Thanks for your help. May be I can not make clear by my writing. The statement which I am going to do is like this "To test such null hypotheses we used the Monte Carlo test, i.e. we chose randomly (5000 times) a value for the current dependent characteristic (richness or actual species pool) in the interval 0 to the maximum possible value of the dependent variable for each observed value of the current independent variable. The maximum observed value was either the calculated size of the regional pool, or the measured size of the actual pool. Each time, we calculated the correlation coefficient r between the independent and dependent variables in order to achieve the empirical distribution of r for the null hypothesis conditions. The empirical probability of cases with a correlation between the two studied variables positive and stronger than that observed in the real data, served as an estimate of the significance level for rejecting the null hypotheses. Thus, by saying that there exists a significant relationship, we mean that the relationship is significantly stronger than expected from our null model". I hope you will figure out and provide me all code for analyses. Thanks. Chitra