The aim of this paper is to determine the effectiveness of cross association in detecting the similarity between correlated geological colum- nar sections. For this purpose, cross association is used to compare several geological columnar sections which are arbitrarily selected from different lo- calities in central and north Jordan. It turns out, for most of the study cases, that the sections which consist of the same rock units (formations) are statistically classified as similar (p-value � .05), while sections of differ- ent rock units (formations) are statistically classified as dissimilar (p-value � .05). Geologists get used to study rocks in outcrops or in drill cores. A sequence of sedimentary rocks may be divided up into a number of lithostratigraphical units usually of various sizes. Superimposed lithostratigraphical units are frequently distinguished from each other based on lithological properties and stratigraphical position. The formation is the fundamental unit of stratigraphy with definite lithological composition or a distinctive interbedded or intergraded succession of rock types that must be differ from the adjacent units above and below. The for- mations in a rock sequence named for some geographical localities that show typ- ical exposures of these formations (known as type sections), are usually traceable or mappable from one exposure to another or from well to well in the subsurface (Krumbein and Sloss, 1963; Doyle et al., 1994). Geologists can draw stratigraphi- cal sections for several outcrops (or cores) in an area, and trace beds from section to section that finally leads to what is known as lithological correlation. This type of correlation basically demonstrates the equivalency of rock units across an area. The matter of correlation between different geological sections is relatively easier if the formation in the vertical rock sequence is not so greatly different in its lithology and thickness from one place to another. However, in many cases