Studies on the preparation of fish silage. 1.

Effect of quality of raw material and type of acid

Item Type article

Authors Jayawardena, K.M.; Villadsen, A.; Guneratne, Q.; Aagaard, J.;

Poulter, R.G.

Download date 22/08/2023 01:29:28

Link to Item http://hdl.handle.net/1834/32889

Bull. Fish. Res. Stn., Sri Lanka, Vol. 30, 1980, pp 17-23.

Studies on the Preparation of Fish Silage

I. Effect of quality of Ra,v Material and ~fype of
By
K. M. JAYAWARDENA*, A. VILLADSEN*, Q. GUNERATNE,* J. AAGAARD*, and R. G. POULTERT

lnttCJ«i!odion
The acid ensilage of fish to produce a protein supplement for feeding to animals is not a new process.
The basic technology was developed more than 50 years ago and substantial quantities are produced
in countries such as Denmark and Poland, mainly for feeding to pigs. In recent years, however, there
has been renewed interest in fish silage as it represents a means of utilizing waste fish and offal in
situations where conventional fish meal production is inappropriate. The process is particularly
suitable for Asian conditions since it can be adapted for use as a cottage industry in the villages or as
a large-scale commercial enterprise. In order to produce fish silage only a minimal amount of training
and equipment is required. There is no need for refrigeration and no pollution or e:ffiuent problem.
Fish silage is prepared by mixing ground or minced fish with an acid. The enzymes naturally
occur.r.i.ng in the fish break down the proteins into smaller soluble units while the acid helps Speed up
the:itr activity and prevents bacterial spoilage. The acid used may be an inorganic acid such as
hydrochloric or an organic acid, such as formic or propionic. Alternatively, a lactic acid-producing
organism and a carbohydrate source can be added. In the latter case the acid produced by
fermentation is then responsible for the preservation. If an inorganic acid is used it is necessary to
neutralise the silage before it is fed to animals.
When correctly prepared fish silage is a grey viscous liquid and has only a slight oily or malty
odour. It can be used in the wet form as a pig feed or alternatively, silage can be dried together with
a filler material such as rice bran and the product used as a protein supplement for chicken or as a
fish feed.
The prospects for fish silage in Sri Lanka (and 7 other Indo-Pacific Fisheries Commission
countries) has recently been reviewed by Disney (1979) and Sumner (1977). These references a1so
contain comprehensive bibliographies relating to work on fish silage. In his review of the prospects for
fish silage in Sri Lanka Disney refers to unpublished data produced by the Institute of Fish Technology,
Sri LaJllka which indicates that the production of fish silage in Sri Lanka is likely to be highly profitable.
However, he points out that there is an urgent need to resolve any outstanding technical problems and
to verify the economic feasibility in practice. To this end a research programme is being undertaken
jointly by the Institute of Fish Technology, Colombo and the Veterinary Research Institute, Peradeniya.
The present paper is the first of a series and describes an experiment in which silver belly
(Leiogt:wthus splendens) of different initial quality were taken and silages produced using different
,c-Oncentrations of hydrochloric and formic acid. The quality and storage life of the various

* Institute of Fish Technology, Crow Island, :f\.1attakkuliya, Colombo 15.

t Seconded from the Tropical Products Institute, 56 f62, Gray's Inn Road, London, England.

17
18 STUDIES ON THE PREPARATION OF FISH SILAGE-I

preparations is reported. Future papers in the series will give details of a study to determine. the
feasibility of producing silage from prawn waste, a study to determine the economics of producing
silage in Sri Lanka and of experimental feeding trials in which silage was fed to broiler and layer chicken.
Trials to be undertaken together with a firm in Mannar using a commercial silage unit capable of
producing 300 tonnes of silage per yea:r will also be reported.

Materials and Methods

Fish Samples
Silver belly (Leiognathus splendens) caught as a by-catch during prawn trawling near Mannar
Island in May 1978 were used. Batch one (15 kg.) designated" good quality", was kept at ambient
temperature (28°C) for 3 hours after capture and then iced. Batch two (15 kg.) designated "bad
quality " was kept at ambient temperature for 12 hours after capture and then iced. Both good and
bad quality batches of fish were kept in ice for 12 hours before being processed into silage.

Preparation of Silage
Three types of silage were produced, hydrochloric acid silage, formic acid silage and silage
made with both hydrochloric and formic acid. Table 1 gives the concentrations of acids used for
each type of silage. The fish samples were comminuted in an electric mincer and 750g of mince was
mixed mechanically with the appropriate acid for 10 minutes. The silages were stored in 2-litre plastic
containers. The strength of the hydrochloric acid used was 12N and the formic acid was a 90% (w/v)
solution.

For the formic acid silages the required concentration of acid was added at the beginning of the
experiment and, although the pH was monitored, no further acid was added during storage. For the
hydrochloric acid silages the pH was brought down initially to the required level and, if required, further
acid was added daily to maintain the pH. Generally the pH became relatively stable after the first
12 days. The procedure for production of silage containing hydrochloric acid and formic acid was
similar to that used for silage made from hydrochloric acid alone. The concentrations of hydrochloric
acid sho>ivn in Table 1 are the total amounts added.

Water Content
Samples (2gj were dried in a convection oven at 105°C for 24 hours. The weight loss was taken
to be due to evaporp_tion of water.

Ash Content
The residue from the water content determination Wp.s ashed in a muffle furnace at 600°C for
24 hours.

Fat Content
This was determined by the methanol/chloroform extraction procedure of Bligh and Dyer
1959).

Crude Protein Content

The micro-kjeldah] procedure was used and analyses were made on 0.2g samples (Pearson,
1970). Total nitrogen was converted to crude protein by multiplying by a factor of 6.25.
K. M. JAYAWARDENA, A. VILLADSEN, Q. GUNERATNE, J. AAGAARD, AND R. G. POULTER 19

Total volatile bases (TVB) and trimethylamine (Tl\tfA) determination

Protein nitrogen in a 25g sample of mB.teria] was precipitated by reducing the pH to 5.2 with a
HC 1 and heating to 70°C. The TVB and TMA content of the non-protein nitrogen fraction was
clete:rmined using the Conway micro-diffusion method (Beatty and Gibbons, 1937).

PH
This was measured directly using a Radiometer 26 pH and a glass electrode.

Results
1he proximate composition of the iced silver belly samples is given in Table 2 along with the
TVB and TMA values. The results are the means of duplicate determinations.

When first produced, the silages made from the good quality fish were very rubbery in texture
whereas those made from had quality fish were much more paste-like in consistency. This occurred
regardless of the type or concentration of acid used. Two to three days after production both batches
of silage were liquid.

The appearance and odour ofth~ various silages at 30, 50 and 115 days after production is given
in Tables 3 and 4. For the good-qmlity fish silages, only the sample acidified with hydrochloric acid
to pH 2.5 kept for 115 days (and is stiH in an acceptable condition after 290 days). Six out of the eleven
samples however, did keep well for shorter periods. For the b:1d quality fish silages, all the samples
kept for 30 days but ;:tgain, only the sample acidified to pH 2.5 kept for 115 days. G~nerally the
samples b~came unacceptable due to putrid off-odours and appearance of moulds or maggots. The
change in pH during the first 41 days of storage in the silages made from formic acid is given in table
5. There was an increase in pH in aU th~ samples. O::J.ly in the silages m:1de from good-quality fish
with 1.5% and 2% formic acid, and possibly in the silage from bad-quaJity fish with 2% formic acid
was the rise excessive.

DISCUSSION
There was no appreciable diff3rence b::tween the proxim:1te composition of the good and bad
quality bJ.tches of fish. The TVB and TMA results indicate that the description of the samples wa'lil
indeed correct, ie. th:: samples stored for 3 h:mrs at 28°C were of g)od quality and those stored for 12
hours were of h1d quality. For temperate water fi3h, eg cod, a TVB value of 30 to 40 mg/lOOg and a
TMA value of 10 to 15mg/1 OOg is taken to indic,1 te that the fish is unsuitable for human food. (ConneH
1975).

The difference in initial texture of the silJ.ge made from good and bJ.d quality samples of fish
must reflbct a diff3rence in the physical nature of the structural proteins in the samples. It is
reasonable to assume that a much greater degree of degradation of these proteins had occurred in fish
held for 12 hours at 28°C compared with fish held at 28°C for only 3 hours. In practice, the initial
rubbery texture of silages prepared from good quality fish meant that it was extremely difficult to evenly
distribute the acid throughout the SJ.'11ple. After 2 to 3 days when the silages made from both good
and bad qu2..lity fish had liquefied, mixing was no longer a problem.

G~neraUy the silages prepared from bad-quality fish remained in an acceptable condition for
longer than th:)se prep:1red from good-quality fish. This is the opposite of wh3.t might be expected
and must refl;:;ct the ease with which the acid may be distributed throughout the silages during the
-.first few days after production.
 STUDIES ON THE PREPARATION OF FISH SILAGE-I

For silage produced with formic acid it appears that a slight rise in pH during storage is to be
expected and will not seriously affect the quality. However, large rises in pH of the silages indicate
putrifaction and should be avoided by increasing the concentration of formic acid and for ensuring
more even mixing.
In Sri Lanka the main use for fish silage is as a poultry feed. For this purpose it is necessary to
have the product in a dried form. In this experiment only the silc.g,:::s in which the pH was reduced to
2.5 or below kept for 115 days. However, an of the silages from the bad quality fish and six of those
produced from good quality fish kept for 30 days. In practice a storage life for wet silage of one
month should be sufficient time to allow it to be mixed with a cheap filler material, such as rice bran or
maize meal, and sun dried. Once the moisture content of the silage/carbohydrate mixture has been
reduced to below 10 %it may then be stored for several months without loss of quality.

Conclusions
1. Silage produced from silver belly held for up to 3 hours at 28°C, initially has a rubbery
texture An even distribution of acid in. such material is di:ffi.cult to achieve and this may reduce the
storage life of the product. Silage produced from fish held at 28°C for 12 hours hp.s more of a
paste-like consistency and mixing is less of a problem.
2. Silages prepared from silver beUy held for 3 or 12 hours at 28°C and in which the pH has
been reduced to 2.5 or below by addition of hydrochloric acid \ViU remain in an acceptable condition
for at least 115 days.
3. Silages which keep for at least 30 days can be produced from silver beHy held for 3 or 12
hours at 28°C by (a) reducing the pH to 3.0 by addition of hydrochloric acid, (b) adding 0.5%fonn.ic
acid and reducing the pH to 3.5 with hydrochloric acid or (c) adding 2.5% form:ilc acid.

ACKNOWLEDGEMENTS
The: help and assistance of the: technical staff of the Institute of Fish Technology is gratefuHy
acknmvledged.

RJEl-iERENCES
BEATI<t, S. A. and GIBBONS, N. E. (1937)
The measurement of spoilage in fish. J. biol. Bd., Canada. 3 (1) : 77-:91.
BLIGH, E. G. and DYER, \V. J. (1959)
A rapid method of total Hpi.d extraction and purification. Can. J. Biochem. Physiol. 37 : 911-917.
CoNNELL, J. J. (1975)
In: Control offish quality. Fishery News (Books) Ltd., England. 179 pp.
DJISNEY,J. G. (1979)
Prospects for fish silage in Malaysia, Sri Lanka, Bangladesh and the Philippines. IPFC Occ. Pap. 1979[1. !FAD
Regional Office for Asia and the Far East, Bangkok, Thailand.
PEARSON, D. (1970)
The Chemical analysis offoods, 6th Edition, Churchill, London, p.l L
SuMNER, J. (1977)
Fish silage production in the IPFC Region--A feasibility study. APHCA Second Sessirm Paper 2171fR.ef. 2,
August 1977, ·Kuala Lampur, Malaysia.
JK. :M. JAYAWARDENAl A. V.JLLADSEN, Q. GUNERATNE, J. AAGAARD A.c'\iD R. G. POULTER 21

TABLE 1
ACID CONCENTRATIONS USED TO PREPARE SILAGE SAMPLES
Hydrochloric Acid
Formic acid
% wfw pH* %wfw
(a) Good quality silver belly :
2.5 10
3.0 5
3.5 6
].5

2.0
2.5
0.5 3.0 6
0.5 3.5 4
0:5 4.0 ~
.)

LO 3.5 4-
l.O 4.0 2

(b) Bad quality silver beHy :

2.() 10
2.5 10
3.0 6
2.0
2.5
3.0
3,5
0.5 2.5 8
0.5 3.0 7
0.5 3.5 5
1.0 3.0 7
1.0 3.5 4
].0 4.(} 3

:11- For snla.ges containing hydrochloric acid, the acid was added until the desired pH was obtained.

TABLE 2

C'HEMIC.A<L COMPOSH10N OF GOOD QUAliTY AND DAD QUAUTir

SILVER BELLY

Sample Water Ash Fat Crude Protein TVB TJVJA

%(wfw) % (wfw) ~~(wjw) % (Yi'/W) mgf100g mgflOOg

Good Quality .. 7-<U 5.0 2.9 18.6 13.6 7.9

Bad Quality 73.5 5.4 3.7 19.2 .. 140.0 27.9
22 STUDIES ON THE PREPARATION OF FISH SILAGJE-1

TABLE 3
r\.PPEARANCE AND ODOUR OF SILAGE MADE FROM GOOD QUALITY
SILVER BELLY
Days of Storage
Sample r------------------~~--------------------~
30 50 115
(a) HCl
pH 2.5 Normal Normal Normal
pH 3.0 Normal Mould Discarded
pH 3.5 Putrid Discarded Discarded

(b) Formic Acid

1.5% Putrid Discarded Discarded
2.0% Putrid Discarded Discarded
2.5% Normal Putrid /Maggots Discarded

(c) HCl and Formic Acid

pH 3.0-0.5% Normal Normal Mould
pH 3.5-0.5% Normal Putrid Discarded
pH 4.0-0.5% Putrid Discarded Discarded
pH 3.5-1.0% Normal Putrid Discarded
pH 4.0-1.0% Mould/Off odour .. Cheesy odour Discarded

N.B.-Normal means a grey viscous liquid with a slight oily or malty odour.

TABLE 4
APPEARANCE Al~D ODOUR OF SILAGE MADE FROM BAD QUALITY
SILVER BELLY

Days of Storage
Sample
30 50 115
(a) HCl
pH 2.0 Normal Normal Normal
pH 2.5 Normal Normal Normal
pH 3.0 Normal Slightly putrid Putrid

(b) Formic Acid

2.0% Normal lV1ould Discarded
2.5% Normal Maggots Discarded
3.0% Normal Maggots Discarded
3.5% Normal Mould Maggots/Off odour

(c) HCl and Forniic Acid

pH 2.5-0.5% Normal Normal Normal
pH 3.0-0.5% Normal Mould Discarded
pH 3.5-0.5% Normal Mould Discarded
pH 3.1)-1.0~~ Normal Mould Discarded
pH 3.5-1.0% Normal Mould Discarded
pH· 4.0-1.0% Normal Mould/Gas Discarded

N.B.-Nornial means a grey viscous liquid with a slight oily or malty odour.
1(. M. JAYAWARDENA, A, VILLADSEN, Q. GUNERATNE, J. AAGAARD, AND R. G. POULTER 23
TABLE 5
pH OF FORlVUC ACID SILAGES

Days of Storage
Formic Acid added to Silage
1 2 4 8 14 19 41
(a) Good qua:ity Silver Belly Silages
1.5% 4.4 4.5 4.6 4.5 .. 5.8 7.0 . . 8.0
2.0% 4.2 4.3 4.3 4.3 4.9 6.5 7.3
2.5% 4.0 4.0 .. 4.1 4.0 3,9 4.2 4.5

(b) Bad quality Silver Belly Silages

2.0% 4.1 4.1 4.4 4.2 4.4 4.4 5.1
2.5% 3.8 4.2 4.1 4.0 4.0 4.2 4.7
3.0% 3.7 3.9 3.9 3.8 4.0 4.0 4.3
3.5% 3.7 3,8 3.7 .. 3.8 3.7 4.0 4.0