James Matthews - conservation reportJAMES MATTHEWS CONSERVATION PRE-DISTURBANCE SURVEY REPORT Ms Vicki Richards Research Officer Department of Materials Conservation Western Australian Museum May 2001 With contributions from Mr Mark Barrett Mr Jon Carpenter Dr Ian Godfrey Mr Kalle Kasi Mr Geoff Kimpton Dr Ian MacLeod Dr Michael McCarthy Mr Adam Wolfe INTRODUCTION In July 2000, staff of the Department of Materials Conservation, Western Australian Museum were invited to be involved in a limited excavation of the James Matthews. Ms Vicki Richards, with assistance from staff of the Department of Materials Conservation and Maritime Archaeology, carried out an on-site conservation pre-disturbance survey on the wreck site concentrating on establishing the state of preservation of the hull remains and their suitability for subsequent recovery, conservation and exhibition. In addition, the Conservation Department was requested to provide an estimated total cost for conserving the wreck. Hence, this interim report is divided into two parts. The first section will discuss the results obtained from the conservation pre-disturbance survey. The second section will briefly discuss the treatments available for a wreck of these dimensions and a very global estimation of the total costs for the conservation of the James Matthews. PART 1 ON-SITE CONSERVATION PRE-DISTURBANCE SURVEY REPORT 2000 JAMES MATTHEWS DATE WRECKED: 22 July 1841 DATES OF INSPECTION 20 September 2000 10, 24, 25, 27, 30, 31 October 2000 1, 2, 16, 17 November 2000 5, 6, 7, 8, 11 December 2000 Dive times for the period of this conservation pre-disturbance survey are reproduced in full in Appendix A. WEATHER AND SEA CONDITIONS The weather over the sixteen days of the survey period followed typical summer patterns. The weather was mostly fine (20-36ºC) with moderate to fresh E/SE winds in the mornings that shifted S/SW by noon/early afternoon and increased in strength towards the late afternoon. A few strong wind warnings were issued over the period of the inspection. Seas were typically slight to moderate (0.5m-1.5m) on a long, low swell (0.3-0.5m). Infrequently there was a slight current flowing in a SE direction. Tides were mixed but mainly diurnal and the minimum low tide and maximum high tide during the survey period was 0.3m and 1m, respectively. The average diurnal range was 0.35±0.17m. The on-site visibility was usually poor, ranging from 0.5 to 1.0m. Occasionally, the visibility increased to 2-3m and on one occasion the visibility was excellent at 5-6m. There appeared to be no connection between the wind and sea conditions and the on-site visibility. Consecutive days with similar sea and weather conditions produced significantly varying in-water visibility. The site is in close proximity to the Cockburn Cement jetty. Large barges deposit loads of calcareous shell and sediment at the end of the jetty where it is dredged and piped into the cement works. The fine particulate material is transferred to the water column on dumping. Consequently, a fine suspension cloud repeatedly moves across the wreck site, dependent on the wind direction and sea conditions, significantly reducing the visibility to less than 0.5m. On one particular day of the survey, the fine sediment was concentrated at the sea water surface, however, after a depth of about 60 cm into the water column the visibility improved. Interestingly, divers working the wreck site transported clear water up into the area of low visibility and “paths” of clear water could be seen trailing after snorkellers. The water temperature at depth increased quite markedly over the three month survey period. The temperature increased from 17ºC measured on the first day of the survey (20/9/00) to 24ºC measured on the last day of the survey (7/12/00). There was no significant temperature gradient throughout the water column (0-3m). The average pH of the sea water was 8.25±0.04 and the average redox potential was 0.142±0.072V. The change in salinity, dissolved oxygen content and temperature of the sea water with depth are shown in Table 1. The average salinity was 34.5ppt. The average dissolved oxygen content compensated for the salinity measured on two consecutive days was 7.88±0.04ppm(S) and 7.73±0.07ppm(S), respectively. Table 1. Salinity, dissolved oxygen content and temperature of sea water on the James Matthews wreck site. Water Depth from Surface (m) Measurements 6/12/00 Measurements 7/12/00 Salinity (ppK) Dissolved Oxygen Content (ppt) Temperature (ºC) Salinity (ppK) Dissolved Oxygen Content (ppt) Temperature (ºC) 0 34.9 7.85 24.3 34.0 7.75 24.4 0.5 34.9 7.85 24.3 34.0 7.80 24.4 1.0 34.9 7.88 24.3 34.0 7.77 24.4 1.5 34.9 7.94 24.3 34.1 7.77 24.3 2.0 35.0 7.91 24.3 33.9 7.68 24.2 2.5 34.9 7.88 24.3 33.9 7.62 24.0 Site Location The James Matthews wreck site lies on the north side of Woodman Point in Cockburn Sound approximately 100m south west of the Cockburn Cement jetty and about 100m off shore (Figure 1). The GPS position is latitude 32º 07.9300`S, longitude 115º 44.6200`E [Kenderdine, 1995]. The site is relatively protected against rough sea conditions in all weathers with the exception of north westerly winds. Description The partial isolation from the prevailing longshore current and ocean swell combined with a narrow tidal range has greatly reduced ocean flushing and sediment interchange between the open ocean and Cockburn Sound. The wind is the most important force driving the water circulation within the Sound. The wind pattern during a typical summer sea-land breeze system with wind speeds in the 0-15cms-1 range causes a slow, well developed anticlockwise gyre of approximately 5-20cms-1. In Cockburn Sound the sublittoral profile is characteristically a wide, gently shelving fringing platform plunging steeply to a flat basin floor consisting of fine carbonate muds. The shoreline from Fremantle to Cape Peron is now largely stable, with its stable forms generally controlled by structures erected at the shore [Department of Conservation and Environment 1979]. The seabed in the vicinity of the wreck is relatively level. The seabed surface comprises of loose calcareous sand. These materials overlay a denser calcareous sandy gravel derived from the weathered lithified Tamala limestone [Environmental Management Services 1994]. There are very sporadic, small patches of sea grass in close proximity (<50m) to the site, however, there are none directly on the wreck site. There are no reefs in close proximity, however, Woodman Point is approximately 200-300m south of the wreck. No rivers or streams discharge into Owen Anchorage and the only source of fresh water is restricted to groundwater except for local stormwater run-off from the mainland discharging across beaches and direct precipitation. Hence, the effect would be minimal. Cockburn Sound has extensive industrial and urban developments on its coastline. During the last thirty years, it has been used as a disposal site for municipal sewage and industrial wastes. Due to these effluents, Cockburn Sound basin sediments are moderately polluted with lead, mercury and zinc which has caused some deterioration of the water quality and contamination of some of the marine biota with these heavy metals. The major inputs of heavy metals to Cockburn Sound were associated with the gypsum effluent from the CSBP fertiliser factory situated towards the southern end of the Kwinana industrial area and sewage effluent from the Woodman Point sewage treatment plant (WPTP) located at the northern edge of the Kwinana industrial area, adjacent to Jervoise Bay. Over a five year period, between 1979 and 1984, the heavy metal input into Cockburn Sound was greatly reduced by the diversion of sewage effluent and changes in industrial practices at Kwinana. However, heavy metals are not biodegradable and they must be removed from the sediment by dilution or dispersal. As Cockburn Sound is partially isolated from longshore sediment and water movements, both dilution and dispersal are likely to be occurring extremely slowly. Therefore, large areas of the Cockburn Sound basin remain contaminated with significant quantities of anthropogenic heavy metals [Monk & Murray 1990; Murphy 1979]. The Woodman Point Wastewater Treatment Plant (WPTP), prior to 1979, was also a major contributor of total nitrogen and ammonia, phosphorus and sulphides to Cockburn Sound. Excessive nitrogen and phosphorus levels cause eutrophication of water bodies and the effects are varied but are always associated with extensive changes in water quality. Sulphides are rapidly oxidised in sea water but unless sufficient mixing with sea water takes place then depletions in dissolved oxygen concentrations can occur. However, this situation produces a local, rather than regional effect [Department of Conservation and Environment 1979]. These effluents have also been associated with eutrophication and damage to the seagrass meadows that are an integral part of the protective sand banks of Cockburn Sound and provide sheltered habitats for marine fauna [Monk & Murray 1990]. The deterioration of seagrass meadows in Cockburn Sound is attributed primarily to the combination of increased phytoplankton blooms and high epiphyte loads. Where epiphyte cover is high, light levels reaching the seagrass were reduced with a consequent decline in seagrass leaf production. WRECK General Observations The James Matthews was a snow brig of 107 tons, believed to have been built in France. It was a copper sheathed wooden vessel, fastened with copper, iron and wooden treenails [Henderson, 1980, 1976]. The wreck lies on a north west-south east axis, at a depth of approximately 1.8±0.4m, covering an area about 25m x 7m x 2m (Figure 2). The vessel lies almost level with little variation in depth between bow and stern on a calcareous sandy seabed. The wreck is mostly buried to a depth of approximately 1.5-2.0m. The wreck lies on the starboard side and has been preserved to the bulwarks by the sand cover [Baker & Henderson, 1979]. The concreted surfaces of the iron structures are quite densely covered with sessile marine organisms including mussels, sponges, barnacles, ascidians, tunicates and a variety of seaweeds. In less concentrated areas of growth, small algal forms are present. Exposed timbers are generally free of marine growth but exhibit extensive damage caused by marine borers. The most prominent fish species are bullseyes and angelfish. Degree of Site Exposure The most exposed areas of the wreck are the iron deck knees on the starboard side, the slate mound towards the stern and the windlass located at the bow. In profile, they rise approximately 0.5-1.5m above the sediment. Astern of the slate mound, many of the structural timbers are exposed to a depth of 0.5m and active marine borers were observed. There are many other areas on the wreck site that are only covered with relatively thin layers of sand. Ballast lies off site, on the port side of the wreck (Figure 2). Seasonal Exposure Exposed wooden structural features are extensively degraded by marine borers and the iron structures are covered in thick, aerobic concretions and heavily colonised indicating that they have been exposed to an aerobic marine environment for an extended period of time. At the time of the survey, there was no evidence of seasonal exposure on this site. That is, in the excavated test trenches, there were no dead marine organisms present on the freshly exposed timbers that would indicate previous exposure to an aerobic marine environment. In addition, these timbers possessed the usual black discolouration associated with wood recovered from anaerobic micro-environments strongly suggesting that this site is not subjected to periodic burial and exposure cycles. Hence, at this point in time, it appears, that in most areas, the site is experiencing its maximum extent of exposure and has been exposed to this depth for some time. Collection of sediment samples for biological, chemical and physical analysis revealed the typical grey to black discolouration at a depth of approximately 20-30cm, indicating an anaerobic environment at about this depth. This implies that the depth to stable sediment is, on average, approximately 25cm. Human Disturbance Four seasons of excavation were carried out on the James Matthews site between 1973 and 1977. At the end of 1973 a triangulation survey of the site was carried out prior to the commencement of a limited excavation in January 1974. This excavation exposed and raised the upper levels of the slate mound in order to ascertain the extent and condition of the site. Very little of the wreck was visible above the sediment/sea water interface prior to the first excavation in 1974. The slate mound and a bundle of large iron lengths and some iron knees were exposed. There was also very extensive coverage of the site with sea grass meadows (Figure 3) [Henderson, 1976]. During the summer 1974-1975 excavation, the area immediately astern of the slate mound was uncovered, the hull recorded and any exposed artefacts were raised. The site was totally exposed during the 1975-1976 summer season, a detailed site plan was completed and small artefacts were raised (Figure 4). A final short excavation was carried out during the summer of 1976-1977. This excavation surveyed and raised the material found outside the main hull structure [Baker & Henderson, 1979; Henderson, 1976; Henderson & Stanbury, 1983]. At the completion of each excavation period the site was reburied to minimise the destructive effects of marine organisms and the physical damage caused by water and sand movement [Baker & Henderson, 1979]. The site appeared to be relatively stable and remained essentially covered with sediment for many years. In 1989, during commercial dredging operations by Cockburn Cement, some timber ribs were recovered and it was noted that there was no sea grass present on the site but it remained essentially buried (Appendix B) [MacLeod, unpub.]. However, a recent visit in early 2000 indicated that there had been extensive scouring and uncovering of the site, especially in the previously excavated hull section astern of the slate. The wreck lies in an area that is regularly traversed by barges that deposit large quantities of shell and sand at the end of the Cockburn Cement jetty. This sediment is then dredged and piped into the factory. The wreck is about 100m south west of this jetty and the effects from propeller wash would be minimal, however, the long term effect of the dredging of deposited shell and sand on sediment transportation in the general vicinity of the wreck site should not be overlooked. In addition, the presence of snagged fishing lines and other modern material suggests local interest in this location for fishing activities. This means that anchors from moored vessels could potentially damage the site. During the 2000 survey period, six test trenches (~2x2x2m) were dredged at various positions on the site (Figure 6) so that the extent of degradation of exposed and buried structural timbers could be ascertained. Probe depths, pH profiles and core samples of selected aerobic and anaerobic timbers were collected. Twelve iron structural features were drilled on two or three separate occasions to measure the corrosion potentials. The sediment was core sampled on and off site for chemical, geological and bacteriological analyses. At the conclusion of the survey the test trenches were backfilled to rebury the hull timbers. Metal Survey Iron The corrosion potentials and pH values of the residual metal surfaces of the major iron structural features were measured and the results are presented in Table 2. The on-site position of the iron material is shown in Figure 2. Table 2. Corrosion potentials of exposed iron material on the James Matthews site. Description of Ferrous Material pH Corrosion Potential (rel. NHE) (V) Depth of concretion + graphitisation (cm) Water Depth (m) Deck knee 1 7.11 -0.216 7.5 2.0 Deck knee 2 7.11 -0.222 7.5 2.0 Deck knee 3 total penetration 15 1.8 Deck knee 4 6.67 -0.093 (unstable) 10 1.9 Deck knee 5 6.89 -0.216 8 1.9 Deck knee 6 7.11 -0.275 2 1.7 Deck knee 7 6.89 -0.209 2.5 1.7 Deck knee 8 6.58 -0.207 2.5 1.4 Hoop 9 7.18 -0.084 9 1.7 Windlass 10 6.71 -0.216 12 1.9 Concretion at bow 11 6.40 -0.234 2 1.5 Star picket 12 8.08 -0.320 2 1.7 The iron structural features exposed to this oxidising marine environment were covered with thick aerobic concretions, heavily encrusted with secondary marine growth. Iron is not biologically toxic and increases the growth rate of encrusting organisms. The concretion acts as a semi-permeable layer on the surface of the iron, effectively separating the anodic and cathodic sites and produces an acidic, iron and chloride rich micro-environment at the residual iron surface. By plotting the measured voltages and the corresponding surface pHs on the Pourbaix diagram for iron at 10-6M in aerobic sea water at 25ºC the thermodynamic stable state of the iron can be ascertained. All iron material measured on this wreck site is actively corroding. The total depth of concretion plus graphitisation ranged from 2-15cm with an average of 7cm. This indicates that corrosion of some of the iron features on-site has been quite extensive. It was extremely difficult to ascertain the depth where the concretion layer ceased and the corroded metal surface began, therefore the annualised corrosion rate could not be calculated. However, the standard corrosion rate for isolated iron in aerobic sea water is 0.1mmy-1 and the wreck has been exposed to the aerobic marine environment for 159 years. From this information, if the iron is corroding at the standard rate, then the depth of graphitisation should be about 2cm. Hence, it could be safely assumed that some of these iron structures with very large total depths of concretion and graphitisation have been corroding at an accelerated rate. The temperature, salinity and dissolved oxygen concentration of the sea water was relatively constant over the survey period of this wreck site. Measurements of dissolved oxygen in the sea water did not change significantly with water depth on this shallow site, therefore, the extensive corrosion of some iron fittings cannot be directly due to dissolved oxygen levels. However, one parameter that does increase markedly with decreasing water depth is the total amount of water movement. Owen Anchorage is sheltered from prevailing ocean currents and swell with the exception of NW winds, but turbulent motion associated with waves caused by traversing barges, sea bed topography and wreck orientation occurs on this site. Some sections of the site have been recently uncovered and scouring has occurred around the iron deck knees. The iron structural features are now exposed to a much greater extent than previously. In shallow depths there is much greater total water movement and the higher the wreck profile above the sea bed, the greater the turbulence. Since corrosion rates are largely determined by the dissolved oxygen flux to the metal surface, increases in overall water movement will increase the flux of dissolved oxygen to the iron object. Consequently, the mean corrosion rate will increase. An increase in the area exposed to the dissolved oxygen flux will also increase the overall corrosion rate [MacLeod 1989, 1988; MacLeod & North 1980]. Corrosion potentials of three star pickets were taken in January 1989 by Dr Ian MacLeod and the measurements are presented in Table 3. Table 3. Corrosion potentials of exposed star pickets on the James Matthews site measured in January 1989 (Appendix B) [MacLeod, unpub.]. Star Picket pH Corrosion Potential (vs NHE) (V) 1 7.17 -0.282 2 8.03 -0.280 3 6.92 -0.299 The pickets were colonised by sponges, tunicates and algae. The water temperature was 23.5ºC. The pH and salinity of the sea water was 8.05 and 34ppt, respectively. The chemical and physical parameters of the sea water environment are similar to those measured during this survey period. The corrosion potentials of the iron structural features measured in 2000 are more positive than the measurements taken in 1989. These more positive potentials may reflect an increase in the overall corrosion rate on the site since 1989, however, the same star pickets measured in 1989 could not be located during this survey for direct comparison. It is very difficult to compare corrosion potentials of different iron features if the metal compositions are not known. Cast iron typically contains 2-6% carbon by weight and has a more positive corrosion potential than wrought iron due to the enobling effect of the carbon. Organic Survey Wood During the initial inspection of the site, extensive marine borer damage was evident on exposed timbers. The extent of degradation of exposed and lightly covered structural timbers varied considerably. The approximate dimensions of the major structural timbers are as follows: frames (18x12cm), outer planking (25x5cm), inner planking (20x4cm) and the keelson (33x20cm). Probe depths of exposed wood ranged from 0.5cm to total penetration of the full width of the timber and this included some frames and sections of the keelson. The average probe depth was 4±3cm. This average depth of penetration indicates that many of the exposed structural timbers are very degraded. Biological, chemical and physical degradation of wood occurs, to some extent, on all shipwreck sites, however, it is proposed that biodeterioration and physical damage are the major mechanisms causing the extensive degradation of these exposed hull timbers. In-situ pH profiles and probe depths of selected aerobic and anaerobic structural timbers in four of the six test trenches were obtained. Samples of these measured timbers were collected for wood identification, maximum water content (Umax) and specific gravity analyses. Only probe depths were obtained for timbers exposed in test trenches 4 and 5. The results of these measurements are presented in Table 4. Table 4. Identification and physical measurements of wood samples recovered from test trenches on the James Matthews. Description Position Average Probe Depth (cm) Wood ID Core Depth (cm) Max Water Content (%) Specific Gravity TEST TRENCH 1 TT1 outer planking 1 aerobic 3 too degraded 0-2.5 103 0.59 2.5-4 106 0.58 4-6.5 88 0.65 TT1 outer planking 2 aerobic 4 white oak? disintegrated 96 0.61 TT1 outer planking 3 anaerobic <0.5 birch 0-1 198 0.38 1-2.5 116 0.55 TT1 keel anaerobic <0.1 beech 0-1 106 0.58 1-2 107 0.58 TEST TRENCH 2 TT2 keelson aerobic 4 white oak 0-2 150 0.46 2-4.5 135 0.49 4.5-6.5 135 0.50 6.5-8.5 97 0.61 TT2 inner planking aerobic 3 (total penetration) white oak 0-1 216 0.35 1-2 257 0.31 TT2 outer planking 2 anaerobic 0.75 elm 0-1 333 0.25 1-2 178 0.41 2-3 208 0.36 3-4 203 0.37 TT2 frame 2 anaerobic <0.2 white oak 0-1 148 0.47 1-2 130 0.51 2-3 152 0.46 3-4 153 0.46 4-5 153 0.45 5-6 169 0.43 TEST TRENCH 3 TT3 inner planking 1 aerobic 3 white oak 0-3 126 0.52 3-5 107 0.57 5-7 111 0.56 7-9 200 0.38 TT3 inner planking 2 anaerobic <0.2 white oak 0-1 123 0.53 1-2 125 0.52 2-3 140 0.48 3-4 147 0.47 4-5 162 0.44 5-6 174 0.42 6-8 166 0.43 TT3 inner planking 3 anaerobic <0.2 white oak 0-1 437 0.20 1-2 147 0.47 2-3 130 0.51 3-4.5 163 0.44 TT3 inner planking 4 anaerobic <0.5 white oak 0-1 453 0.19 1-2 358 0.24 2-3 484 0.18 TEST TRENCH 4 TT4 inner planking 1 aerobic 2 TT4 frame 1 aerobic/ anaerobic 2.5/ 0.5 TT4 frame 2 aerobic 1 TT4 frame 3 aerobic 2 TT4 frame 4 aerobic 2 TT4 timber 5 aerobic 1.5 TEST TRENCH 5 TT5 inner planking 1 aerobic 2 TT5 frame 1 aerobic 2 TT5 frame 2 aerobic 5 TT5 frame 3 aerobic 4 TT5 outer planking 1 aerobic 5 TEST TRENCH 6 TT6 frame 1(1) aerobic 5 white oak 0-2 133 0.50 2-4 108 0.57 4-6 169 0.42 TT6 frame 1(2) aerobic 4 white oak disintegrated 151 0.46 TT6 frame 1(3) anaerobic 0.5 white oak 0-1 246 0.32 1-2 155 0.45 2-3.5 144 0.47 3.5-5.5 130 0.51 5.5-6.5 132 0.50 TT6 outer planking 2 aerobic 0.75 white oak 0-1.5 367 0.23 1.5-3 196 0.38 3-4 185 0.40 TT6 outer planking 5 anaerobic <0.5 white oak 0-1 280 0.29 1-2 143 0.48 2-3 167 0.43 3-4 235 0.33 Maximum moisture content is an easily measured quantity which may be related to specific gravity and thus to the extent of degradation of the wood. It is universally used as an indicator of wood deterioration and is the basis of a classification scheme. Waterlogged timbers may be classified as follows: Class 1 (>400)-an extremely degraded, extensive outer surface with very little solid core; Class II (185-400)-a degraded outer surface with a thin, partially degraded area and a considerably larger solid core; Class III (<185%)-a very thin degraded outer surface layer overlying an extensive undegraded core. It should be noted that many of the wood samples had been subjected to extensive depredation by marine borers and calcium carbonate lined the bore holes. Maximum water content is determined by measuring the waterlogged mass, drying it to constant weight at 105ºC and applying the standard formula [Pearson, 1987 (p.66)]. Due to the presence of calcium carbonate, the maximum water contents for samples attacked by marine worms would be lower than their real values. That is, these samples would be more degraded than the maximum water contents would indicate. Wood samples collected from a variety of timbers from different locations on the wreck site were identified by transmission microscopy. All samples recovered from the inner planking, frames and the keelson were identified as white oak. The wood species of the outer planking varied but white oak, birch and elm were identified. Interestingly, the keel sample was identified as beech. The full report is reproduced in Appendix C. The extent of deterioration of timbers in each of the six test trenches will be described separately. The site positions of each test trench are shown in Figure 6. Test Trench 1 The site positions and pH profiles of the structural timbers in test trench 1 are shown in Figure 7 and 8, respectively. The average depth to the cessation of marine borer attack in test trench 1 was 45cm. In general, the plots of pH versus core depth of the timbers in test trench 1 (Figure 8) follow a typical sigmoidal relationship. That is, the pH of the wood near the surface is high then as the depth into the timber increases there is a sharp and rapid decrease in pH that tends to plateau with increasing core depth. The higher pH measured on the wood surface, slightly more acidic than sea water, is indicative of the pH being controlled by the buffering capacity of the sea water. More importantly, this maximum pH denotes the area of greatest deterioration. As wood degrades in a marine environment by physical, chemical and biological means, its polysaccharide content is reduced and spaces created within the wood structure are then filled with alkaline sea water. This initially occurs in the outer more exposed areas of the timber. Hence, the normally acidic nature of the wood becomes progressively more alkaline with increasing degradation due to the inward diffusion of sea water. The pH then rapidly decreases as the depth increases into the wood core indicating a decrease in the extent of degradation. The pH will reach a minimum denoting the area of least deterioration where the wood is less waterlogged. The overall decrease in the pH of the wood core is an indication of the inherent acidity of wood. The innermost wood is still waterlogged, albeit to a lesser extent than the outer surfaces, therefore the pH will be more alkaline than the standard pH of seasoned, modern, undegraded wood of the same species. The wood species of outer planking 1 was not identified as the sample was too degraded with extensive wood borer damage (Table 4). Outer planking 2 was tentatively identified as white oak but was also very deteriorated. The maximum water contents of these timbers indicated that they were relatively undegraded, however, from visual examination of the samples, it was obvious that this was not the case. The presence of large quantities of calcium carbonate in the wood borer holes had increased the dry weight and artificially decreased the maximum water contents. Specific gravity of wood is related to the maximum water content and thus to the state of degradation of the wood. Decreases in the specific gravity of wood samples compared to standard values for undegraded wood of the same species are caused by the decrease in the amount of cellulose present in the waterlogged wood. Calculations of specific gravity are based on the dry weight of wood samples and hence, the presence of calcium carbonate will similarly affect this measurement. Therefore, using the maximum moisture content or specific gravity as an indicator of the state of wood degradation cannot be applied to timbers with extensive marine borer attack. However, the pH profiles provide a good indication of the state of degradation irrespective of the extent of wood borer attack. The pH values of the profiles for outer planking 1 and 2 are considerably higher than those values measured for the other timbers, indicating that they are particularly degraded throughout the total width of the timber. These timbers were exposed on the wreck site and subjected to the aerobic marine environment where considerably more biological and physical damage would occur. The pH profile of the outer planking 2 was anomalous. Outer planking 2 was partially buried and the pH profile was recorded directly above the sediment line. The largest populations of bacteria are at the sea water/sediment interface. Therefore, the lower pH value recorded on the surface of outer planking 2 may be due to microbial and fungal deterioration producing acidic metabolites and wood by-products that could be accumulated near the surface. The pH then increases dramatically as the timber is traversed internally. This increase in pH indicates that this timber is extremely degraded. Furthermore, comparisons of the pH profiles between outer planking 1 and 2 show that outer planking 2 is considerably more degraded than outer planking 1 due to the increase in biodeterioration at the sediment/sea water interface. Outer planking 3 and the keel were identified as birch and beech, respectively. The maximum water contents, specific gravities and the pH profiles indicate that they are in good condition with a very thin degraded outer surface (<0.5cm) layer overlying an essentially undegraded core. The keel was in such good condition that the drill bit would not penetrate the total width of the timber. The more acidic interior of outer planking 3 could represent differences in the wood species and not differences in the extents of degradation. These timbers were totally buried prior to the dredging of the test trench and have been subjected to an anaerobic environment. It is well known that wood recovered from deoxygenated environments is usually well preserved because the wood is predominantly protected from extensive physical and biological deterioration. The majority of the surface degradation of the anaerobic timbers would have occurred before the wreck remains were buried, however, slow biodeterioration of the wood surface may continue under anoxic conditions due to the presence of sulphate reducing bacteria which utilise the wood as a nutrient source. However, the keel was covered with copper alloy sheathing and the surface of outer planking 3 was impregnated with copper corrosion products that would have inhibited biodegradation. Two pH profiles were measured on the keel to check the reproducibility of the technique which is subject to contamination by sea water and the inherently diverse nature of wood. It is obvious that both pH profiles are in good agreement. The coring and subsequent measuring techniques are reproducible within a range of less than 5% of the mean pH value for each depth interval. This range of error is quite acceptable considering the relatively crude drilling and measurement procedure and the fact that the pH measurements are taken underwater. However, obtaining reasonably consistent results does depend heavily on the experience of the operators. Test Trench 2 The site positions and pH profiles of the structural timbers in test trench 2 are shown in Figure 9 and 10, respectively. The average depth to the cessation of marine borer attack in test trench 2 was 20cm. In general, the plots of pH versus core depth of the timbers in test trench 2 (Figure 10) are similar to those observed from test trench 1 (Figure 8). The pH of the wood near the upper surface is high then as the timber is traversed vertically there is a decrease in pH that tends to plateau with increasing depth. The higher pH values near the surface of the timbers denote the area of greatest deterioration. The subsequent decrease in pH is indicative of a decreasing extent of degradation towards the interior of the timbers until the pH reaches a minimum that represents the area of least deterioration within the timbers. All wood samples were identified as white oak (Table 4) with the exception of the outer planking 2 that was elm. The keelson and the inner planking were exposed and extensively teredo damaged, therefore, their maximum water contents and specific gravities cannot be used as indicators of their state of preservation. The keelson is the most degraded of the timbers in this test trench. The pH values of the profile for the keelson are much higher than those measured for the other timbers, denoting more extensive degradation throughout the entire width of the timber. The surface of the inner planking is very degraded to a depth of about 1.5cm and then the pH decreases dramatically to a minimum of 7.18 at a depth of 2.5cm representing the area of least degradation. After this minimum, there was a sharp turning point where the pH began to increase rapidly, indicating a dramatic increase in the extent of deterioration as the opposite side of the timber is approached. The total width of this timber is about 4cm. From the pH profiles it is obvious that these timbers are very degraded and have been subjected to an aerobic environment for an extended period of time. Both the frame 2 and the outer planking 2 were subjected to an anoxic micro-environment prior to the dredging of test trench 2. The frame was identified as white oak and the maximum water contents and specific gravities of the core sample sections indicated that the wood was in good condition. The pH profile represented a timber possessing a relatively degraded outer surface, about 2cm thick, overlying an essentially large, undegraded core. The outer planking 2 was identified as elm. The maximum water contents and the specific gravities of the core sample sections indicated that this timber was considerably more degraded than the white oak frame. The pH profile supported these measurements. One of the reasons for the apparent increase in the extent of degradation of this timber may be that it only has a total width of 4cm and this would allow easier and more rapid penetration of sea water into the wood structure. Another possibility is that elm is more permeable and more susceptible to biodeterioration than oak. However, the outer planking was better preserved in comparison to the keelson and inner planking exposed to an entirely aerobic marine environment. Test Trench 3 The site positions and pH profiles of the structural timbers in test trench 3 are shown in Figure 11 and 12, respectively. The average depth to the cessation of marine borer attack in test trench 3 was 20cm. The pH profiles of these timbers in test trench 3 (Figure 12) follow the same basic trend as the other timbers in test trench 1 and 2. Therefore, the same explanation for the decrease in pH from the outer surfaces to the inner regions of the wood will apply to these timbers. However, these profiles differ from the others in one respect. After the minimum pH is reached denoting the area of least degradation there is a turning point and the pH begins to increase again. This indicates a gradual increase in the extent of wood deterioration as the interior side of the timber is approached. In addition, the final pH of the timber is considerably more acidic than the pH measured on the upper, more degraded surfaces. Hence, the extent of deterioration of the outer surfaces is significantly greater than the interior surfaces. This is not unexpected as the upper surfaces of the timbers would be more exposed to the degradative effects of the local micro-environment. These timbers were all identified as white oak, therefore, direct comparisons between the extents of degradation of the timbers is made easier. The inner planking 1 was exposed to the full ravages of the aerobic marine environment and was extensively damaged by marine borers. Consequently, the pH profile indicates that it is quite deteriorated throughout the total width of the timber. The other three inner planking timbers had been subjected to a primarily anaerobic environment under the sediment and were correspondingly less degraded than the aerobic wood. The inner planking was numbered in ascending order as the test trench was traversed from the surface to a maximum depth of 0.5m. Hence, inner planking 4 was buried to a greater extent than inner planking 2. Comparisons between the maximum water contents, specific gravities and the pH profiles of the anaerobic timbers indicates that the extent of degradation of these timbers increases with increasing burial depth. This is an unusual situation but the total width of the timbers decreases from inner planking 2 (9cm) down to inner planking 4 (3.5cm). Therefore, if these timbers were subjected to exactly the same micro-environment then the thinner timber would be more degraded than the wider planks. Interestingly, there was some teredo attack evident on the sample obtained from planking 3 but none on the samples collected from planking 2 and 4. This means that at some point in time, planking 3 must have been subjected to an aerobic environment as wood borers need well oxygenated environments to survive. The fact that there was no obvious biological damage on the other timber samples may just be a result of non-representative sampling. It is also possible that there was some form of biological protection near the timber sampling positions for the wood to be spared the depredation of marine organisms. From these measurements and observations it may be possible that these timbers have been subjected to an aerobic environment for at some point in time and then reburied. Test Trench 4 The site positions of the structural timbers in test trench 4 are shown in Figure 13. Only probe depths were measured on these timbers and the average is presented in Table 4. The average depth to the cessation of marine borer attack in test trench 4 was 25cm. Probe depths give an indication of the depth to undegraded wood. That is, the extent of the degraded outer surface. Most of the timbers had been exposed to the aerobic marine environment for an extended period of time as teredo damage was extensive on all timbers. These timbers were, on average, extremely degraded to a depth of about 2cm. The section of frame 1 that was subjected to an essentially anaerobic micro-environment was significantly less degraded (<0.5cm). Test Trench 5 The site positions of the structural timbers in test trench 5 are shown in Figure 14. Only probe depths were measured on these timbers and the average is presented in Table 4. The average depth to the cessation of marine borer attack in test trench 5 was 20cm. Again, these timbers were extensively deteriorated by marine organisms. The average probe depth was 4cm. These timbers are more degraded than the timbers measured in test trench 4 which would suggest that they have been subjected to the ravages of the open ocean for a longer period of time. Test Trench 6 The site positions and pH profiles of the structural timbers in test trench 6 are shown in Figure 15 and 16, respectively. The average depth to the cessation of marine borer attack in test trench 6 was 20cm. The pH profiles of these timbers in test trench 6 (Figure 16) follow the same basic trends as the other timbers in test trench 1,2 and 3. However, complete pH profiles for all timbers were not obtained as the diver operating the drill was inexperienced and large lengths of the timber core were traversed without measurement due to the drill operator applying excessive pressure. Hence, the interpretation of the profiles will be tentative. The aerobic areas of the cant frame 1 and outer planking 1 and 2 had extensive wood borer damage. All timbers in test trench 6 were identified as white oak. Three separate pH profiles of the cant frame 1 were measured along its length because this timber extended into the sediment. The first two positions were measured where the timber had been subjected to an essentially aerobic environment and the third position in the area previously buried in the sediment. The pH profiles for the first two positions indicated that the aerobic section of the timber was extremely degraded throughout the entire timber width. The profile for the section of the timber subjected to an anoxic micro-environment was in much better condition, however, it did seem to have a more progressive decrease in the extent of deterioration as the core depth increased than other anaerobic timbers. The maximum water contents and specific gravities of the outer planking indicated they were more degraded than other outer planking measured on this site. The pH profiles also supported these observations. Sediment The average pH of the sediment at a depth of about 6.5cm was 8.01±0.10. The average redox potential of the sediment at a depth of approximately 14cm was –0.019±0.068V. These pH measurements indicate that the sediment at these depths is slightly basic in nature. However, the average redox potential indicates that the sediment is neither oxidising nor reducing in nature. This indicates that the zero-Eh interface coincides with the sea water/sediment interface. This means that on the sediment surface, aerobic bacteria have removed all the free oxygen from the interstitial water and CO2 accumulates and the pH will increase accordingly. Conversely, aerobic oxidation of organic material in the surface sediment will produce acidic metabolites and by-products and the pH will subsequently decrease. Therefore, the average pH of the surface sediment is a compromise between opposing biological equilibria. Many core samples of the sediment were collected near each test trench site and from control areas 25 and 50m off the main wreck site. The core samples revealed the typical grey to black discolouration at a depth of approximately 20-30cm, indicating an anaerobic environment at about this depth. This implies that the depth to stable sediment is, on average, approximately 25cm. These sediment samples are currently being subjected to very extensive chemical, geological and biological analyses and the results are still pending. The results from these analyses will assist in understanding the physico-chemical interactions between the local anaerobic micro-environments, the biota and the wreck site. Hence, the effect of burial on wooden vessels may be better understood, since full burial of wreck sites is one of the classic means of managing these underwater cultural heritage sites. CONCLUSIONS The James Matthews site is a typical, open circulation, oxidising marine environment. Due to the shallow nature of the site (~2m) the water temperature ranges quite markedly with the seasons. In summer the temperature can increase to 24-25ºC and this would significantly increase biological activity on this wreck site, subsequently increasing the degradation rate of exposed organic materials. Another disadvantage of the shallow conditions is the increase in total water and sand movement with decreasing water depth. This would increase the amount of physical damage caused to exposed structural features on the site. However, the site is relatively protected against rough sea conditions in all weathers with the exception of north westerly winds. The seabed in the vicinity of the wreck is relatively level and comprises of loose calcareous sand. The wreck is mostly buried to a depth of approximately 1.5-2.0m. The depth to stable sediment is, on average, approximately 25cm. At the time of the survey, there was no evidence of seasonal exposure. The site was first inspected in the 1970s and the site was extensively buried and covered in sea grass meadows. In 1989, it was noted that there were no sea grass beds on site but the wreck was still well buried. In 2000, the site was considerably more exposed and there were very sporadic, small patches of sea grass in close proximity (<50m) but there were none directly on the wreck site. This lack of sea grass would increase sediment mobility and result in a decrease in wreck coverage. In addition, the dredging of shell and sand occurring at the end of the Cockburn Cement jetty is probably contributing to the increasing extent of wreck exposure. More geological analysis of the sediment is needed to ascertain the reason for the lateral translation/dispersal of the sediment from the wreck site. It is most likely a combination of physical, chemical and biological reasons for the demise of the seagrass beds and subsequent sediment transportation/scouring of the site. It is also difficult to ascertain the long term effect that the gross disturbances caused by the 1970’s excavation seasons had on the degradation of the vessel itself. However, much of the sea grass covering the site was removed during the earlier excavations and/or smothered by the backfilling process. It seems likely that the sea grass on the wreck site never fully recovered from this interference and this would have lead to an increase in the extent of exposure of the vessel remains. However, at this point in time, it appears, that in most areas, the site is experiencing its maximum extent of exposure and has been exposed to this depth for some time. The most exposed areas of the wreck are the iron deck knees, the slate mound and the iron windlass. They rise approximately 0.5-1.5m above the sediment. The iron material measured on this wreck site is actively corroding. On average, the wooden structural timbers are exposed to a depth of about 20cm but the stern section has been exposed to a depth of approximately 40cm. There are many other areas on the wreck site that are only covered with relatively thin layers of sand. Exposed timbers exhibit active and extensive marine borer damage and the average probe depths indicate that they are extremely degraded. Generally, the pH profiles indicate that exposed timbers possess thick, extensively degraded outer surfaces and are particularly degraded throughout their total width. The extent of degradation of buried timbers varied considerably, but generally, they were in good condition with a relatively thin degraded outer surface overlying an extensive, undegraded solid core. In 1989, buried white oak timbers from the James Matthews were disturbed and subsequently raised during commercial dredging by Cockburn Cement. The average maximum water content was calculated at 121% which is in relatively good agreement with the moisture contents calculated for the anaerobic timbers sampled during this survey. However, due to the small number of analyses carried out in 1989, it is difficult to ascertain if there has been any significant increase in the extents of degradation of the buried timbers in the last 11 years. However, there is no doubt that the increase in site exposure has significantly increased the extent of degradation of the exposed hull structure. Aerobic biodeterioration and mechanical degradation are the major causes of aerobic wood degradation. Aerobic bacteria and organisms, such as marine worms, cannot survive under anoxic conditions hence, with the removal of the sediment, biological degradation and physical abrasion would increase dramatically. More importantly, deterioration will continue if steps are not taken to alleviate or, at least reduce the major degradative forces acting on this site. In conclusion, the results of this on-site conservation pre-disturbance survey indicate that the exposed hull remains of the James Matthews are in relatively poor condition but most of the wreck remains are buried and these structural timbers are in a good state of preservation. The rate of degradation would have accelerated dramatically with the significant decrease in sediment coverage over the past few years. Therefore, a synergistic maritime archaeological and conservation management plan must be devised to significantly reduce the continued deterioration of this historic shipwreck site. REFERENCES Australian Admiralty Chart 1972, Gage Roads and Cockburn Sound AUS 117, The Hydrographic Service, Perth. Baker, P. & Henderson, G. 1979, ‘James Matthews excavation. A second interim report’, The International Journal of Archaeology, vol. 8, pp. 225-244. Department of Conservation and Environment 1979, Cockburn Sound Environmental Study 1976-1979, Report No. 2, (eds J. Hutchinson & L. Moore), Department of Conservation and Environment (WA), pp. 1-103. Environmental Management Services 1994, Notice of Intention for Dredging the Approaches to the Magnetic Treatment Facility at HMAS Stirling, Western Australia, AGPS, Canberra. Henderson, G. 1976, ‘James Matthews excavation, summer 1974. Interim report’, International Journal of Archaeology, vol 5, pp. 245-251. Henderson, G. 1980, ‘Unfinished voyages. Western Australian Shipwrecks 1622-1850’, University of Western Australia Press, Nedlands. Henderson, G. & Stanbury, M. 1983, ‘The excavation of a collection of cordage from a shipwreck site’, The International Journal of Archaeology, vol. 12, pp. 15-26. Kenderdine, S. 1995, ‘Shipwrecks 1656-1942: A guide to historic wreck sites of Perth’, Western Australian Maritime Museum, Fremantle. MacLeod, 1988, 'Conservation of corroded concreted iron', in Corrosion - A Tax Forever. Proceedings of Conference 28. Perth 21-25 November 1988, vol. 1, Australasian Corrosion Association, Perth, pp. 2-6.1-2-6.9. MacLeod, I.D. 1989, 'The electrochemistry and conservation of iron in sea water', Chemistry in Australia, vol. 56, pp. 227-229. MacLeod, I.D. unpublished, Conservation Work Book, p. 88. Department of Materials Conservation, Western Australian Maritime Museum, Fremantle. MacLeod, I.D. & North, N.A. 1980, '350 years of marine corrosion in Western Australia', Corrosion Australasia, vol. 5, pp. 11-15. Monk, R. & Murray, F. 1990, Heavy Metal Contamination in the Marine Sediments of the Cockburn Sound Region, School of Biological and Environmental Sciences, Murdoch University, Murdoch. Murphy, P.J. 1979, Cockburn Sound Environmental Study. Technical Report on Industrial Effluents, Report no. 6, Department of Conservation & Environment, Perth. Pearson, C. (ed.) 1987, Conservation of Marine Archaeological Objects, Butterworths, Sydney. PART 2 ESTIMATED COSTS FOR THE CONSERVATION OF THE JAMES MATTHEWS. From the results obtained from the on-site conservation pre-disturbance survey it appears that the exposed sections of the James Matthews are very degraded but most of the wreck is buried and in excellent condition. Therefore, it should be possible to raise the wreck and successfully conserve the wooden hull remains intact. During the conservation and drying processes the vessel would need to be on public display. In 1989, some buried white oak timbers from the James Matthews were recovered during commercial dredging operations and these timbers were used in natural freeze drying experiments carried out in the Antarctic by Ambrose et al. (1994). Despite their sound condition, without impregnation with polyethylene glycol and freeze drying there was substantial surface damage with extensive cracks, exfoliation and checking of the deteriorated outer surface of these timbers. Hence, the wooden hull remains of the James Matthews will need to be impregnated with polyethylene glycol to prevent cracking, warping and splitting of the structural timbers prior to drying by slow dehumidification. Two basic methods of impregnation were contemplated for the conservation of the James Matthews. The first was continuous spraying of the hull with aqueous solutions of polyethylene glycol. The second method was treatment with aqueous solutions of polyethylene glycol by total immersion in a large tank. Continuous spraying was the stabilisation method applied to the extensive remains of the Wasa shipwreck located in the Wasa Museum in Stockholm, Sweden. The Mary Rose, which is located in Portsmouth, England, is currently being conserved by a similar spraying method. On contacting Dr Mark Jones, the Head of Conservation at the Mary Rose Trust and the head conservator of the Mary Rose, he suggested that the spraying method should be avoided as it would take considerable time (15-20 years) and was fraught with many technical difficulties. Dr Jones has successfully conserved a Dover Bronze Age boat by the total immersion technique and recommended this method of stabilisation for the James Matthews. The Bremen Cog, located at the Schiffahrtsmuseum in Bremerhaven, Germany has also been successfully preserved by the total immersion method. Communications with the Head of Conservation of this project, Dr Per Hoffmann, suggested that this method would be the most appropriate for the James Matthews. Comments and estimated costs from these two international experts form the basis of this estimated cost analysis for the conservation of the James Matthews. It is envisaged that the treatment would take approximately ten years. Copies of the communications are reproduced in full in Appendix D. Estimated Cost Analysis Projected Major Expenditure (based on a ten year treatment period) Estimated Cost Construction of a public viewing area and impregnation tank (380m3) $600 000-$1.5M Plumbing costs, including piping, filters, pumps, pressure release valves, temperature sensors, heaters, refrigeration, controls, building management systems $150 000-$300 000 Polyethylene glycol 400 and 4000 $750 000-$1M Other chemicals (biocides, floccing agents) $150 000 Running costs (plant maintenance/electricity/gas) $500 000 Labour costs (3 persons) $1.3M Drying Costs $600 000-$1M TOTAL $4.05M-5.75M The estimated total cost over a ten year treatment period for the total immersion method is approximately $4.05 to $5.75M AUD. The initial construction and setting up costs for the total immersion treatment would be substantial (approximately $1.32M-$3M), however, on going costs are less than for the spraying method as less plant maintenance is involved and the treatment time is effectively halved. References Ambrose, W.R., Neale, J.L. & Godfrey, I.M. 1994, 'Antarctic freeze drying of waterlogged timbers: A feasibility report', in Proceedings of the 5th ICOM Group on Wet Organic Archaeological Materials Conference. Portland/Maine, 16-20 August 1993, eds P. Hoffmann, T. Daley & T. Grant, ICOM, Committee for Conservation Working Group on Wet Organic Archaeological Materials, Bremerhaven, pp. 231-252. APPENDIX A DATE DIVERS DIVE TIME (min) ACTIVITY 20/09/00 IMG(2) 75 probe depth 20/09/00 VR/IDM (1) 94 cp 10/10/00 VR/IMG (3) 90 cp 24/10/00 VR/IMG/EW (4) 120 TT1 pH & sample 25/10/00 VR/GK (5) 105 TT1 pH & sample 27/10/00 VR/PL (6) 85 TT1 pH & sample 27/10/00 VR/JG (7) 65 TT2 pH & sample 30/10/00 VR/IMG (8 & 9) 145 TT2 & TT3 pH & sample 31/10/00 VR/MM (10&11) 90 TT3 pH & sample 1/11/00 VR/MM (12&13) 70 TT4 & TT5 probe depth 1/11/00 VR/AW (14) 105 TT6 pH & sample 2/11/00 VR/GK (15) 55 TT3 pH & sample 2/11/00 VR/GK (16) 55 TT6 pH & sample 16/11/00 VR/JC/MB (17) 120 geol. sediment 17/11/00 VR/JC (17) 140 cp 5/12/00 VR/JC (19) 200 bact. sediment 6/12/00 VR/JC (21) 80 chem. sediment 6/12/00 VR/JC (21) 90 chem. sediment 7/12/00 VR/JC (22) 90 chem. sediment 8/12/00 VR/GK/MB/Mohan The West (23) 50 geol. sediment 8/12/00 VR/GK/Mohan The West (23) 50 geol. sediment 11/12/00 VR/GK (23) 110 geol. sediment 11/12/00 VR/MM (23) 50 geol. sediment TOTAL VR 2059 TOTAL IMG 340 TOTAL JC 720 TOTAL IDM 94 Note: The total dive times are those of the conservation staff of the Department of Materials Conservation. APPENDIX B [hand written note - not able to reproduce for web] APPENDIX C James Matthews Wood Identification Report Seventeen (17) samples were provided by Vicki Richards for identification. The samples were cut and then polished to a 1200 grit finish and then examined using a low power microscope (x18). Recognizable features were used to identify the wood. Sample 1 (TT2 frame 2): Sample is a Quercus species (white oak) – usual features of rays wider than pores, ring porous, tyloses etc Sample 3 (TT2 keelson): Sample is a Quercus species (white oak) – usual features of rays wider than pores, ring porous, tyloses etc Sample 4 (TT1 outer planking 1): Features are difficult to detect but the wood is likely to be white oak. Sample 8 (TT1 keel): Features identified included: Rays of 2 distinct widths Nodes on rays Distinct growth rings Very small pores, numerous, radial multiples less than four The size of the rays indicate that this sample is a Fagus species (as the ship is of European origin it is likely to be Fagus sylvatica – European beech) Sample 10 (TT1 outer planking 3): Features: Rays are narrower than pores Growth rings are distinct Pores are numerous, small, in radial multiples less than 4, diminish in size slightly across the growth ring and oval-shaped. Of the timbers that possess the above features (Acer, Betula Nothofagus, Populis, Salix and Chloroxylon) the best match is with Betula species (birch). Sample 11 (TT3 inner planking 1): Sample is a Quercus species (white oak) – usual features of rays wider than pores, ring porous, tyloses etc Sample 24 (TT2 outer planking 2): Features: Rays are narrower than pores Soft tissue in tangential arrangement Ring porous The sample is an Ulmus species (elm). Sample 40 (TT3 inner planking 4): As this sample is highly degraded it was not easy to identify – one small area was noted in which ring porosity was observed. Rays are of 2 distinct widths. Tyloses were not abundant. A larger piece was examined – the presence of small pores across the growth ring indicate that the sample may be red, rather than white oak. However the extensive deterioration of the sample makes it difficult to clearly differentiate between these 2 species. The sample is therefore a Quercus species (either red or white oak). Sample 41 [TT6 forward frame 1(1)]: This sample was badly affected by teredo attack – it was possible to note the presence of rays of 2 distinct widths and a hint of ring porosity. This sample was substituted with Sample 30 (3) [forward frame 1 (3)].This sample was not much better in that it was not possible to observe the presence of any large pores. Very small pores were noted in flared, radial arrangements (typical of the latewood pores in white oak). Because of the observed presence of some large pores in sample 41, it is likely that this sample is a Quercus species (white oak). A further sample, 31 (2) [TT 6 forward frame 1 (3)] was supplied by Vicki Richards. Examination of this piece confirmed the tentative identification – definitely white oak. Sample 43 A (TT6 outer planking 5): Sample is a Quercus species (white oak) – usual features of rays wider than pores, ring porous, tyloses etc Samples 2, 13, 31, 36, 42, 43 and 44: These samples were in quite good condition. They were examined and found to be Quercus species (white oak). There were differences in the features however. For instance, Sample 2 must have been from oak that had a very rapid growth (large growth rings present), sample 43 was a very young sample of oak (the rays were radially oriented rather than parallel) and sample 44 was very slowly grown (tight growth rings). Dr Ian Godfrey Head of Department of Materials Conservation Western Australian Museum APPENDIX D Dear Vicki Sounds a very interesting project that you are involved with. Based upon my experiences at the Mary Rose Trust I would suggest you tank the wreck of the James Matthew. The treatment of the Dover Bronze Age Boat within a tank followed by freeze drying was a most enjoyable experience whereas the treatment of the hull of the Mary Rose by a spraying technique has resulted in many sleepless nights. Joking aside, the conservation of a large wreck by spraying will take much longer and is fraught with technical difficulties. I will try and visit you all in Western Australian to discuss in detail the Mary Rose conservation programme. Costings (estimates) 1. and 2. Construction of a public viewing area and enclosure. The Mary Rose ship hall was made of a fabric supported by an anodised aluminium frame. This spans across an historic dry dock which houses the wreck of the Mary Rose. The roof is double skinned with a thermal layer sandwiched between the outer and inner fabric. Cost of such a fabric structure with air conditioned public viewing enclosure would be between £200,000 - £500,000 dependent upon fabric specification. It is worth approaching a company that specialises in fabric buildings. 3. Spray jets, piping, filters, pumps, heaters, refrigeration, controls, building management systems etc will cost between £50,000 to £100,000. Quality of pipework (stainless steel more costly than polypropylene), pumps and building management system.. Don't forget safety features such as pressure relief valves, pressure and temperature sensors, and liquid level sensors. 4. PEG 200 or 400 is approximately £600 / metric ton, PEG 4000 is slightely dearer at £750 / metric ton. 5. Other chemical (biocides, anti-foaming agents etc. about £5000 per annum 6. Running costs (plant maintenance and electricity) ( approximately £10,000) 7. Labour costs (£130,000 per annum for 7 staff) 8. Drying costs ( dependent upon volume around the hull and degree of control). Estimate cost of air drying equipment and controls £200,000 to £350,000. These cost are based upon the approximate dimensions of the James Matthew. With best wishes Mark Jones ‹ James Matthews Dampier › View the discussion thread.