Preparation & Measurement of Fish Samples

Sprat, Sardines, Pilchards & Anchovy

Because of the small size of sprats, the sampling procedure is different to that of other fish, such as herring or mackerel. It is necessary to calibrate using larger numbers of fish, and interpret the results as a batch average (batch mean).

Before taking Fatmeter readings, the fish should first be lightly cleaned of any slime or excess water. We recommend that Fatmeter readings are taken utilising a SINGLE READING on EACH of a large number of fish. The Fatmeter measuring head is then applied to the skin of the intact fish at the recommended position as indicated in the RECOMMENDED MEASUREMENT SITES section of this manual. Groups of fish are selected using the Fatmeter, to lie within the following indicated range of fat/oil content:

13.7 to 15.1% 15.2 to 16.8%

16.9 to 18.7%

18.8 to 20.9% 21.0 to 23.3%

23.4 and over

The increasing size of these ranges is due to the logarithmic nature of the instrument’s response. Sufficient numbers of fish are taken, in each range, to fill a 600 ml beaker. This is normally between 20 and 30 fish, depending on size. The mean Fatmeter reading is then taken for each batch.

The fish then have head and tail removed, are split and gutted, and the ’butterfly’ fillets so formed in each batch, are minced together in a blender (eg Magimix). Thorough mixing is mandatory if a truly homogeneous sample is to be presented to the Foss-Let. This is especially true if the samples are to be stored, since, in high oil content fish, some separation of oil will occur.

After completion of Foss-Let analysis, the average of the results should be directly comparable with the average of the results achieved using the Fatmeter.

Eels

Select FOUR eels of similar size and weight at random from a batch. Wipe of any excess slime or water from the skin, then take one reading on each side of the fish. The instrument sensor head is placed on the skin of the intact fish just behind the vent (anal outlet) along the middle line. Please ensure that the sensor head is in good contact with the body of the fish (not easy with small eels).

Take the eight readings with the Fatmeter. The resultant average should accurately reflect the overall fat/oil content of the FOUR eels used for the test.

To accurately compare the Fatmeter results to Foss-Let laboratory analysis requires that you carefully follow our procedures in the preparation of the fish samples.

Firstly, cut off the head, and remove the gut. Then remove the skin. ALL of the remainder of the fish, including bones, fins, flesh, etc. are utilised for the Chemical Analysis. All FOUR prepared carcasses are minced in the blender for approximately two minutes. This procedure normally produces more mince than is required for the Foss-Let determination.

If the mince has been stored for any length of time, it should be thoroughly mixed before extracting the sample for the Foss-Let. This is particularly important with fish of known high fat/oil content, since the oil will tend to separate from the solid material of the mince.

The mean value of the fat content of the mince of the four fish is then compared to the mean value of the fat content derived from the Fatmeter readings.

Salmon & Trout

Adult Salmon and Trout are relatively large fish. Before taking Fatmeter readings, the fish should first be wiped to remove any slime or excess water. The Fatmeter sensor head is then applied to the skin of the intact fish on each side, just above, and parallel to the lateral line, at the positions indicated in your measurement chart.

Where the fish is between 100-500gms FOUR fish should be chosen at random from a batch. Fatmeter readings are then taken from BOTH sides of each of the four fish. The Fatmeter will then display the average fat content of the fish tested.

Where the fish size is in excess of 500 gms each fish should be measured individually, with EIGHT READINGS taken (four per side of the fish). The measurement sites should be carefully selected as per instructions in this manual.

The fish are then filleted, the belly walls removed and the excess dorsal fat deposit also removed. The skin is peeled off, care being taken that no flesh is removed with the skin. The remaining fillets of the fish are then minced together in a blender for approximately two minutes. If the mince has been stored for any length of time, it should be thoroughly mixed before extracting the sample for the Foss-Let. This is particularly important with fish of known high fat/oil content, since the oil will tend to separate from the solid material of the mince.

The mean value of the fat content of the mince of the four/one fish is then compared to the mean value of the fat content derived from the Fatmeter readings.

Herring

Eight fish are selected at random from a batch of fish. Remove excess water from the skin of the fish and Fatmeter readings obtained from one side of each fish. The instrument sensor head is placed on the skin of the intact fish just under the dorsal fin and above the lateral line, parallel to the backbone on the thickest part of the fish, as per measurement chart.

Please take care to keep the central conductor of the sensor head away from the belly cavity and above the strip of brown flesh which runs from head to tail just under the skin in the middle of the fish.

However, where larger Herring need to be measured….eg. above 500 gms, then use only FOUR similar sized fish for the test. A measurement should be taken from BOTH SIDES of these larger herring samples. After taking the eight readings with the Fatmeter, the resultant average should accurately reflect the overall fat/oil content of the FOUR herring used for the test.

The fish are then filleted and the belly walls and fins removed. The fillet skins are lightly cut into approximately 10 mm squares to allow proper mixing in the Blender. (It is difficult to strip the skin from a herring without removing at least some of the underlying flesh). All sixteen fillets are minced in the blender for approximately two minutes. This procedure normally produces more mince than is required for the Foss-Let determination. If the mince is to be stored, it should be thoroughly mixed before extracting the sample for the Foss-Let. This is particularly important with fish of known high fat content, since the oil will tend to separate from the solid material of the mince.

The mean value of the fat content of the eight/four fish is then compared to the mean value of the fat content derived from the Fatmeter readings.

Mackerel

The recommended sampling procedure is similar to that for herring, with the sensor head being placed approximately in the positions shown in measurement charts provided…ie. directly under the dorsal fin.

However, where larger mackerel…..eg. above 500 gms, are being measured and assessed with the Fatmeter, it is necessary to utilise only FOUR fish of equal size for the test. A measurement should be taken from BOTH SIDES of these larger mackerel samples. We have found that there can be up to 8% difference from one side to the other of a large mackerel. Having taken the eight readings with the Fatmeter, the resultant average should accurately reflect the overall fat/oil content of the FOUR mackerel used for the test.

With mackerel, it has been found necessary to remove the skin prior to blending. This can be done more easily than with herring, although care is still required to make certain that all the flesh is retained on the fillet. The problem of oil separating from the mince is more acute than is the case with herring because of the higher values of oil content commonly encountered.

Thorough mixing is mandatory if a truly homogeneous Foss-Let sample is to be presented. As with herring, the mean value of fat content obtained from the eight fish using the Foss-Let is compared to the mean value of fat content derived from the Fatmeter readings taken on the same eight fish.