Estimation of population mean in successive sampling using information on auxiliary character