The objective of this paper is to study the economic history of crises in Argentina. Following Prebisch´s conjecture, we look for regularities in the behavior of key macroeconomic variables in the neighborhood of the crises. We perform parametric and non-parametric analysis in order to determine whether the Argentinean crises respond to the predictions of first, second or third generation models. Firstly, we identify crises episodes throughout the Argentine history from 1885 to the present. We apply the Eichengreen, Rose and Wyplosz (1994) methodology to sort crises from noncrises periods, and we distinguish among deep crises (crashes), mild crises and turbulent episodes. Secondly, we split the sample in crises and non-crises years and carried out graphical analysis, and not parametric tests in order to establish weather key macroeconomic variables, namely public expenditure, GDP, external debt, exports, imports and the current account deficit change significantly before, during and after crises. We report the two-sample Kolmogorv-Smirnov test of equality of distributions and the Kruskal-Wallis test of equality of population. Since the previous studies are intrinsically univariate, we also performed a regression analysis, estimating a logit model. It is found that fiscal variables have an important role in determining the probability of crisis. The domestic macroeconomic effects, measured by GDP growth and real M3 growth are also very strong. An appreciation of the currency (lagged once) also increases the probability of a crisis. We can also see that an impairing of external conditions make a crisis more probable.