- Print Friendly - Help

• 17.24 Species Testing
• 17.25 Inspection of Product Damaged During Storage
• 17.26 Random Sampling

17.24 Species Testing

17.24.1 Introduction

Laboratory testing for species identification is designed to verify that product packed in licensed premises is true to label.

Samples are collected from meat export slaughterhouses, meat packing houses and export stores on a random basis, and analysed using an iso-electric focusing technique.

Equipment for taking frozen samples consists of a cordless electric drill, drill-battery recharger, 15 mm auger bits and a steel tool box.

17.24.2 Random Selection of Premises

Premises at which samples are to be taken are determined by Head Office using a computer. Each premises is given a weighting depending on its production of boneless product.

The inspector in charge of each premises is to ensure that the sample is taken when notified.

17.24.3 Procedure

(a) Select a carton of frozen boneless product and securely position for drilling.

It is preferable that cardboard and plastic is not mixed with the sample, therefore for:

(i) Face opening cartons, prise up one corner of a flap, make a V shape cut through the inner flap and plastic liner, and push these back out of the way.

(ii) Box and lid cartons, the V shape cut is made directly through the carton and plastic liner, and the product exposed.

(b) Drill firmly down into the product using the forward mode. The frozen meat shavings are collected, facilitated by the reverse mode. Two drillings are needed for a 30 gm sample. Avoid fatty samples as the test involves the separation of muscle protein.

(c) Place the shavings in a plastic bag and identify with a standard buff label showing:

- species recorded on the carton

- MAF carton seal number

- commercial description of product on the carton

- premises number

- date of sample collection.

(d) Reseal the V shape hole with official MAF adhesive tape.

17.24.4 Laboratory Despatch

The 30 gm sample is sent, preferably with the routine fat samples, to:

Meat Monitoring Service
Wallaceville Animal Research Centre
Private Bag
Upper Hutt

All samples are kept frozen and must be sent each week regardless of what other samples are sent. The regular AHL specimen boxes are used in all cases.

17.25 Inspection of Product Damaged During Storage

17.25.1 Introduction

These procedures deal with the direct or indirect damage to, or contamination of, product by such things as:

- fire
- volatile gases (burning structural and other materials, repairs to glass fibre/reinforced polyester cladding, gas torches)
- unauthorised use of aromatic chemicals.

The provisions of the Food and Drug Regulations 1973 are used to supplement the inspection authority of the Meat Act 1981.

17.25.2 General Procedures

(a) Ask the company or owner to nominate a liaison officer, and make sure that the inspection procedures are clearly explained to him. Deal with the liaison officer only.

Where loss adjusters (insurers) are called in by the company, again deal with one person who can represent all the adjusters involved.

(b) Detain all product in the affected areas (Ag.M.74 on doors) and establish the approximate extent of the damage. The detention is to be confirmed in writing and a copy filed along with all other correspondence relating to the incident.

(c) Notify the company or manager, in writing, that all grossly affected product is condemned. Supervise its disposal.

(d) Use a sampling plan for the rest of the suspect product.

17.25.3 Inspection Plan

The company or owner is given two options:

(a) 100% inspection of suspect product, or

(b) Lot by lot sampling (Inspection by Attributes)

This method involves random statistical sampling of a predetermined 'lot' comprising carcasses or containers.

(i) A 'lot' may be any number of carcasses/containers specified by the company or owner.

(ii) Tables of probability are used to determine the acceptance of the lot. These probabilities are based on sample sizes less than 10% of the lot sizes.

(iii) The acceptable levels are:

- minimum 95% probability and maximum 5% defective level

- minimum 99% probability and maximum 1% defective level

if the defect is a risk to human health.

(iv) Assuming that no defects are permitted in the sample, then the minimum sample size is:

- 60 for the 95% p, 5% d

- 300 for the 99% p, 1% d.

It is not possible to reduce these sample sizes without reducing the probability and making the test unacceptable. This means that for lots less than 600, the sample size will be greater than 10% of the lot. This is acceptable as the probability is raised above 95% thereby making the test more stringent but the results more accurate.

17.25.4 Collection of Specimens

(a) Select carcasses or containers on a random basis and randomise the presentation of specimens to the taste panel - refer 17.25.5.

Use of random tables - refer 17.26.

(b) Take a specimen (approx. 5 gm) from the flap of the carcass or from a corner of the container. The specimen should include fat wherever possible.

Specimens from these locations will have had a greater surface area exposed to contamination and will have been the first affected by temperature change.

17.25.5 Taste Panels

(a) A number of taste panels may be set up, of about five persons per panel. Each panel should be relieved after 10-15 specimens as the senses of taste and smell become deadened. At least one hour's rest will be necessary before the senses recover.

(b) The specimens are prepared and analysed as follows:

(i) Boil each specimen for a few minutes in a test tube plugged with cotton wool stopper.

(ii) Remove the stopper and note the odour. Taste the water and give the sample a brief chew.

Abnormal specimens should be readily distinguishable.

17.26 Random Sampling

17.26.1 Introduction

The purpose of random selection is to remove the bias which occurs when investigators use their own judgement. Random selection implies that every item in the population has an equal chance of appearing in the specimen. However, it does not necessarily follow that the specimen will be a 'true' cross-section of the population, as chance will sometimes produce an unusual and therefore unrepresentative result. The smaller the specimen, the more likely that error will occur.

17.26.2 Use of Random Tables

Random tables can be found in most statistics books. A table is shown to illustrate the following example.

e.g. cold store fire

lot size determined by owner to be 800 carcasses.
60 carcasses from each 800 lot will be sampled.

(a) A position on the table is chosen arbitrarily e.g. column 31 row 25.

The numbers may be read across the row or down the column.

(b) The numbers are read in groups of 3 as the lot comprises 800 (3 digits) carcasses. Numbers outside the range 1-800, and duplicates are discarded, until 60 different numbers are obtained.

These carcasses are then sampled.

Reading across the row:

(c) A new set of 60 is established for each lot. me numbers may continue where the previous set finished or a new starting point may be determined.