A chronic feeding study was carried out in mice with two chemically modified potato starches, hydroxypropyl distarch phosphate (HP-starch) and starch acetate (AC-starch), and with lactose and sodium alginate. Each of the materials was fed to a group of 75 male and 75 female mice for 89 wk. The dietary level of the test products was gradually increased until the diets contained (by weight) 55% HP-starch, 55% AC-starch, 55% lactose or 25% alginate. The control diet contained 55% pregelatinized potato starch. Each of the four test materials caused increased water consumption, distinct caecal and colonic enlargement, a slightly increased incidence of intratubular nephrosis and, with the exception of AC-starch, also slightly lower body weights. An increased incidence of gastric trichobezoars was observed in mice fed either the modified starches or lactose. The occurrence of concrements in the renal pelvis along with slight urinary changes, such as increased amounts of amorphous material in the urine and increased urinary Ca content, in mice fed HP-starch, AC-starch or lactose was regarded as an effect of little, if any, toxicological significance. Alginate fed at 25% (w/w) of the diet was nephrotoxic to mice, as shown by extremely high water consumption, high urine production, urinary incontinence, high pH and low specific gravity of the urine, increased level of blood urea nitrogen, increased kidney weights, distension of the renal calyx and the high incidence of dilated distal tubules. Caecal and colonic enlargement and changes in urinalysis were found to be reversible and had completely or largely disappeared within 2–5 wk of the cessation of the treatment in wk 87. The incidence of intratubular calcinosis or of concrements in the pelvic space was not reduced during the recovery period. The study did not provide any evidence of carcinogenicity of the products tested.