BATCH QUALITY

Before each fish delivery, a quality check of fingerlings must be performed at the hatchery. The size of fish must be verified and the size variation quantified. This procedure is indispensable in order to ensure that the net where the fish will be stocked has a mesh size suitable for the size of the fish (see Table 23), so that no fish will become trapped or escape from the cage.

Fish size

Fish size variation can be quantified by sampling the fish and weighing them one by one, and then calculating the "coefficient of variation" (CV). This parameter is a standardized measure of variation. The CV is defined as:
CV = 0/p
where,
? = standard de viation.
? = mean weight
This value can alternatively be expressed as a percentage (CV x 100), which is referred to as the relative standard deviation (RSD).
Low values of RSD (e.g. between 3 and 10 percent in breams) usually indicate homogenous batches. High values of RSD (more than 20 percent) indicate a large size variation, i.e. fish that range in size from small to large, and the presence of specimens that are possibly too small for the mesh size of the net. A high RSD also indicates that the feed pellet size - determined by the mean weight of the fish - may be not suitable for all of the fish that make up the batch.

Therefore, the size of the feed pellets provided to the fish needs to be carefully evaluated.

Disease

The quality check of the fish sample must also include:
• Quantification of the percentage of deformed fish is obtained by sampling (Plate 96). A sample of 100-200 fish is collected from the tank, anaesthetized and inspected to determine the percentage of deformed fish. Usually, a deformity level of 3 percent is an acceptable limit. There is a large variety of fish deformities, but the most common usually relate to the fishes' skeletal structure. In the field, a preliminary examination can be performed, but a closer analysis with an X-ray machine is recommended in order to reveal any possible hidden deformities. A pathological survey is conducted by sending a sample to a laboratory. A sample of about 30 fish is collected (one sample from each tank, if the batch is divided into more than one tank) and sent to a laboratory for fish pathology diagnosis.
Fish should be inspected for viral, bacterial and parasitological diseases. This is a very important inspection, which should be conducted a few days before the fish are delivered to the farm. This will permit the evaluation of any possible hidden diseases, and will minimize both the risk of disease outbreak in the batch and the introduction of the pathogen into the environment, which could result in the possible transmission of diseases to fish populations in the wild (Colorni, 2002).