Aspergillus niger is a widely used expression host for homologous and heterologous protein production in biotechnological processes. In order to increase product yields, a thorough optimisation of these cultivation processes is necessary. Considering mRNA as the key molecule, which transports the genetic information between DNA and protein production side, the quantification of product specific gene expression provides useful information about product formation already on the level of transcription. Quantitative reverse transcription polymerase chain reaction (qRT-PCR) is a powerful tool to obtain data about gene transcription. However, using this technique the choice of an appropriate reference system is a crucial aspect to provide optimal data normalisation. A prominent approach is the use of so called housekeeping genes as internal references. However, validation of the usability of these reference genes is the fundamental step before starting with qRT-PCR experiments. Adequate reference genes for A. niger have not been published so far. Therefore, 10 possible candidate genes from different functional classes were selected and their applicability as internal references validated. Transcript levels of these genes were compared in sets of 9, 41 and 19 samples from diverse cultivations of A. niger. Under the chosen experimental conditions, the genes act, sarA and cox5 have been identified as genes with the most stable gene expression. The three reference genes were used to normalise qRT-PCR data for glaA gene expression which showed a high correlation with glucoamylase production in continuous cultivations.