The NTP chose to initiate studies in fish as an exploration of alternate or additional models for examining chemical toxicity and carcinogenicity. The use of small fish species in carcinogenicity testing offered potential advantages as a bioassay test system, including significant savings in cost and time over rodent studies. Large numbers of small fish could be easily maintained in a limited area. The two species chosen for study were guppy (Poecilia reticulata) and medaka (Oryzias latipes), both of which are hardy, easily maintained, and have a low occurrence of background lesions. The three chemicals chosen for study in fish had already been studied by the NTP in rodents, permitting a comparison of results between the two models. Two of the chemicals used (2,2-bis(bromomethyl)-1,3-propanediol and 1,2,3-trichloropropane) were mutagenic and multisite carcinogens in rats and mice. The third chemical, nitromethane, was nonmutagenic with a more modest carcinogenic response in rodents. Male and female guppies and medaka were exposed to 2,2-bis(bromomethyl)- 1,3-propanediol (greater than 99% pure), nitromethane, (greater than 99% pure), or 1,2,3-trichloropropane (99% pure) in aquaria water for up to 16 months. OVERALL STUDY DESIGN: Groups of approximately 220 guppies (two replicates of 110) were maintained in aquaria water containing nominal concentrations of 0, 24, 60, or 150 mg/L 2,2-bis(bromomethyl)-1,3-propanediol; 0, 10, 30, or 70 mg/L nitromethane; or 0, 4.5, 9.0, or 18.0 mg/L 1,2,3-trichloropropane. Groups of approximately 340 medaka (two replicates of 170) were maintained in aquaria water containing 0, 24, 60, or 150 mg/L 2,2-bis(bromomethyl)-1,3-propanediol; 0, 10, 20, or 40 mg/L nitromethane; or 0, 4.5, 9.0, or 18.0 mg/L 1,2,3-trichloropropane. The overall study durations were 16 months for all guppy studies, 14 months for 2,2-bis(bromomethyl)-1,3-propanediol-exposed medaka, and 13 months for nitromethane- and 1,2,3-trichloropropane-exposed medaka. Ten guppies and 10 medaka from each group replicate were sacrificed at 9 months for histopathologic analysis. Approximately one third of the remaining fish from each group were placed in chemical-free water at 9 months and constituted a stop-exposure study component. The remainder of the fish were exposed for the duration of the study and constituted the core study component. A stop-exposure component was added to determine if stopping the exposure at 9 months and transferring to chemical-free aquaria might allow for better survival and tumor development. The sex of guppies and medaka was determined at histopathologic analysis. 2,2-BIS(BROMOMETHYL)-1,3-PROPANEDIOL - 16-MONTH STUDY IN GUPPIES: 2,2-Bis(bromomethyl)-1,3-propanediol was chronically toxic to guppies in the 60 and 150 mg/L core and stop-exposure groups. Due to mortality, exposure of core study animals in the 150 mg/L group was terminated on day 443, after approximately 64 weeks on study, and fish were maintained in 2,2-bis(bromomethyl)- 1,3-propanediol-free water in the exposure system until the end of the study at 69 weeks. Nominal exposure concentrations of 24, 60, and 150 mg/L provided actual aquaria water exposure concentrations of 20.0, 53.5, and 139.0 mg/L 2,2-bis(bromomethyl)- 1,3-propanediol, respectively. There were no treatment-related differences between the control and exposed groups in body weights or lengths. At 9 months, hepatocellular adenomas occurred in one 24 mg/L male and in one 150 mg/L male. In the core study, the incidence of hepatocellular adenoma or carcinoma (combined) in 150 mg/L males was greater than that in the controls; multiple adenomas occurred in two 150 mg/L males and in one 150 mg/L female. Cholangioma occurred in a small number of exposed males and females. In the stop-exposure study, incidences of hepatocellular adenoma (including multiple) and of hepatocellular carcinoma were greater in 150 mg/L males than in controls. One cholangioma and one cholangiocarcinoma occurred in the 150 mg/L female group. 14-MONTH STUDY IN MEDAKA: Exposure to 2,2-bis(bromomethyl)-1,3-propanediol did not result in any significant reduction in survival, although the mortality of fish was somewhat greater in the 60 and 150 mg/L core study groups than in the control and 24 mg/L groups. After reallocation, mortality of medaka in the 60 and 150 mg/L core groups was slightly increased over the corresponding stop-exposure groups. Nominal exposure concentrations of 24, 60, and 150 mg/L provided actual exposure concentrations of 19.4, 56.9, and 137.8 mg/L 2,2-bis(bromomethyl)- 1,3-propanediol, respectively. Core study animals in the 60 and 150 mg/L groups were significantly larger, in both body length and weight, than control group fish. In the core study, the incidence of hepatocellular adenoma or carcinoma (combined) was increased in 150 mg/L males. Cholangiocarcinomas occurred in a few exposed males and females, with all but one occurring in 150 mg/L fish. One cholangioma occurred in a 150 mg/L female, and one occurred in a control female. In the stop-exposure study, incidences of hepatocellular adenoma or carcinoma (combined) were marginally increased in the 150 mg/L group of males and in the 60 and 150 mg/L groups of females as compared with controls. Cholangiocarcinoma occurred in one male and one female in the 150 mg/L groups and in one control female. NITROMETHANE - 16-MONTH STUDY IN GUPPIES: Although the cause of death could not be confirmed in many cases, mortality in the 70 mg/L groups appeared to indicate that this level of nitromethane exposure was chronically toxic. This is confirmed by the similar survival rate of guppies from all treatments following removal from treatment aquaria and placement in stop-exposure. Due to the high mortality of fish in the 70 mg/L core study groups, these fish were removed from treatment (day 396) and fixed for histological analyses after approximately 57 weeks on study. The controls and other exposed groups were sacrificed at 70 weeks. Nominal exposure concentrations of 10, 30, and 70 mg/L provided actual exposure concentrations of 9.9, 28.7, and 66.4 mg/L nitromethane, respectively. There were no treatment-related differences between the control and exposed groups in body lengths or weights. 13-MONTH STUDY IN MEDAKA: Nitromethane in the aquaria supported a substantial microfaunal growth which, without frequent cleaning, affected water quality and treatment concentrations. To maintain acceptable water quality and treatment concentrations potentially affected by the rapid microfaunal growth, the study aquaria were brushed once and siphoned three times each day. Due to this frequent activity, a number of fish probably died due to mechanical injury. Unfortunately, the cause of death could not be confirmed in many cases; the mortality from this activity is believed to have been approximately uniform among treatments and should not have affected the comparison of survival between treatments. Based on mortality in this study and the previous life-span evaluation, the life phase of this study was terminated approximately 13.5 months after hatching. Nominal exposure concentrations of 10, 20, and 40 mg/L resulted in actual exposure concentrations of 9.3, 20.8, and 41.7 mg/L nitromethane, respectively. No differences between control and exposed groups were found in body lengths or weights at the 9-month interim evaluation. Due to mortality, unequal numbers of fish were distributed among the core study and stop-exposure aquaria at 9 months. Differences in lengths and weights were found at 13 months. The biological significance of this finding is unknown. At 9 months, a single cholangiocarcinoma occurred in a 40 mg/L male. Hepatocellular adenomas occurred in two 20 mg/L males and in one 40 mg/L female. In the core study, one cholangioma occurred in a 20 mg/L male, and cholangiocarcinomas were seen in a few exposed males, but none occurred in control males. 1,2,3-TRICHLOROPROPANE - 16-MONTH STUDY IN GUPPIES: The survival of exposed guppies was less than that of the control group at 9 months. Reduced survival was evident at 6 months in the 18.0 mg/L groups and at 7 months in the 4.5 and 9.0 mg/L groups. Survival was significantly reduced in the 18.0 mg/L core study group within 1 month of the 9-month interim evaluation, and mortality in this group was 42.6% between 9 months and study termination. Nominal exposure concentrations of 4.5, 9.0, and 18.0 mg/L resulted in actual exposure concentrations of 4.4, 8.8, and 18.2 mg/L 1,2,3-trichloropropane, respectively. Guppies in the 18.0 mg/L core study group were significantly longer and weighed more than the controls. Fish in the 18.0 mg/L stop-exposure group also weighed more than the controls. Mortality of fish during the study resulted in unequal numbers of individuals distributed to core study and stop-exposure aquaria at 9 months. This appears to have influenced the length and weight of fish measured at study termination (i.e., the smaller tank population allowed the fish to grow more). Observed differences in weight and length between controls and 18.0 mg/L fish was most likely an artifact of the reduced numbers of fish in the 18.0 mg/L aquaria. At 9 months, multiple hepatocellular adenomas occurred in one 4.5 mg/L male, and one hepatocellular adenoma occurred in a control male. In the core study, increased incidences of cholangiocellular (bile duct) and hepatocellular neoplasms occurred in exposed groups of males and females. Cholangioma and cholangiocarcinoma were seen in several exposed males and females. In the stop-exposure study, increased incidences of hepatocellular neoplasms occurred in 18.0 mg/L males and increased incidences of cholangiocellular (bile duct) neoplasms occurred in 18.0 mg/L females. (ABSTRACT TRUNCATED)