Methods Of Plant Breeding

METHODS of PLANT BREEDING BY HERBERT KENDALL HAYES Professor and Chief of Division of Agronomy and Plant Genetics, College of Agriculture, University of Minnesota AND FORREST RHINEHART IMMER Professor of Agronomy and Plant Genetics, College of Agriculture, University of Minnesota FIRST EDITION FIFTH IMPRESSION McGRAW-HILL BOOK COMPANY, INC. NEW YORK AND LONDON METHODS OF PLANT BREEDING COPYRIGHT, 1942, BY THE McGRAW-HiLL BOOK COMPANY, INC. PB1NTED IN THE UNITED STATES OF AMEK1CA All rights reserved. This book, or parts thereof, may not be reproduced in any form without permission of the publishers. COMPOSITION BY THE MAPLE PRESS COMPANY, YORK, PA, PRINTED AND BOUND BY COMAC PRESS, INC., BROOKLYN, N. Y, PREFACE Plant breeding is an applied science that is carried out effici ently only through the application of other basic plant sciences. The rapid increase in knowledge of genetics since the rediscovery of Mendels laws of heredity in 1900 and the application of these laws to plant breeding were essential steps in the development of plant breeding as a science. The contributions of cytogenetics in recent years have furnished, in many cases, a clear picture of genetic relationships based upon differences and similarities of chromosome morphology, structure, arid function. Many eco nomic plants are polyploids, and a knowledge of chromosome numbers, pairing behavior in crosses, and gene differences among related species and varieties is essential in building new varieties of plants with the characters desired by the grower and con sumer. Physical and chemical methods of inducing changes in chromosome number and structure and of inducing gene changes are being developed. Satisfactory technicsfor inducing poly ploidy in species and hybrids are available for certain types of plant breeding problems. In order to evaluate a variety, it is necessary to compare it with varieties of known performance. The comparisons made by the plant breeder are extensive, and frequently only a few replications can be grown. The development of adequate statistical methods has aided greatly in making reliable com parative trials. Experimental methods of making reliable comparisons are one of the tools of the plant breeder. Methods have been devised in many cases for differentiating quality, for a determination of the relative value of different characters, including chemical properties, that make it possible, under conditions of controljjadf pollination to select for the characters desired. In problems of breeding for disease resis tance, a knowledge of the genetics of the pathogen is as essential as that of the crop plant itself. With each individual plant, information regarding available varieties, their characters, and vi PREFACE their wild relatives furnishes a basis for the combination of genes desired by the breeder. For diseases caused by pathogens it is equally important to know the probable mode of origin of new strains of the organism, and the number, distribution, and genetic nature of the strains present in the region where the crop plant is to be grown. The subject matter presented in Methods of Plant Breed ing has been used in both undergraduate and graduate courses at the University of Minnesota. The undergraduate course is taught only to junior and senior students. The graduate courses are given for the purpose of teaching standardized methods of breeding for particular categoriesof breeding problems and to present the current viewpoint when the most desirable method of breeding is not so well known. This is with the belief that each of the various methods of hybridization, including the pedigree method of selecting during the segregating generations, the bulk method with self-pollinated plants, the backcross method and convergent improvement, has certain advantages and disadvantages that make it desirable under some conditions and less desirable for other breeding problems...