To discriminate European eel (Anguilla anguilla, Linnaeus 1758) developmental stages on the basis of body colour, we observed the pigmentation pattern of skin and several biological characteristics of eels collected in the low course of River Tiber (Rome, Italy). A total of 454 individuals have been assigned to a determined stage (Yellow, Intermediate and Silver) and examined to measure five external parameters (total length, weight, pectoral fin length, vertical and horizontal eye diameter). A subsample of 229 eels has been sacrificed in order to collect liver and gut weight and to determine age from otholit observation. A supervised regression technique, Partial Least Squares (PLS) analysis, has been used to develop a model explaining the co-variation between measured parameters, and the three developmental stages. A good discriminant model was obtained for both datasets: the first PLS analysis, using five variables, yields a correct classification of 78 % of the individuals, the second, with eight input variables, yields a correct classification of 79,4 % of the individuals. Yellow and Silver eels are well discriminated by the model, while Intermediate eels don’t constitute an outlined group. The method of staging European eels on the basis of skin colour appeared to be reliable and easy to adopt during field surveys.
Capoccioni, F., Costa, C., Menesatti, P., Ciccotti, E. (2007). Eel silvering stage based on Pls classification, 6, 11-12 [Dl B-51.702-03.].
Eel silvering stage based on Pls classification.
CICCOTTI, ELEONORA
2007-01-01
Abstract
To discriminate European eel (Anguilla anguilla, Linnaeus 1758) developmental stages on the basis of body colour, we observed the pigmentation pattern of skin and several biological characteristics of eels collected in the low course of River Tiber (Rome, Italy). A total of 454 individuals have been assigned to a determined stage (Yellow, Intermediate and Silver) and examined to measure five external parameters (total length, weight, pectoral fin length, vertical and horizontal eye diameter). A subsample of 229 eels has been sacrificed in order to collect liver and gut weight and to determine age from otholit observation. A supervised regression technique, Partial Least Squares (PLS) analysis, has been used to develop a model explaining the co-variation between measured parameters, and the three developmental stages. A good discriminant model was obtained for both datasets: the first PLS analysis, using five variables, yields a correct classification of 78 % of the individuals, the second, with eight input variables, yields a correct classification of 79,4 % of the individuals. Yellow and Silver eels are well discriminated by the model, while Intermediate eels don’t constitute an outlined group. The method of staging European eels on the basis of skin colour appeared to be reliable and easy to adopt during field surveys.Questo articolo è pubblicato sotto una Licenza Licenza Creative Commons