The paper presents a statistical method for testing the significance of incongruence between phylogenetic data sources. The authors propose a test based on the null hypothesis of congruence, using a measure of incongruence, D, which is calculated as the difference in tree lengths between combined and separate matrices. This measure is used to assess whether the observed incongruence is statistically significant. The method involves random partitioning of data to estimate the null distribution of the measure. The test is applied to compare different data matrices, such as morphological and molecular data, and has been used in various studies. The authors also address concerns about the use of combined matrices and the potential for one type of data to overwhelm another, arguing that the test effectively identifies incongruence regardless of data size. The method is implemented in a program called 'arn' that automates the congruence test. The approach is compared to the Mann-Whitney U test, which is used to assess differences between two samples. The paper concludes that the proposed method provides an effective way to test for incongruence between phylogenetic data sources.The paper presents a statistical method for testing the significance of incongruence between phylogenetic data sources. The authors propose a test based on the null hypothesis of congruence, using a measure of incongruence, D, which is calculated as the difference in tree lengths between combined and separate matrices. This measure is used to assess whether the observed incongruence is statistically significant. The method involves random partitioning of data to estimate the null distribution of the measure. The test is applied to compare different data matrices, such as morphological and molecular data, and has been used in various studies. The authors also address concerns about the use of combined matrices and the potential for one type of data to overwhelm another, arguing that the test effectively identifies incongruence regardless of data size. The method is implemented in a program called 'arn' that automates the congruence test. The approach is compared to the Mann-Whitney U test, which is used to assess differences between two samples. The paper concludes that the proposed method provides an effective way to test for incongruence between phylogenetic data sources.