2352 WWW.CROPS.ORG CROP SCIENCE, VOL. 53, NOVEMBER– DECEMBER 2013 RESEARCH S election is the cornerstone of plant breeding and is practiced across all stages of sugarcane improvement (Skinner et al., 1987). As a clonally propagated crop, the irst episode of selection in sugar- cane improvement programs occurs among seedlings planted from true seed of desired crosses. Referred to as the seedling stage (here- after referred to as Stage I), this is the only stage to be established from true seed. Subsequently, the seedlings are appraised either as individuals (mass selection) or as families (family selection). The second stage of selection (hereafter referred to as Stage II) occurs when individual seedlings selected in Stage I are clonally propa- gated through stem cuttings and evaluated in row plots. Family selection in sugarcane involves positive selection of whole populations of seedlings based on data derived from fam- ily plots (Kimbeng and Cox, 2003). Family selection in Stage I is widely practiced to diferent extents in Australia (Hogarth et al., 1990; Jackson et al., 1995a, 1995b; Cox and Stringer, 1998; Kim- beng et al., 2000), the United States (Milligan and Legendre, 1990; Chang and Milligan, 1992a, 1992b; Tai et al., 2003), India (Shanthi Evaluating Sugarcane Families for Yield Potential and Repeatability Using Random Coeicient Models M. M. Zhou,* C.A. Kimbeng, S. Andru, T. L. Tew, M. J. Pontif, and K. A. Gravois ABSTRACT Sugarcane (Saccharum species hybrid) is a clon- ally propagated crop but during variety develop- ment, the irst stage of selection occurs among seedlings established from true seed. All sub- sequent stages are established from vegetative stem cuttings and evaluated as clones. During family selection, whole families of seedlings are selected or rejected based on the family mean performance. Individual seedling selection is restricted to selected families. Confounding due to “seed type” can occur when the relative performance of families or individual genotypes differ between the seedling and clonal stages. The objective of this study was to demonstrate the use of random coeficient models (RCMs) for estimating the potential of sugarcane families to produce elite progeny. Data were collected from 17 families evaluated as seedlings in 2002 and clones in 2003. The elite families selected using RCM analysis produced higher mean cane yield in both the seedling and clonal stages and a higher proportion of high yielding clones than those selected using analysis of covariance and analysis of variance. The slope was the most dis- criminating parameter for family repeatability and was associated with high within-family variability. M.M. Zhou, C.A. Kimbeng, and S. Andru, School of Plant, Environmental and Soil Sciences, LSU AgCenter, Baton Rouge, LA 70803; T.L. Tew, Sugarcane Research Unit, ARS, USDA, Houma, LA 70360; M.J. Pontif and K.A. Gravois, Sugar Research Station, LSU AgCenter, St. Gabriel, LA 70776; M.M. Zhou, South African Sugarcane Research Institute, P. Bag X02, Mt Edgecombe 4300, South Africa; M.M. Zhou, University of the Free State, PO Box 339, Bloemfontein 9300, Republic of South Africa. Received 24 Jan. 2013. *Corresponding author (Marvellous.Zhou@sugar.org.za). Abbreviations: ANCOVA, analysis of covariance; RCM, random coeicient model. Published in Crop Sci. 53:2352–2362 (2013). doi: 10.2135/cropsci2013.01.0052 © Crop Science Society of America | 5585 Guilford Rd., Madison, WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.