Purpose : The image quality of magnetic resonance angiography (MRA) varies according to the imaging techniquesapplied and the parameters affected by blood flow patterns, as well as by the shape of the blood vessels. Thisstudy was designed to assess the influence on signal intensity and its distribution of the geometry of thesevessels, the imaging parameters, and the concentration of contrast media in MRA of stenosis and aneurysm models.Materials and Methods : MRA was performed in stenosis and aneurysm models made of glass tubes, using pulsatileflow with viscosity and flow profile similar to those of blood. Slice and maximum intensity projection (MIP)images were obtained using various imaging techniques and parameters ; there was variation in repetition time, flip angle, imaging planes, and concentrations of contrast media. On slice images of three-dimensional (3D)time-of-flight (TOF) techniques, flow signal intensity was measured at five locations in the models, and contrastratio was calculated as the difference between flow signal intensity (SI) and background signal intensity (SIb)divided by background signal intensity or (SI-SIb)/SIb. MIP images obtained by various techniques and usingvarious parameters were also analyzed, with emphasis in the stenosis model on demonstrated degree of stenosis, severity of signal void and image distortion, and in the aneurysm model, on degree of visualization, distortion ofcontour and distribution of signals. Results : In 3D TOF, the shortest TR (36 msec) and the largest FA (50$^{circ}$)resulted in the highest contrast ratio, but larger flip angles did not effectively demonstrate the demonstrationof the peripheral part of the aneurysm . Loss of signal was most prominent in images of the stenosis modelobtained with parallel or oblique planes to the flow direction. The two-dimensional TOF technique also causedsignal void in stenosis, but precisely demonstrated the aneurysm, with dense opacification of the peripheral part.The phase contrast technique showed some distortions in the imaging of stenosis, and partial opacification of ananeurysm. Contrast enhanced imaging offered no advantages in the imaging of the stenosis, but was excellent fordemonstration of the aneurysm. Conclusion: This study demonstrates a spectrum of MRA images of stenosis andaneurysm model according to variation in imaging parameters and the concentration of contrast media.