In the present study, prediction equations for biomass were generated from easily measurable parameters such as diameter at breast height (dbh) and total height. Cross validation was carried out between two localities to test whether equations developed for one locality could be used for other localities. Two localities were compared within India and a further comparison was made with teak growing in Costa Rica, Central America. The result showed that best-fit models developed for one zone cannot be used for other zones. The authors agree with Wang et al. (1995) that errors in biomass estimation can be reduced to a minimum only by employing site-specific equations. Hence, there is an imperative need to develop predictive equations on a regional basis in order to improve their accuracy.