Abstract: Objective To establish pancreatic cancer model in Sprague Dawley（SD） rats ,and study the distribution and the counts of myofibroblast(MF) in pancreatic cancer and non-cancerous pancreatic tissues.Methods Directly implanted DMBA into pancreas parenchyma of SD rats and established TSA intervening group and control group. The carcinogenesis of rats executed within 3-5 months were inspected by HE stain and microscope for the study of pathomorphological changes.And myofibroblast(MF) was stained by Heidenains method for the counts under microscopy of high fields. Results (1)The prevalence of pancreatic cancer in experimental group was 48.7%(18/37), including 17 pancreatic ductal adenocarcinoma and 1 fibrosarcoma. The prevalence of pancreatic cancer in intervering group was 33.3%(12/36), including 11 pancreatic ductal adenocarcinoma and 1 fibrosarcoma.The prevalence of pancreratic carcinoma was significantly in experimental group and in intervening group than that in control group (0/10,P<0.05 or P<0.01).The maximal diameters of tumor mass of experimental group was higher than that of intervering group (P<0.05).(2) The counts of MF were significantly higher in pancreatic cancer of experimental group and intervering group than those in non-cancerous pancreatic tissues of experimental group and intervering group (P<0.01). The counts of MF were significantly higher in moderately or severely atypical ductal hyperplastic tissues than those in normal tissues or mild atypical ductal hyperplastic tissues in non-cancerous pancreatic tissues(P<0.05). The counts of MF were significantly higher in non-cancerous or cancerous pancreatic tissues of experimental or intervering group than those in control group (P<0.05 or P<0.01). Conclusions Pancreatic MF might have an important effect on the carcinogenesis and progression of pancreatic cancer.