Development of methods based on determining expression of individual genes resulted in the need for large amounts of high quality RNA preparations. It is widely accepted that in intact rRNA the 28S and 18S band ratio must be 2:1. It is not quite clear what is the main cause of lower rRNA bands intensity ratio. It is difficult to isolate RNA with 2:1 28S/18S ratio from RNase-rich and some tumor tissues. At the same time this requirement may be excessive and RNA preparations with lower 28S/18S rRNA ratio may be quite adequate for most techniques of determining gene expression. As demonstrated in this study, the level of a particular RNA may be reliably determined by RT-PCR even in a total RNA that is usually considered as degraded (28S to 18S ratio as low as 0.4), provided that random primer is used in RT. In contrast, the use of the oligo(dT) primer in RT-PCR may lead to underestimation of specific mRNA level in the degraded RNA samples, depending on the distance of amplified fragment from the poly(A) end. A criterion based on average degradation level of a number of reference genes is suggested to discriminate specific RNA degradation from random and unspecific ones.