Dehydropyrrolizidine alkaloids (DHPAs) are a large group of globally important plant-derived pro-toxins that can contaminate or are naturally present in animal feed and the human food supply as well as herbal supplements. Their bioactive metabolites are potentially hepatotoxic, pneumotoxic, genotoxic and carcinogenic. Due to the difficulty in obtaining sufficient quantities of purified DHPAs, toxicity studies have largely relied on single intraperitoneal injections in rodent models, and carcinogenicity studies have been limited to a small handful of the hundreds of isolated DHPAs. To assess the relative toxicity of structurally diverse DHPAs in a more biologically relevant manner, male California White chicks were dosed orally with 0.01, 0.04, 0.13, or 0.26 mmol of seven different DHPAs and three DHPA N-oxides kg-1 bodyweight for 7 days. DHPAs were grouped in relation to their toxicity based on clinical, serum biochemical, and histopathological evaluations as well as tissue adduct accumulation rates. Using the same model, a reduced extract from comfrey, a commonly used DHPA containing herb, was compared to its two major constituent DHPAs, intermedine and lycopsamine. Based on the same parameters, the comfrey extract was more toxic than pure lycopsamine or intermedine. Addressing the need for a more sensitive carcinogenicity model, male heterozygous p53 knockout mice were treated with riddelliine 5, 15 or 45 mg kg–1 bodyweight day-1 by oral gavage for 14 days, or given a long-term treatment of riddelliine 1 mg kg-1 bodyweight day–1 in pelleted feed for 12 months. Exposure to riddelliine increased the odds of tumor development in a dose-responsive manner (odds ratio 2.05 and Wald 95% confidence limits between 1.2 and 3.4). The most common neoplasm was hepatic hemangiosarcoma, which is consistent with previously published lifetime rodent studies. The results of this research demonstrate that the California White chick model is sensitive for comparison of DHPA toxicity, and data obtained from this research can be used to validate previous DHPA toxicity research. It also demonstrates that comfrey toxicity may have been previously underestimated. The heterozygous p53 knockout mouse model is beneficial for further investigation of comparative carcinogenesis of structurally and toxicologically different DHPAs and their N-oxides.