Objective: Collected samples of Gracilaria (and any other seaweeds with commercial potential) are to be preserved using standard methods, and be despatched overseas for appraisal of phycocolloid properties.
Samples of fresh seaweed were transported to the laboratory at Naduruloulou Aquaculture Research Station (MAFF) as soon as was practicable. For samples collected from Suva Harbour, transportation to MAFF was done the same day. Samples from Kubuna Waters were kept fresh in sacks held in the sea at Dromuna Village (Kaba) for up to 3 days before transportation. The day after arrival at the laboratory, samples were thoroughly handwashed under freshwater hoses to remove salt and dirt, and sorted according to species. Plants were, by preference, sundried on 0.5 mm-mesh black plastic netting; this was done for 6 hr per day on two consecutive days. If the weather was rainy, the seaweeds were dried in a kerosene-fired fishpellet kiln at 40–60°C for 24 hr. Samples from Serua were air-dried in bright sunshine within 24 hrs of collection. As soon as drying was complete, the plants were placed in airtight plastic bags and stored in an airconditioned room at 22–23°C. Seaweed samples were divided into smaller airtight plastic bags and wrapped into parcels for delivery to analytical laboratories. The minimum sample size required by NZ Industrial Research Ltd (and also by most other laboratories) for initial screening of phycocolloid properties is 10 g dry weight, however 50 g was submitted to ensure that there would be enough for testing. A minimum of 500 g is usually needed for food ingredient testing.
Priority for initial screening of phycocolloids was given to comparison of the three Gracilaria species collected from Serua. Firstly, Gracilaria species are one of the most important sources of seaweed phycocolloid worldwide. Secondly, the three Gracilaria species from Serua were all collected from the same place on the same day. In this way possible variations in phycocolloid comparisons due to environmental factors could be eliminated, and so make us more confident that observed differences between the species are in fact genetic.
Dry matter content
The drip-dry fresh weight of 5 subsamples (between 50–250 g) for each species was recorded, then these samples were oven-dried for 24 hr at 60°C. The dried samples were reweighed, and the mean % dry matter content was calculated. These values (and the standard deviations, s) are given in Table 1. Note the very low values for Acanthophora; this is not encouraging for use of this species as a source of phycocolloid. It must also be remembered that dry matter content differs between seasons. This data probably represents maximum values, since dry matter content is usually highest during the off-season for plant growth.
Five 250 g subsamples of G. maramae (Nasese) and A. spicifera (Salia Reef) from the same batch were also sundried for a period of 6 hours. Comparison with the oven-dried samples showed only a small difference in moisture content between sun-dried and oven-dried plants (see Table 1). This means that, in good weather, Gracilaria, Hypnea and Acanthophora plants can all be thoroughly sun-dried in the space of a single day compared with 3–4 days for Kappaphycus (Sam Mario, pers. comm.). This will be a big advantage for commercial utilisation of these species, as Fiji's Kappaphycus industry was always constrained by a shortage of rack space for drying, and by the need to be constantly alert for rain over the 4 day drying period.
Table 1. Dry matter content of seaweeds collected
| Species | Sun-dried | Oven-dried | ||
|---|---|---|---|---|
| Gracilaria maramae | 17.4% | (s = 0.34) | 16.0% | (s = 0.03) |
| Gracilaria edulis | - | 14.3% | (s = 0.32) | |
| Hypnea pannosa | - | 15.9% | (s = 1.30) | |
| Acanthophora spicifera | 6.40% | (s = 0.56) | 4.50% | (s = 1.91) |
Production of any new commodity must be market-driven. It is useless to invest in producing anything that cannot be sold for sufficient price. For new seaweeds that may have potential as sources of phycocolloids like agar or carrageenan, their phycocolloid properties need to be evaluated. In particular, the response from commercial companies must be the primary factor to determine whether further work should be carried out on the commercialisation of seaweeds in the Pacific island countries.
In general, it is the commercial companies who are best placed to make judgements about the commercial quality of phycocolloids, because they hold the best information about specifications for particular uses (such uses are many and varied). For this study, dried samples of three Gracilaria species from Serua (G. maramae, G. edulis, and G. arcuata var. snackeyi) were submitted to Kadoya and Co. Ltd, a commercial seaweed importer in Japan with a particular interest in sources of Gracilaria.
No carrageenan-bearing seaweeds were considered for analysis, because the species found (Hypnea pannosa, Acanthophora spicifera, Laurencia spp., Gelidiella sp.,) were all of low commercial interest internationally as carrageenan sources, except for one (Kappaphycus alvarezii, formerly known as Eucheuma cottoni) which is of high interest but has already been thoroughly studied in terms of carrageenan extract and its market value is well known. Some of the Fiji carrageenophytes will be of academic interest, however these are better followed up as topics for study by post-graduate chemistry students at the University of the South Pacific.
An independent view of phycocolloid quality was obtained from Industrial Research Ltd (IRL), a chemical laboratory in New Zealand. It was thought important that testing of Fiji seaweed samples should also be done by a research laboratory that is not a part of the commercial seaweed industry, in addition to testing by prospective purchasers of seaweed. While the response from commercial companies is the most important indicator to determine whether a market exists for a seaweed or not, there is potentially a conflict of interest when the company interested in buying seaweed (and setting its price) is the same company that holds all the information about its phycocolloid quality.
It is also important that we do not create a (possibly lasting) bad impression about Fiji's seaweeds by sending away samples that turn out to be of poor quality as sources of phycocolloid. There is a risk of this happening if we dispatch samples that we do not yet know anything about. To maintain any interest in Fiji phycocolloids, there should be some preliminary tests to screen samples and ensure that nobody's time is wasted by dispatch of samples that turn out to be worthless. Commercial companies often receive samples from people, and they will give a higher priority to samples that are accompanied by encouraging results from preliminary tests
Evaluation of seaweeds as commercial sources of phycocolloid is not an easy task. During the present survey, the authors have found firstly that there is a wide variety of different phycocolloid tests, which can be expensive to carry out. Secondly, the scientific literature contains little of commercial relevance when comparing test data with results from other seaweed sources. The situation is eloquently summed up by Armisen and Galatas (1987), and the following points can be made as a warning for the unwary:
The existing literature on phycocolloid evaluations has largely come from academic scientists who use a wide range of extraction methods, who are unfamiliar with specification requirements or different phycocolloid grades of commercial users, and who measure only the most elementary parameters (usually % yield and gel strength) from small samples of untypically-clean seaweed;
Test results are as much a function of extraction methods as of phycocolloid quality. This means that absolute values for phycocolloid parameters are elusive, and relative values only have meaning when comparing results obtained by the same method in the same laboratory (refer to the discussion of the Hurtado-Ponce and Umezaki (1988) data in section 1.2 above as an example of this);
There is a multitude of different industrial uses for phycocolloids, and each carries its own set of specifications. These specifications may be commercially sensitive and not generally available except by direct approach to particular companies. The wide range of uses raises the danger that a phycocolloid source may be rejected as being of no use, when a use for which it may be ideal has been overlooked. It also poses the problem of just which phycocolloid parameters ought to be measured, and what values they ought to be compared with. There is no coherent system for industry outsiders to match the multitude of phycocollid sources against the multitude of uses.
Proper phycocolloid evaluations are much longer and more complicated processes than those usually published in the scientific literature. They begin with small-scale laboratory experiments on about 400 – 500 g (dry weight) of seaweed, divided into batches for various different extraction conditions. If results are promising, then pilot-plant runs are performed on about 1 kg batches (as many as 5 batches) in conditions similar to the factory process. It is on this pilot-plant extract that complete analyses are made, and suitability in practical applications are evaluated. For this, data and experience is needed of the actual specifications required by different markets for phycocolloids and for particular practical applications. Such data and experience is not often found outside of the commercial extraction companies themselves.
Armisen and Galatas (1987) also provides some advice for countries evaluating their phycocolloid sources for the first time. Firstly, they recommend that, as soon as the abundance of seaweeds has been estimated (even approximately), the quantity and quality of agar in the seaweed should be evaluated in terms of its practical use before starting any operations for seaweed production. Secondly, a range of representative samples should be collected from different geographical locations and in different seasons. This is because the results of tests on seaweed collected from only one location or one time of the year could be atypical of the country as a whole.
There is a wide range of possible tests that can be made on phycocolloid samples (depending on the desired end use), and some of the tests are expensive to carry out. The most efficient approach is to short-list samples based upon relatively simple tests of their physical properties. These shortlisted samples can then become the subject of more detailed tests of chemical properties and of suitability as food ingredients.
Samples can be short-listed as to quality by working through the following range of preliminary tests, which are in two groups, physical tests and chemical tests:
| AGAR | CARRAGEENAN |
| Does the extract gel? | Does the extract gel? |
| Agar yield (% of dry wt) | Carrageenan yield |
| Gel strength | Gel strength with K or Ca added |
| Appraisal of extract colour | Appraisal of extract colour |
| Gel elasticity | Can extract be used “semi-refined”? |
| Does it gel in high sugar concentations? | Is the gel brittle? |
| Chemically, is it agar? | Is it kappa, iota etc? |
| Any unusual substitution groups? | Does the extract need alkali treatment? |
| Melting temperature | |
| Setting temperature | |
| Is it useful without alkali treatment? |
Any sample that looks promising after these preliminary tests can then be selected for detailed Food Ingredient Testing work or tests for New Extract Uses. Examples include tests of sugar and milk reactivity, gelling temperature, melting temperature, viscosity, gel texture, and so on. These further tests would reveal whether the extracts meet specifications for various food-industry purposes. The ease of extraction (amounts of chemicals required, ease of filtration, etc.) are also very relevant considerations.
Although detailed commercial specifications for phycocolloid quality have not been found in the scientific literature, many publications have provided data for the more elementary quality tests. These include moisture content, phycocolloid yield, gel strength, and melting and gelling temperatures, and levels of the main chemical substitution groups.
Most relevant to Fiji is Gracilaria agar data provided by Nelson et al. (1983) and Hurtado-Ponce and Umezaki (1988), because several of the species investigated are also recorded as being found in Fiji (though one must bear in mind the confusion that applies in Gracilaria taxonomy generally). This data is summarised in Table 3. However the data will only give a rough guide as to which species may be commercially important in Fiji, because (i) only some of the more basic parameters have been measured, (ii) there are varietal differences in phycocolloid within a species between geographical locations, (iii) there are differences between seasons and culture conditions, and (iv) to compare Fiji species against this data, we would have to exactly duplicate their extraction methods.
Table 2. Overseas agar data for Gracilaria species found in Fiji (from Nelson et al., 1983; Hurtado-Ponce and Umezaki, 1988)
| Species | Country | Date | % yield | gel strength (g) | gelling temp. | melting temp. |
| G. coronopifolia | Philippines | 24/3/86 | 49 | 126 | 27 | 88 |
| G. edulis | Philippines | 24/3/86 | 36 | 136 | 55 | 75 |
| G. eucheumoides | Philippines | 10/4/86 | 34 | 127 | 34 | 77 |
| G. verrucosa | Philippines | 16/3/86 | 24 | 266 | 52 | 84 |
| G. sp. | Philippines | 17/3/86 | 26 | 464 | 41 | 77 |
| G. coronopifolia | Taiwan | 10/12/81 | 18 – 32 | 110 – 195 | 39 – 41 | - |
| G. edulis | Taiwan | 10/11/81 | 32 | 130 | 39 | - |
| G. verrucosa | Taiwan | 10/8/81 | 21 – 24 | 145 – 170 | 38 – 40 | - |
| G. edulis | Micronesia | 2–8/81 | 20 – 71 | 78 – 100 | 30 – 44 | - |
| G. lichenoides | Micronesia | 1–9/81 | 28 – 36 | 240 – 340 | 34 – 37 | - |
| Species | Country | Date | Viscosity | % Sulphate | % Pyruvate | % Anhydrogalactose |
| G. coronopifolia | Taiwan | 12/81 | 3.2 – 3.8 | 3.9 – 4.3 | 0.1 | 38 – 42 |
| G. edulis | Taiwan | 11/81 | 3.8 | 3.7 | 0.2 | 37 |
| G. verrucosa | Taiwan | 8/81 | 3.6 – 3.7 | 4.5 – 5.2 | 0.1 | 40 |
| G. edulis | Micronesia | 2–8/81 | 2.1 – 3.3 | 5.0 – 5.2 | 0.1 | 36 – 40 |
| G. lichenoides | Micronesia | 1–9/81 | 3.6 – 4.3 | 2.5 – 3.1 | 0.1 | 37 – 44 |
Armisen and Galatas (1987) provide benchmark levels for the more basic agar parameters from the commercial extractor's point of view. According to them, a good food grade agar will have a moisture content of 18% or less, gel strength between 600 – 1100 g cm-2 (normally 700 – 800) measured by the Nikan-Sui method, and a bacterial count below 10,000 per g. Lead content must be below 5 ppm and arsenic below 3 ppm. A bacteriological grade agar must have higher gel strength (1000-plus), gelling temperatures of 32 – 36°C, melting temperature of 85 – 86°C, a lack of hydrolysis by bacterial exoenzymes and lack of any bacterial growth-inhibitors. There are also specifications for clarity, levels of residual salts, and syneresis (“weeping” of water from the gel) (Moss and Doty, 1987).
Other things must also be taken into account, such as interactions with culture media components like meat extracts, sugars, salts etc. However there are no universal or detailed specifications available for this, since they vary for different bacteriological uses. Armisen and Galatas (1987) say that these are usually confidential to the manufacturers, however Moss and Doty (1987) state that product description and specification bulletins ought to be freely available directly from manufacturers upon request.
Japan Agricultural Standards for imported agar seaweeds (published in JETRO, 1980) list several inspection steps to assess seaweed quality. These include subjective judgements of dried-seaweed texture and colour, dryness (less than 20% water content is desirable), and percent purity (proportion of algae to foreign matter like sand, shell etc) which should be 60% or more. JAS standards for processed agar are given in the table below.
Table 3. Japan Agricultural Standards for processed agar (from JETRO, 1980)
| Criterion | Special grade | 1st grade | 2nd grade | 3rd grade |
| Colour, lustre | Lusterous white | Yellowish-white | Reddish-white | Light yellow/red |
| Gel stength | >600 g cm-2 | > 350 g cm-2 | >250 g cm-2 | >150 g cm-2 |
| Water content | 22% or less | 22% or less | 22% or less | 22% or less |
| Crude protein | 1.5% or less | 1.5% or less | 2% or less | 3% or less |
| Hot-water insolubles | 0.5% or less | 2% or less | 3% or less | 4% or less |
| Crude ash | 4% or less | 4% or less | 4% or less | 4% or less |
Matsuhashi (1987) points out that these JAS standards were set by government rather than by industry, for the purpose of regulating imports into Japan. He warns that this, and diversification of product uses since, means that conventional quality evaluations based upon official standards may only give a very rough indication of the required physicochemical properties for practical uses. For example, a gel strength test will show whether an agar is likely to be roughly of commercial use, but its quality for any particular use will not be known until ease of extraction, gel texture, flexibility, melting temperature and reactivity with other food ingredients are known.
Even so, researchers such as De Castro (1993) have used such standards to evaluate new Gracilaria species. De Castro tested G. changii, G. heteroclada and G. coronopifolia for agar yield, gel strength, gelling temperature, melting temperature, viscosity, moisture content and ash content, and compared results against values in the Japanese Specifications for Processed Agar (JSPA), United States Pharnacopoeia (USP) and Food Chemical Codex (FCC) as well as against a preparation of commercially-available agar gel. G. changii and G. heteroclada scored well, but G. coronopifolia (also found in Fiji) scored poorly in both yield (about 19%) and gel strength (about 160 g cm-2). The author concluded on this basis that two of the species tested had high potential as raw materials for agar.
The above remarks about difficulties in meaningful phycocolloid evaluations have used agar for specific examples, but the general conclusions are also applicable to carrageenans.
To overcome the problems so far experienced in finding commercially-relevant information with which to evaluate phycocolloid quality, two approaches are possible.
One approach is to ask any commercial company who tests samples of Fiji seaweeds to also provide previously-obtained results for other seaweed sources with known commercial value, extracted by the same methods and characterized by the same tests. This will rely upon the goodwill and cooperation of the companies involved.
The other approach is to request copies of some commercial standards or specifications for a selection of product uses, against which the results of any independent tests may be compared. Again, this will rely upon cooperation from commercial companies.
The choice of company is also important. Some companies may be on the look-out for phycocolloids with new properties for specially uses, while others just want a good-gelling phycocolloid with high yield to use in general food or industrial uses. Any special properties in your seaweeds may simply go unrecognised by the latter type of company. This is why independent analysis is important to really know what is in your seaweeds. But at the end of the day, if companies don't want to buy your seaweed, no one can force them.
The phycocolloids present in the three species of Gracilaria from Serua (G. maramae, G. edulis and G. arcuata var. snackeyi) were tested by Industrial Research Ltd, and the physical properties plus a detailed account of the chemistry of these agar extracts has now been published (Falshaw et al. 1999). IRL's conclusion was that each of these seaweeds holds some promise as a candidate for commercial agar production; and hence for aquaculture. The alkali-modified agars from all three species had reasonable gel strengths (>300 g cm-2 for 1% gels), good clarity and low colour and were obtained in acceptable yields. The G. arcuata v. snackeyi agar had a higher melting point (ca. 100° C) than conventional agars (normally around 85° C), which is a useful property for the food industry and is only the second time that a high melting point agar has ever been found in the Gracilaria genus.
The same samples analysed by Falshaw et al. were submitted to Kadoya & Co Ltd, whose response was that phycocolloid content was good but the gel strength was lower than their existing sources in South America. They would consider buying Fiji Gracilaria only to blend into other supplies, and only if South American sources became scarce. Kadoya & Co. did not notice that one of the three species from Serua had a high melting point agar, neither did they make any comment about it after the results of the independent tests by Falshaw et al. were brought to their attention. It would appear that the particular seaweed markets they supply do not require this feature.
While at least one of the Fiji Gracilaria species has interesting properties with commercial potential, it is not an easy task to create a market for a new phycocolloid. Industry users are mostly locked-in to existing suppliers in countries with a proven track record of phycocolloid properties and continuity of supply. Smaller niche markets exist for specially phycocolloids, but it is difficult for an industry outsider to match the plethora of end uses to a particular phycocolloid (or vice versa). However the industry “insiders” do scan the academic literature to find out what new developments are occurring in phycocolloid properties and sources. Perhaps the publication of our data for Fiji seaweeds by Falshaw et al. (1999) will be the best advertisement for native Fijian seaweeds, which hopefully may pay off at some time in the future.
For now, however, it appears that there is no immediate export market for these seaweeds as sources of phycocolloid. Any utilization of Gracilaria (or Hypnea) seaweeds in Fiji would need to be justified for some other reason, such as the food market locally or overseas, as food for aquacultured molluscs like greensnail, or as a wastewater treatment (for example in the nutrient-rich effluent from prawn farms or sugar mills).
