The hypermodified DNA base beta-D-glucosylhydroxymethyluracil, also called J, is a naturally occurring DNA modification. J was initially detected by 32P-postlabeling in Trypanosoma brucei and was recently also found in several other eukaryotic parasites. To use 32P-postlabeling as a method to quantitate the absolute levels of J in DNA we have tested the postlabeling efficiency of J using various synthesized standard oligonucleotides containing J. It is known that modified nucleotides, especially bulky ones, are often partially recovered by postlabeling and they are poor substrates for some of the enzymes used. We found that on average only 50% of J is recovered, which shows that the amount of J in T. brucei DNA has been twofold underestimated. Experiments with a short oligomer and defined pyrimidine tracts showed that the incomplete recovery of J is caused at least in part by resistance of J-containing DNA to degradation by micrococcal nuclease.