Observation strategies for morphological characterisation of forages

Several different methods using different numbers of plants and observations have been used for characterisation of germplasm. In this study, two different observation strategies for characterisation were tested in order to determine the minimum number of observations and plants per accession whilst still obtaining accurate results using eight forage legume species, eight forage grass species multiplied from seeds and eight clonally propagated forage grass species. In most cases, 10 observations, with 1 observation per plant would be sufficient to reach an acceptable error percentage. Observations taken on a plant basis with one observation per plant appears to be the most efficient way of scoring morphological characters. If less than 10 plants per accession are available, the reduced number of plants can be compensated by taking more observations per plant. An acceptable error percentage can often not be reached when the number of plants is less than four or five, even if many observations per plant are taken.