Application Data Sheet 1

Title
Comparison of Distell Fatmeter and Foss-Let methods of lipid analysis.

Product
Atlantic Salmon.

Data Source
Major UK fish processor.

Sampling Method
Four measurements made on each side of the fish above the lateral line. Distell’s measurement methodology for…Salmon-1 Please refer to measurement instructions section for further details. Foss-Let analysis on the trimmed section taken from below the dorsal fin.

Statistics
Number of fish : 97
Maximum lipid content (Foss-Let) : 12.2%
Minimum lipid content (Foss-Let) : 3.8%
Regression equation, y = a + bx : a=0.704; b=0.906 (y is Fatmeter result, x is Foss-Let)
Correlation coefficient (r) : 0.758
Standard error : ±1.32

Comparison of Foss-Let and Distell methods for Lipid Measurement

Comments
Although the correct measurement method was used for the Fatmeter readings, the Foss-Let analysis was carried out on a section taken from below the dorsal fin. Distell’s calibration…Salmon -1 is representative of… The 2 x Trimmed Fillets of the fish, and using only the section under the dorsal results in a poor correlation between the Fatmeter and laboratory analysis.

Conslusion
Always follow Distell’s recommended calibration methodology rigorously if you wish to achieve a high correlation with laboratory analysis.

Application Data Sheet 2

Title
Comparison of Distell Fatmeter and Foss-Let methods of lipid analysis.

Product
Atlantic Salmon.

Data Source
Major UK fish processor.

Sampling Method
Four measurements made on each side of the fish above the lateral line. Distell’s measurement methodology for… Salmon-1 Please refer to measurement instructions section for further details. Foss-Let analysis on the trimmed section taken from below the dorsal fin.

Statistics
Number of fish : 97
Maximum lipid content (Foss-Let) : 12.2%
Minimum lipid content (Foss-Let) : 3.8%
Regression equation, y = a + bx : a = 0.926; b = 0.835 (y is Fatmeter result, x is Foss-Let)
Correlation coefficient (r) : 0.748
Standard error : ±1.26

Comparison of Foss-Let and Distell methods for Lipid Measurement

Comments
Although the correct measurement method was used for the Fatmeter readings, the Foss-Let analysis was carried out on a section taken from below the dorsal fin. Distell’s calibration…Salmon –1.. is representative of… The 2 x Trimmed Fillets of the fish, and using only the section under the dorsal results for the laboratory analysis results in a poor correlation between the Fatmeter and laboratory analysis.

Conclusion
Always follow Distell’s recommended calibration methodology rigorously if you wish to achieve a high correlation with laboratory analysis.

Application Data Sheet 3

Title
Comparison of Distell Fatmeter and Foss-Let methods of lipid analysis.

Product
Atlantic Salmon.

Data Source
Major UK fish processor.

Sampling Method
Four measurements made on each side of the fish at points directly beneath the dorsal fin (Mowi Section). Distell’s measurement methodology for… Salmon-2 .
Please refer to measurement instructions section for further details.
Foss-Let analysis on… Correctly prepared “Mowi Section”.

Statistics
Number of fish : 97
Maximum lipid content (Foss-Let) : 12.2%
Minimum lipid content (Foss-Let) : 3.8%
Regression equation, y = a + bx : a = 0.066; b = 1.038 (y is Fatmeter result, x is Foss-Let)
Correlation coefficient (r) : 0.878
Standard error : ±0.964

Comparison of Foss-Let and Distell methods for Lipid Measurement

Comments
The correct measurement method was used for the Fatmeter readings, and the Foss-Let analysis was carried out on the correctly prepared “Mowi Section”. Distell’s calibration…Salmon –2.. is representative of… The Mowi Section of the fish, and by preparing the proper sample for the laboratory analysis has resulted in a high correlation between the Fatmeter and laboratory analysis.

Conclusion
Always follow Distell’s recommended calibration methodology rigorously if you wish to achieve a high correlation with laboratory analysis.

Application Data Sheet 4

Title
Comparison of Torry Fatmeter and Foss-Let methods of lipid analysis.

Product
Atlantic Salmon.

Data Source
Torry Research Station.

Sampling Method
Four measurements made along the length of the fish, above the lateral line. Distell’s measurement methodology for… Salmon-1 .
Please refer to measurement instructions section for further details.
Foss-Let analysis on... Correctly prepared... 2 x Trimmed Fillets.

Statistics
Number of fish : 22
Maximum lipid content (Foss-Let) : 11.4%
Minimum lipid content (Foss-Let) : 0.11%
Regression equation, y = a + bx : a = 0.297; b = 1.06 (y is Fatmeter result, x is Foss-Let)
Correlation coefficient (r) : 0.915
Standard error : 1.33

Comparison of Foss-Let and Distell methods for Lipid Measurement

Comments
The correct measurement method was used for the Fatmeter readings, and the Foss-Let analysis was carried out on the correctly prepared “2 x Trimmed Fillets”. Distell’s calibration…Salmon –1.. is representative of… The 2 xTrimmed fillets of the fish, and by preparing the proper sample for the laboratory analysis has resulted in a high correlation between the Fatmeter and laboratory analysis.

Conclusion
Always follow Distell’s recommended calibration methodology rigorously if you wish to achieve a high correlation with laboratory analysis.

Application Data Sheet 5

Title
Comparison of Distell Fatmeter and Bligh and Dyer methods of lipid analysis.

Product
Herring.

Data Source
Technological Laboratory of the Danish Ministry of Fisheries.

Sampling Method
Technological Laboratory of the Danish Ministry of Fisheries. Eight measurements made in the centre of each of EIGHT fish, above the lateral line. Distell’s measurement methodology for… Herring-1 .
Please refer to measurement instructions section for further details.
Bligh and Dyer analysis on… Correctly prepared…16 x Trimmed Fillets..

Statistics
Number of fish : 57
Maximum lipid content (B and D) : 24.06%
Minimum lipid content (B and D) : 3.3.%
Regression equation, y = a + bx : a = 0.797; b = 0.867 (y is Fatmeter result, x is B and D)
Correlation coefficient (r) : 0.884
Standard error : ±2.08

Comparison of Bligh & Dyer and Distell methods for Lipid Measurement

Comments
The correlation obtained with this method was good despite the fact that the standard procedure was not followed. Improvements gained by averaging 8 x 8 samples were lost by using the Bligh and Dyer method for fat measurement. This method is known to correlate poorly with Foss-Let Chemical Method.

Conclusion
As Distell has correlated their calibrations against Foss-Let Method, then best comparison is gained by using Foss-Let Method during Laboratory analysis.

Application Data Sheet 6

Title
Comparison of Distell Fatmeter and Foss-Let methods of lipid analysis.

Product
Herring.

Data Source
Torry Research Station, UK Ministry of Agriculture Fisheries and Food.

Sampling Method
One measurement made just above and parallel to the lateral line on each fish. Distell’s measurement methodology for… Herring-1 .
Please refer to measurement instructions section for further details.
Foss-Let analysis on… Correctly prepared…2 x Trimmed Fillets..

Statistics
Number of fish : 137
Maximum lipid content (Foss-Let) : 21.5%
Minimum lipid content (Foss-Let) : 0.16%
Regression equation, y = a + bx : a = 0.069; b = 0.984
Correlation coefficient (r) : 0.955
Standard error : ±1.87

Comparison of Foss-Let and Distell methods for Lipid Measurement

Comments
The correlation obtained with this method was very high despite the fact that the standard procedure was not followed. Individual fish rather than eight fish were measured. For the average of eight fish the results would be even better.

Conclusion
Always follow Distell’s recommended calibration methodology rigorously if you wish to achieve a high correlation with laboratory analysis.

Application Data Sheet 7

Title
Comparison of Distell Fatmeter and Soxhlet Methods of Lipid Analysis.

Product
Atlantic Salmon.

Data Source
Major French producer.

Sampling Method
Four measurements made along the length of the fish, above the lateral line. Distell’s measurement methodology for Salmon-1 .
Please refer to measurement instructions section for further details.
Soxhlet analysis carried out on the actual parts of the flesh measured by the Fatmeter, pooled for each side of the fish and compared with the average meter reading for those four places.

Statistics
Number of fish : 32
Maximum lipid content (Soxhlet) : 21.0%
Minimum lipid content (Soxhlet) : 2.25%
Regression equation, y = a + bx : a = 0.811; b = 0.759 (y is Fatmeter result, x is Foss-Let)
Correlation coefficient r : 0.85
Standard error : ±2.7

Comparison of Soxhlet and Distell methods for Lipid Measurement

Comments
The correlation obtained with this method was good despite the fact that the standard procedure was not followed. The low b-value probably arises from this difference in method. Although the correlation coefficient is high, the large standard error is a reflection of the presence of several outliers.

Conclusion
Always follow Distell’s recommended calibration methodology rigorously if you wish to achieve a high correlation with laboratory analysis.

Application Data Sheet 8

Title
Comparison of Distell Fatmeter and Soxhlet methods of lipid analysis.

Product
Sea bass, trimmed fillets.

Data Source
IFREMER, Nantes.

Sampling Method
Three uncalibrated measurements made along the length of the fish, both sides, above the lateral line, using the Salmon-1 calibration…at the following points…
1. just behind the gills,
2. directly beneath the dorsal fin,
3. directly behind the dorsal fin,

Not using Distell methodology.
Soxhlet analysis on carried out on 2 x Trimmed Fillets

Statistics
Number of fish : 29
Maximum lipid content (Soxlet) : 16.6%
Minimum lipid content (Soxlet) : 2.4%
Regression equation, y = a + bx : a = 0.234; b = 0.783 (y is Fatmeter result, x is Foss-Let)
Correlation coefficient (r) : 0.80
Standard error : ±1.86

Comparison of Soxhlet and Distell methods for Lipid Measurement

Comments
The data here was obtained on Sea Bass, for which the meter had not been calibrated. They were also obtained in a manner that was not the recommended procedure for any species. This was because of the fact that this was one of the first studies of this species, and the measurement method was designed to be optimal for the fish shape and size. Both these factors explain the low value of b.

Conclusion
Distell has since developed a calibration for Sea Bass, and the optimum methodology, as described in themeasurement section of this handbook, will allow a high correlation between Fatmeter and laboratory method.

Application Data Sheet 9

Title
Comparison of Distell Fatmeter and Soxhlet methods of lipid analysis.

Product
Sea bass, trimmed fillets.

Data Source
IFREMER, Nantes.

Sampling Method
Three uncalibrated measurements made along the length of the fish, both sides, above the lateral line, using the Salmon-1 calibration…at the following points…
1. just behind the gills,
2. directly beneath the dorsal fin,
3. directly behind the dorsal fin,

Not using Distell methodology.
Soxhlet analysis carried out on 2 x Untrimmed Fillets

Statistics
Number of fish : 29
Maximum lipid content (Soxlet) : 16.6%
Minimum lipid content (Soxlet) : 2.4%
Regression equation, y = a + bx : a = 1.818; b = 0.675 (y is Fatmeter result, x is Foss-Let)
Correlation coefficient (r) : 0.82
Standard error : ±1.72

Comparison of Soxhlet and Distell methods for Lipid Measurement

Comments :
The data here was obtained on Sea Bass, for which the meter had not been calibrated. They were also obtained in a manner that was not the recommended procedure for any species. This was because of the fact that this was one of the first studies of this species, and the measurement method was designed to be optimal for the fish shape and size. Both these factors explain the low value of b.

Conclusion
Distell has since developed a calibration for Sea Bass, and the optimum methodology, as described in the measurement section of this handbook, will allow a high correlation between Fatmeter and laboratory method.

Application Data Sheet 10

Title
Comparison of Torry Fatmeter and Foss-Let methods of lipid analysis.

Product
Smoked Salmon Mince (Atlantic Salmon).

Data Source
Distell.

Sampling Method
Please refer to measurement instructions section for further details. Foss-Let analysis on… carried out on 300 gms of the blended salmon mince, measured by the Fatmeter.

Statistics
Number of samples : 20
Maximum lipid content (Foss-Let) : 23.95%
Minimum lipid content (Foss-Let) : 7.7%
Regression equation, y = a + bx : a=-1.164; b=0.924 (y is Fatmeter result, x is Soxlet)
Correlation coefficient (r) : 0.914
Standard error : ±1.0

Comparison of Foss-Let and Distell methods for Lipid Measurement

Comments
The data was obtained using FOUR Smoked Salmon Fillets from one Processor.

Conclusion
A new, dedicated calibration had recently been preparedfor the above product. Samples supplied by the processor. However, the high correlation achieved is highly dependent upon the smoking and salting regimes of the individual processor. Each processor is likely to have a different smoking and salting technique, which could result in poor correlation between Fatmeter and laboratory results in some cases.

Application Data Sheet 11

Title
Comparison of Torry Fatmeter and Foss-Let methods of lipid analysis.

Product
Bluefin Tuna.

Data Source
Distell.

Sampling Method
Four fillets cut from each of four fish from the central section of the fish and eight readings taken on each section. Distell’s measurement methodology for… Bluefin - 1 .
Please refer to measurement instructions section for further details.
Foss-Let analysis carried out using all four fillets.

Statistics
Number of samples : 16
Maximum lipid contents (Foss-Let) : 12.8%
Minimum lipid content (Foss-Let) : 1.6%
Regression equation y = a + bx : a=0.177; b=0.964 (y is Fatmeter result, x is Soxlet)
Coefficient correlation (r) : 0.892
Standard error : ±1.73

Comparison of Foss-Let and Distell methods for Lipid Measurement

Comments
The samples were collected from fish between 5 - 10 Kg in weight, with the Fatmeter measurements taken thru’ the skin.

Conclusion
Fatmeter Measurements on whole tuna carcass can be affected by the skin thickness, and bone platelets. With the larger tuna fish, great care is necessary, to ensure that meter readings are taken from areas of the fish where skin thickness does not affect the measurements.