Because the developing lung is changing over time, quantification of miRNA expression requires careful selection of endogenous control genes according to the studied period of development. Because some developmental events are delayed in male lungs compared with female lungs, the sex has also to be considered. The selected samples covered four developmental stages extending from the end of the pseudoglandular stage 15) to the end of the alveolar stage 30). This developmental period includes lung maturation and alveolarization, which are respectively related to respiratory distress syndrome and bronchopulmonary dysplasia, two major diseases frequently observed in cases of preterm birth. One pool per sex per litter and three litters per time point were analyzed. Because the use of multiple control genes is highly recommended for normalization of RT-qPCR data, five putative endogenous snoRNA control genes were selected. These snoRNAs were subjected to a non-exhaustive expression study with adult mouse tissues by Wong et al. and sno202 was proposed as normalization gene because it showed the highest abundance and least variability across the 12 tested tissues. In this study, RT-qPCR was performed to quantify expression of sno135, sno142, sno202, sno234, and sno251. The results were expressed as mean Cq, which is the standard name for Ct or Cp according to the Realtime PCR Data Markup Language guidelines. The gene to gene differences between the Cq Staurosporine values were quite similar for all the tested developmental time points. The most expressed gene was sno202 for both sexes at all the tested developmental stages, which is consistent with the study of Wong et al. performed on adult mouse tissues, including the lung. sno251 showed the higher variation across the different developmental stages, while Cq values of sno234 were the most stable from stage to stage. Several softwares were developed to analyze the expression stability of reference genes, the most largely used being geNorm, NormFinder and BestKeeper. They are used here. geNorm calculates the stability value M based on the arithmetic mean of all pairwise variations to determine the stability of control genes; the lower the M value, the higher the stability. NormFinder estimates the overall expression variation of the candidate normalization genes, as well as the intra-group and the intergroup variations. Again, decreasing stability values indicate increasing gene expression stability. The two programs determine also the best pair from a panel of control genes. geNorm proceeds by stepwise exclusion of the gene with the highest M value, and a new M value is calculated for the remaining genes, ending with a combination of the two most stable genes. The ranking of genes vary during this process. geNorm also provides the optimal number of reference genes required for normalization. NormFinder selects two best genes with minimal combined inter- and intra- group expression variation.