KEEL AND TARSUS LENGTH MAY PROVIDE A GOOD PREDICTOR OF AVIAN BODY-SIZE

Body size is a key measure in many studies. The first principal compon ent using various skeletal measures (skeletal PC1) is commonly conside red as the best indicator of body size. In practice, external body mea sures are used to predict body size, because skeletal measures are har d to obtain in the field. Hitherto, however, external measures could n ot explain more than 40-60% of the variation in skeletal PC1. Here we use a sample of 36 Serin Serinus serinus specimens to obtain an extern al measure of body size which shows a high correlation with skeletal P C1, has low measurement error, and is easy to measure in the field. St epwise multiple regression of skeletal PC1 (dependent variable) on dif ferent external measures revealed keel and tarsus length as the best s ubset of predictors. The two variables accounted for 76% of the total variance in skeletal PC1. This was a substantial improvement over the use of traditional measures (e.g. tarsus, wing and tail length), which only accounted for 63%. We confirm therefore the previous suggestion of using external keel length as a good predictor of overall body size , especially when it is used together with tarsus length. The sum of t he two standardized variables can be used as a simple and reliable pre dictor of avian body size.