oscopy, false positive results are not uncommon with double contrast radiology, and up to 29% of polyps may be missed, although this figure falls to under 10% for polyps over 1 cm in diameter.'3 The presence of diverticular disease makes the diagnosis of associated lesions more difficult,'0 14 some authorities recommending that colonoscopy should be performed in all patients with symptomatic diverticular disease. 14 Another important group comprises patients with overt or occult rectal bleeding with no cause found on rigid sigmoidoscopy or barium examination: lesions may be shown in 27-81% of such cases, including tumours in 3-1 1%.4 15 Accurate cost comparisons in our hospitals are difficult to make, but the greater necessity for colonoscopy after an initial barium study would suggest that a policy of performing endoscopy initially is likely to work out less costly. For this reason and for reasons of accuracy and expediency and to avoid multiple investigations total colonoscopy may be the preferred initial investigation for those with suspected colonic disease. We have shown that radiology is not a prerequisite for adequate colonoscopic examination. We thank Professor 0 James and Dr R Lendrum for allowing us to include their patients in the study. 1 Tikkanen MJ. Immunogenetic polymorphism of apolipoprotein B in man. Studies with a monoclonal anti-Ag(c) antibody. Am HeartJ 1987;113:462-7. 2 Viikari J, Akerblom HK, Nikkari T, et al. Atherosclerosis precursors in Finnish children and adolescents. IV. Serum lipids in newborns, children and adolescents. Acta Paediatr Scand 1985;318 (suppl): 103-9. 3 Berg K. DNA polymorphism at the apolipoprotein B locus is associated with lipoprotein level. Clin Genet 1986;30:515-20. 4 Talmud PJ, Barni N, Kessling AM, et al. Apolipoprotein B gene variants are involved in the determination of serum cholesterol levels: a study in normo-and hyperlipidaemic individuals.