Much work has been and is still devoted to applications of nucleic acid measurements, and especially the RNA/DNA ratio, in marine biology, but several papers show a variability that makes this ratio probably unsuitable to some field applications. Five rearing experiments were carried out with Dover sole (Solea solea L.) larvae, three of them in the same standard conditions at 19-degrees-C and two others at temperatures closer to those found under field conditions (12 and 13-degrees-C). The temperature effect on growth is in accordance with general patterns. However, during ontogenetic development, the RNA/DNA ratio exhibits a very strong variability which is not reproducible in different experiments performed under the same conditions. The main consequence is that no relationship is found between this ratio and observed instantaneous growth rates. In starved larvae, the RNA/DNA ratio rapidly reaches low values, which are dependent on temperature and age, but in the same range as in fed larvae during the first days after hatching. These results are discussed in relation to the literature that advocates the use of this ratio as an index of nutritional status or growth of fish larvae. It is suggested that great care should be taken in using the RNA/DNA ratio, which cannot be considered as a reliable index of growth.