Evaluación del uso de las medidas externas estándar en los análisis morfométricos de mamíferos

Abstract

THIS IS A LACK OF UNIFORMITY ABOUT WHETHER TO INCLUDE STANDARD EXTERNAL MEASUREMENTS (TOTAL LENGTH, LENGTHS OF HIDFOOT, TAIL, AND EAR) IN MORPHOMETRIC ANALYSIS OF MAMMALS; WHEN SUCH MEASUREMENTS ARE INCLUDED, THE REASONS FOR CHOOSING ONE OVER THE OTHERS IS NOR USUALLY STATED. THIS STUDY ANALYZED WHETHER EXTERNAL MEASUREMENTS OF A SAMPLE SATISFY THE ASSUMPTIONS COMMONLY USED IN STATICAL ANALYSIS: UNIVARIATE AND MULTIVARIATE NORMALITIES, AS WELL AS THE CORRELATION AMONG THESE VARIABLES. DIFFERENCES BETWEEN MEASUREMENTS TAKEN BY DIFFERENT PREPARATORS AND THE VARIANCE DUE TO THIS FACTOR, AS WELL A VARIANCE- COVARIANCE MATRIX HOMOGENEITY, ARE ALSO ANALYZED. THIS STUDY USES A SAMPLE OF PEROMYSCUS MELANOTIS (RODENTIA, MUROIDEA) THAT WAS RECENTLY USED IN A GEOGRAPHIC VARIATION ANALYSIS. IN GENERAL, THE RESULTS SHOW THAT THESE VARIABLES DO NOT SATISFY THE ASSUMPTIONS OF THE TESTS APPLIED TO THEM AND IN FACT, THERE ARE SIGNIFICANT DIFFERENCES AMONG MEASUREMENTS TAKEN BY DIFFERENT PREPARATORS WHICH PRODUCE CONSIDERABLE VARIANCE WHITHIN SAMPLE. THESE RESULTS AND THEIR IMPLICATIONS SHOW THE EXTERNAL MEASUREMENTS ARE NOT RECOMMENDABLE FOR MORPHOMETRIC ANALYSIS.