Purpose: The purpose of this retrospectiv study was to improve MRI-imaging of the pancreas. The first step was to determine the anatomy of the pancreas, especially the transversal diameter in relation to age and gender, using morphometric measures. The second aspect of the work was to answer the question, if Mangafodipir, a primary liver and pancreas specific contrast agent, produced a measurable enhancement of signal intensity (SI) and, derived from this, of signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). The last aspect addressed the dynamic, contrast enhanced MR imaging (DCE-MRI). The purpose here was to analyse the effect of secretin on pancreas perfusion. To achieve that, a computer based processing was done using a specific software designed for this purpose. Test subjects and methods: 142 pancreas MRI-scans of 64 subjects were retrospectively analysed from 2003 to 2007. MRI-scans were performed either with or without the contrast agents Gadolinium or Mangafodipir. The sizes of the pancreas, SI-, SNR- as well as CNR-values were measured on the basis of these MRI-scans. DCE-MRI-analyses were performed with or without secretin application. SI-curves and, based on computer based functions parameter, also perfusion parameter were obtained from DCE-MRI-sequences. All results were statistically appraised using a double sided, paired students-T-test. Results: The analysis showed a mean size for the caput pancreaticus of 27,1 ± 4,5 mm. Thereby it was bigger than the corpus with 21,6 ± 5,6 mm and than the cauda with 22,6 ± 5,3 mm. The size from pancreatic regions was independent of age. Men pancreases were in average bigger than women pancreases (caput: 27,7 ± 3,8 mm versus 26,5 ± 4,9 mm). The SNR from caput was about 76, independently of the contrast agent used. After application of Mangafodipir, the CNR of the Caput amounted 26, in contrast to 17 after application of Gadolinium. These differences were significant. DCE-MRI-analyses with and without secretin showed no significant differences to the perfusion parameter. But the caput of the pancreas showed a trend to a better perfusion after application of secretion. Conclusion: Using the liver and pancreas specific contrast agent Mangafodipir leads to a better contrast enhancement between pancreas parenchyma and muscle tissue in the MRI-representation of the pancreas. Pancreas lesion can be possibly better detected with this method. Hence, Mangafodipir-enhanced MRI represents a potential diagnostic procedure for pancreatic tumors. Application of secretin by DCE-MRI does not seem to be suitable in the diagnosis of pancreatic tumor and diseases. Changes in perfusion in several pancreatic regions can be showed using DCE-MRI determined perfusion parameters and hence, it facilitates the diagnosis of pancreas diseases.