource of Darling Harbour was found in the Aboriginal middens that adorned the shoreline and also within extensive naturally formed ancient shell beds in the intertidal sand flats near the head of the harbour. The boat ramp and the remains of shell processing, while not conclusively linked archaeologically as contemporary, may both be evidence for the temporary or informal use of this land and its resources for lime production. Shell from the shell beds may have been dug out at low tide and loaded on to boats that could be floated to the shore with the incoming and high tide. The timber boat ramp may have been used to haul the laden vessels as close to the lime burning sites as possible. It was not possible to definitively link this use of the area to a particular owner or occupant of the land. What is clear from the archaeological evidence is that these activities pre-date Cooper & Levey’s purchase and mill complex development in 1825. They may be associated with the use of the land by James Smith, who appeared to advertise this land for sale in 1822.

The analysis of the shell and pollen revealed evidence for the changes to the environment resulting from the drastic changes in land use after British settlement in 1788. Land clearance, urban development, industry and manufacturing irrevocably altered the landscape and the harbour environment. Foreshore resources exploitation, pollution and sedimentation in the harbour caused a change in the natural shell species range. Shell from the Darling Quarter site confirmed that mud oysters died out and rock oyster became dominant, adapting to the changed environment in the harbour in the first few decades of the colony. Pollen from native species were in the majority in the sampled harbour sand deposits. Given the nature of the intertidal sand flat, these pollens may have been introduced prior to 1788 or may have been blown in from nearby areas still covered by native vegetation. Pollens from exotics, or introduced species, such as weeds and cereals, that indicate the changing local vegetation growth pattern and the new use of the area for flour milling, were also found within the sampled harbour sands.

There was very little evidence for harbour pollution from human waste and rubbish during this early phase of use. Ceramic, glass and animal bone fragments are typical as archaeological evidence of refuse disposal and rubbish dumps. Very few such artefacts were found within the foreshore deposits at this site. Just one of the environmental samples taken from the intertidal sands contained the human sewerage indicator Cloacasporites sydneyensis. This indicator features more in samples from later phases of use, reflecting an increased urban population and affect on the natural environment.

Land clearance and development caused erosion that led to a reported increase in sedimentation and shoaling in the harbour. This change in the harbour or foreshore environment was not obviously identifiable in the archaeological record. The sand bank material (context 8446 in Area 6 and 9570 and 9571 in TT60 in Area 9) identified as possibly forming a barrier that enclosed a freshwater lagoon-type feature could be interpreted as being the result of such changes in the natural harbour depositional processes. However, the 1802 plan of Sydney records the land adjacent to the site as being “open forest land”. This suggests that it was not cleared, or at least not fully cleared at this time, and therefore not likely to have been subject to undue or excessive erosion resulting in increased harbour sedimentation. In fact, the land surrounding the site was one

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 178 of the last waterfront areas in the southern part of the eastern harbour shore to be occupied, and developed, with some reclamation and construction not beginning until the mid 1820s, and intensive reclamation works not until the 1830s and 1840s. It may be the case that accelerated erosion and sedimentation within the foreshore and harbour was not occurring in this location, and at this early stage in the sites development.

Figure 3.4.32: Interpretation of the early foreshore activity including Brooks’ early property and field boundaries, a timber boat ramp and evidence for shell lime burning in Copper & Levey’s property. Evidence for a low sandbank during this phase may indicate that the freshwater lagoon still existed by the early 1820s. Harper’s 1822 plan has been used in the background, and this depicts the location of Brooks’ slaughterhouse and also illustrates that there was no recorded development on Cooper & Levey’s or Ramsay’s properties. Extract from Cryerhall, Plan 9.4 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 179

3.5 Barker’s Mill, Jetty and Mill Pond 1825 to 1870s

3.5.1 Introduction Barker’s mill was a significant business on the 19th-century Darling Harbour waterfront and one of its biggest employers. The mill played a key role in the development of Sydney’s food production and manufacturing industry. Its importance lies not only in being one of the earliest and most successful steam-powered flour mills in New South Wales, but also in its ownership and management by Thomas Barker. The mill was powered by the third imported steam engine in the colony; the first being at Dickson’s mill to the south of Barker’s, and the second at John Raines Darling Mills in Parramatta. During the late 1820s and 1830s Barker’s mill was a successful and lucrative flour mill, processing grains from the hinterland and supplying flour to a rapidly growing Sydney population. Following changing economic conditions in the 1840s, the mill diversified into textile production.1 Through the later 19th century, the mill complex was developed and adapted in response to both technological changes, business diversification, economy and market needs.

Thomas Barker owned and was closely involved with the mill until his death in the 1870s. Barker was not only a prominent manufacturer and engineer, but also a public figure and philanthropist in 19th-century Sydney. Barker arrived in Sydney in 1813 with his guardian John Dickson.2 He began his engineering and manufacturing career as an apprentice in Dickson’s flour mill. By the late 1820s, Barker had created greater competition in the milling industry with his construction of two windmills and the purchase of the newly built steam-powered flour mill just north of Dickson’s. In conjunction with developing his mills, he acquired large acreages of land where he grazed sheep and cattle. In the mid 19th century he became involved and well-known in public affairs. He was one of the early promoters of railways, involved at board level in a number of banking institutions, a member of the Legislative Council, active in the promotion of education and among other appointments, was a council member of the Sydney College and a trustee of the Sydney Grammar School, he was a founder of the Destitute Children’s Asylum and also a member of the Sydney Female Refuge Society.3

Previous archaeological investigations have recorded remains of Barker’s mill. During the major redevelopment of Darling Harbour in the 1980s, a surviving section of the original 1820s stone wall of Barker’s mill was dismantled and stored by the Darling Harbour Authority.4 Excavations in 2004 for the Cross City Tunnel uncovered further buried remains of the mill building.5 Sandstone footings, stone flagging, a water tank and a mill stone were among the remains recorded during these works.6

The site of Barker’s mill complex (Area 9) as depicted on the 1865 plan is mostly outside the Darling Quarter site; except for a section of the southern corner of the yard (Figure 3.5.1). The mill complex was initially much larger, and included a mill pond, and a jetty and wharf area that provided access to the harbour (Figure 3.5.2). The Darling Quarter excavation uncovered archaeological remains of these early 19th-century components of the mill complex. The southern section of the mill yard and pond was excavated as two separate areas; Area 9 Mill Yard and Area 9 Mill Pond.7 A section of the former jetty and wharf area was within the site, and was excavated as part of Area 6.8

1 Casey & Lowe 2006 Appendix 4: 9, History of Barker’s Mill written by Rosemary Annable. 2 Walsh 1966a. 3 Walsh 1966a. 4 Johnson & Parris 2008: 24. 5 Undertaken by Casey & Lowe in 2004. 6 Casey & Lowe 2006. 7 Trench Reports by Harrop for Area 9 Mill Yard (Vol 2, Section 7.6) and Cryerhall for Area 9 Mill Pond (Vol 2, Section 7.7). 8 See Area 6 Trench Report by Cryerhall et al (Vol 2, Section 7.2). ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 180

Figure 3.5.1: Barker’s mill (Area 9) as illustrated on the 1865 plan. The mill and yard is mostly located outside the site (dashed red) and basement excavation (purple). City of Sydney Archives, C&L additions.

Figure 3.5.2: Barker’s mill complex at the time of his purchase illustrated on Hallen’s 1828 sketch. The sketch shows the locations of the mill building, ponds and jetty. The property boundary is indicated with the dashed blue line. The site is outlined in red and the basement excavation in purple. SRNSW Reel 2628, C&L additions.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 181

Archaeological evidence of mid to late 19th-century changes to the mill complex was also recorded. In the 1840s, saltwater was drawn to the mill engines via a brick drain. Barker reclaimed land to the west and northwest of the mill, and as it was sold-off, a new boundary wall was constructed to delineate the mill yard from the adjoining developments. Also around the 1840s, measures were taken to manage excess water from the mill pond with the construction of an overflow drain and other flood management features. The mill pond did eventually become redundant and was backfilled by the 1860s as reticulated water became more widely available in the 1850s.9

The archaeological results associated with Barker’s mill, jetty and mill pond are described in detail in the Area 9 Mill Yard Trench Report, Area 9 Mill Pond Trench Report and Area 6 Trench Report (Vol 2, Sections 7.6, 7.7 and 7.2 respectively). The detailed site plans for these areas are Plans 10.7 to 10.16, 10.38 to 10.44, 10.61 and 10.62 (Vol 4, Section 10).

3.5.2 1825 Mill Development by Cooper & Levey Thomas Barker did not undertake the initial development and construction of the mill complex. The construction of the main mill building, mill pond, and associated wharf and jetty was undertaken as a joint venture by business partners Daniel Cooper and Solomon Levey. In 1825 Cooper & Levey purchased the large parcel of land between Brooks’ allotment to the south and George Ramsay’s to the north. Most of this original land grant (Area 9) is located to the east of the site (Figure 3.5.3). Within the site was a stretch of the original shoreline (part Area 9 and part Area 5) and associated intertidal sand flats (part Area 5 and part Area 6).

Figure 3.5.3: Cooper & Levey’s original land acquisition in 1825 (green boundary) as recorded on Harper’s 1822 plan with the site area divisions (blue lines) and the basement outline (purple) and site boundary (red). SRNSW with C&L additions.

9 Aird 1961: 5-6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 182

Cooper & Levey were the first of the site’s early grant holders to undertake extensive development of their waterfront allotment. To the east of the site and on terra firma, Cooper & Levey constructed the large five-storey sandstone mill building, and imported a steam engine to power the mill. A mill pond, or water reservoir, was constructed in the southern corner of the mill complex, close to the original shoreline (Figure 3.5.2, Figure 3.5.4). At the same time, the foreshore was transformed to provide wharfage and a jetty for the mill complex. The construction of the mill pond and the jetty involved substantial earthworks consisting of excavation for the mill pond and reclamation for the wharf leading to the jetty. This development appears to have been completed by mid 1827 when Thomas Barker purchased the mill and lands from Cooper & Levey.

Figure 3.5.4: Graphic showing the general location of the archaeological remains of the northeast jetty wall, surfacing and mill pond remains within the basement (purple outline). Also depicted is the modern culvert cutting through the southern part of Area 6 and 9, excavation areas (green), bedrock at high tide and a number of key test trenches. Extract from Cryerhall, Plan 9.5 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 183

Archaeological remains of the northeast part of the jetty structure were excavated and recorded (Figure 3.5.4). The jetty remains consisted of timber formwork for sandstone walling, bulk fills and partial surfacing. These were located within Area 6 and Area 9 (Vol 4: Plan 10.8, 10.11 to 10.15, 10.38 to 10.41, 10.60). The remains within the site represented the junction of the jetty with the original shoreline, and the made-ground that formed the associated wharf or yard area. The jetty extended west from the original shoreline and beyond the basement excavation area and therefore was not investigated as part of this project (Figure 3.5.8). It is likely the jetty structure projecting into the harbour consisted of timber piles and decking. The southern side or extent of the jetty from historic plan overlays was once located between the boundary of Area 6 and Area 7. However there were no archaeological remains, as the area was truncated by a large and deep modern culvert (Figure 3.5.8).

Archaeological remains of the southwest part of the mill pond were within the basement excavation area (Figure 3.5.4, Figure 3.5.9). These consisted extensive earthworks in the form of a clay bank, a substantial timber revetment and sediments within the pond. The remains of the jetty, with associated reclamation and surfacing, and the mill pond were recorded and described in the various trench reports and other documents as Phase 4 (mid 1820s to 1830s). The mill complex is also referred to in these reports as Barker’s mill (including the jetty and mill pond), even though Cooper & Levey were responsible for its initial development and construction. Thomas Barker purchased the newly constructed mill in 1827 and was involved in its operations until his death in 1873.

3.5.3 Cooper & Levey Background10 Cooper & Levey’s business partnership was by the late 1820s one of the largest and most successful in New South Wales, and so its origins deserve some brief comments. Daniel Cooper and Solomon Levey both arrived in New South Wales as convicts, in 1816 and 1815 respectively, and begun to establish themselves as successful merchants in their own right by 1820.11 The earliest reference to ‘Cooper and Levey’ appears in November 1820, when George Ison was found guilty of ‘embezzling various articles, the property of Messrs Cooper and Levy of Sydney’.12 However, the corresponding report in the Sydney Gazette does not refer to Levey, and identifies Robert Cooper as the victim of the crime, which occurred from the brig Campbell Macquarie, on which Ison was employed in June 1820.13 Yet this apparent discrepancy is resolved when it is realised that Robert Cooper, Daniel Cooper and Solomon Levey were all co-owners of the Campbell Macquarie, at least by 1822, if not earlier.14 It seems quite possible that at this time ‘Cooper and Levey’ referred to a partnership between Robert Cooper, Daniel Cooper and Solomon Levey.

Further references to ‘Cooper and Levey’ occurred over the next few years, suggesting that this partnership continued. Between April and August 1821, Cooper & Levey pursued a writ of fieri facias against Thomas Clarkson,15 implying that he had incurred unpaid debts against them. In October 1823, Solomon Levey refuted claims that Cooper & Levey were diluting gin sold to FE Forbes, possibly from the ship Actaean.16 Cooper & Levey may also have had business interests in Van Diemen’s Land at this time, as they pursued a writ of fieri facias in December 1823 against Espie & Clark, on property at Bagdad, north of Hobart.17 Again in December 1823, FE Forbes

10 This section has been written by Nick Pitt, Archaeologist, Casey & Lowe. 11 Bergman 1964, 1967 and 1968; Davidson 1966. 12 SRNSW Colonial Sec. Papers, Reel 6023, X820, p 19. 13 Sydney Gazette 25 November 1820, p 2(3). 14 Knight 1966; Bergman 1964: 403, citing Bigge Appendix, vol. 129, pp. 62712-4. The Campbell Macquarie had been advertised for sale by Joseph Underwood in Sydney Gazette 19 June 1819, p 2(3). 15 Sydney Gazette 14 April 1821, p 2(2), 2 June 1821, p 1(1), 9 June 1821, p 1(2), 23 June 1821, p 2(1), 14 July 1821, p 1(2), 21 July 1821, p 2(3), 4 August 1821, p 2(3). 16 Sydney Gazette 30 October 1823, p 2(3). 17 Hobart Town Gazette 20 December 1823, p 1(3). ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 184 claimed to have lent ‘Robert Cooper and Levy’ £372.18 According to Solomon Levey, defending himself against allegations of smuggling, Daniel Cooper, Robert Cooper and Levey in partnership bought 375 rolls of tobacco worth £9000 from Aspinall, Brown and Co. in May 1824. They then proceeded to sell this tobacco to individuals and traders in amounts ranging from 1 to 228 rolls.19 From these scant references, it appears that Cooper & Levey were involved in a range of general trading at this period, importing goods, or buying them wholesale and selling them on to settlers and traders. This picture is supported by a ledger of Cooper & Levey, covering 1823 to 1825, partially published in the 1960s.20 This ledger lists accounts for a large number of individuals prominent in the colony at the time, mostly buying a range of goods, from paint and iron, to striped shirts, to rice and salt.21

However, Cooper & Levey became particularly prominent after Daniel Cooper and Solomon Levey purchased the Lachlan and Waterloo Company in 1825, also known as the Lachlan and Waterloo Mills or the Waterloo Warehouse. This acquisition was announced in February 1826, with the contract finalised on 28 June 1825, with Cooper & Levey paying £4700 in Spanish Dollars at 5/- each.22 As part of the process of Cooper & Levey acquiring the Waterloo Warehouse, both merchants ceased to do business on their own, at their own premises, instead moving their concerns to the Waterloo Warehouse premises on George Street, Sydney.23 On 5 May 1826, Daniel Cooper and Solomon Levey finally signed a formal deed of co-partnership, each bringing £30,000 of joint stock into the concern.24

Daniel Cooper had been a co-owner of the Waterloo Company, a water flour mill named the ‘Waterloo Mills’ by Governor Macquarie, since it was constructed in 1820 by a partnership consisting of William Hutchinson, George Williams, William Leverton and Daniel Cooper himself.25 By April 1821, this partnership had grown to also include Samuel Terry and Thomas William Winder.26 Although this same advertisement strenuously asserted that the concerns of the Lachlan and Waterloo Flour Mills did ‘not extend to any other Branch of Trade or Merchandize whatsoever’,27 by May 1821 the company’s warehouse moved from the corner of George and King Streets to the corner of George and Market Streets, and the first mention was made of their promissory notes.28 By December of that year, the company was seeking tenders from stonemasons and bricklayers to erect substantial warehouses on their George Street premises.29 In June 1822, the Waterloo Company began issuing notes in dollar amounts,30 which were being

18 Sydney Gazette 1 January 1824, p 4(1). 19 The Australian 23 December 1824, p 4(1). 20 McCredie 196?; Note that Bergman 1964: 407 and the editor of McCredie 196? considered this ledger to be of the Waterloo Company, which was bought by Cooper and Levey in 1825. However this judgement seems influenced by a belief that the partnership of Cooper and Levey only formed in 1825 (Bergman 1964: 406), and although the words ‘the waterloo here commenced the firm of Cooper and Levey’ appear on folio 77 of the ledger (McCredie 196?: 24), a definite judgement does not appear possible without re-examining the original ledger, which could not presently be located in major archives. 21 See reproduced ledger pages in McCredie 196?. 22 Sydney Gazette 24 February 1825, p 4(3) and The Australian 24 February 1825, p 1(3); Bergman 1964: 406, citing Wentworth Papers Miscellaneous, p 160 (Mitchell Library). 23 The Australian 7 April 1825, p 4(4), 13 June 1825, p 4(4); Sydney Gazette 28 April 1825, p 1(5). 24 Solomon Levey estate papers A 5541, No. 1, Deed of copartnership betwen Daniel Cooper and Solomon Levey, Merchants, 5 May 1826, W. C. Wentworth (Mitchell Library). (Transcript available at http://www.sl.nsw.gov.au/discover_collections/society_art/jewish/business/levey/index.html (accessed 27/02/2012). 25 Sydney Gazette 30 September 1820, p 3(2). 26 Sydney Gazette 7 April 1821, p 4(1). 27 Sydney Gazette 7 April 1821, p 4(1). 28 Sydney Gazette 26 May 1821, p 1(2). 29 Sydney Gazette 15 December 1821, p 4(3). 30 Butlin 2002 [1968]: 158; Sydney Gazette 23 August 1822, p 2(2). ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 185 forged by the following August.31 The Waterloo Company continued trade with diversified milling, banking and general trading interests after it was purchased by Cooper & Levey.

Cooper & Levey were clearly associated with their allotment at Cockle Bay from July 1825, when their steam engine arrived from Liverpool, by the ship Lalla Rookh and it was reported that the engine ‘is to be put into motion in Cockle-bay, below the old burial-ground, with as little delay as possible’.32 However, the importation of this steam engine appears to have been a well planned affair. An engineer, William Lowe, came out with his wife and two children on a five-year contract, accompanying the steam engine.33 The steam engine itself appears to have been imported by Warham Jemmett Browne following the order of Daniel Cooper,34 and as Browne had left in July 1824, before returning on the Lalla Rookh,35 it seems likely that Cooper had ordered the steam engine a little over a year in advance. This suggests that Daniel Cooper alone, the Waterloo Company, or Cooper & Levey acting together, may have had an association with the allotment before July 1825.

In any case, Cooper & Levey quickly built a mill to utilise the engine. In July 1825, Daniel Cooper requested ‘One Miller, One Carpenter, One Blacksmith, One Stone Mason or Bricklayer, Twelve Strong Bodied Labourers’ to help erect the steam engine.36 Construction was well under way by October 1825, when a new building on Cockle Bay belonging to Cooper & Levey was described as ‘conspicuous’ by the Sydney Gazette.37 The mill became operational in December 1825.38 In September 1826, The Monitor printed a detailed description of Cooper & Levey’s Cockle Bay premises: Our reporter gives us the following description of Messrs. Cooper and Levey's new stores and steam engine, recently erected in Cockle Bay. —“The building which is of stone, is raised to the height of five floors or stories, each capable of containing two and three thousand bushels of grain with convenience, leaving ample room for working, &c. The engine while it performs the operation of grinding and dressing at the same time, and smutting when required, conveys the water from a spacious newly-cut reservoir into the boiler, by means of a spindle and wheels, which is constructed to work as high as the roof of the building. Manual labour is thus saved in the conveyance of the wheat to and from the hopper. A circular chain with additional links attached at intervals, passed from the ground to the top, through a trap-door in each story; this being in continual motion, similar to the chain of a smoke-jack, either draws up the wheat or lowers it, or both at the same time if necessary. With one pair of stones, one thousand bushels of wheat can be ground, but the engine is capable of turning three pair. A commodious wharf is about being constructed at the back of the building, for which the locality of the situation offers many advantages”.39

31 Sydney Gazette 23 August 1822, p 2(2). 32 Sydney Gazette 7 July 1825, p 2(3); cf The Australian 7 July 1825, p 4(4), which also located the prospective site of the engine as being Cockle Bay. 33 Sydney Gazette 7 July 1825, p 2(1), 2(3); The Australian 7 July 1825, p 4(4); 1828 Census, National Archives Copy, HO 10/25, p 140; Casey & Lowe 2002: 6-7. 34 Sydney Gazette 31 October 1825, p 4(1), 7 July 1825, p 2(1), 27 October 1825, p 2(4). 35 Sydney Gazette 10 June 1824, p 3(2), 17 June 1824, p 1(2); 15 July 1824, p 2(1), 7 July 1825, p 2(1). 36 SRNSW Colonial Sec. Papers, Reel 6062, Item 4/1782, p 73. 37 Sydney Gazette 6 October 1825, p 3(4); cf Solomon Levey’s memorial dated 21 October 1825, which also mentions the building under construction, SRNSW Colonial Sec. Papers, Reel 6062, 4/1782, p 107. 38 Sydney Gazette 19 December 1825, p 1(4); The Australian 29 December 1825, p 3(2). 39 The Monitor 15 September 1826, p 6(3). ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 186

A jetty associated with Cooper & Levey’s mill appears in historical plans from around 1828-30.40 The ‘wharf’ referred to in The Monitor report would have been useful for Cooper & Levey, as it would allow grain to be brought for milling by water. It would have also been useful in providing wharf facilities for vessels owned by Cooper & Levey in partnership or individually. Solomon Levey had interests in several ships in the 1820s, including the schooner Mary, the brig Campbell Macquarie, a former government sloop HMS Snapper and a locally built schooner Elizabeth which was active in sealing expeditions.41

Cooper & Levey sold their Cockle Bay mill to Thomas Barker in mid-1827. Thomas Barker had arrived in Sydney in 1813 as an apprentice to John Dickson, who constructed the first steam flour mill in Sydney.42 In 1824, Barker requested a town allotment in a memorial, and received the reply that he could have any vacant allotment of his choice.43 He appears to have chosen a corner block on the southwest corner of Sussex and Bathurst Streets, adjoining the allotment where Cooper & Levey established their mill. In any case, Barker was based at this location by September 1826, when advertised that he would buy wheat or sell flour ‘at his Residence in Sussex-street, next to Messrs. COOPER and LEVEY's Steam Engine’.44 A reported robbery in December 1826 also confirms that Thomas Barker was ‘living near the Steam Engine’ at the time.45 By 1826, Barker had also begun to develop his own milling interests, with the completion of a windmill at Darlinghurst.46 The first report of Cooper & Levey selling their mill to Barker was in May 1827, when The Australian reported the sale for £5000.47 The actual conveyance in the old system land records was dated 16 June 1827,48 while a schedule of deeds in Thomas Barker’s legal papers lists Barker taking out a mortgage from Cooper & Levey for £5000 plus interest on 18 June 1827.

Before the sale of Cooper & Levey’s mill and steam engine was finalised, Daniel Cooper received a formal grant of the land where it was situated, in which it was described as: All that Piece of Parcel of Land, lying and situate on the West Side of Sussex Street, in the Town of Sydney containing by Admeasurement 482 Rods, Bounded on the North by Raine & Ramsays Allotment to the South boundary of Barkers Allotment. And by that Allotment to the Street in the East by Sussex Street 575 links, on the South by Brook’s Allotment to the water and on the West by the Water of Cockle Bay.49

Soon after Barker received a formal lease for the 76 perches of land at the southwest corner of Sussex and Bathurst Streets, which he was already occupying. A parish map of the Parish of St Andrew (Figure 3.5.5) shows Daniel Cooper’s grant, as well as property boundaries for surrounding allotments. This map has no date, but it is likely to have been annotated between mid-1826, when Darling Harbour was named,50 and mid-1831, when the land was granted again to Thomas Barker.51

40 SRNSW Surveyors’ Field Books No. 347, A Hallen, 2/5195, Reel 2628. 41 Bergman 1964: 403-404. 42 Walsh 1966a. 43 SRNSW Colonial Sec. Papers Fiche 3076, Item 4/1836A, No.32, pp.151-152; Reel 6012, Item 4/3510, p 656. 44 The Monitor 15 September 1826, p 1(1). 45 The Australian 6 December 1826, p 3(4). 46 Walsh 1966a; SRNSW Colonial Sec. Papers, Fiche 3119, Item 4/1840B, No.30, pp. 189-96; Fiche 3269, Item 9/2740 p.4; Sydney Gazette 24 June 1826, p 3(4); The Monitor 15 September 1826, p 1(1); SRNSW NRS 13836; Item 7-449, Reel 2561. 47 The Australian 16 May 1827, p 3(2). 48 Casey & Lowe 2002: 5, citing Bk 11 No 420 Conveyance dated 16 June 1827 (Land & Property Information). 49 SRNSW NRS 13836; Item 7/482, Reel 2704, grant dated 1 June 1827. 50 Sydney Gazette 24 May 1826, p 2(2), 14 June 1826, p 2(3); The Monitor 16 June 1826, p 5(2). 51 Sydney Gazette 31 May 1831, p 1(2); Sydney Gazette 12 April 1832, p 2(2); Attested copy of 1831 land grant, dated 19 October 1831, in Barker papers A 5398/3, No. 29, Abstract of title of Thomas Barker Esq to Land in Bathurst and other Streets Sydney (1847). (Mitchell Library); Register of Land Grants and Leases, SRNSW NRS 13836, Item 7-473, Reel 2700. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 187

Figure 3.5.5: Detail of Map of the Parish of St Andrew, AO Map 272. No date, but likely to have been annotated between mid-1826 and mid-1831. NSW LPI Historical Land Records Viewer

3.5.4 The Jetty An article on Cooper & Levey’s mill dated to September 1826 refers to the ‘commodious wharf’ that was in the process of being constructed.52 The jetty structure was first illustrated on Ambrose Hallen’s sketch dated between 1828 and 1830 (Figure 3.5.6). It is also recorded on the 1833 city survey (Figure 3.5.7). The jetty stretched from the original shoreline westward into the harbour. Hallen’s sketch includes the measurements of the jetty and a later sketch plan of the area, dated to 1853 though likely drawn earlier, records the length of the jetty as 5.40 chains (356.4 feet or 108.6m).53 The remains of the northeast junction jetty where it met the rocky shoreline were found within Area 6 (Figure 3.5.8, Figure 3.5.9, Vol 4: Plan 9.6, Plan 10.8, Plan 10.60). This part of the jetty was constructed with sandstone walling within timber formwork (Figure 3.5.10, Figure 3.5.11). Behind the walling, bulk sand and clay-rich fills were used to raise the ground level and create an intermediate area between the land and the timber jetty that projected far into the harbour. This area was essentially a wharf, and aimed to provide stable, level, open yard space for stockpiled goods, either off-loaded from or awaiting loading onto ships berthed at the jetty. Patchy remains of the surfacing of this large reclaimed area were recorded in Area 6 and Area 9.

52 The Monitor 15 September 1826, p 6(3). 53 Sketch plan showing southern end of Darling Harbour and adjacent streets, 1853, Dixsons Map Collection, Cb 85/14. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 188

Figure 3.5.6: Ambrose Hallen’s sketch of the mill complex including the jetty dated to 1828-1830, when the mill was owned by Thomas Barker. The sketch notes survey details about the jetty and the mill buildings. Note that the building illustrated to the north of the jetty is an enlarged version of the mill that is actually located directly east of the jetty. The inset shows a detail of the jetty sketch. A Hallen 1828-1830 FB 347 Reel 2628 SRNSW.

Figure 3.5.7: Detail from the 1833 survey showing the jetty and how it connected to the original shoreline. This survey also contains later additions (written and drawn in red in the 1880s). Thomas Barker owned the mill and lands by this time. . Historical Atlas of Sydney, City of Sydney Archives.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 189

Figure 3.5.8: Location of the archaeological remains of the northeast part of the jetty and mill pond overlaid on the 1833 survey showing the full extent of the jetty. The basement is outlined in purple and the site boundary is in red. Tumbalong Park is now located over the western end of the jetty. City of Sydney Archives. Extract from Cryerhall, Plan 9.6 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 190

Figure 3.5.9: Extract from the detailed excavation plan of Area 6 and Area 9 Phases 1 to 4. This shows the location of the jetty wall, the mill pond and the test trenches and sections that record the surfacing and jetty infill. It also shows the location of the earlier boat ramp that was buried by the jetty fill during its construction in the mid 1820s. Extract from Plan 10.60 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 191

Figure 3.5.10: Extract of the detailed plan of Area 6 showing the jetty wall, formwork, jetty fills and the wall’s relationship with the bedrock above high tide level. The red arrows and numbers indicate the location of section and profile drawings. Vol 4: Plan 10.8.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 192

Jetty surface Jetty wall

Reclamation

Harbour floor

Jetty fill

Figure 3.5.11: View of the jetty wall remains from the shore with remains of the jetty in-fill and surfacing. The timber slipway was revealed below the jetty fill. The harbour floor is to the right with reclamation (Phase 5) overlying. View to west. Scale 1m.

3.5.4.1 Formwork Timber formwork (context 8460) was used in the construction of the northern sandstone wall of the jetty (Figure 3.5.12, Figure 3.5.13 and Vol 4: Plans 10.8, 10.12). Almost 20m of formwork was recorded in plan, and exposed in more detail in three test trenches (TTs 28, 29 and 30) along the northern side of the jetty wall (Locations of TTs in Vol 4: Plan 10.8). The formwork consisted of closely placed timber planks driven into the harbour sands (Figure 3.5.14). The planks were roughly formed by longitudinally-split logs up to 1.5m in length. Bark had been removed but otherwise the logs were not milled, and in section they were either rectangular or triangular. The base end of each piece of timber was roughly hewn and shaped to a point (Figure 3.5.16). Generally, the timbers were driven into the sand with the cut side facing towards the water. In TT28 there was evidence for horizontal brace timbers on the external face of the shoring (Figure 3.5.15).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 193

Figure 3.5.12: Exposed timber formwork (context 8460) with the sandstone jetty wall (context 8366) beyond. View to the south. Scale 1m

Three timber samples were sent for species identification.54 Two of these (#162 and #166) were identified as Eucalyptus pilularis, or blackbutt. This hardwood was abundantly available in the early years of settlement, and is likely to have grown within one kilometre of the site.55 The availability of this species at the time of the jetty construction suggests that locally sourced timbers are likely to have been used.56 Also identified as part of the formwork was Eucalyptus oblique (#165), commonly known as messmate or messmate stringybark. Like blackbutt, messmate is a hardwood which is used today in construction and flooring.57 Messmate does not naturally grow in the Sydney area, but is endemic to the Tablelands districts of New South Wales.58

The top of the formwork was located within the tidal range, between RL 0.07m and -0.32m. In general, the tops of the timbers were just above the upper layer of harbour sand. However in TT30 for example, several layers of sands and silts were deposited over the timbers during the lifespan of the jetty (these are discussed below). At the eastern end of the jetty, closest to the shore, the top of the jetty wall would have been about 800mm above the top of the formwork, and at the western limit of excavation, the wall would have been about 1.3m above. The timber formwork enabled the sandstone wall of this part of the jetty to be built from below the water table. A construction trench though harbour sands and below high and into low tide levels would require a retention system to avoid trench wall collapse. It would appear that the timbers were driven into the sand first, and then the trench for the sandstone wall excavated. The timber formwork provided a stable trench and also restricted the amount of tidal water entering during the construction process.

54 Identification by Dr Jugo Ilic, KnowYourWood. Vol 3, Section 8.8. 55 Benson & Howell 1990: 42-44. 56 Kuiters 2010: 9, Vol 3, Section 8.8. 57 Bootle 2005: 287, 310. 58 Bootle 2005: 310. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 194

Figure 3.5.13: Detail of the top of the formwork (8460) and the stone wall behind. The right of the frame shows the amount of harbour deposits and accumulation against the wall. View to the south. Scale 1m.

Figure 3.5.14: View of the harbour-side of the jetty wall with the top of the timber formwork exposed. The oyster shells adhering to the stone just above the top of the formwork indicate that the high tide water levels rose to cover the formwork. View to southeast. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 195

Figure 3.5.15: Horizontal timbers used as extra bracing on the outside of the formwork. These were revealed below some harbour sand in TT28. View to the south. Scale 1m.

Figure 3.5.16: Group photo of the timber samples from context 8460 retrieved during bulk excavation. Note the pointed ends which would have aided the timbers to be driven into the sand. From left to right: sample #163 (24B), #162 (24E), #165 (24A), #164 (24F), #159 (24C) and #166 (24D). Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 196

3.5.4.2 Sandstone Jetty Wall The north jetty wall, or seawall, (context 8366) was constructed within timber formwork from below the water table (Figure 3.5.17, Figure 3.5.18). About 20m of the wall was within the basement excavation area (Vol 4: Plan 10.8). The remains began 6.5m west of the shoreline (bedrock 8474) and were located at around the high tide level (RL 0.5m). The easternmost point of the wall did not survive well, and consisted of just a linear arrangement of rubble sandstone. The collapse and damage of the shoreward part of the wall had occurred during the 19th century, and likely during the reclamation of this area in the late 1830s. The underlying geology meant that the foreshore sand deposits in this area were quite stable and firm, and this meant the landward end of the wall had been built directly onto the sand deposits.

As the jetty wall extended further into the intertidal zone, towards deeper harbour waters, it became more substantial (Figure 3.5.19). The timber formwork and more formal walling began about 10m southwest from shore and at about the median tide height (RL 0m). From this point the wall was constructed using a combination of neatly cut and faced rectangular sandstone blocks, rubble and roughly cut blocks (Figure 3.5.20, Figure 3.5.21). The neatly cut and finished sandstone was arranged in formal courses and formed the wall-face on the harbour side. The rubble and roughly cut sandstone was not formally coursed and acted as reinforcement on the interior side of the wall. In cross-section, the wall had a wider base with a narrower top (Figure 3.5.17). This form gave the wall strength to retain the bulk fills deposited behind it to create the working surface for the large wharf area attached to the jetty (Figure 3.5.22, Figure 3.5.23).

A machine-excavated trench centrally located within the wall revealed that the lower course was constructed with neatly dressed blocks aligned in a single row on the northern side. Every third stone was oriented perpendicular to the orientation of the jetty wall, creating a saw-toothed face on the southern side. The southern side was then packed with roughly-cut blocks, at least doubling the width of the wall. This method was reproduced in the upper courses, with smaller elements used on the southern side. The jetty wall was at least three courses deep at the shoreward end and is likely to have been considerably deeper as the wall extended into the harbour (Figure 3.5.23). The junction between the wall and the timber part of the jetty was not within the excavation area but it may have extended a further 10m to the west, corresponding to the junction illustrated on the 1833 survey and the 1828 sketch (Figure 3.5.24, Figure 3.5.25). This junction is likely to represent the point where it was no longer possible to construct sandstone walling due to the greater depth of the water.

Though the foreshore sloped westward towards the harbour the wall coursing remained level and horizontal. Above the low water mark the dressed sandstone blocks had been removed, probably salvaged for recycling prior to the later 1830s reclamation works. With the surviving courses there was no apparent bonding agent, and the same was the case for the internal rubble buttress walling. Tool marks created a pleasant pattern and texture on the face of the dressed stones. Within the tidal range, at around RL -0.3m, or just above low tide levels, was a line of rock oysters adhered to the wall (Figure 3.5.21). The typical habitat of Rock oyster is within this tidal range, and their presence on the sandstone wall further demonstrates this species adaptability to the changing natural environment.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 197

Figure 3.5.17: East-facing section (Section #30) through the sandstone jetty wall constructed within timber formwork. It shows how the stone face of the jetty wall was previously removed. Also illustrated are the harbour sands, low tide mark and the infill and surface level of the jetty. Extract from Vol 4: Plan 10.12.

Figure 3.5.18: East-facing sections (Section #28 and 29) through the sandstone jetty wall, showing also the timber formwork and harbour sands. Extract from Vol 4: Plan 10.12.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 198

Figure 3.5.19: Remains of the northern jetty wall with ashlar stone construction on the harbour-face and rubble sandstone on the interior, visible in the background. View to west. Scale 1m.

Figure 3.5.20: Detail showing the stones facing the harbour neatly cut and with chisel mark surfacing on the face. View to south. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 199

Figure 3.5.21: The jetty wall consisted of ashlar blocks on the harbour face. These were robbed-out before the wall was covered by the reclamation fills in the 1830s. On the internal side of the wall were large rough-cut sandstone blocks. View to the south. Scale 1m.

Figure 3.5.22: The remains of the jetty wall showing the rough-cut sandstone used in the interior abutted by the black occupation deposit over the jetty surface. The harbour-facing stones were robbed-out before being covered with the later reclamation fills, as seen in the section where orange clays cover the wall remains, behind the scale. View to east. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 200

Figure 3.5.23: Recording the jetty remains. The jetty wall was solid sandstone with wooden timber formwork along the harbour side. Grey coarse sands formed the deposits on the harbour-side of the wall. This photo also illustrates the deepening wall courses further into the harbour zone. View to west.

Figure 3.5.24: The junction of the stone wall and the timber part of the jetty is indicated with the blue arrow and is located outside the basement excavation area (purple). 1833 plan with Cryerhall additions ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 201

Figure 3.5.25: Hallen’s c.1828 sketch of the jetty illustrates the stone seawall part of the jetty attached to the timber structure projecting into the harbour (arrow points to the junction). North is at the top. SRNSW.

3.5.4.3 Jetty Infill59 A large amount of fill was used to reclaim and build up a surface level behind the jetty wall (to the south). These fills also sealed the remains of the timber boat ramp discussed in the previous section (Section 3.4.4.2). These bulk fills were sampled and recorded mostly in test trenches and in sections throughout the northern part of Area 6 (Vol 4: Plans 10.8, 10.13, 10.14. 10.15). Around 100 separate and distinguishable dumps of fill were identified within the test trenches and sampled areas. The majority of these were not assigned context numbers but were described in detail in the field on the sections drawings (see also plans in Vol 4, Section 10). A small number of fills and groups of fills were assigned context numbers (Table 3.5.1). The individually recorded jetty fills within the test trenches and sections have been described in detail in the Area 6 Trench Report (Vol 2, Section 7.2).

Table 3.5.1: Summary of report illustrations and site plan references with associated context for the jetty infill material. Detailed Plan Vol 4: Section 10 Site Plan Reference Contexts Plan 10.13 Section 23 None assigned. Section codes only Section 25 8426, 8395, 8396 Section 27 8479, 8480, 8477 Plan 10.14 Section 20 8491 Section 24 8372, 8373, 8374, 8385 Plan 10.15 Section 16 None assigned. Section codes only Section 21 None assigned. Section codes only Not report graphic Section 26 8487, 8486, 8466, 8465, 8464, 8463, 8462 No report graphic Near TT22 8470, 8488, 8351, 8472, 8473

59 The infilling of the jetty to create the jetty for the mill was also the first phase of reclamation at the Darling Quarter site. This and the artefacts from the fill material are also discussed in the next section - Section 3.6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 202

The material used to raise the ground level behind the jetty wall was predominantly sand of various colours and consistencies, but mostly grey and pale brown medium to coarse-grained (Figure 3.5.27, Figure 3.5.28). Many fills were also mottled with silty material, clay chunks, differing colours of sand such as white or buff, with inclusions of twigs and other organic matter (Figure 3.5.26). Rubble sandstone of all sizes was also present in some cases. A typical example of a one of these jetty fills is context 8375 recorded in Section 24 (Vol 4: Plan 10.14) and sampled (sample #77) for pollen analysis. This context represented several bands of sands ranging from pale to dark grey and some with inclusions of blackish silty sand (Figure 3.5.27). Context 8375 displayed tip lines from south to north and was located just to the south of the jetty wall. The tip lines are also illustrated in Figure 3.5.29.

All the sandy fills appeared to be redeposited natural sands, and very likely had been dredged or excavated from the surrounding harbour foreshore. Pollen analysis from context 8375 appears to confirm this.60 The sample consisted of 39% casaurina, 30% raspwort (native herb) and 11% eucalypt. Also within the sample was evidence of swamp selaginella and fern, species common to damp organic rich soils such as swamp environments. The microflora lacks definite exotic pollen taxa such as cereal.61 The presence of two large flour mills, first Dickson’s in 1815 and then Barker’s from the mid 1820s, in the immediate vicinity of the site has resulted in cereal pollens being present in all the post-1800 sampled contexts but context 8375. The absence of the cereal pollens in this instance can be explained by this context being a redeposited natural (pre-1788), dredged or excavated from the harbour floor and used within the 1825 construction of the jetty.

Other types of material used included crushed and rubble sandstone, and dense yellow to orange clays. The local sandstone bedrock gives rise to dense orange and yellow clays, and these too were redeposited as fills for the jetty construction. A typical example of a clay fill is context 8462 in TT22 and Section 36 (Figure 3.5.30). This east-to-west tipped fill consisted of bands of yellow and reddish orange clay with frequent inclusions of decaying ironstone (forming reddish colour) and lenses of pale grey sands. Context 8395 serves to represent the crushed sandstone type fills. This context was recorded in Section 25 and was located directly above the bedrock platform 8448 (Figure 3.5.31). It consisted of yellowish brown coarse sand with crushed sandstone fragments.

There was a notable absence of artefacts from the test trenches through the jetty fills. Overlaying the bedrock 8448 and the timber ramp were two buff-coloured sand and crushed sandstone fills; contexts 8395 and 8426. A small number of ceramic and glass fragments were retrieved from both fills. The glass (2 MIC) and ceramic (2) from contexts 8395 are consistent with the mid 1820s construction date for the jetty.62 The ceramic from context 8426 consisted of a fragment of a fine earthenware blue transfer-printed poe and breakfast cup also dating from c.1830.63 This is a slightly later date than the construction of the jetty, and it is likely do to mixing of later reclamation fills during the mechanical excavation of the fill. The lack of artefacts is not surprising considering the origin of the fills as redeposited natural material from the surrounding harbour foreshore and terrestrial locations; both relatively undeveloped and sparsely occupied at this time.

The fills were deposited over the intertidal sandstone platform and sand flat. From the areas tested and sectioned, the fills were deposited from the south and tipped northward toward the jetty wall (see Figure 3.5.29). Also noted were tip lines from east-to-west (Figure 3.5.30). Testing in the southern part of the jetty was not possible as this area contained a large diesel tank and required extensive remediation of the surrounding contaminated soils. Therefore it was not possible to gain

60 See Macphail 2010 pollen analysis report in Vol 3, Section 8.6. 61 Macphail 2010: 20, Vol 3, Section 8.6. 62 Casey & Lowe artefact database. 63 Ward 2011: 24, Vol 3, Section 8.1. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 203 any information about how this part of the jetty was reclaimed. The fill created a large flat yard area that attached the land to the timber part of the jetty projecting into deeper harbour waters. In the east area the depth of the fill was as little as 200mm over the bedrock, and this increased to about 1.2m in depth at the western limit of the basement excavation. The surface of the jetty is discussed in the next section (Section 3.5.4.4)

Figure 3.5.26: Detail of the east-facing section of TT21 (Section #21) showing the bands of sand mottled with decayed clay that formed the infill behind the jetty wall. The red arrow points to a band of crushed whitish yellow sandstone and shale (context 8439) that formed the raised surface of the jetty. View to west. Scale 1m.

Figure 3.5.27: Section #24 through the redeposited grey harbour or foreshore sands used to fill-in behind the jetty wall and raise the ground level above the high water level. View to southwest. Scale 1m. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 204

Figure 3.5.28: Southern end of Section #20 showing the grey sands and crushed sandstone fills used to raise the ground level within the jetty. The base of the section and in the foreground is the top of the naturally deposited sands of the foreshore just above the mean tide level (around 0.1m) and the yellow arrow points to the material that formed the jetty surface about 1.2m above. View to west. Scale 1m.

Figure 3.5.29: Extract from combined section #20 and #24 drawing to illustrate the tip lines of the individual fills and a detailed description of each. East-facing section. Extract from Plan 10.14 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 205

Figure 3.5.30: North-facing section (Section #36) of an east-west test trench (TT22) through the jetty infill showing orange clay-rich fills as well as grey sands. These fills were tipped from east to west. View to southeast. Scale 1m.

Jetty fills

8448 8446 8444

Figure 3.5.31: Crushed sandstone and coarse yellow sand was also used as jetty infill. Section #25, west- facing, with bedrock and naturally deposited foreshore sands below the jetty fill. View to east. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 206

3.5.4.4 Jetty Surface A large area of land was reclaimed from the foreshore during the construction of the jetty (Vol 4: Plans 10.8, 10.38, 10.60). The reclamation for the jetty created a yard area between the main mill complex and the jetty projecting into the harbour. To make this new land more usable as a wharf, and protected it from weathering and erosion, it was capped with semi-formal surfacing. This surface material was recorded in a number of test trenches and sections across Area 6. It was also exposed in plan in a small area on the western edge of excavation, and another small area on the southern boundary with Area 7. The surface appeared in Section #s 16, 20, 21 and 27 (Vol 4: Plans 10.13. 10.15).

Over most of the Area 6 (north, west and east), the surface was recorded with two context numbers; context 8439 (including context 8440) and context 8478. These numbers represent the majority of the surfacing evidence. Most of the surfacing consisted of a layer of very compact crushed white sandstone within a white clay matrix (Figure 3.5.26, Figure 3.5.32, Figure 3.5.33). It was also recorded as being formed with crushed pink sandstone (context 8435 in section #20, Figure 3.5.34). The surfacing material varied in thickness from 10mm to 210mm. The level of the surface at its eastern limit (context 8478) was RL 1.38m (highest recorded level) where it existed over fill that overlay bedrock. At the western limit of the excavation, further into the harbour zone, it was at RL 0.88m (lowest recorded level). These levels demonstrate that the jetty surface had a gradual slope from the shore westwards into the harbour.

Truncated by numerous 20th-century services and piles was a small patch of dark brown silty material (context 8339) over the surface of compacted stone and fill. This was identified as an accumulated deposit over the jetty surface. It contained frequent inclusions of wood chips and what appeared to be sawdust. It was located toward the western extent of the jetty remains within Area 6 and represents the accumulation of material over the jetty surface between the late 1820s and late 1830s, prior to Barker’s extensive reclamation works. Within this deposit was a British halfpenny, George IV, dated between 1825 and 1827 (cat #80445).64 The coin was very worn and it may have been lost quite soon after the jetty construction (c.1827) and been subject to continuous trample from the wharf traffic to and from the mill.

Located to the southeast of the jetty wall remains, was a small patch of paving (context 9265) that once formed part of the initial jetty surface (Figure 3.5.35). It was recorded during the excavation of Area 9 (Vol 4: Plans 10.38, 10.60). The paving was a much more formalised surface material than recorded in Area 6. Though it was not physically linked to the other surface material, the pavers were laid directly over the bulk sand fills used to reclaim the land. The paving was also cut by a later drain (context 9204) that was dated to the 1840s and Barker’s ownership of the mill. The pavers were small flat and irregular-shaped pieces of sandstone. The surface level was at RL 0.88m, a similar level to that recorded further to the northwest in Section 16. The paving did not exist to the north and it was not clear if it ever did. During the excavation, this area was extremely soft and ‘boggy’ at this level, creating great difficulty in using the machine in this area. It appeared that the bulk sand fills had not been consolidated to any great extent prior to further land filling in the 1830s and 1840s by Barker. It may be that if more formal surfacing, such as pavers, had existed here, they may have been removed during later 19th-century development works.

Accumulated over the pavers was a thin wash of dark brown silt material (context 9266). One interesting artefact from this material was a lead token or counter with a pelleted starburst relief design on the single die-struck obverse (cat #80334).65 Similar lead tokens or counters have been found in small numbers on many pre-1840s sites in Sydney and Parramatta. Their exact date of

64 Casey & Lowe artefact database. See also Vol 3, Section 8.2, Miscellaneous Report. 65 Casey & Lowe artefact database. See also Vol 3, Section 8.2, Miscellaneous Report. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 207 manufacture is not known. These items always have a limited range of simple poorly-rendered designs, many of which are variants of a spoked wheel. Their purpose is not fully understood and it is thought that they may have (also?) been used as work jettons given each day and then exchanged for real wages perhaps at the end of the week.66 It is likely this token was dropped by a worker at Barker’s mill.

Figure 3.5.32: Compacted crushed white sandstone surface 8478 (red arrow) in Section 27. This was located on the eastern side of Area 6 where it joined with Area 9. In the foreground is the gentle sloping weather-eroded sandstone shore 8474. The sand and crushed sandstone fills below were deposited to ‘fill in’ behind the jetty wall and raise the levels above the high water mark. View to south. Scale 1m.

Jetty Figure 3.5.33: Section of the Wall jetty surface excavated in plan. The compacted crushed white sandstone and clay deposit (context 8439) that formed the surface (arrowed) overlay the jetty infill 8440 that also formed part of the surface (below the scale). Black silt accumulated over this surface (top of photo). The jetty wall is on the left of the photo. View to northeast. Scale 1m.

66 Stocks 2013, Vol 3, Section 8.2. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 208

Figure 3.5.34: Section #20 showing a compacted layer of crushed pinkish sandstone (blue arrow), context 8435, that formed part of the surfacing of the jetty. It overlies deep sand and sandstone fills that ‘filled in’ the jetty and raised the surface above the harbour floor (foreground and base of section) and above the water. View to west. Scale 1m.

Figure 3.5.35: To the south of Area 6, near the boundary with Area 7, was a small area of rough sandstone paving 9265. This surface was located above redeposited sands, similar to the jetty fills recorded in the north of the Area associated with the jetty wall. It was cut by the mill pond overflow drain (1840s?) and sealed by levelling fills that were introduced to the area after reclamation works in the 1840s. View to east. Scale 1m. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 209

3.5.5 The Mill Pond A large mill pond was created as part of Cooper & Levey’s mill development in the southwest of the complex (Figure 3.5.6). It was located above the high tide mark and is at its closest to the harbour shoreline in the southeast corner. Its southwest corner was excavated through naturally deposited sand and silts, and bulk fills appeared to have been used to reclaim land behind the jetty wall. The construction of the mill pond in this part of the mill complex was largely determined by the location of the creek mouth. Also, the natural attributes of this general location had previously facilitated the formation of a lagoon that at one time filled with freshwater from the creek. The archaeological evidence for such a landscape feature is discussed in Section 3.2.5.5. It was not possible to fully investigate the mill pond structure or its relationship with the surround natural landform due to the presence of potential acid sulphate soils (PASS).

The mill pond is first recorded on Hallen’s 1828-1830 sketch of the area, but is not included on the 1833 survey (compare Figure 3.5.6 and Figure 3.5.7). On Hallen’s sketch the pond is located in the southwest of the mill complex, just to the east of the jetty, and above the high tide level (Figure 3.5.36). The sketch also illustrates the pond as two separate structures located to the southwest of the mill building and it is fed by a stream from the east and to the west, and perhaps a watercourse or drain is depicted exiting into the harbour. There is also a channel or path drawn on the plan from the western side of the lower pond heading north toward the western side of the mill building. No information is given regarding the mill-pond structure apart from dimensions. The sides of the pond are drawn with straight edges.

Figure 3.5.36: Hallen’s field book from 1828-1830 showing two separate mill ponds (red arrows) fed by a stream from the east (blue arrow) and perhaps with an exit channel to the west and maybe a path or a drain/channel to the mill building. SRNSW.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 210

The northwest corner of the southern pond, as illustrated in Hallen’s sketch, was within the excavation area and was included as part of the excavation of Area 9 (Figure 3.5.37; Vol 4: Plan 10.38). The archaeological evidence from the site provided further detail about how the pond was constructed and what it may have looked like. The construction of the mill pond and the jetty both required large-scale earthworks and it would appear from the archaeological evidence that they were undertaken concurrently. The western edge of the mill pond was built into the bulk sand fills used to create the raised jetty surface. A timber revetment, consisting of long timber planks driven into the jetty fill and natural foreshore sands, formed the northern and western pond edge. Surrounding this revetment was a clay bank. This provided extra support to the exterior of the pond. The western edge was further consolidated and protected from the tidal influence by a deep and narrow clay-filled cut that would have acted as an impermeable barrier to the tidal fluctuations. Located at the southern extent of the excavation area, and part of the mill pond structure, was a brick and timber structure that may have served as a base for a pumping system, such as the one referred to in the 1826 description of Cooper & Levey’s mill.67

Figure 3.5.37: Approximate location of the mill pond taken from the 1855 survey and overlaid onto the 1833 survey to show its relationship with the jetty. The yellow arrow points to the part of the mill pond that was within the excavation of Area 9. Much of the mill pond should survive to the east, beneath Harbour Street. City of Sydney Archives with additions by Cryerhall.

67 The Monitor 15 September 1826, p 6(3). See also Section 3.5.3. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 211

3.5.5.1 Timber Revetment The main structure of the mill pond consisted of timber walling (context 9251) (Figure 3.5.38, Figure 3.5.39). It was constructed with of a series of individual timber planks driven into the ground, as demonstrated in TT59 to the north (Vol 4: Plans 10.38, 10.39, 10.60). On the western side in TT69, it appeared that the timber planks were placed within a clay-lined cut (Vol 4: Plans 10.38, 10.40, 10.60). The timbers were placed quite closely together and were between 40mm and 60mm apart. The planks were between 1m and 2m in length and on average 200mm in width, and 100mm in breadth. The top of the timber walling was exposed between RL 1.4m and 1.6m, level with or about 200mm above the surrounding ground level. In TT69 the base of the timbers were recorded at around RL -0.7m.

The secondary row found on the interior of the northern wall may be evidence of repair or consolidation work (Figure 3.5.40, Figure 3.5.41, Figure 3.5.41). Between the double row was a fine grey silty clay packing (context 9260) that may have acted as a waterproofing agent. The mill pond was constructed through ground that was created during the reclamation works for the jetty construction, and as such consisted of bulk fill material of varied stability; sands, crushed sandstone and clays. The timber walling therefore formed a revetment around the exterior of the pond that prevented the sides from collapsing

Analysis of three samples taken from the timber revetment identified the wood used as Eucalyptus saligna (sample #134), Eucalyptus siderophloia (sample #147) and Eucalyptus pilularis (sample #171) (Figure 3.5.43). All tree species are hardwood eucalypts endemic to the Sydney region. Eucalyptus saligna or Sydney blue gum is common along the coast of New South Wales, as well as parts of Queensland.68 It grew naturally in the Sydney area and was particularly heavily forested in the north western suburbs now known as Pennant Hills, Epping and Ryde from the first few years of European settlement, and by 1823 was being obtained from as far out as Annangrove and Kenthurst.69 Eucalyptus siderophloia (grey ironbark) and Eucalyptus pilularis (Blackbutt) were also readily available in the Sydney area at the time of the mill pond’s construction, and were particularly abundant in the Turpentine-Ironbark Forest, which is likely to have extended to within 1km of the site.70

All three tree species were identified as suitable for construction soon after British settlement and continue to be used for similar purposes today.71 Blue gum timber was historically used in flooring, construction, and ship building, and grey ironbark was recognised as suitable for naval purposes.72 The lining of the mill pond was therefore made of timber which was likely to have been obtained relatively locally, and which was typically used for such purposes. This is consistent with the use of both blackbutt and grey ironbark in the earlier timber-built boat ramp (context 8377), and the contemporary the use of blackbutt as formwork (context 8460) for the jetty wall (context 8366).73

68 Boland et al 2006: 294. 69 Hudson & Henningham 1986: 3, 10. 70 Benson & Howell 1990: 42-44. 71 Hudson & Henningham 1986: 5; Bootle 2005: 252, 278, 287. 72 Hudson & Henningham 1986: 5, 8. 73 Kuiters 2010: 14, Vol 3, Section 8.8. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 212

Figure 3.5.38: Extract from the detailed plan of Area 9 showing the mill pond remains. Vol 4: Plan 10.38 ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 213

Figure 3.5.39: The timbers formed a revetment on the northwest corner of the mill pond, the red arrow indicates the north wall and the blue the west wall. View to west. Scale 1m.

Figure 3.5.40: The timbers forming the northern part of the mill pond revetment looking towards the yard area of the mill. There was a double row of timbers along this section. View to northwest.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 214

Figure 3.5.41: Detail of timbers within the northern wall showing the mix of cut and un-worked tops. View to north from within the pond. Scale 1m.

Figure 3.5.42: The row of timbers on the northern wall which had collapsed into the pond. View to west. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 215

Figure 3.5.43: Timber samples from the mill pond revetment 9251. Note the pointed ends but otherwise unworked. These samples were identified as Blackbutt. Scale 1m.

3.5.5.2 Clay Lining and Clay Bank The timber revetment on the western side of the pond was constructed within a cut that was clay- lined, as demonstrated at the eastern end of TT69 (Figure 3.5.44, Figure 3.5.45; Vol 4: Plans 9.7, 10.40). The ground on the eastern side was excavated, or cut to create the mill pond. The cut was at least 1m in depth and was through the surrounding sandy reclamation fills and natural foreshore sands. The side of the cut was then lined with grey and greenish-coloured clays with sand inclusions (contexts 9290 and 9296). The use of clay and clay-like material to line channels and reservoirs to make them watertight is known as ‘puddling’. ‘Puddled clay’ is usually made by mixing heavy clay with some water to make it more malleable. Sand or grit can also be added to this mix. The puddled clay is then used to line the interior of the structure. The mill pond was excavated through ground that consisted of sandy reclamation fills with foreshore sand flat deposits below, and the clay lining recorded in TT69 illustrates this process of puddling.

There was no evidence for a clay-lining on the northern side. However in TT59, the natural sand flat level was sealed by an extensive deposit of dense greenish clay, context 9466 (Figure 3.5.46). This clay would have functioned in the same way and was likely to have been deposited as part of the puddling process.

Surrounding the outside of the timber revetment was a clay bank or bund (Figure 3.5.47, Figure 3.5.48). The top of the bank material was located between RL 1.35m and 1.6m; around the same as the surrounding ground level. The bank was made from several bulk fills of mostly dense plastic orangish and pinkish clays recorded as contexts 9464 and 9465 on the northern side, and as 9289, 9288 and 9237 on the western side. It survived best on the northern side where it was up to 3.5m in width. In TT59, the depth of the bank material was recorded as being up to 600mm. It was formed over a general bulk fill clay-rich layer context 9466 that sealed the natural ground, located at 1.2m below the top of the bank. On the western side not much of the bank survived as it had been truncated by modern services (Figure 3.5.49, Figure 3.5.50). It was recorded in TT69 as being at least 6m in width and up to 750mm in depth.

The materials used to form the bank consisted of the heavy and dense natural clays that are found in association with sandstone bedrock, and therefore could have been easily sourced locally. Such clays are highly impervious to water and may have been used to enclose the mill pond as part of the puddling process, while also providing further stability and buttressing to the structure. The clays appeared mixed and had a sand content and they may have been processed prior to being used in the same way as the puddling clay lining of the mill pond. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 216

Figure 3.5.44: Machine-excavated test pit at the eastern end of TT69 to investigate the construction of the mill pond. Outside the timber planks forming the mill pond revetment wall was a clay-lined cut, indicated by the dashed yellow line. View to southeast.

Figure 3.5.45: North-facing section of TT69 showing the clay-lined cut and the timber revetment. Highlighted in orange are the clay fills that formed the external clay bank. These were deposited over mixed natural harbour sands that were either the remains of a naturally deposited sandbank or fills associated with the jetty reclamation process. Extract of Cryerhall, Plan 9.7, Plan 10.40 (Vol 4).

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 217

Figure 3.5.46: Detail of the east-facing section of TT59 located on the north side of the mill pond (left of picture). The arrows indicate the dense clay fill, context 9466, located above the natural harbour sands on this side of the pond. This material was an extensive deposit rather than the lining or fill of cut as seen in TT69 on the western side of the mill. This clay fill appeared to have been a temporary surface level, as indicated by the sandstock brick pads used prior to the formation of the clay bank (orange and pink clay above). View to west. Scale 1m.

Figure 3.5.47: Clay fills forming a bank around the northern side of the mill pond. The line of the clay is indicated by the yellow dashed line and the mill pond with a blue dashed line. View to west. Scale 1m. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 218

Figure 3.5.48: Clay bank on the northern side of the mill pond during excavation. This photo shows the surviving height of the bank, in the section on the left and right of the photo. View to north. Scale 1m.

Figure 3.5.49: Photo of the western side of the mill pond and the remains of the clay bank truncated by modern services such as the yellow pipe and the brick junction pit. The extent of the clay bank is indicated with the dashed red line. View to west. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 219

9286

Figure 3.5.50: Eastern end of TT69 showing the clay fills 9289, 9287, 9288 and 9237 (yellow lines) that formed the bank overlying the clay lining 9290 (blue line) and abutting the timber revetment 9251. The bank is cut by a later clay filled feature 9286. View to south. Scale 1m.

3.5.5.3 Brick and Timber Structure Connected to western side of the mill pond were the truncated remains of a timber and brick structure (context 9241) (Figure 3.5.51, Figure 3.5.52; Vol 4: Plan 10.38, 10.41). Contemporary with the mill pond construction, it has been interpreted as perhaps being a foundation pad for a pump that brought water from the mill pond to the boilers. Overall, the structure measured 4.5m east- west by a minimum of 2.75m north-south (Figure 3.5.53). It was built into the clay bank material and consisted of a timber-frame inlaid with sandstock bricks to form a pad or foundation-type structure.

The frame consisted of north-south and east-west timbers creating almost square sections. The individual timbers were mostly planed square and ranged in size. The north-south timbers were laid on top of the east-west and nailed or bolted where they joined. The east-west timbers were cut at the point of intersection to allow the level placement of the north-south timbers. As the structure was truncated, it was difficult to determine the original overall plan, layout and dimensions, or that of the individual elements. Enough survived to indicate that it consisted of possibly six north-south timbers laid on possibly four or five east-west timbers, thus creating five columns (north-south) with three rows, each compartment 1.25m by 1m and in total 5.5m east- west by c.3m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 220

Figure 3.5.51: Structure 9241 (surviving extent indicated with dashed red lines) was located on the western edge of the mill pond revetment. It was constructed with a timber frame inlaid with sandstock bricks. This photo shows it buried beneath a modern brick structure. View to the west. Scale 1m.

Figure 3.5.52: View of timber and brick floor structure 9241 from the south. It was truncated by a modern service pipe and brick-built culvert, and later sandstone structure 9240 was built over it. Photo taken looking north, the mill pond is located just to the right (arrow). Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 221

Figure 3.5.53: Extract from the detailed site plan of the mill pond showing the timber and brick flooring context 9241 adjoining the western side of the pond. Vol 4: Plan 10.38.

Each compartment was inlaid with sandstock bricks tightly set on their ends (Figure 3.5.54). There was no overall pattern to the layout as some were placed in rows side-to-side and some face-to- face. The gaps between the bricks were packed with a greyish brown ‘mud mortar’, consisting of silty clay with inclusions of grit and sand. The surface created by the bricks was worn in places, reflecting variations in the quality of the bricks. Three samples of bricks were taken from 9241 (sample #s 140, 141, 142) and are dated between 1800, perhaps earlier, to the 1850s (Robyn Stocks pers. comm.). This date range fits nicely with the mill pond construction date in the late 1820s.

The structure was attached to the western revetment wall of the mill pond and associated with two other structural elements; a sandstone wall (context 9250) and a timber ramp (context 9262) (Figure 3.5.55). Structure 9241 was located 9.5m south of the northwest corner of the pond where there was a break in the revetment. A sandstone wall abutted both the brick pad structure and the timber revetment (Figure 3.5.56). At a height of two courses over a footing, the top of the wall was at a similar height to the revetment. It only survived for a stretch of 1.2m heading west as it was truncated by a 20th-century brick-built culvert. The wall was on the same alignment at the timber frame of 9241 and formed a 90º angle with the pond revetment 9251.

Attached to the eastern end of structure 9241 and projecting into the mill pond was a timber ‘ramp’ 9262 (Figure 3.5.57). This was constructed using 10 timber planks laid side-by-side. The northern edge had a timber on its side forming an edge. The planks formed a ramp-like structure into the mill pond and it was at least 1m in length (Figure 3.5.58). It was set at a 45º angle and was nailed or ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 222 bolted to the timber frame of the floor 9241 (Figure 3.5.59). Behind the ramp was the clay lining, context 9263, a pale yellow and grey sandy clay. Mill pond sediment 9249 accumulated against the ramp.

Structure 9241 was clearly associated with the mill pond. It’s method of construction created a solid and durable structure that would have been capable of supporting a heavy weight, such as a piece of machinery. The mill pond acted as a reservoir and supplied the boilers with fresh water to convert to steam to power the mill engine. This water needed to be transported to the boilers and the structure 9241 may have been associated with a pumping system by providing a strong base for a large pump.

Figure 3.5.54: Detail of the sandstock brick surface inlaid within the timber frame. These were tightly set and placed on their ends with ‘mud mortar’ filling the gaps. Scale 1m.

Figure 3.5.55: Two further structural elements, sandstone wall 9250 and timber 9250 ramp 9262, were associated with floor structure 9241 and these were located at the point where it joined with the mill pond revetment (yellow dashed line). 9262

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 223

Figure 3.5.56: Detail of the internal face of sandstone walling 9250 and its junction with the timber revetment 9251. View to the north. Scale 1m.

Figure 3.5.57: Timber ramp 9262 attached to the eastern element of the timber frame of 9241. Note the timber forming the northern edge (arrow). The ramp rested on packing clay 9263 and mill pond sediment had accumulated around it. View to the northwest.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 224

Figure 3.5.58: East-facing section of the mill pond and timber ramp, also showing the relationship with sandstone walling and the brick and timber flooring. Extract from Vol 4: Plan 10.41.

Figure 3.5.59: North-facing section through the brick and timber floor 9241 and the timber ramp 9262. Extract from Vol 4: Plan 10.41.

3.5.6 Cooper & Levey’s Mill Yard As part of the mill complex development, Cooper & Levey also built a large five-storey sandstone mill building, and imported and erected a steam engine to power the mill. The mill building was located to the north of the mill pond, on the original property boundary with Ramsay’s land to the north. Hallen’s 1828-1830 sketch records two annexes on either side of the main building (Figure 3.5.2). The western annex is later recorded on an 1847 as ‘the boiler and engine house’ (Figure 3.5.62). The sketch also records a long building to the east of the mill pond that is identified as a ‘smithy and dye house’ on the 1847 plan. It is not known whether Cooper & Levey were responsible for the construction of the eastern annex or the structure to the east of the mill pond, or whether these were some early additions by Thomas Barker following his purchase of the mill in 1827. Remains of the mill building were recorded during two previous archaeological investigations. In 1986, substantial remains of the southwest corner were exposed and recorded ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 225 during the Darling Harbour redevelopment.74 During the construction of the Cross City Tunnel in 2002, further archaeological remains of the main mill building were recorded.75 The site of the main mill buildings and most of the mill complex was located to the east outside the Darling Quarter site.

The northwest corner of the mill yard was within the site boundary. This part of the yard was to the east of the shoreline and above the high tide level. Several test trenches in Area 5 (Boiler House) and Area 9 (Mill Yard) revealed that the bedrock was located at RL 2m in the northeast of Area 5 and sloped west and southwest to between RL 1m, and was around RL 1.2m in Area 9. As part of the overall landscaping and earthworks that created the mill pond, jetty and wharf, fills were used to prepare and level the mill yard. These fills were similar to those recorded overlying bedrock and in associated with the jetty and wharf surfacing further west in Area 6. Above the fills was an informal surface of compacted dark brown to black silty sand. The surface level of the mill yard was recorded between RL 1.5m and 1.8m. This surface level sloped westward and connected with the surface of the wharf and jetty. No further archaeological features could be firmly associated with the construction of the initial Cooper & Levey mill complex.

3.5.7 Reconstructing Cooper & Levey’s 1825 Mill Development Cooper & Levey made a considerable investment in the construction of the mill complex and associated wharf and jetty facilities (Figure 3.5., Figure 3.5.62, Vol 4: Plan 9.8). The property they acquired consisted of gently sloping land with a rocky shoreline and an extensive intertidal sand flat. Other natural attributes included a creek that flowed through the site. This emptied into the harbour, where perhaps the underlying geology and natural foreshore environment supported the formation of a lagoon-type feature.76 The substantial mill building was constructed adjacent to the shoreline on firm sandstone bedrock, and perhaps using sandstone quarried on site or nearby.

A large mill pond, initial in two parts, was constructed above the high tide level at the mouth of the creek and the junction between the land and the foreshore. The mill pond was created to dam freshwater to supply the steam engine that powered the flour mill. It is likely that the natural lagoon or pond-type feature was exploited in the excavation to create the mill pond. Excavation through natural clays and bedrock would have been likely to construct most of the mill pond. However, the western part of the pond was constructed within the sandy foreshore. Harbour sands had been deposited, either naturally or during earthworks associated with the mill and/or jetty construction, over the natural sand flat and thereby creating a ground level above the high tide mark. The western edge of the pond was cut through this sandy material. The large cut was lined with puddling clay as part of the process to make the pond watertight and it is likely the entire interior of the pond was lined. The western edge was revetted by a substantial timber wall consisting of long planks and poles. Around the exterior was a bank of dense and compacted clays that further reinforced the pond structure at this structurally vulnerable point. Perhaps the clays excavated during the construction of the mill or eastern part of the mill pond were used to form this bank.

An extensive intertidal sand flat abutted the property’s shoreline. This meant the site had no immediate access to deep harbour water, and the transport link this would have provide. To provide the mill complex with this facility, Cooper & Levey built a long jetty projecting into the harbour. As part of the jetty construction, extensive reclamation of the foreshore was also

74 Casey & Lowe 2006. 75 Archaeological work undertaken by Casey & Lowe in 2002. Casey & Lowe 2002; Casey & Lowe 2006. 76 Evidence for the natural landform and attributes such as a freshwater lagoon are discussed in Section 3.2.5.5 and 3.2.5.6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 226 undertaken. The reclaimed land joined the jetty to the land proper, and also formed a large, open wharf or yard space about 500mm above the high tide level to process and store raw materials and finished products. The northern limit of the jetty yard area was formed by a substantial sandstone wall built within timber-pile formwork. The wall was constructed the intertidal bedrock platform into the sand flat. Behind the well-finished sandstone blocks that formed the outer face of the wall was a substantial buttressing of courses of large rubble and roughly cut sandstone blocks. The jetty wall also acted as a retaining wall for the jetty reclamation fills. Though the southern limit of the reclaimed jetty area did not survive, it is likely that was also a sandstone or timber pile wall built from the rocky shoreline across the sand flat.

Bulk fills of harbour sands, crushed sandstone and clays were used to reclaim the land between the defined limits of the jetty yard area. These fills were deposited from east-to-west, or moving from the land towards the harbour, and also from south-to-north. The sand flat greatly facilitated this reclamation process as it was formed with reasonably compact alluvium that created a stable base below the tidal range from which the land was reclaimed. The sand flat was also extensive; stretching at least 50m from the shoreline, therefore a large area could easily be reclaimed. As the sand flat was intertidal, bulk fills could be more easily transported and deposited by barge at high tides or by cart at low tides. Gradually the ground level was raised above the high tide level by dumping and consolidating the fills. Redeposited harbour sands were the most common fill type and these may have been sourced locally, from localised informal dredging, upcast from the jetty wall construction trench and excavated material from the mill pond excavation. The crushed sandstone and clay deposits were also likely to have been sourced locally, and perhaps may have been excavated material from the construction of the mill building and pond.

The main part of the jetty structure projecting into the harbour was not within the excavation area but it was likely constructed using timber piles and decking. It was likely that ballast (sandstone rubble) was dumped into the deep harbour water to anchor the piles and provide structural stability for the timber jetty. The further into the harbour, the deeper the water became, and therefore the construction of sandstone walling would have become challenging, if not prohibitive, expensive and unnecessary. The sandstone walling formed not only a seawall or retaining wall for the reclaimed land to protect it from erosion, but also a quay wall for smaller vessels than those using the jetty to berth. The sandstone wall is likely to have continued southwest for a further 10m beyond our excavation limits and the low tide level, where it joined with the east-west aligned timber structure of the jetty proper. The surface of the reclaimed land forming the wharf or yard area was mostly informal and consisted of compacted clay and sandstone fragments. To the south, a small patch of irregular sandstone paving indicated that at least in some areas the surfacing was more formal. The variation in the quality of the surface may have been a response to differing functions or activities within the wharf and yard.

Within the main mill yard, Cooper & Levey constructed a substantial five-storey sandstone mill building that included a basement level. This housed the steam engine they imported from England, the boiler room, mill stone and various granary or storage levels. The building was located on the northern boundary of the mill complex, just to the north of the mill pond. Other buildings may have also been built during the initial construction phase, such as the long building to the east of the pond later identified as a smithy and dye house.

The construction of the long jetty into the harbour involved extensive reclamation of the foreshore leading up to it. It was likely that the reclamation and introduction of the new projecting structure created an obstruction in the harbour that affected the natural dynamics and depositional patterns, resulting in increased shoaling, formation of sandbanks and shallowing of the water, around the jetty and reclaimed land. A comparison of the location of the shoreline recorded on the 1822 Harper plan, before the jetty was constructed, and the 1833 detail survey after it was built,

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 227 indicates that a certain amount of shoaling had occurred to the north and south of the jetty at its junction with the land. The shoreline in these areas appeared to have been extended further westward by the deposition of sands above the high water level; a natural process that may have been altered and accelerated by the reclamation and construction of the jetty.

Figure 3.5.60: Reconstruction of the Cooper & Levey mill development 1825 with Harper’s 1822 plan as the background. The location of the mill buildings, mill pond and creek have been extrapolated from Hallen’s 1828-1830 sketch, an 1847 plan of Barker’s mill from the University of Sydney archives and the 1855/6 detail survey. The jetty location has been traced from the 1833 detail survey. The archaeological data informed the location of the low tide mark, the jetty wall, mill pond revetment and interpretation of the external bank. A comparison of the shoreline depicted on Harper’s 1822 plan and the 1833 detail survey led to the interpretation of shoaling to the north and south of the jetty. Cryerhall, Vol 4: Plan 9.8. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 228

3.5.8 Thomas Barker’s Mill Purchase, 1827 Cooper & Levey sold the land and mill complex to Thomas Barker in mid-1827. Barker had been apprenticed in 1813 to John Dickson, who constructed the first steam-powered flour mill in Sydney. In 1824 Barker was granted land at the southwest corner of Sussex and Bathurst Streets, just north of Cooper & Levey’s land. Barker had developed his own milling interests and in 1826 he established a windmill in Darlinghurst. Barker took a mortgage from Cooper & Levey for £5000 plus interest on 18 June 1827 for the land and mill. So by the end of 1827 Thomas Barker had purchased the ready-made mill and acquired a large land holding on the harbour south of Bathurst Street in the process (Figure 3.5.2). Though the initial investment and mill construction was undertaken by Cooper & Levey, it was Thomas Barker’s management and further investment that created a long and successful milling business.

3.5.9 Barker’s Mill 1827 to 1879 A detailed history of Barker’s mill was undertaken as part of the assessment for the Cross City Tunnel project.77 The mill continued to be used for processing grain after Barker purchased it. In 1847 a portion of the mill was leased to John Walker for the milling of woollen textiles.78 The textile operation was taken over by Barker in 1852, when it was one of the two largest of its kind in NSW.79 The 1865 Sands Directory listing for the lot corresponding with Barker’s mill includes ‘Barker & Co., millers and Ebsworth O.B., cloth factory’. Barker sold the grain mill to his nephews and the tweed mill to Ebsworth in 1868, but re-purchased the latter in 1870 following Ebsworth’s death. In 1870 Barker is the only listing in the Sands Directory for the property.80 The tweed factory was housed in the older portion of the mill which burnt down in 1872.81 The textile mill resumed operation under the direction of John Vicars & Co until the early 1890s.82 Meanwhile, the grain mill continued as before, until the mill complex was sold by Barker in 1879.83

A number of archaeological features and events associated with mid to late 19th-century development of the mill were recorded in Area 9 Mill Yard and Area 9 Mill Pond and detailed descriptions are provided in the trench reports. A change in the functionality of the mill’s steam engine by the 1840s is indicated by the construction of a substantial brick drain that is identified on an 1842 plan as being a ‘drain for (?) of saltwater to engine’ (Figure 3.5.61). This would suggest that saltwater was being used to make steam to power the engine by this time. However, the mill pond containing freshwater remained throughout the 1840s and 1850s. It was eventually backfilled between the mid-1850s and mid-1860s. The backfilling of the pond is likely to have occurred after the mill was connected to the city’s reticulated water service.

Also after the 1850s, a major change to the layout and facilities of mill complex occurred. In the 1830s Barker had reclaimed large areas of the foreshore, creating new land and wharfage. While Barker had sold off a large portion the land and waterfront in the early 1840s, he retained a land corridor and access to his jetty (through Area 6) (Figure 3.5.62). New goods transport options were created with the construction of a railway between Sydney and Parramatta, Goulburn and Bathurst in the early 1850s. A goods line linking Darling Harbour with the Sydney Terminus was completed in 1855. Thomas Barker played a key role in the early development of the railways, and the transport of raw materials (grains and wool) to his mill was via rail after the 1850s. The large track of land connecting the mill with the jetty was no longer needed for the mill operations and by the 1860s it

77 A detailed history of Barker’s Mill can be found at http://caseyandlowe.com.au/pdf/city/barkers.pdf. 78 Johnson & Parris 2008: 24-25. 79 Casey & Lowe 2006: 5. 80 Casey & Lowe 2006: Appendix pp. 17-18. 81 Casey & Lowe 2006: 6. 82 Johnson & Parris 2008: 25. 83 Casey & Lowe 2006: 6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 229 was being leased. In the 1870s it was sold to timber merchants Miller & Harrison, thereby finally severing the once crucial physical connection between the mill and the waterfront. Historic plans dating to the 1850s and 1860s record the changes to the layout of the mill complex (Figure 3.5.63, Figure 3.5.64). There were also archaeological remains of these changes; footings for a new boundary wall, and the remains of a toilet and a cesspit.

Figure 3.5.61: A plan from 1842 advertising Barker’s sale of 108 allotments. This plan also illustrates the main mill buildings complex and the purple arrow points to the drain or channel to bring saltwater to the engine. The mill is still connected by a land corridor, to the south of Thomas Street, to Barker’s jetty. The eastern site boundary is indicated with the red dashed line. Image rotated, north to the left. ML SP 811.1773/94.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 230

Figure 3.5.62: Detail from a c.1847 plan of Barker’s mill identifying the various buildings and features within the complex. The addition of a ‘Cloth Factory’ is noted. The eastern limit of the Darling Quarter site is indicated with the dashed line. A north- south boundary fence or wall on the western side of the mill building has been illustrated on this plan. Sydney University Archives, 1847.

Figure 3.5.63: The extent of Barker’s mill complex outlined with the dashed blue line on the 1855 plan. The site (dashed red) and basement (purple) are also indicated. The mill pond is still in existence, as is the land corridor to the waterfront. City of Sydney Archives, C&L additions.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 231

Figure 3.5.64: By the mid 1860s the pond had been backfilled and the size of the mill complex (dashed blue) reduced as the land connecting it to the wharf was leased and in the 1870s sold. The site boundary is in red and the basement excavation area in purple. City of Sydney Archives, C&L additions.

3.5.10 Mill Yard Developments, Post-1840 Following extensive reclamation of the foreshore to the west and northwest of the mill complex in the 1830s, Barker offered 108 allotments for sale in the early 1840s. The allotments were for both residential and industrial waterfront developments. Barker also sold-off sections of the original land grant, such as the Bathurst and Sussex Street frontages. Barker’s land sale reduced the extent of the mill yard and its waterfront access. By the end of the 1840s, the mill yard had been delineated from the surrounding development by a boundary wall (Figure 3.5.62). Footings for part of the western boundary wall were recorded in Area 9 Mill Yard (Figure 3.5.66; Vol 4: Plan 10.38).

During reclamation process the land around the mill was built up and further consolidated. The mid 19th-century surface level was raised by around 300mm. In the mill yard this was between RL 1.7m and 1.8m. The new surface level, like the first 1820s level, was graded and reduced in level heading toward the harbour. Throughout Area 6 the surface level was recorded at RL 1.5m, around 20m west of the mill boundary wall, and RL 1.3m at the very western limit of excavation.

Sewerage and stormwater management systems were either updated or newly installed during the reclamation and levelling process, and as development continued throughout the 1840s. The remains of one of these drains was found in association with the mill pond. The mill pond was located and designed to catch water from a natural watercourse. The two freshwater channels were illustrated on historic plans and identified on the 1847 plan as ‘drains’ (Figure 3.5.62). By the 1840s saltwater was also being used for the steam engine, lessening the need for freshwater from the pond. Dating to the 1840s, a sandstone and brick drain was built from the western edge of the pond towards the harbour. It functioned as an overflow drain for the pond.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 232

3.5.10.1 Boundary Wall Footing The archaeological remains of the western boundary wall consisted of two courses of sandstone footing (context 9419) (Figure 3.5.65). Just over 5m of this footing survived. The courses were stepped and the lowest course contained a single row of roughly dressed sandstone blocks that were up to 700mm in width. The second course was constructed with smaller roughly dressed sandstone blocks. There appeared to be no mortar bonding the footings. A large modern service trench truncated the wall remains, though it is location and extent is illustrated on historic plans and is indicated on the site plan (Figure 3.5.66; Vol 4: Plans 9.9, 10.38).

Figure 3.5.65: Section of the western boundary wall footing (red arrow) built during the 1840s to delineate the mill from the surrounding new developments. The sandstone wall and paving on the left are the remains of a later mill yard structure. View to the south. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 233

Figure 3.5.66: Extract from an interpretive plan of Area 9 and Area 6, main features associated with development of Barker’s mill in the 1840s. J Miskella, A Cryerhall, Vol 4: Plan 9.9.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 234

3.5.10.2 Pond Overflow Drain Located to the west of the mill pond were the remains of an east-west aligned stone drain 9204 (Figure 3.5.66, Figure 3.5.67, Figure 3.5.68; Vol 4: Plans 10.38, 10.42, 10.60). Though the physical link between the drain and the pond was truncated by modern services it has been interpreted as being associated with the mill pond and likely functioned as an overflow drain. During the 1840s and 1850s the pond still functioned as a reservoir for freshwater channelled from east of the mill yard. Though the reliance on water from the mill pond lessened during these decades, it continued to collect water and in the process silted-up. In order to manage high water levels and flooding events after heavy rains, this drain was built to direct the water underground to the harbour. The 1855 plan illustrates an outlet from the pond on the western edge (Figure 3.5.69). An overlay of the archaeological remains onto this plan suggests that the drain continued from this point below ground to the harbour (Figure 3.5.70; Vol 4: Plan 9.10).

Overall around 15m of this overflow drain was exposed during the excavation of Area 9 Mill Pond (Figure 3.5.66; Vol 4: Plan 10.38). About 12m to the east of this, a 4.5m section of overflow drain was exposed in TT23 during the excavation of Area 6 (Vol 4: Plans 10.9, 10.60). The fall in levels from the most easterly point of the drain at the western edge of the mill pond to the most westerly point exposed within the excavation limits was from RL 1.04m to 0.18m. The terminus of drain was not found, nor expected to be within the basement excavation. It was likely to have exited into the harbour at the wharf located a further 80m or so to the west from the western limit of the drain remains.

Figure 3.5.67: View of the overflow drain (arrow) from within the mill pond. The area between the drain and mill pond was truncated by modern services. View to the west. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 235

Figure 3.5.68: Sandstone overflow drain near the western edge of the mill pond flowing downwards towards the harbour. View to the west. Scale 1m.

Figure 3.5.69: The 1855 plan of the mill complex (dashed blue) and the basement excavation (purple) showing the mill pond with an outlet on the western side, and the insert is a detail of this feature (red circle). City of Sydney Archives, C&L additions.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 236

Figure 3.5.70: Overlay of the archaeological remains of the overflow drain (Figure 3.5.65) and the 1855 plan of the mill pond showing the relationship with the drain (green arrow) and the outlet (blue arrow). Extract from Cryerhall, Vol 4: Plan 9.10.

Sections of the overflow drain were constructed using different techniques (Vol 4: Plan 10.38). Though it did not survive, the part of the structure that connected to the mill pond was probably an open drain, as indicated on the 1855 plan. The first section of the drain remains to the east was built with large sandstone blocks laid end-to-end with a channel chiselled out, leaving rough tooling marks (Figure 3.5.71, Figure 3.5.72). This section was a minimum of 4.25m in length. The individual blocks were on average 900mmm in length, 600mm in width and 400mm in depth. The channel was 300mm in width and 200mm in depth and had a concave profile. It is likely that the drain capping had been removed previously. The base of the drain channel fell from RL 1.04m in the east to RL 0.68m over a length of 4.25m. As the drain dropped in level to the west, it survived in better condition and the capping material for the next section remained in situ. The construction technique also changed.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 237

Figure 3.5.71: Detail of the eastern section of the overflow drain. The individual segments were sandstone blocks with a channel chiselled out, leaving the surfaces covered with tooling marks. The capping for this section of drain did not appear to survive. View to the west. Scale 1m.

Figure 3.5.72: View of the section of the eastern part of the overflow drain. View to the south. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 238

The next section of drain, about 7.5m in length, was constructed with sandstone blocks and timber (Figure 3.5.73, Figure 3.5.74). The base of the drain was formed with timber planks (context 9253) (Figure 3.5.75). Large rectangular sandstone blocks formed the sides and it was capped with short timber planks. The base timbers were laid in an east-west orientation and were a minimum of 1.7m in length and c.250mm in width. One timber also had notches cut out in two places, indicating that it was a re-used plank. The planks were laid end-to-end along the base of this section of drain. The large rectangular sandstone blocks were well-cut and had tooling marks on all visible faces. They were laid on their side and placed end-to-end to form the drain sides. The blocks when in place were on average 800mm in length, 300mm in width and 400mm in depth. The external width of this section of drain was just over 1m with an internal channel width of 350mm. The drain was capped with shorter sections of timber laid perpendicular to the drain (or in a north-south alignment). These timbers had been somewhat displaced and many were not in situ. On average the planks were c.1m in length and 200mm in width. These were also likely re-used timbers. The timber capping would have been visible from ground level and the sandstone blocks were buried below ground. The fall in this section of drain was from RL 0.68m in the east to about RL 0.5m.

Figure 3.5.73: Section of the drain with stone sides and timber capping. View to the east. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 239

Figure 3.5.74: View from the overflow drain back toward the mill pond. This shows the sandstone and timber section of the drain after excavation. View to the east. Scale 1m.

Figure 3.5.75: Detail showing the re-used timber at the base, the large sandstone blocks with tooling marks and the timber plank capping. View to the west. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 240

The next section of the drain consisted of alternating segments of two sandstone ‘dish’ sections on top of each other forming an enclosed channel, followed by a segment of timber base and sandstone sides capped with timber planks (Figure 3.5.76, Figure 3.5.77). Only 4m of this part of the drain was investigated as it was truncated (to the west) by later activity. Its external width varied from 500mm to 1m with each segment; however the internal channel width remained consistently between 300mm and 350mm. The fall within the drain to its western extent was to about RL 0.4m (just a little below high tide level).

Figure 3.5.76: The last section of the overflow drain before it was truncated to the west consisted of alternating segments of two dish drain stones on top of each other with sandstone and timber- capped ones. View to the east. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 241

Figure 3.5.77: Removal of the timber plank capping and sediment within the drain allowed for a closer look at how this section of the drain was constructed. Timber formed the base of the sandstone and timber-capped segments. View to the west. Scale 1m.

A test trench (TT23) excavated in Area 6 exposed a section of brick drain (context 8430) in an east- west orientation (Figure 3.5.66, Figure 3.5.78; Vol 4: Plan 9.9, Plan 10.9). This drain was on the same alignment as, and appeared to be the western continuation of, the mill pond overflow drain. This section of the drain was roughly east-west aligned, with an interior base RL of 0.18m. It was hexagonal in profile, with an interior width of 400mm and height of 300mm (Figure 3.5.79). The bricks were sandstock with rectangular frogs. The bricks from this section of drain were sampled (sample #17 and #188) and have been dated to c.1840 to c.1870, with a likely date around the 1850s (Robyn Stocks pers. comm.).

The change in materials from stone to brick may be the result of two phases of drain construction; the brick being an extension reflecting separate phases of reclamation or ground consolidation. The phases of reclamation are discussed and illustrated in Section 3.6. The brick section of the drain was located within the boundary between Barker’s land and Brooks’ to the south (Murphy’s Wharf) and though the surrounding land had been reclaimed during the 1840s, this ‘no-man’s-land’ was not reclaimed until the early 1850s (Section 3.6.9.5).

Several sections excavated through the sealed parts of the overflow drain revealed the nature of the fill. Directly above the timber base (context 9253) was deposit 9252. This consisted of 200mm of fine dark grey to black clayey silt with visible organic content. Above this was context 9207; a dark grey waterlogged silty sand up to 300mm in depth. These fills represent both the accumulation of silt within the drain over its use and lifetime, and also accumulation after the mill pond went out of use. Pollen sample (#152) was analysed and found to contain cereal, pine,

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 242 dandelion, eucalypt, raspwort, samphires, sedge, trilete fern spores (fungal spores).84 Also present were the human waste indicators, Cloacasporites, confirming that the drain was transporting raw sewage towards the end of its useful 'life' in the 1850s.85

Figure 3.5.78: The brick section of the overflow drain was recorded in Area 6 as context 8430. View to the east. Scale 1m.

84 Macphail 2010: 37; Section 8.6, Vol 3. 85 Macphail 2010: 37; Section 8.6, Vol 3. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 243

Figure 3.5.79: Detail of the drain after capping was removed. It was hexagonal in profile. Scale 1m.

3.5.10.3 Other Mill Pond Water Management Features Located at the western edge of the mill pond and associated with the overflow drain was another sandstone structure (context 9240) (Figure 3.5.80). The top of the structure appeared to correlate to the general surface level around the mill pond during the 1840s and 1850s. The even surface created from tightly fitted large flat rectangular sandstone blocks which were aligned to slope from east-to-west (Figure 3.5.81). The sandstone structure was built over the brick and timber foundation pad (context 9241) once it became redundant, and probably functioned as part of the overflow system for the pond.

Also associated with the 1840s alterations to the mill pond was a deep dense clay-filled feature (context 9286) (Figure 3.5.82; Vol 4: Plans 9.7, 10.38, 10.40). The linear cut was about 700mm in width and up to 2.1m in depth from the contemporary surface level, with a base level of RL -0.8m (300mm below low tide level) (Figure 3.5.83). Located 3m from the western pond exterior, the cut appeared to be an attempt at preventing water leakage from the mill pond. This side of the pond was built into the intertidal sand flat and sandy reclamation fills of the 1825 jetty. While the pond’s timber revetment and clay bank would have provided a certain degree of impermeability, increased sedimentation within the mill and decrease in use of its water may have led to an increase in volume and a rise in the water level.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 244

Figure 3.5.80: View of sandstone structure 9240 (blue arrows) in relation to the mill pond wall (in foreground) and the overflow drain (yellow arrow). It overlies the earlier phased brick and timber floor 9241. View to the west. Scale 1m.

Figure 3.5.81: Stone structure 9240 survived in two sections, possibly deliberately separated, which sloped from east to west. It overlay the earlier brick floor 9241. View to the north. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 245

Figure 3.5.82: View of feature with clay fill 9286 cutting through the brick flooring 9241. This is located 3m to the west of the mill pond revetment and was in a similar alignment. View to the north. Scale 1m.

Clay Bank

9286

Jetty Infill

Harbour Sands

Figure 3.5.83: At the eastern end of TT69 feature 9286 was investigated. It was cut through the clay bank material, jetty infill and the natural harbour sands below. View to the south.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 246

3.5.10.4 Spending a Penny at the Mill The historical and archaeological evidence suggest that in the 19th century there was an absence of toilet facilities for the workers and occupants within the Darling Quarter site. There was no archaeological or cartographic evidence for toilets prior to reclamation and after the intensive development of the site in the 1840s. One can only assume that the harbour played a key role in providing convenience for the mill workers, and labourers and engineers involved in the reclamation and maritime infrastructural works. Archaeological evidence from the Steam Mill Street residential development (Area 8, see also Section 3.7) to the northwest of Barker’s mill, indicated that it initially included just three cesspit toilets for seven houses (Vol 4: Plan 10.31). A further three plumbed toilets were added after the 1860s; though two of these were within one property and appeared to be shared facilities with access from the rear and side lanes (Vol 4: Plan 10.33)

Density, overcrowding and the presence of noxious industries such as tanning and slaughtering resulted in poor sanitary conditions for the working class of the city, especially around Darling Harbour, and from the 1840s it was the subject of a number of inquiries.86 The condition of the houses was mixed, with home owners being the most conscientious about cleanliness and order. Even in some of the rented houses which were in ‘good order’, such as those belonging to local employers Mr Barker and Mr Monk, the state of the closets or privies was ‘highly offensive’ or ‘disgusting’.87 In the 1870s members of the Sewage and Health Board undertook personal inspections of many houses and business premises.88 Among the places included in the inspections were Duncan Street, Barker’s Lane, Steam Mill Street and the tweed manufactory. Golden’s Buildings in Duncan Street, just to the north of Barker’s mill, were perhaps the worst as they had no back door or yard, and the whole community shared a single toilet. This was located between two of the houses and opened directly onto the street and was: In such an exposed situation as to be a continual offence to decency. No one can enter or leave it without being seen by the tenants of all the houses opposite, besides the employees of the flour mill on the corner, where there are a number of men at work. Some of the women told us that their husbands were so ashamed to make use of that closet in the face of so many persons that they went away and found accommodation elsewhere. No less than twenty-two individuals have to make use of it.89

Toilets are not usually annotated on historic plans but can be identified, and quite easily for those located at the rear of residential properties. The small square or rectangular structures within large industrial complexes, either adjoining buildings or isolated within large yard spaces, are not as easily assumed to be toilet facilities. The layout and functions of buildings within industrial spaces are not as familiar or straightforward in their interpretation. Archaeological investigation can provide evidence regarding the original use and evolution of these structures. Within the four main 19th- century industrial properties of the site: Barker’s mill (Area 9), Murphy’s Wharf (Area 7), Miller & Harrison’s Timber Yard and Sawmill (Area 6) and PN Russell & Co. Engineering Works (site of Brodie & Craig’s Builder’s Yard until 1859) and Carriage Works (Areas 4 and 5), a number of potential toilet structures can be indentified on the 1855 and 1865 plans (Figure 3.5.84, Figure 3.5.85).

On the 1855 there is one possible toilet structure in Area 4 (Brodie & Craig’s Builder’s Yard at the time) and one possible in Area 5 (Travers’ Wharf in the 1850s). There are more in Area 7 (Murphy’s Wharf) with four potential toilet structures; this may be reflecting the fact that a number of businesses were located around the wharf. There are at least seven potential toilet sites within

86 Johnson & Parris 2008: 90. 87 Casey & Lowe 2008b: 58. 88 Sydney City and Suburban Sewage and Health Board, ‘Eleventh Progress Report’, LAV&P 1875-6 (5): 605-609 & 619. 89 Sydney City and Suburban Sewage and Health Board, ‘Eleventh Progress Report’, LAV&P 1875-6 (5): 605-609 & 619. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 247

Barker’s mill on the 1855 plan. Significant changes to the industrial complexes had occurred by the time the 1865 plan had been produced and the number of potential toilets structures decreases. For example in Area 4, quite a number of new and large buildings had been constructed during the redevelopment of the site for the PN Russell & Co. foundry in 1859. There are no obvious toilet structures illustrated within the yard and it is likely that the redevelopment for the foundry included plumbed toilets within the buildings. On the 1865 plan, the Area 5 (PN Russell & Co. Carriage Works site in the 1860s) potential toilet is in the same place. The Area 7 (Murphy’s Wharf) numbers remain the same, but some in different locations. Barker’s mill has less with five potential toilet structures, mostly located around the periphery of the yard. Plumbed toilets are likely to have been installed as part of the main mill building additions in the 1850s and 1860s.

Archaeological remains of just one from a potential of four mid 19th-century toilets was excavated in Area 7. This consisted of the truncated remains of an unlined cesspit dug into the ground with the remnants of a primary fill, and several backfill deposits dating to the mid to late 19th century. This feature is discussed in more detail in Section 3.10.4. The remains of two of a possible three toilet structures within the southwest corner of Barker’s mill yard were excavated (Vol 4: Plan 10.38). Located at the very southwest corner and perhaps corresponding to a structure first recorded on the 1865 plan were the remains of a brick cesspit (context 9495) (Figure 3.5.86). This structure is located on the western side of the mill boundary (Figure 3.5.66). This land was still part of the mill complex up to the early 1860s when it was leased to a timber merchant. Though not recorded on the 1855 plan, the type of sandstock bricks and lime mortar used to construct the cesspit suggest that it may have been built in the 1850s and therefore initially part of Barker’s mill complex.90 The pit was fairly typical in form, being rectangular with external measurements of 2.1m by 1.3m and a depth of just over 1m. There was no primary fill deposit present and the pit was backfilled with a mixture of industrial and demolition-type waste (context 9503). The backfilling of the pit occurred when plumbed facilities became more available from around the 1860s. This structure did not appear to have been upgraded to a plumbed toilet by either Dent or Miller & Harrison.

Evidence for a plumbed toilet (context 9453) was found closer to the mill buildings (Vol 4: Plan 10.38). It had a post-1860s date and was likely to have been part of a lumber store addition to the mill complex in the 1870s (first illustrated on the 1880 plan). The toilet structure consisted of brick flooring and walls (Figure 3.5.87). The sandstock bricks used in its construction were reused and a rendered surface placed above the brick floor is post-1860s in date.91 The plumbing for the toilet appeared to have been a replacement or later addition as the sewerage pipe trench cut through fill deposits that post-dated the construction of the toilet (Figure 3.5.88). The glazed ceramic pipe was reddish brown in colour and 120mm in diameter. The western termination splayed to a diameter of 160mm, 400mm short of the edge of the toilet. The gap was occupied by two bricks, which seemed in situ as part of the drainage system, and may have been a remnant of an earlier drain.

90 Miskella 2012: 39, Vol 3, Section 8.11. 91 Miskella 2012: 42, Vol 3, Section 8.11. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 248

Figure 3.5.84: Illustration of the location of potential toilet structures (green arrows) on the 1855 plan with the site boundary (dashed red), basement excavation (purple) and public domain excavation areas (yellow). The industrial complexes are defined with the dashed blue line. City of Sydney Archives, Cryerhall additions. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 249

Figure 3.5.85: Illustration of the potential toilet structures (pink arrows) on the 1865 plan. Site boundary (dashed red), basement excavation (purple) and public domain excavation areas (yellow). The industrial complexes are defined with dashed blue lines. City of Sydney Archives, Cryerhall additions. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 250

Figure 3.5.86: Sandstock brick cesspit in Area 9 dating from the 1850s. The cesspit was likely built before Barker leased this part of the mill yard to Dent in the 1860s. Scale 1m.

Figure 3.5.87: The remains of the c.1870s toilet structure in Area 9. The red arrow indicates the remains of the sandstone wall associated with the lumber store illustrated on the 1880 plan. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 251

Figure 3.5.88: Photo of the toilet structure in Area 9 including the remains of the sewerage system. View to the west. Scale 1m.

3.5.11 Evidence of Operational Changes at the Mill Steam-powered engines were developed in Britain in the 18th century, mostly for use in the mining industry to pump water from mine shafts. In the 1770s, Scottish engineer James Watt had made significant technological advances and design changes to improve steam engine efficiency, though still only producing power in one direction, pulling. Further technological advances by Watt and other engineers by the early 19th century meant that consistent power could be applied in two directions, and a dual-action piston could be used to drive machinery in mills and factories. These advances in steam power freed mills and factories from geographical and other constraints. They no longer needed to rely on an ample water supply, wind or horses for power.

Mills established in the early days of the colony were both wind and water-powered. The location of a mill was more determined by a sites natural attributes in providing a power source rather than its ease of connection to transport for raw materials and products. Darling Harbour had the potential to provide industry with excellent transport facility but was lacking in natural attributes conducive to power production. Importation of steam engines from Britain in the early 19th century, by entrepreneurial manufacturers such as Dickson and Cooper & Levey (Barker’s mill engine), enabled the industrial development of Darling Harbour beyond that of shipbuilding and goods or materials storage.

A constant water supply was still necessary for a steam-powered mill. Both Dickson and Barker had freshwater reservoirs at their mills for use in the boilers to create steam and also presumably for other general manufacturing purposes and processes. Dickson dammed the head of the harbour to retain freshwater from the several creeks exiting at this point. Barker’s mill pond was also situated at the exit point of a freshwater source. Water used to create steam in the first decade or so of Barker’s mill was probably drawn from this freshwater source as saltwater could not be used for

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 252 boilers as it led to a build up of salt as the water was boiled-off.92 The use of saltwater required frequent shut-down of the boiler for cleaning and removal of the over-salted water. However, by the early 1840s, Barker’s mill was utilising Darling Harbour and its endless supply of saltwater for use in his engines. A drain was built from the harbour and saltwater was pumped to the boilers in the mill (Figure 3.5.89). This addition may reflect a technological advancement or a change in the capacity of Barker’s steam engine system to utilise saltwater. The use of saltwater may also have become possible at the time when Barker expanded his mill to include more than one boiler and engine. The expansion of the mill is recorded on the 1842 plan with annexes either side of the original building. These are later annotated on the 1864 lease plan as both containing engines and boilers (Figure 3.5.90). With two sets of engines, more regular servicing and cleaning could be undertaken without halting production completely.

The impetus for the switch from freshwater to saltwater may have also been due to increased water requirements as Barker’s milling business succeeded and expanded in the 1830s. The mill pond retained freshwater originally from a natural creek that channelled water and stormwater. Reliability may have become an issue as the volume of water would naturally decrease in the summer months or over dry spells. The watercourse also flowed down through the city to the harbour, and with increased development, its water was likely to have become more polluted. Stagnation and increased sedimentation would have rendered the mill pond water less useable, as it could not be readily used in manufacture of consumable products and it would also damage the internal workings of the boiler system.

Figure 3.5.89: The saltwater drain is illustrated on the 1842 subdivision plan and the inset is a detail from an 1847 reproduction of the plan that provides a clearer version of the annotation of the drain as ‘Drain for the supply of salt water to the Engine’. ML SP 811.1773/94.

92 Forester 2008: 10. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 253

Figure 3.5.90: Detail from the 1864 lease of part of Barker’s mill to Ebsworth (delineated with heavy lines around the textile part of the mill and yard). It shows the saltwater pipe (arrowed) and the layout of the mill containing two boilers and engines in annexes on either side. Claim no. 1080, City of Sydney Archives.

Figure 3.5.91: The mill pond had been backfilled by the time the 1865 survey had been undertaken (right) but had still be present in 1855 survey (left). City of Sydney Archives.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 254

3.5.11.1 Saltwater Drain Remains of the saltwater drain (context 8344) were present in Area 6 and Area 9 Mill Yard (Vol 4: Plans 10.9, 10.38, 10.59, 10.60). Though buried below ground, the alignment of the channel was recorded on the 1842 and 1847 subdivision plans for Barker’s land sale (Figure 3.5.89). A lease dated to 1864 also identifies the saltwater pipe (Figure 3.5.90). It connected to the southwest corner of the mill building and joined with the harbour to the southwest at the boundary with Brooks’ land. The drain would have been around 80m in length and it was recorded in detail in three areas. In Area 6 about 12m was exposed in plan and was recorded as part of Section #17 & 26, and in Area 9 it was recorded in TT70 and as part of Section #46. The top of the drain was recorded at RL 1.2m in TT70; the closest part to the mill building. This level fell as it headed toward the harbour and was recorded in Area 6 as being around RL 0.8m with a base level of RL -0.4m.

The construction cut for the drain was through bedrock near the mill and also through the first reclamation fills for Cooper & Levey’s jetty, dated between 1825 and 1828 (Figure 3.5.92, Figure 3.5.96). Following construction, it was then sealed by levelling fills associated with Barker’s reclamation project dating to the later 1830s (Figure 3.5.93, Section #17 & 26; Vol 4: Plan 10.59). The structure was originally built entirely of sandstock bricks bonded with a sandy shell mortar. An arched brick roof sprung from the double-brick walls (Figure 3.5.95). The sandstock bricks were flat and have been dated between 1800 and 1860.93 The structure was deep and narrow with a channel up to 1.8m deep and only 170mm wide at the top. The interior of the brick channel contained vertical wooden planks that appeared to shore the walls against collapse. The planks were installed in pairs at intervals of 2m along the drain. Within Area 6 were the remains of a 450mm square manhole or access point constructed from sandstock brick.

The drain walls flared somewhat with depth (up to 40mm on each side) (Figure 3.5.97), however this appeared to represent partial collapse after the removal of the capping, rather than a design detail. The brick arched capping was removed, either as a repair or for brick-salvage, later in the century. This event is represented by the re-cut (context 8354) after a series of levelling fills (such as context 8361) that sealed the drain structure (context 8344) and its original construction cut (context 8352) (Vol 4: Plan 10.59).

A section of the drain was excavated in Area 6 and was found to contain layered deposits of sediment. At the base was 200mm of black and dense sticky clay (context 8453). Above this was 100mm of mixed silt and sand (context 8452). It lacked the sticky quality of the deposit below and was lighter in colour. This was followed by 300mm of very dark, grey to black silty accumulation (context 8451). It was very dense with no visible inclusions and a strong organic smell. Oyster shells adhered to the walls of the drain at this point, marking the mean tide level, and indicating that tidal water flowed through the channel for extended periods of time. The final deposit (context 8358) had built up to within 370mm of the top of the drain. It was a dark brown-grey sand and silt sediment with a malodorous organic content.

Though the fall in levels suggest that it was a straight-forward drain to direct stormwater/sewerage from the mill to the harbour, the structure, according to the 1842 plan functioned as a channel for pumping seawater to a saltwater boiler for the operation of a steam engine in Barker’s mill. A beam engine would have sufficed to draw the water through the drain. These pumps had been in use in the early part of the century in Britain, and were successful in pumping water from flooded shaft mines in Cornwall.94 They were built on large and small scales and could be adapted accordingly. The depth of the channel may have been enough to ensure the channel contained water at least during high tides if not always. As proof of this there were several oyster shells adhering to the

93 Miskella 2012: 17, Vol 3, Section 8.11. 94 Cocks 2005: 243. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 255 inner wall of the drain. Its base in the section excavated in plan on the eastern edge of Area 6 was at RL -0.35m or just above the low tide level (at RL -0.5m).

Figure 3.5.92: The mill saltwater drain (8344) as exposed in the eastern part of Area 6, close to the boundary with Barker’s mill complex (Area 9). It was cut through the late 1820s jetty infill. Note the manhole in the foreground. View to the south.

Figure 3.5.93: Detail from Section #17 & 26 (Vol 4: Plan 10.59) showing the saltwater drain and construction cut through the jetty fill (red dashed lines) and the sealed by levelling fills dated from the 1840s (after the main reclamation phase in the late 1830s). Later re-cuts for repair or brick-salvage are highlighted in green.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 256

Figure 3.5.94: Continuation of the brick saltwater drain northeast of the manhole. The brick arch capping survived to a greater degree and the drain was wider than the section to the south. View to the southwest. Scale 1m.

Figure 3.5.95: TT70 in Area 9 exposed a section of the brick drain 8344. The brick arch survived, although it had collapsed slightly. The yellow dashed lines indicate the construction cut through the jetty fills. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 257

Figure 3.5.96: Brick saltwater drain 8344 at the western edge of Area 9 running into Section 46. Here it was cutting jetty-related fill and bedrock. View to the east. Scale 1m.

Figure 3.5.97: Detail of the saltwater drain 8344 structure in Section 17.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 258

3.5.11.2 Mill Pond Sedimentation and Backfilling The mill pond was constructed between 1825 and 1827, and acted as a water reservoir for the mill complex until it finally became redundant and backfilled in the early 1860s. One of the last depictions of the pond is in a painting by William Christie dated to 1859 (Figure 3.5.98). The mill pond did not appear to function perfectly as a freshwater reservoir. Soon after construction it required additional water and flood management features. The pond was subject to sedimentation from the offset as the watercourse(s) that fed it flowed down the hill bringing with it eroded materials, sewerage and pollutants from the surrounding urbanised environment.

Only the southwest corner of the pond was within the excavation area and its investigation was limited due to the presence of PASS (see Section 3.1.5). Large-scale excavation below the top of the waterlogged sediments was not possible. However, some information regarding the nature of the later deposits within the pond, prior to backfilling in the early 1860s, was gained from two small test trenches. A 1.8m sample core for pollen analysis (sample #211) was also taken from the top of the pond sediment at RL 1.4m. A large machine-excavated trench through the backfill material allowed for a detailed record of the post-1860s fills that sealed the pond remains (Figure 3.5.102).

The first test trench (2m by 1m) within the mill pond was placed next to the timber revetment (Figure 3.5.99, Figure 3.5.100; Vol 4: Plan 10.38). It was excavated to the top of the mill pond accumulated deposits (context 9249). This consisted of dark grey clayey silt with frequent bands of fine whitish sand and silt. Inclusions of organic matter, twig and general woody detritus, was also noted within this material. Context 9249 represents the last sediment accumulation within the mill pond prior to it going out of use and being backfilled. Due to excavation limitations (PASS) this material could not be examined in detail. This sedimentation was found in TT57, also below industrial fills 9248 and other backfill material (Figure 3.5.101; Vol 4: Plan 10.40). It was generally located at RL 1.3m, meaning it was on average 300mm below the top of the revetment.

Figure 3.5.98: Painting by William Christie dated to 1859 showing a view looking west with Barker’s mill on the right, and the mill pond in the middle ground. In the background is Darling Harbour and Pyrmont with its new railway, complete with a steam locomotive. The painting depicts the mill pond water level as being somewhat lower than the surrounding ground level. SLNSW, ML 1426.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 259

Figure 3.5.99: View of the mill pond and the top of the sediment 9249 on the interior side of the revetment 9251. View to the northwest. Scale 1m.

9248

9249

Figure 3.5.100: A test trench on the interior side of the western revetment exposed the top of the sediment 9249 below the industrial waste fill 9248. View to the northwest. Scale 1m.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 260

9248

9249

Figure 3.5.101: Detail of the test trench section showing the top of the material filling the mill pond. The grey clayey silt 9249 was sealed by imported industrial waste fill 9248. View to the north. Scale 1m.

9249

Harbour sands

Figure 3.5.102: A machine-excavated extension to the eastern end of TT69 revealed the extent of the sediment against the western edge of the pond (red arrows). View to the southeast.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 261

The extension to TT69, excavated by machine, allowed for a photograph and quick recording of this sediment in section (Figure 3.5.103). It was 200mm in depth at the western edge of the pond abutting the timber revetment. The base of the sediment dipped to 900mm about 2m east, creating a concave profile and presumably it deepened further eastwards into the pond. It was a dense, waterlogged material and appeared banded with silts, black organic matter and other grey clayey silts. It overlayed context 9297; a deposit consisting of grey sand and organic material interpreted as either natural harbour sand or redeposited harbour sand as part of the infilling process for the jetty.

Pollen analysis was undertaken of a sample core obtained from the mill pond sediments 9249.95 The core was taken further into the interior of the pond than the test trenches. It was taken from around RL 1.4m and to a depth of around RL-0.4m (just above low tide level). Though the base of the sample appeared to have intersected with natural foreshore deposits they still contained cereal pollens. The fossil pollen data points to four broad phases of sediment accumulation dated from the 1820s to the 1860s.96 Each of the phases is represented by bands of silt and sands relating to both discrete and episodic depositional events. Native, cereal and exotic, including weed, varieties, fossil pollens are represented throughout the accumulated deposits.

Of particular note is the presence of the human waste indicator, Cloacasporites. Trace occurrences of Cloacasporites in the lowest level sampled from the mill pond sequence confirm that raw sewage almost certainly was being deposited on the foreshore at this location in the 1820s and around the time of its construction.97 Relative abundances of Cloacasporites in the late 1830s to 1850s core sample sections in the mill pond sequence are lower than those recorded in the Areas 7 and 8 cesspit samples (contexts 8037 and 8737). Nonetheless, these values are remarkably high given the aquatic depositional environment and confirm the discharge of raw sewage into the mill pond during the 1840s and 1850s. Such 'episodes' may have been due to natural events, such as storms flushing sewage down the inflowing watercourses, or simply reflects opportunistic behaviour by local residents.98 Cloacasporites is also recorded from the sediments filling the overflow drain (context 9204).

The backfilling of the mill pond had begun by the early 1860s and appeared to be complete by the production of the 1865 survey (Figure 3.5.91). There seemed to be at least two phases of backfilling in the archaeological record. The first phase of backfilling consisted of industrial waste material. The second phase was distinguished by a multitude of thin bands of artefact-rich clays, silty clays and industrial-waste deposits with distinct tip lines from east-to-west. A possible third phase consisted of thicker bulk deposits of silty clay and industrial waste. The several discrete working surfaces within the first two phases of material indicate that backfilling may have occurred over some length of time; though it cannot be determined from the deposits alone whether this was days, weeks, months or years. Historical records of the mill’s development in the 1860s and 1870s support this phased process of backfilling rather than a single short event.

95 See Pollen Report by Macphail in Vol 3, Section 8.6. 96 Macphail 2010: 41, Vol 3, Section 8.6. 97 Macphail 2010: 57, Vol 3, Section 8.6. 98 Macphail 2010: 58, Vol 3, Section 8.6. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 262

Figure 3.5.103: TT57 was excavated through the mill pond backfill and into the last sediment accumulation 9249, located below the dashed yellow line. The fill between the scale and the dashed yellow line is the first phase of backfill consisting of industrial waste context 9206. The mill pond wall is indicated by the red arrow. View to the southwest. Scale 1m.

In the late 1860s, Thomas Barker sold both the flour and textile mill. Ebsworth, who had been managing the textile mill since the early 1850s, bought it with a mortgage from Barker in 1868. However, following Ebsworth’s death in 1870, Barker & Co. bought back the mill. During the 1870s, the textile mill, also recorded as tweed factory or tweed mill, was modernised with new machinery and a horizontal steam engine. The mill complex was also expanded to include all of the original mill building and new additions to the south. A flour mill still operated but within a much smaller complex, and the management of the tweed factory was taken over by John Vicars. This development is illustrated by comparing the 1865 and 1880 plans (Figure 3.5.91 and Figure 3.5.104. The large yard space created by filling in the mill pond became part of the tweed factory and was used as a tenter ground (area used for drying newly manufactured cloth) (Figure 3.5.104).

The initial backfill of the early 1860s consisted of industrial waste (contexts 9206 and 9248). Context 9206 was recorded at the base of TT57 (Figure 3.5.103). It was up to 600mm in depth and consisted of level bands of dark grey silt with industrial waste consisting of cinders, coke and slag nodules. This fill was loose and waterlogged at the base. It overlaid 9249; the last sediment accumulation in the mill pond. Context 9206 also contained sandstock brick, glass, ceramic, leather and metal and is outlined below in Section 3.5.11.3.99 It was located generally between RL 1.3m and RL 2m. There were no discernible tip lines and it is this that differentiates these fills from those discussed above.

99 See also the Artefact Overview, Section 4 and the Artefact Catalogue in Vol 6, and the various specialist reports, Vol 3, Section 8 for detailed information regarding the artefacts from the first phase of backfilling. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 263

Located along the western perimeter of the mill pond and in the area opened-up in plan was context 9248. This was also an industrial waste fill and it consisted of ferrous-stained (orangey brown) clinker, coke and slag nodules, of loose compaction below a firm crust (Figure 3.5.99). It was a minimum of 300mm in depth and also sealed the last sediment accumulation 9249 within the mill pond (Figure 3.5.100). This material is part of the same event described by context 9206, that being part of the initial phase of backfilling of the mill pond in early 1860s.

It is likely that the source of the industrial waste used for this first backfilling event was from the mill complex or from the nearby industries. Such material was the waste product from furnaces; used to heat boilers, in blacksmith’s forges, and larger foundries and engineering workshops. The use of industrial waste as a levelling material was common and widespread throughout the site from the mid to late 19th century. The industrial waste also contained artefacts associated with domestic waste (discussed below).

The first phase of backfilling with industrial waste had raised the ground level within the pond over the level of the structural elements. At the western edge of the pond the ground level after this backfilling was at around RL 1.6m and at the very eastern limit of the excavation, further into the pond, it was at RL 2m (similar to ground level across site at this time). Following this, further backfilling and levelling of the mill pond was undertaken, and the ground level was raised to around RL 2.8m. The large section excavated through this material (TT57) revealed many layers of clay-rich and industrial-waste fills dumped and had been spread out from east to west (Figure 3.5.102, Figure 3.5.105; Vol 4: Plan 10.44). Similar levelling events occurred in all the properties of the site throughout the mid to late 19th century.

However, the second phase backfills within the mill pond differed from those used across the rest of site, insofar as they contained significantly more artefacts. Artefacts from the machine excavation of the test trench were collected as context 9201. The individual fills were numbered and recorded in more detail during the section drawing process and artefacts were also collected. The context numbers and descriptions of the individual fills are presented in Area 9 Mill Pond Trench Report: Table 10.1 (Vol 3: Section 7.7). The artefacts were mostly household or domestic in nature and included wine/beer bottles, condiment and food jars, consumption-related tablewares, and hygiene-related items, and the leather shoe pieces and clerical artefacts found may indicate trade-related waste disposal (outlined in Section 3.5.11.3 below).100 The artefacts offer a broad date range, though they are generally supportive of an 1860s to 1870s date range for the main backfilling inferred from historic plans and records.

This final phase of bulk fills relates to general levelling across the site rather than specifically mill pond backfilling. This phase is represented by contexts 9212 and 9213, and is dated to the later 19th century (post-1880).

100 See also the Artefact Overview, Section 4 and the Artefact Catalogue, Volume 6, and the various specialist reports, Section 8 for detailed information regarding the artefacts from the first phase of backfilling. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 264

Figure 3.5.104: The southern part of the mill yard is identified on the 1880 plan as a ‘Tenter Ground’ for the Vicars Tweed Factory. The red circle indicates the former location of the mill pond, and the arrow the section of it within the excavation area. City of Sydney Archives.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 265

Figure 3.5.105: Detail from north-facing section of the mill pond backfill dated from the early 1860s. Extract from Vol 4: Plan 10.44.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 266

3.5.11.3 Municipal Waste as Backfill From the mid 19th century fills were used to level and consolidate the landform created by reclamation throughout the site Waste product from the surrounding industries, consisting of clinkers, coke, ash, and slag nodules, was readily available and fills consisting of this material dominated the type used to re-level and re-surface the vast yard spaces. Included within these thin layers of compacted industrial waste were artefacts associated with household rubbish. The industrial waste material may have been collected and stored specifically for re-use as levelling fills. The presence of household rubbish items within the earlier phases of industrial waste levelling fills in the 1840s and 1850s may be indicative of a local waste management system, or be the result of more clandestine waste disposal and dumping. In general, the artefacts consisted of small fragments of ceramic, glass, metal, bone (not all collected) and this may be the result of re- deposition or indicate some form of processing (such as crushing) prior to being mixed with the industrial waste and used as part of the levelling fills.

Sydney was incorporated as a city in 1842, and the municipal council became responsible for providing a range of services for the city such as lighting, street upgrades, water and sewerage services, and rubbish collections. However, the development and management of these municipal services was haphazard and strained by increasing urbanisation and population growth throughout the latter half of the 19th century. Municipal waste, defined here as general household, street and local trade waste collected by the council and various contractors and dumped at the City Common, was also being used by private landholders as infill, at their request.101 Municipal waste collection and disposal was not always a priority and it seemed to take the outbreak of plague in 1900 to alert the city authorities to the health hazards associated with rubbish accumulation and inadequate disposal practices. An exchange between the council’s nuisance inspector and a member of the Health Board outlines the amount and type of rubbish collected in one week and how it was disposed of in the 1870s. A total of 1690 loads (in carts) were collected and these consisted of 1,007 loads of street sweepings, 432 loads of household rubbish, 25 loads of refuse from the city markets, 36 loads from earth closets and 190 loads of street refuse from gully shafts, 542 dead animals (dogs, cats, rats and fowls) were also collected, and were disposed of in the City Common, large holes in Harris Street, Macquarie Street, some went to the Domain and to Wynyard Square.102

The surrounding area was highly urbanised by the time the mill pond was backfilled in the early 1860s. Densely developed streets of terrace housing flanked Barker’s mill and the other industrial premises and depots such as the P N Russell & Co. Engineering Works and Murphy’s timber and coal yard. Interdispersed amongst these were numerous small businesses involved in local trade such as butchers, grocers and bootmakers. This type of local community would have generated a considerable and increasing amount of rubbish and waste throughout the latter half of the 19th century. The material used to backfill the mill pond was almost certainly sourced from the local area as industrial, household and trade waste are all represented in the archaeological record and artefact assemblage. The practice of municipal waste disposal within privately held land, and the increasing density of the local residential population, is reflected in the notable increase in the ratio of household rubbish to industrial waste in the mill pond backfill when compared to earlier and subsequent general yard levelling events.

The final sediment within the mill pond, prior to formal backfilling, was context 9249. This deposit also included a number of artefacts that may represent dumping within the mill pond during the final phase of its use but more likely they were items that sunk or were pressed into the sediment during the initial backfilling when fills consisted mostly of industrial waste; contexts 9248 and 9206. As such, artefacts from these three contexts represent items found within the first phase of

101 Fitzgerald 1992: 262. 102 Fitzgerald 1987: 76. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 267 backfilling dated to the early 1860s. The fills contained a relatively large number of artefacts; context 9206 contained a total of 203 items (415 fragments), 14 bone and 3 shell, context 9248 contained 30 items (51 fragments), 3 bone and no shell, and context 9249 contained 70 items (135 fragments), 15 bone and 4 shell, with the three contexts together then containing a total of 303 items (601 fragments), 32 bone and 7 shell.103 These artefacts, while from fills that consisted of industrial waste product they are generally indicative of household rubbish, and do not reflect industrial activities (Figure 3.5.106 to Figure 3.5.110).104 In fact, only four items can specifically be identified with the general function of industry; these being the leather off-cuts (4) from context 9206 and 9249.

Overall the artefacts within the three contexts are identified with functions and activities that generally take place within the household sphere, including the consumption of beverages (58), such as aerated water, beer/wine, champagne, gin/schnapps and ginger beer; the storage, preparation, serving and consumption of food (111), such as bowls, breakfast cups, children’s mugs, cups, dishes, jars, oil/vinegar bottles, pickle/chutney bottles, plates, platters, saucers, slop bowls, sugar bowls and tureen/vegetable dish; household items associated with maintenance and furnishings (10), such as blacking bottles, candle stick, matting, ornament, trivet and vase; personal items associated with clothing and hygiene (45), such as boots, button, ewers, galoshes, perfume bottle, poes, shoes, toothbrush box and wash basin; pharmaceutical items (8), such as patent medicine bottles and vials; and the recreational items (11) are all tobacco pipes. Two children’s mugs, both in context 9206, were associated with children. One has a black transfer print and features the remains of a verse on the exterior body, ‘If you.../’have your.../’done go if not s...’ (#73192), and the other is a clobbered design featuring the remains of a scene showing children playing marbles (#73126).105 There were a number of fragments of worn-out leather shoes and galoshes from the contexts that may be interpreted as discard from a local shoe maker, however, considering the context, some of these shoe/boot fragments were more likely discarded by local residents as part of their household rubbish (Figure 3.5.106).106

103 See Tables 77, 78, 79, Artefact Overview, Section 4. 104 Table 80, Artefact Overview, Section 4. 105 R Ward, Artefact Overview, Section 4, and Artefact Catalogue, Vol 6. 106 The Miscellaneous Report (Vol 3, Section 8.2) interprets the leather shoes/boots and galoshes from these fills as being discarded from a shoe maker due to the presence of 4 leather off-cuts in two of the fills. However, they may also have originated from a local resident as part of their general household waste, and perhaps more likely given the wear and tear noted on the items and the predominance of artefacts associated with household rubbish. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 268

Figure 3.5.106: Selection of shoes/boots from contexts 9248, 9249 and 9206. These were likely discarded as rubbish in a domestic context, collected with as municipal waste, and dumped with the initial backfill of the pond, that Left: Suede lace-up boots (9249/#8400), upper right: turned outsole (9248/#8397), lower right: duck-billed shoe/boot (9206/#8395). S. Kuiters. Scale 10cm.

From these three fills there were 14 identified ceramic patterns, on 41 individual items, in a range of colours, including blue, black, green and purple transfer print, along with blue flow, black flow and green flow.107 The majority of the patterns are commonly recognised on other Sydney historical excavations, especially ‘Albion’, ‘Asiatic Pheasants’, ‘Cable’, ‘Chantilly’, ‘Corsina’, ‘Fibre’, ‘Kulat’, ‘Rhine’ and ‘Willow’, with the ‘Rhine’ (9), ‘Asiatic Pheasants’ (6) and ‘Willow’ (6) patterns the most commonly represented. Four of the patterns did not begin manufacture until the 1860s, ‘Brussels’, ‘Cable’, ‘Chantilly’ and ‘Kulat’, and with the time lag between purchase and discard then considered, confirms that they were disposed of from post-1860 events. This is also reflected by the 21 artefacts that feature manufacturer, supplier or pattern marks, on stoneware bottles, plates, breakfast cups, glass aerated water bottles, glass patent bottles and tobacco pipes. The marks all feature long date ranges, with nine of the items continuing well into the 20th century.

The faunal remains identified in the three contexts 9206, 9248 and 9249 were very small, with just 32 fragments in total. Sheep are the dominant species (17), followed by cattle (5), Chicken (5), pig (2), dog (1), unidentified bird (1) and unidentified mammal (1). The small number of bones made any further analysis uninformative.108 This was also true of the shell, with just seven shells identified, five of these being Sydney rock oysters and the remaining two identified as a Sydney cockle and a hooded rock oyster.109

107 See R Ward 2010 Ceramics Report: Vol 3, Section 8.1 and Artefact Overview: Vol 1, Section 4. 108 Fillios 2010, Vol 3, Section 8.4. 109 See Table 83 and 84 in the Artefact Overview, Section 4 for faunal and shell remains from the fills. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 269

Figure 3.5.107: Selection of artefacts from the final sediment (context 9249) and the first phase of backfilling (context 9206) within the mill pond. Household rubbish was included with the industrial waste material used as backfill. Clockwise from top left: beer/wine bottle 9206/#75189, lens 9206/#80312, porcelain button 9249/#80330, polished wood and steel ink pen 9206/#80313, knife 9206/#80313, iron trivet 9206/#68763, tinware bowl 9249/#68759, C. Crop, London with 'J' 'L' spur mark 9206/#80320, pipe stem C. Crop, London 9206/#80319. Gallery2. Scale 1cm.

Figure 3.5.108: Condiment bottles and patent medicine bottle from the first phase of backfill (context 9206). Left: sauce bottle #75207, right: patent medicine bottle #75222. Gallery2Scale 10cm.

______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 270

Figure 3.5.109: All ceramic items from 9249, which included a penny ink bottle (#58632-58635 and #73246- 73257). Gallery2Scale 10cm.

Figure 3.5.110: Selection of ceramic patterns from 9206. Back row: Canton Views pattern plates in blue transfer print (#73156-#73158) and green transfer print (#73168). Middle row: Rhine pattern cups in black transfer print (#73184; #73185), blue transfer print (#73147; #73148) and green transfer print (#73162).Front row: Asiatic Pheasants pattern cups in black transfer print (#73182) and blue transfer print (#73142), Gallery2. Scale 10cm.

The second phase of backfilling within the pond was distinguished by the type of material used and the tip lines of the fills. More silty clays and other materials were used than just industrial waste product. Artefacts were collected during the machine-excavation through the backfill and were collected as context 9201. Several more contexts numbers were assigned to this phase during the recording of the test trench section (contexts 9214 to 9231). Though differing in the types of ‘soil matrix’, the second-phase fills also contained many artefacts originating from household rubbish. The artefacts from all these contexts were dealt with together, though the overwhelming majority of the 686 items came from context 9201.110 Context 9201 contained a total of 651 items (1,420 fragments), 54 bone and 10 shell, context 9211 contained 1 item (1 fragment), with no bone or shell.

110 See Table 85, Artefact Overview, Section 4. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 271

The artefacts from the second phase of backfilling are consistent with the types of artefacts contained in bulk household waste disposal deposits (Figure 3.5.111 to Figure 3.5.113).111 Only two items are specifically identified with the general function of industry, both leather off-cuts found in context 9201, and may have originated from a local shoe/boot maker.

Overall the artefacts in these backfills are associated with functions and activities that occur within the household sphere, with items identified with food (312 items) and beverage (188 items) being particularly dominant. Together these two general functions are associated with 500 items, together representing 72.9 per cent of artefacts recovered from this phase of backfilling. The most commonly identified single shape within this entire backfilling assemblage were the beer/wine bottles (154), with the remaining 34 beverage-related bottles comprising of alcohol, aerated water, champagne, ginger beer and stout. Food-related items were the most commonly featured, with a range of 27 individual shapes identified within the 312 items. These included items associated with preparation (bowls, dishes), storage (bottles, fish paste/meat paste jar, demijohn, jars, oil/vinegar bottles, oil bottle, pickle/chutney bottles, sauce bottles, vinegar bottles), serving (bowl/jug, dishes, jugs, platters, stemware, tureens) and consumption (bowls, plates, tumblers, breakfast cups, cups, milk/cream jug, saucers, slop bowls, small plates, teapot). Out of these 312 food-related items, the most commonly identified were associated with consumption of meals and beverages (219), with plates (84), saucers (53) and cups (45) the most commonly recognised shapes.

Other examples of household-related activities represented within these backfills include those associated with clerical activities (ink bottles, pen, penny ink bottles, slate pencil), household maintenance and decoration (blacking bottles, figurine), personal hygiene and clothing (ewer, ointment jars, perfume bottles, poes, toothbrush boxes, wash basins, boots, galoshes, shoes), pharmaceutical (various bottles, jars) and recreational (tobacco pipes).112 There were a number of other leather shoes and galosh fragments in context 9201. This context represents at least 20 separate fills or tipping events that contained many artefacts identified with household rubbish, and as was the case for leather items from the earlier phase of backfill, these other shoe leather and galosh fragments are likely to have been discarded as part of general household rubbish (Figure 3.5.111).113

In general the date range of the artefacts from context 9201 et al was very broad; ranging from items beginning manufacture in the late 18th century, to those beginning production in the mid- 19th century and continuing into at least the mid 20th. However, two items, both in context 9201, were identified as beginning manufacture in the 20th century. A blue flow saucer was recovered with a transfer-printed basemark featuring a scroll cartouche with ‘CHUSA(N)’ / ‘D.P. & Co.’, the initials belonging to the Diamond Pottery Co and dating between 1908 and 1935. The second item was a base fragment from a glass bottle featuring the embossed trademark logo of the Australian Glass Manufacturers and dating from 1930.114 The wide range of dates found here is understandable given that the mill pond was backfilled by a series of levelling events and not a single dumping episode. The excavation methodology also resulted in the mixing of artefacts from all levelling events. The inclusion of early 20th-century items, while intrusive to the phase they have been associated with, does demonstrate that the process of levelling and land consolidation with fills that contained general household rubbish continued beyond the 19th century.

111 See Table 86, Artefact Overview, Section 4. 112 Details on artefacts from R Ward 2013, Artefact Overview, Section 4. 113 The Miscellaneous Report interprets all the leather items from the mill pond backfill as being discarded stock from a shoe maker. As the leather items were not all from one context or individual filling events, and included with items that are undoubtedly domestic in nature, it is more likely that the shoes (all or some) are from household rubbish deposits. 114 Casey & Lowe artefact database. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 272

Figure 3.5.111: Selection of shoe/boots from context 9201, second phase of backfilling representing at least 20 separate filling events. Top right: vulcanised rubber galosh #8369, top left: child’s lace-up shoe (#8370), bottom: suede ankle boot (#8373). S. Kuiters. 10cm scale.

From the second phase of backfilling a total of 32 ceramic patterns, on 123 individual items, were identified.115 Thirteen of these patterns were also found in the first phase of backfilling. The patterns feature mainly on food-related items, particularly plates (50), followed by dishes (13), cups (12), saucers (9), bowls (8), platters (8), breakfast cups (5), tureens (4) and slop bowls (2). The few non-food items are hygiene shapes, such as ewers (2), poes (2) and wash basin (1), or unidentified (6). Many of the patterns are commonly seen in assemblages from other Sydney sites.116 These include ‘Albion’, ‘Asiatic Pheasants’, ‘Cable’, ‘Chantilly’, ‘Corsina’, ‘Fibre’, ‘Forest’, ‘Gem’, ‘Kulat’, ‘Palestine’, ‘Peacock’, ‘Rhine’, ‘Spangle’ and ‘Willow’, with the ‘Willow’ (26), ‘Albion’ (20) and ‘Rhine’ (13) patterns were the most commonly represented. Six of the patterns did not begin manufacture until the 1860s (‘Antique’, ‘Cable’, ‘Chantilly’, ‘Dulcamara’, ‘Kulat’ and ‘Peacock’), one until the 1870s (‘Brooklyn’) and one did not begin production until 1908 (‘Chusan’).

115 See Table 87, Artefact Overview, Section 4. 116 R Ward 2010: Vol 3, Section 8.1. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 273

Figure 3.5.112: Selection of glass bottles from the second phase of backfill (context 9201) within the mill pond. Left to right: Champagne bottle (#75176), vinegar bottle (#75181), pickle/chutney bottle (#75178), and Worchester sauce bottle (#75167). Gallery2. Scale 10cm.

Figure 3.5.113: A selection of ceramic from the mill pond backfill (context 9201). First row (left to right): Albion (#72964); Alma (#72894); Antique (#72982); Asiatic Pheasants (#72943); Berlin (#73056); Brooklyn (#72892); Cable (#72887). Second row: Canton Views (#72990); Chantilly (#73044); Chusan (#73061); Citron (#73043); Continental Views (#72975); Corsina (#72978); Dulcamara (#73038); Eton College (#72976). Third row: Fibre (#72912); Forest (#72950); Gem (#72948); Genevese (#72993); Kulat (#72973); Palestine (#72949); Peacock (#72971); Rhine (#72868); Rose (#73060). Bottom row: Royal Cottage (#72972); Spangle (#72889); Sydenham (#72914); Versailles (#72890); Vine (#73024); Whampoa (#73062); Willow (#73022). Gallery2. Scale 10cm.

Faunal remains were identified in just two of the contexts associated with this second phase of mill pond backfilling, in contexts 9201 and 9230.117 A total of 58 bone fragments were recovered, with 54 in context 9201 and just 4 in context 9230. Sheep were the most dominant species (39), followed by cattle (16), chicken (2) and pig (1). All major body parts were represented with both the cattle and sheep bone having cut and saw marks. Rather than originating from household waste, analysis of the bone in context 9201 considered it to be consistent with either commercial or

117 Fillios 2010, Vol 3, Section 8.4. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 274 industrial waste, such as an abattoir, butcher, tanner or soap/candle maker. This is due to the relatively high frequency of appendicular elements (limbs) compared to other skeletal parts.118 These same two contexts also contained the only shell found in this backfilling phase, with a total of just 11 Sydney rock oysters identified, with ten in context 9201 and just the one in context 9230.

The artefacts from the final phase of backfilling represented by contexts 9212 and 9213 were not separated from the preceding phase (9201 et al) during artefact analysis. Interpreting these contexts as being part of later 19th-century or early 20th-century levelling events was the result of analysis for the synthesis of results (Section 3) and is based on the RLs and presence of early 20th- century artefacts in the pond backfill assemblage.

The artefacts from the two separate phases of backfilling, while overwhelmingly domestic in nature, also include trade-related items such as waste product from a shoe/boot maker, butcher and/or tanner. The artefacts and their inclusion in the mill pond backfill are associated with household rubbish disposal practices and municipal waste management in Sydney in the late 19th century. It is likely that waste from the local community, such as the residents of Steam Mill Street (Area 8), and the various small local businesses was collected by the council and disposed of within the nearest available hole in need of filling, such as the redundant mill pond. The presence of a number of near whole or whole items may also indicate a swift disposal. Further research, analysis and comparison with other assemblages from the reclamation and levelling fills, cesspit backfills and other rubbish dumps may reveal more information regarding industrial and municipal waste, its disposal, secondary uses and deposition patterns within the site, and within the wider context of the city of Sydney.

3.5.12 Discussion of Results The 1820s and 1830s were a time of great entrepreneurial opportunity in the colony. During the early decades of the 19th century, New South Wales evolved from its origins as a penal colony and staging post, and fast became a valuable asset to the British Empire in terms of trade and economy.119 Initially, exporting high-demand products from the cattle industry, sealing and whaling was a main driver in the burgeoning local economy. However, it was the increasing demand for wool to supply the looms of industrialised England that provided major stimulus to the colony’s economic growth, especially from the 1830s. Following a decade of focus and investment in a public building program by Governor Macquarie, new policies introduced by Governor Brisbane in the early 1820s aimed to encourage free trade. New incentives, such as land grants, were offered to capitalists.120 The pastoral industry, supported by continued assignment of convict labour, was expanding and the duties on the colony’s important export, wool, were lowered by Brisbane.121

Expansion of the primary industries, economic and trade growth, and a steadily increasing population of free settlers, government staff and convicts, led to development and growth of secondary industries, the professional and merchant class.122 Sydney was the colony’s main urban centre. It had the largest population; at the beginning of the 1820s it was estimated to be around 7,000 and by the end of the 1830s it was almost 30,000.123 The town functioned as a local trade port and a clearing port for exports, and it was also the location for most of the colony’s early local industries.124 The establishment of the Government’s Market Wharf by Governor Macquarie in

118 Fillios 2010, Vol 3, Section 8.4. 119 Edwards 1978. 120 Edwards 1978: 37. 121 Edwards 1978: 37. 122 Edwards 1978: 43. 123 Aplin 2000: 58. 124 Wotherspoon 1978: 12. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 275

1810 shifted the geographical focus for commercial and trading activities to the southern urban fringe and the eastern foreshore of Darling Harbour (Cockle Bay). Shortly after in 1815, the colony’s first steam-powered flour mill was established on the harbour by John Dickson. Located at the head of the harbour and close to the market district, Dickson’s mill was at the forefront of private enterprise and development on the harbour. The subsequent private development of port facilities, local manufacturing industries and commercial businesses on or connected to Darling Harbour was crucial to the colony’s economic growth in the first half of the 19th century.

The archaeology of Barker’s mill, jetty and mill pond can be viewed as the material remains of this spirit of entrepreneurial opportunity at the advent of an evolving and diversifying colonial economy. Cooper & Levey were responsible for the initial construction of the mill. By the end of 1825 their investment had successfully transformed the large waterfront land grant to include a large stone mill building powered by an imported steam engine, a wharf and a freshwater reservoir. The business partnership of Cooper & Levey was one of the most successful in the colony in the 1820s and they serve to exemplify its entrepreneurial fervour. Both arrived as convicts; Solomon Levey in 1815 and Daniel Cooper in 1816, and by the early 1820s they had established themselves as successful merchants.125 Owning or part owning a number of ships, they were involved in trading goods in the local and intercolonial markets.126 Newspaper reports of Cooper & Levey defending themselves in court over accusations of diluting gin after its sale and smuggling tobacco indicate that their enterprises may have been somewhat dubious and opportunistic, though this perhaps is in keeping with the colony’s pervading entrepreneurial ethos of the time.127

Cooper & Levey while involved in the expansion of the flour milling industry, were also land speculators and property developers. In 1820 Daniel Cooper was a partner in the Lachlan and Waterloo Flour Mills which was acquired by Cooper & Levey in 1825.128 At the same time, they were investing in the construction of the mill at Darling Harbour. Regardless of whether Cooper & Levey intended this to be a speculative development destined for quick sale or an expansion of their own milling enterprise, it further demonstrated their business acumen and ingenuity. The value of waterfront land at Darling Harbour was well recognised and often referred to in newspaper land sale advertisements. The site was located near the markets and commercial trade area, it had a supply of freshwater, and though it was harbour-fronting, its natural attributes were not initially conducive to the port facilities necessary for industrial development. The site’s potential was only realised by Cooper & Levey’s development. Identifying a growth market, they constructed a flour mill to supply a product that was in high demand. The flour mill was powered by a steam engine, and produced flour that was sold at welcome competitive prices: The Steam Engine of Messrs Copper and Levey erected in Cockle Bay, for grinding, wheat, .&c. has been in active service now upwards of a fortnight.— Nothing can more powerfully show the spirit of enterprise which actuates the merchants of this Colony, than undertakings of this magnitude. Nor can a greater proof be given of the rapid progress the Colony is making, than the successful and spirited manner in which they are carried into effect. We cannot, indeed, help speaking of them in terms of wonder and of pleasure, at witnessing their effects, and at the same time, sources of wealth in a community, as it were of but yesterday's growth. This Steam Engine was only imported last June in the Lalla Rookh, and the stupendous pile of

125 Bergman 1964, 1967 and 1968; Davidson 1966. 126 Cooper & Levey appeared to have business interests in Hobart in the early 1820s. They pursued a writ of fieri facias in December 1823 against Espie & Clark, on property at Bagdad, north of Hobart. Reported in the Hobart Town Gazette 20 December 1823, p 1(3). 127 In October 1823, Solomon Levey refuted claims that Cooper & Levey were diluting gin sold to F E Forbes, possibly from the ship Actaean. Reported in Sydney Gazette 30 October 1823, p 2(3). In 1824, it was reported that Solomon Levey was defending himself from allegations of smuggling in relation to Cooper & Levey’s trading of tobacco. Reported The Australian 23 December 1824, p 4(1). 128 Sydney Gazette 24 February 1825, p 4(3) and The Australian 24 February 1825, p 1(3); Bergman 1964: 406, citing Wentworth Papers Miscellaneous, p 160 (Mitchell Library). ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour 276

building which contains it, has been erected since that period —a satisfactory proof of the means and wealth of the Colony. (The poor of the Colony will have great reason to rejoice at, the additional facility thus given to the grinding of wheat, as it must tend to make the price of flour and of bread bear a more just proportion to that of wheat. "Money will make money," says the proverb, and Messrs. Cooper and Levey seem to understand the force of it. It certainly is much to their credit, that in exemplifying the proverb, they render themselves, not only useful members of society, but benefit those who need it most — the Poor.129

Their development also showed a degree of creativity and resourcefulness in transforming the expansive intertidal sand flat into a suitable wharf facility for the vessels that transported grains from the hinterland to the mill. Reclamation and jetty construction turned this well-situated property into a desirable and in-demand property type, which was sold to Thomas Barker a year and half after it became operational. Cooper & Levey’s mill complex development laid the foundations, so-to-speak, for one of the more successful, significant and well-known local manufacturing industries, Barker’s mill.

Thomas Barker’s background and social standing was quite a contrast to that of Cooper & Levey. In 1813, he arrived in Sydney with his guardian John Dickson and was apprenticed in his mill.130 By the time he purchased Cooper & Levey’s steam-powered mill, he was already involved in the milling industry in his own right, being the owner of a windmill in Darlinghurst.131 Not only was he a manufacturer and engineer, he was a pastoralist who had acquired large acreages of land for grazing sheep and cattle. During the 1830s, these two facets of his business were very successful given the market, both local and export, demand and positive economic conditions of that decade. They were likely to have provided him with the capital and resources to survive the following decade of cyclical economic depressions. Also during the 1830s, Barker invested in large-scale reclamation of the mill’s foreshore as land speculation, though profit from this may have taken some time to materialise given the slow sales and subdivision failures during the poor economic conditions of the 1840s.

Barker was also later to become quite heavily involved in public life, banking and the development of railways. His ability to diversify his business interests was perhaps the main reason for his success and stature. The expansion of the mill by the 1840s to include two engines and its diversification in the mid 1840s to include textile production showed a level of both risk and optimism. These changes were detectable in the archaeological record. The mill pond that supplied freshwater became inadequate as it silted and required flood management measures such as the overflow drain. To secure a more constant supply of water, saltwater was drain from the harbour to the boilers, and as there were now two, production in the mill would not completely stop during the resulting more frequent engine system shut-downs to deal with the adverse affects of using salinated water.

The archaeological remains of the mill complex are associated with the early entrepreneurs and industrialists that shaped and drove the colony’s economy in the 1820s and 1830s. The combination of opportunity and ingenuity led to the transformation of the natural landscape to provide space and facility for one or the more significant industrial businesses of the 19th century: Barker’s Mill.

129 The Australian 19 December 1825, p 2. 130 Walsh 1966a. 131 Bergman 1964. ______Casey & Lowe Archaeological Investigation 2008-2010 Darling Quarter, Darling Harbour