In this investigation we applied the techniques of lung sound mapping and time-expanded wave-form analysis to four common diseases that involve the lung: interstitial pulmonary fibrosis (IPF), chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), and pneumonia (Pn). Twenty subjects were studied in each group. We also studied 15 subjects without evidence of lung disease. Differences in timing, character, and location were observed, which allowed separation among these groups. Multiple logistic regression models were created and tested by the bootstrap method. Regression models correctly classified 68 and 79% of subjects. Area under the receiver operating curve ranged from 0.96 for IPF and CHF to 0.80 for COPD. We conclude that auscultatory differences exist among common pulmonary conditions and that statistical models based on auscultatory data perform well in predicting diagnostic categories.