Volume 45 No. 2 June 2013

SABRAO

JOURNAL of BREEDING and

GENETICS

ISSN 1029-7073

VOL. 45 NO. 2 JUNE 2013

CONTENTS

SABRAO Journal, Regional Secretaries and Editorial Board...............................i Lee J-Y, Kim Y-T, Kim H-J, Lee J-H, Kang C-S, Lim S-H, Ha S-H, Ahn S-N, Kim Y-M. Characterization of the HMW-GS 1dx2.2 gene and its protein in a common Korean wheat variety.............159 Sholihin. Performance of promising clones of cassava (Manihot Esculenta Crantz.) for early maturity on some locations over years in Indonesia.........................................169 Azad MAK, Mazumdar MNN, Chaki AK, Ali M, Hakim ML, Mamun ANK, Hase Y, Nozwa S, Tanaka A, Koike A, Ishikawa H, MA Azam.. Photoperiod-insensitive mutants with shorter plant height identified in the M1 generation of rice irradiated with carbon ion beams……………………....179 Shahbazi H, Aali E, Imani AA, Zaeefizadeh M. Inheritance of cell membane stability under heat and osmotic stresses in bread wheat.........................187

Chauhan JS, Singh KH, Mishra DC. AMMI and bi-plot analyses to identify stable genotypes of Indian mustard (Brassica Juncea L.) for oil and seed meal quality characters............................................195 Singh VV, Rajoria A, Chauhan JS, Meena ML, Kumar S. Interrelationships among morphological and seedling characters in F5 progenies of Indian mustard (Brassica juncea L.)...........................................203 Zdravkovic J, Ristic N, Girek Z, Pavlovic S, Pavlovic N, Pavlovic R, Zdravkovic M. Understanding and overcoming seed dormancy in eggplant (Solanum melongena L.) breeding lines................................211 Bahrani A, Madani A, Madani H.. Evaluating the tolerance of bread wheat genotypes for post-anthesis water stress: water use efficiency and stress tolerance indices................................................221 Malambane G, Jaisil P, Sanitchon J, Suriharn B, Jothityangkoon D. Evaluation of genetic variation among finger millet

(Eleusine coracana L. Gaertn) accessions using RAPD markers............................231 Sharma A, Devi J. Genetic divergence in French bean for pod yield-related traits...................................................240 Tempeetikul V, Techawongstien S, Lertrat K, Techawongstien S. Inheritance of pungency in Thai hot pepper (Capsicum Annuum L.).........................................248 Rauf Y, Subhani A, Iqbal MS, Tariq M, Mahmood A, Shah MKN. Screening of wheat genotypes for drought tolerance based on drought related indices..........255 Suriharn B, Lertrat K. Relationship Between elevation levels and yields of tropical waxy corn genotypes...............264 Rebica T, Kumar M, Sharma PR, Noren KS, Datt S. Combining ability analysis for seed yield and component traits in pea under the foot hills of Northeast India..276 Iftikhar R, Hussain SB, Khaliq I, Smiullah. Study of inheritance for grain yield and related traits in bread wheat (Triticum aestivum L.)........................................283 Ramesh M, Arunakumari J, Prashanth Y, Ranganatha ARG, Dudhe MY. Population

improvement for seed yield and oil content by using working germplasm in sunflower (Helianthus annuus L.).........................291 Htoon W, Kaewpradit W, Jogloy S, Vorasoot N, Toomsan B, Akkasaeng C, Puppala N, Patanothai A. Responses of peanut (Arachis Hypogaea L.) genotypes to N2-fixation under terminal drought and their contributions to peanut yield........296 Htoon W, Kaewpradit W, Jogloy S, Vorasoot N, Toomsan B, Akkasaeng C, Puppala N, Patanothai A. Relationships between root traits and nutrient uptake and nitrogen fixation in peanut under terminal drought...............................................311 Sennoi R, Jogloy S, Saksirirat W, Banterng P, Kesmala T, Patanothai A. evaluation of seedling and adult plant stages resistance to Sclerotium Rolfsii Sacc. in Jerusalem artichoke (Helianthus Tuberosus L.).....324 Senamontry K, Lertrat K, Suriharn B.. Response to five cycles of modified mass selection for ear length in waxy corn....332 SABRAO Board ………………...............x Instructions for authors ………............xii

SABRAO THE SOCIETY FOR THE ADVANCEMENT OF BREEDING RESEARCH IN ASIA AND

OCEANIA Visit our website at: http://www.sabrao.org/

i

SABRAO JOURNAL OF BREEDING AND GENETICS ISSN 1029-7073

The SABRAO Journal of Breeding and Genetics is an international journal for plant breeding and genetics research, and is the official publication of the Society. Its objective is to promote the international exchange of research information on plant breeding, by describing new findings, theories, and/or achievements of a basic or practical nature. It also provides a medium for the exchange of ideas, news of meetings, and notes on personal and organizational achievements and developments among the members of the Society. Research articles, short communications, methods, reviews, tutorials, commentaries and opinion articles will be accepted or invited for publication. Scientific contributions will be refereed and edited to international standards. The journal mainly publishes articles for SABRAO members and it is strongly preferred that at least one author should be a current member of the society. From January 2012, there is a US$50 publication fee FOR ALL ARTICLES (including SABRAO members), which must be paid before publication (after acceptance of the article). The publication fee for non-members is US$100. This requirement is used to pay for journal printing costs and to maintain the website.

ADVERTISEMENTS

Advertising will be accepted from Universities offering courses of potential interest to students from SABRAO countries and from book companies or computer software suppliers whose products promote the aims of the Society. Relevant international conferences may also be advertised. Prices are available on application to the Editorial Board.

SABRAO WEBSITE http://www.sabrao.org This website will contain information about the society, information about current officers and regional secretaries, upcoming congresses, and issues of the SABRAO Journal of Breeding and Genetics. In order to improve access for authors and researchers, reprints of journal articles will be uploaded as soon as the journal issue is published.

ii

REGIONAL SECRETARIES Regional Secretaries are elected by the members in each Region. They play an indispensable role in the operations of the Society by: • notifying members of Society announcements, e.g. from the Secretary-General; • recruiting new members; • collecting the annual subscriptions and transferring these to the Treasurer after deducting

expenses; • distributing the Journal issues to paying members if these are delivered to a region in bulk; • organizing other activities, such as local chapter newsletters and meetings, e.g., for the

induction of new members; • keeping books of account and sending an audited statement to the Treasurer annually; and • providing the Secretary-General with a list of financial members in their region each year. In 2013, the Regional Secretaries are as follows: AUSTRALIA Dr. Mandy Christopher Leslie Research Centre, 13 Holberton Street PO Box 2282, Toowoomba Queensland 4350 Email: [email protected] BANGLADESH Dr. Abul Kashem Chowdhury Professor DepartmentinGenetics and Plant Breeding Patuakhali Science and Technology University Patuakhali-8602, Bangladesh Email: [email protected] CHINA (PEOPLES’ REPUBLIC OF) Prof. Cheng Xuzhen Institute of Crop Sciences Chinese Academy of Agricultural Sciences 30 Bai Shi Qiao Road, Beijing 100081. Email: [email protected] INDIA Dr. Ramakrishnan M. Nair AVRDC - The World Vegetable Center Regional Center for South Asia ICRISAT Campus, Patancheru 502 324 Hyderabad, Andhra Pradesh Email: [email protected] INDONESIA Dr. Ismiyati Sutarto Horticulture/Agriculture, CRDIRT-BATAN Jl Cinere, Pasar Jumat, Jakarta 12440. Email: [email protected]

iii

JAPAN Prof. Kazutoshi Okuno Laboratory of Plant Genetics and Breeding Science University of Tsukuba, Tennodai 1-1-1, Tsukuba, 305-8572. Email: [email protected] KOREA Dr. Kyu-Seong Lee, Reclaimed Land Agriculture Research Division NICS, RDA 570-080 #457 Pyeongdong-ro, IKSAN, Jeollabuk-do. Email : [email protected] MALAYSIA Dr. Abdul Rahim Bin Harun Malaysian Nuclear Agency Bangi 43000, Kajang, Selangor. Email: [email protected] PAKISTAN Prof. Naqib Ullah Khan Department of Plant Breeding and Genetics The University of Agriculture, Peshawar 25130 Khyber Pakhtunkhwa Email: [email protected] PHILIPPINES Prof. Teresita Borromeo Department of Agronomy, University of the Philippines Los Baños, College, Laguna. Email: [email protected] SRI LANKA Prof. D.P.S.T.G. (Thilak) Attanayaka Faculty of Agriculture and Plantation Management Wayamba University of Sri Lanka Makandura, Gonawila (NWP) Email: [email protected] TAIWAN, REPUBLIC OF CHINA Dr. Hsun Tu Rural Development Foundation, 5F, 7, Section 1, 4 Roosevelt Road, Taipei 100. Email: [email protected] THAILAND Dr. Kamol Lertrat Department of Plant Science and Agricultural Resources Faculty of Agriculture,Khon Kaen University Khon Kaen 40002, Thailand Email: [email protected] USA/CANADA. Dr. Georgia Eizenga USDA-ARS Dale Bumpers National Rice Research Center 2890 Hwy. 130 East, Stuttgart, AR 72160 Email: [email protected]

iv

VIETNAM Dr. Bui Chi Buu Institute of Agricultural Sciences for Southern Vietnam 121 Nguyen Binh Khiem, District I, Ho Chi Minh City. Email: [email protected]

v

SABRAO EDITORIAL BOARD

The SABRAO Editorial Board was established in 2012. By establishing an editorial team co-ordinated by the Editor-in-chief, it is hoped that the efficiency, content and quality of the journal will dramatically improve. The main duty of associate editors will be processing manuscripts for publication in the journal. This involves finding reviewers, communicating with corresponding authors, following up completed evaluations of manuscripts, checking revisions are thoroughly done, and editing/formatting. Each Associate Editor will be acknowledged as the “communicating editor” for the relevant article when it is published.

Other minor duties include being a contact point for SABRAO members in their respective countries, providing new ideas for the journal (e.g. topics for special issues, ideas for website etc.), and assisting in the preparation and compilation of special issues and conference proceedings. Associate Editors: Dr. Sang-Nag Ahn Professor Department of Agronomy, College of Agriculture & Life Sciences Chungnam National University, Daejeon 305-764 REPUBLIC OF KOREA Email: [email protected] Area of expertise: QTL mapping, molecular genetics and breeding of rice Dr. CN Neeraja Principal Scientist, Biotechnology Unit Crop Improvement Section Directorate of Rice Research, Rajendra Nagar, Hyderabad – 500030 INDIA Email: [email protected] Area of expertise: molecular genetics and breeding Dr. C. Ravindran Assistant Professor Krishi Vigyan Kendra Agricultural College and Research Institute, Tamil Nadu Agricultural Univeristy, Madurai, Tamil Nadu, 625107 INDIA E-mail: [email protected] Area of expertise: breeding and genetics of horticultural and fruit species Dr. Naqib Ullah Khan Professor Department of Plant Breeding and Genetics Khyber Pakhtunkhwa Agricultural University Peshawar 25130 PAKISTAN Email: [email protected] OR [email protected] Area of expertise: plant breeding and quantitative genetics

vi

Dr. Sathiyamoorthy Meiyalaghan (Mei) Scientist Plant & Food Research Private Bag 4704, Christchurch, 8140 NEW ZEALAND Email: [email protected] Area of expertise: genomics and molecular breeding Dr. Cheng Xuzhen Professor Institute of Crop Sciences, Chinese Academy of Agricultural Sciences (CAAS) Beijing 100081 CHINA Email: [email protected] Area of expertise: plant breeding in pulses Dr. Ramakrishnan M. Nair Vegetable Breeder - Legumes AVRDC - The World Vegetable Center ICRISAT Campus, Patancheru 502 324 Hyderabad, Andhra Pradesh INDIA Email: [email protected] Area of expertise: plant breeding and genetics research in pulses and pasture legumes Dr. Sivananda V. Tirumalaraju Research Associate II Soybean Breeding, Genetics and Genomics Program Department of Plant Science South Dakota State University Brookings, SD 57006 USA Email: [email protected] Area of expertise: plant breeding (peanut, soybean and canola), molecular breeding, molecular marker technology Dr. Arbind K. Choudhary Principal Scientist (Plant Breeding) Indian Institute Pulses Research (ICAR) Regional Research Centre UAS Dharwad 580 005, Karnataka INDIA Email: [email protected] Area of expertise: Genetics and breeding of legumes

vii

Dr. Akshaya K. Biswal Plant Breeding, Genetics and Biotechnology (PBGB) International Rice Research Institute (IRRI), Los Baños, Laguna 4031 PHILIPPINES Email: [email protected] Area of expertise: plant molecular genetics, plant molecular biology Deputy Editor-in-Chief Dr. Sanun Jogloy Department of Plant Science and Agricultural Resources Faculty of Agriculture, Khon Kaen University Khon Kaen 40002 THAILAND Email: [email protected] Area of expertise: plant breeding, quantitative genetics, physiological traits Editor-in-Chief Dr. Bertrand (Bert) Collard International Rice Research Institute (IRRI) Los Banos, Laguna 4031 PHILIPPINES Email: [email protected] Alternative: [email protected] Area of expertise: plant breeding and genetics, QTL analysis, molecular breeding

viii

MESSAGES FROM THE EDITOR-IN-CHIEF MOVING TO AN ELECTRONIC JOURNAL SYSTEM From 2013 onwards, the journal will completely move to an electronic format with open access. All articles will be published as pdf files on the website. This was a unamnimous decision made during the SABRAO General Meeting held in January 2012 based on providing greater access for the journal and due to the financial status of the society. Limited printing of hard copies will still continue mainly for institutions or libraries by subscription. NEW SCOPE Plant breeding has changed considerably in the last 20 years. The Editorial Board has decided that from 2013 onwards, the scope of SABRAO J. Breed. Genet. will focus on specific topics of breeding and genetics research that are of direct practical relevance to plant breeders. Classical quantitative genetics research will be considered in the context of how useful the research is to breeders. Authors conducting research in the following topics will be encouraged to submit their articles to the journal:

• Molecular breeding (e.g. marker assisted selection) • QTL mapping and validation • Genetic diversity analysis – primarily using DNA markers • Use of agronomic, morphological or physiological traits in selection • Multi-environment trial analysis • Germplasm evaluation • New methods (e.g. phenotyping methods) of broad interest to breeders • Classical quantitative genetics investigating genetic control of simple or oligogenic,

trait heritabilities, combining ability

Other topics may be submitted after consultation with the Editorial Board. A survey of SABRAO members was conducted in November to December 2012, providing useful feedback. All respondents were generally satisfied with the current scope of the journal. One exception was tissue culture and transformation which several respondents considered beyond the scope of the journal. Many members indicated a preference to see more articles involving molecular breeding. Several excellent suggestions for review articles or special issues were also made. The Editor-in-Chief sincerely thanks members who completed the survey, and welcomes any feedback or suggestions in the future.

ix

CALL FOR SOCIETY MEMBERS TO BE REVIEWERS FOR OUR JOURNAL The SABRAO journal continues to receive a large number of articles. Reviewers play a critical role for the journal by evaluating and editing manuscripts. Interested society members - especially on topics involving quantitative genetics and genetic diversity - are encouraged to register as a potential reviewer for manuscripts submitted to the journal by emailing the Editor-in-Chief. NEXT SABRAO CONGRESS – Indonesia 2015 The next SABRAO congress will be held in Indonesia in 2015 around June (dates and venue to be confirmed). Details will be circulated to members and posted on the website. ACKNOWLEDGEMENTS The Editor-in-Chief would sincerely like to thank the many reviewers for their time and effort. A list of all reviewers will be indicated on the website. Thanks to the new publishing team: Copy Editor: Tess Rola Senior Assistant Editor: Lea Licong Assistant Editors: Ludy Nicar, Marlyn Rala and Ms. Yves Caisip (IRRI) The efforts of the web manager for the SABRAO website, Ella “Kaye” Domingo are also gratefully acknowledged.

SABRAO Journal of Breeding and Genetics 45 (2) 159-168, 2013

CHARACTERIZATION OF THE HMW-GS 1DX2.2 GENE AND ITS PROTEIN IN A COMMON KOREAN WHEAT VARIETY

JONG-YEOL LEE1, YEONG-TAE KIM1, HYO-JUNG KIM1, JUNG-HYE LEE4, CHON-SIK KANG2, SUN-HYUNG LIM1, SUN-HWA HA1, SANG-NAG AHN3* and YOUNG-MI KIM1*

1National Academy of Agricultural Science, RDA, Suwon, 431-707, Korea 2National Institute of Crop Science, RDA, Suwon, 431-707, Korea

3Department of Agronomy, Chungnam National University, Daejeon, 357-764, Korea 4Gyeonggi-Do Agricultural Research and Extension Services, Yeoncheon, 486-833, Korea

*Corresponding authors’ email addresses: Young-Mi Kim: [email protected]; Sang-Nag Ahn: [email protected]

SUMMARY The high-molecular-weight glutenin subunit (HMW-GS) 1Dx2.2 from the Korean wheat cultivar Uri was identified and characterized by one and two dimensional gel electrophoresis, and by liquid chromatography electrospray ionization-tandem mass spectrometry (LC-ESI MS/MS). The amplified coding region of this gene was also cloned and sequenced. Molecular characterization of the 1Dx2.2 gene revealed an open reading frame of 2919-bp encoding a polypeptide of 973 amino acids. In addition, the number of cysteines is four compared to that of a typical x-type HMW-GS which has 4 cysteines. A sequence comparison of the 1Dx2.2 allele in Uri with previously characterized 1Dx2.2 alleles revealed 8 single nucleotide polymorphisms, among which four SNPs were found to be non-synonymous. To confirm the authenticity of the 1Dx2.2, extracted glutenin proteins were separated by two-dimensional gel electrophoresis and analyzed by mass spectrometry. Three spots were observed in the 1Dx2.2 position and were identified as HMW-GS 1Dx2.2 by LC-ESI MS/MS and a MASCOT database search. We successfully purified 1Dx2.2 to a single band by fast protein liquid chromatography (FPLC). This is first report on the molecular cloning and characterization of the HMW-GS 1Dx2.2 in a Korean wheat variety and purification of its protein product. Keywords: Triticum aestivum L, HMW-GS, 2DE, FPLC, wheat quality improvement

Manuscript received: May 12, 2013; Decision on manuscript: May 31, 2013; Manuscript accepted: May 31, 2013. © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013

Communicating Editor: Bertrand Collard

INTRODUCTION High molecular weight glutenin subunits (HMW-GSs) play an important role in determining the dough viscoelastic properties and bread-making quality of bread wheat (Triticum aestivum L.) (Shewry et al., 1992). It is known that HMW-GSs are encoded at the Glu-1 loci (Glu-A1, Glu-B1 and Glu-D1) on the long arms of homoeologous group 1 chromosomes, which comprise two genes, x-

and y-type HMW-GS (Shewry et al., 2003). The primary structure of a HMW glutenin subunit is composed of four regions, the signal peptide, N- and C- terminal domains, and a central repetitive region (Shewry et al., 1995). In the N- and C-terminal regions, the cysteine residues which form inter- and intra-chain disulfide bonds are highly conserved in both number and position, and the inter-chain disulfide bonds are known to stabilize the glutenin polymers. The repetitive domains of the subunits comprise highly

RESEARCH ARTICLE

SABRAO J. Breed. Genet. 45 (2) 159-168

160

repeated sequences, based on a combination of two or three short peptide motifs.

Allelic variations in HMW-GS genes influence dough properties. The HMW-GS pair Dx5 + Dy10 has been found to be associated with a higher dough strength compared with the HMW-GS pair Dx2 + Dy12 (Payne, 1987; Odenbach and Mahgoub, 1988; Shewry et al., 1992; Lu et al., 2004). The molecular mechanisms underlying this phenomenon, however, remain to be elucidated.

Subunit 1Dx2.2 is widely distributed in Korean wheat varieties and advanced lines, being present in 17 (65%) of the established 26 Korean wheat cultivars (Park et al., 2011). Furthermore, its presence is associated with poor quality and inferior noodle color and lower loaf volume and harder crumb firmness despite similar protein contents with commercial flours of higher quality (Park et al., 2005, 2006). 1Dx-type HMW-GSs have been cloned and sequenced in common wheat varieties and their relatives (Margiotta et al., 1993; Buonocore et al., 1996; D’Ovidio et al., 1994, 1996; Wan et al., 2005; Zhang et al., 2007; Ren et al., 2008; Fang et al., 2009). 1Dx subunits show size variations within their repeat domains and this phenomenon is due almost entirely to the insertion and duplication of blocks of high repeat sequences (Wan et al., 2005), and to unequal homologous recombination and illegitimate recombination contributing to the duplication and deletion of large fragments present in Glu-D1 alleles (Zhang et al., 2007).

To elucidate the properties of different glutenin subunits in dough, in vitro re-oxidation experiments that incorporate purified subunits into base flour have been performed and the dough mixing properties have been conducted by 2 g mixograph (Schropp et al., 1995; Sissons et al., 1998; Bekes et al., 1994, 1995; Shewry et al., 2003). For the functional study of HMW-GSs, it is very important to develop methods to purify large amounts of individual HMW-GS proteins.

In this paper, we report the isolation and characterization of the HMW-GS 1Dx2.2 gene and protein from the Korean wheat cultivar Uri by one and two dimensional gel electrophoresis and mass spectrometry. In addition, we report

the successful purification of 1Dx2.2 protein by FPLC. MATERIALS AND METHODS Plant materials The Korean wheat cultivar Uri was kindly provided by the National Institute of Crop Science, Suweon, Korea. It is an early-maturing, semi-dwarf, lodging-resistant and soft white winter wheat and was derived from the cross Geuru/Ol in 1981. Wheat flour from the Chinese Spring (1Bx7 + 1By8 and 1Dx2 + 1Dy12) and Jokyoung (1Ax1, 1Bx7 + 1By8 and 1Dx5 + 1Dy10) cultivars were used as standards for HMW-GS identification. Extraction of glutenin mixtures for one and two-dimensional SDS-PAGE A glutenin fraction was prepared using the procedure of Singh et al. (1990) and was precipitated with acetone containing 15% (v/v) TCA. Protein quantification for extracted HMW-GSs was determined using the Bradford method. Electrophoresis and mass spectrometry HMW-GSs were identified by one-dimensional SDS-PAGE. For two-dimensional SDS-PAGE analysis of HMW-GSs, the glutenin pellets obtained from precipitation with acetone containing 15% (v/v) TCA were dissolved in 350 µl of rehydration buffer containing 7M urea, 2M thiourea, 2% CHAPS, 0.5% IPG buffer added onto an 18 cm IPG strip (pH3-11, GE healthcare). After in-gel rehydration for 15 h, IEF was carried out for a total of 80 kVh. The IPG gel strips were equilibrated with 6 M urea, 75 mM Tris-HCl (pH 8.8), 29.3% glycerol, 2% SDS and 1% DTT for 15 min and then incubated with the same buffer containing 2.5% iodoacetamide instead of DTT for 15 min. For the second dimension, SDS-PAGE was performed using a 10% gel. After electrophoresis, the gel was stained with Coomassie Brilliant Blue R-250. Protein spots were excised from the gel and in-gel digested with trypsin. Tryptic peptides were subsequently

Lee et al. (2013)

161

analyzed using reversed phase capillary HPLC directly coupled to a Finnigan LCQ ion trap mass spectrometer (LC-MS/MS) (Zuo et al., 2001). Proteins were identified by MS/MS ion search using the MASCOT search engine (Matrix science). Extraction and FPLC purification of HMW-GSs Extraction and purification of HMW-GSs were performed using a modification of the procedure of Wang et al. (2006). Briefly, flours from the wheat cultivar Uri was defatted with 0.5 M NaCl and monomeric glidians were removed by extraction with 50% (v/v) propan-1-ol. The glutenin proteins were then extracted with 50% (v/v) propan-1-ol containing 5% (w/v) 2-mercaptoethanol at 60 °C.

In order to extract only HMW-GS mixtures, the propan-1-ol concentration was increased to 60% (v/v) and the precipitate was collected by centrifugation and discarded. The supernatant was allowed to stand overnight at 4°C, and the precipitated protein was collected by centrifugation, dissolved in 8 M urea and dialyzed against 1% (v/v) acetic acid for 30 h and freeze-dried. To check the extraction of only HMW-GSs and removal of LMW-GSs and glidians, SDS-PAGE was performed at the last step of the extraction. HMW-GSs were then separated using the hydrophobic chromatography method described by Wang et al. (2006). The chromatography was run on a resource PHE analytic column (6.4 mm i.d. x 30 mm, particle size 15 μm, bed volume 1 ml) using the AKTA-FPLC system (Amersham Pharmacia Biotech). The columns used were first equilibrated with buffer A (0.05 M Tris–HCl containing 4 M urea and 0.45 M (NH4)2SO4, pH 8.0). Extracted HMW-GS mixtures were treated with 6 ml of 8 M urea containing 5% (v/v) 2-mercaptoetanol and the dissolved mixtures were then loaded onto the columns. HMW-GS subunits were eluted using a linear gradient buffer A and B (0.05 M Tris–HCl containing 4 M urea, pH 8.0) at a flow rate of 0.5 ml/min at a room temperature. HMW subunits of fractions eluted from the column were confirmed by SDS-PAGE. The subunits were then pooled and dialyzed against 1% acetic

acid for 60 h at 4 °C. The dialyzed HMW subunits were finally freeze-dried.

Genomic DNA preparation, PCR amplification, and cloning and sequencing of the 1Dx2.2 gene Genomic DNA was prepared from young leaves via the CTAB method described by Liang et al. (2010). In order to amplify the whole ORF of 1Dx2.2 gene by genomic PCR, primers were constructed according to the DNA sequences of HMW-GS 1Dx5 (X12928) and 1Dx2 (X03346) as follows: P1, 5’-ATGGCTAAGCGGTTAGTCCT-3’; P2, 5’-CTATCACTGGCTGGCCGAC-3’. PCR reactions were performed using high fidelity PrimeSTAR HS DNA polymerase (Takara) in a total volume of 50 µl containing 25 µl 2 x PrimeSTAR GC buffer, 0.2 mM of dNTPs, 10 pmol of each primer, 100 ng of genomic DNA, and 1.25 U PrimeSTAR HS DNA polymerase. The amplification program was as follows: denaturation at 98 ℃ for 3 min; 35 cycles of 98 ℃ for 10 s, and 68 ℃ for 3 min; and a final extension step at 68 ℃ for 5 min. PCR products were cloned into the T-Blunt vector (SolGent). The nucleotide sequence was obtained by producing sets of overlapping subclones by restriction enzyme and primer walking. DNA sequencing was performed by GenoTech Corp (Daejeon, Korea) and the final nucleotide sequences for 1Dx2.2 ORF were determined from the sequencing results of three independent clones. Comparative analysis of the 1Dx2.2 genes The nucleotide acid sequence and deduced subunits of 1Dx2.2 that was previously identified, accession number AY159367 (Wan et al., 2005), JF736015, and the 1Dx2.2 gene identified this study were aligned using Lasergene version 7 (DNASTAR).


162

RESULTS SDS-PAGE, 2 DE and LC-ESI MS/MS analysis SDS-PAGE of HMW-GS proteins from the hexaploid wheat genotypes Jokyoung, Uri and Chinese Spring showed the presence of a very slow migrating 1Dx2.2 protein of the Uri variety when compared with Jokyoung and Chinese Spring (Figure 1). To confirm authenticity of this 1Dx2.2 product and further characterize the HMW-GSs expressed in Uri, two-dimensional gel electrophoresis and LC-ESI MS/MS analysis were conducted. The two-dimensional electrophoretic profile showed three spots at the position 1Dx2.2, two spots in the position corresponding to 1Bx7, one spot in the 1By8 position and two spots in the 1Dy12 position (Figure. 2). Each of these spots was digested and subjected to LC-ESI MS/MS. The molecular weight, pI value, and a MASCOT score value for each spot are shown in Table 2. Three spots in the 1Dx2.2 position were identified as the HMW-GS 1Dx2.2 and the other spots were also identified as the HMW-GSs in each position. Spots corresponding to 1Dx2.2 showed more acidic properties than other HMW-GS spots.

Isolation and characterization of the HMW-GS gene 1Dx2.2 Genomic DNA from the wheat cultivar Uri was used as a template for the amplification of the HMW-GS gene 1Dx2.2, using the primers P1 and P2 as it is known that HMW-GS genes contain no introns. A PCR product of about 2.9 kb was cloned and its sequence was analyzed using overlapping subclones prepared by restriction enzyme cutting and primer walking. The cloned PCR product contained an ORF of 2,913 bp encoding a peptide of 973 amino acids. This sequence has been deposited in the GenBank database of NCBI under accession no. JX112756. The amino acid sequence of the subunit 1Dx2.2 has a typical x-type HMW-GS structure when compared with other x-type HMW-GSs. It comprises a signal peptide of 21 amino acids, an N-terminal domain of 89 amino acids, a repetitive domain of 819 amino acids, and a C-terminal domain of 42 residues. Four cysteine residues were identified, three in the N-terminal domain and one within the C-terminus (Figure. 4). The central repetitive domain of 1Dx2.2 comprises 23 nonapeptides (GYYPTSPQQ), 86 hexapeptides (PGQGQQ) and 24 tripeptides (GQQ).

Table 1. Positions of SNPs identified in the 1Dx2.2 s gene. Accession no. 9 235* 393 651 1248 1259* 1348* 2129* AY159367 G A G A A T C A JF736015 A G A G T C T T Uri 1Dx2.2 G G G G A C C A * Restriction enzyme site. Table 2. LC- ESI MS/MS identification of HMW-GSs from wheat cultivar Uri. Spot numbers are indicated in Figure 2.

Spot no.

Protein description

MW (Da) exp.

pI exp.

NCBI accession no.

No. of peptides

MASCOT score

1 HMW-GS 1Dx2.2 102990 5.61 gi26515125 3 92 2 HMW-GS 1Dx2.2 102990 5.61 gi26515125 3 80 3 HMW-GS 1Dx2.2 102990 5.61 gi26515125 3 139 4 HMW-GS 1Bx7 85182 8.68 gi71084277 6 384 5 HMW-GS 1Bx7 85182 8.68 gi71084277 6 302 6 HMW-GS 1By9 75656 8.64 gi22090 5 180 7 HMW-GS 1Dy12 70824 7.64 gi121452 3 105 8 HMW-GS 1Dy12 70824 7.64 gi121452 2 143

Lee et al. (2013)

163

Figure 1. SDS-PAGE analysis of HMW subunits in wheat cultivars Jokyoung (lane 1), Uri (lane 2) and Chinese Spring (lane 3).

Figure 2. 2D gel electrophoresis of glutenin fractions from wheat cultivar Uri for LC-ESI MS/MS identification. Individual spot numbers are indicated.

Figure 3. (a) FPLC profile of the HMW-GS mixture from wheat cultivar Uri. (b) SDS-PAGE patterns of the combined elution fractions in the FPLC profile. In fraction 3, a single band of 1Dx2.2 was detected. 1, 2, 3 and 4 correspond to the peak numbers in (a).


164

Figure 4. Multiple alignment of the deduced amino acid sequences of 1Dx HMW-GS genes including 1Dx2.1 (AY517724), 1Dx2.2 (JX112756), 1Dx2.2* (AJ893508), 1Dx2 (X03346) and 1Dx5 (X12928). Signal peptide, N-terminus, repetitive domain and C-terminus are indicated. Cysteines are boxed.

Lee et al. (2013)

165

Figure 5. Multiple alignment of 1Dx2.2 nucleotide sequences. Non-synonymous substitutions are indicated.

Detection of SNPs in three 1Dx2.2 variants The nucleotide sequence of the 1Dx2.2 gene in the Uri cultivar was compared with two reported 1Dx2.2 genes, AY159367 (Wan et al., 2005) and JF736015 and a total of 8 SNPs were detected as shown in Table 1 and Figure 5. One SNP each in the signal peptide (at position 9) and in the N-terminus (at position 235) was detected and the other 6 SNPs were identified in the repetitive domain. Of these 8 SNPs, 5 were non-synonymous and resulted in amino acid substitutions: 235 at position 79 (valine -

methionine), 1248 at position 416 (glutamine - histidine), 1259 at position 420 (proline - leucine), 1348 at position 450 (proline - serine) and 2129 at position 710 (glutamine - leucine). Four SNPs resulted in changes to a restriction enzyme site: at position 235 Ban I, at position 1259 Msp I, at position 1348 PspG I, StyD4 I and at position 2129 Bfa I. Purification of HMW-GSs by FPLC 1Dx2.2 was purified from flour of the common wheat cultivar, Uri, using a hydrophobic


166

Resource column. SDS-PAGE of the preparations showed a high level of purity (Figure 3). Various elution conditions were tried using different combinations of pH, hydrophobicity, flow rate, and linear gradients. The elution procedure showing the highest purity level for HMW-GS was found to be a linear gradient of 0.45 M (NH4)2SO4 concentration in buffer from 100% to 0%. Figure 3b shows the SDS-PAGE patterns of the fractions and the HMW-GS mixture extracted before running. Peak 1 was a mixture of 1Dy12 and 1By8. Peak 2 was traces of the 1D2.2 subunit. Peak 3 was pure 1Dx2.2 subunit. Peak 4 contained the 1Bx7 subunit with traces of the 1D2.2 subunit. DISCUSSION We have here identified and characterized a HMW-GS 1Dx2.2 subunit by SDS-PAGE and LC-ESI MS/MS, and PCR amplified the coding sequence from the wheat cultivar Uri.

Alleles encoding the 1Dx subunit showed a large amount of size variation through segment deletions or insertion in repetitive domains, and this makes it possible to develop PCR-based markers that can distinguish these alleles. PCR markers have been developed for the detection of 1Dx2 and 1D5 (Ahmad, 2000), and it is also likely that markers can be readily developed for the identification of 1Dx2.2, 1Dx2 and 1Dx5 (Figure 4). Comparison of the Uri 1Dx2.2 sequence with that of two other 1Dx2.2 subunit genes showed 8 nucleotide polymorphisms (Wan et al. 2005). Four SNPs among these variations cause changes in restriction enzyme sites, and the development of genotype-specific CAPs markers using these SNPs is under way (data not shown). These markers could be useful in selecting specific alleles for wheat quality improvement in a breeding program, assuming they are polymorphic in breeding material.

We identified 8 protein spots by 2-DE: three, two, one and two spots at the expected positions for 1Dx2.2, 1Bx7, 1By8 and 1Dy12, respectively. In addition, LC-ESI MS/MS analysis showed that the spots matched with the corresponding proteins previously characterized.

These results suggest that the wheat genome contains at least three copies of 1Dx2.2, two copies of 1Bx7, one copy of 1By8 and more than two copies of 1Dy12. In addition, 1Dx2.2 is most likely to be acidic protein, which is a typical feature of 1Dx HMW-GSs.

Previously, Bietz and Simpson (1992) described the successful purification of a HMW-GS by using a combination of size exclusion and ion-exchange chromatography. However, our initial attempts using these techniques were not successful as they were time-consuming and labor intensive. Among the many established protein purification techniques, the hydrophobic column chromatography was found previously to be a simple and easy to use method for purification of a HMW-GS from flour (Wang et al., 2006).

The subunits 1Dx2.2 + 1Dy12 are widely distributed in most of the known Korean wheat cultivars and advanced lines and show inferior processing properties compared with commercial noodle wheat flours (Park et al., 2006, 2011). Although the molecular basis for the association of 1Dx2.2 with poor quality wheat has not been clearly determined, previous studies have indicated a positive relationship between the size of the HMW-GS and the dough strength. Indeed, subunits 1Dx2.2 and 1Dx2.2* have been associated with a greater dough strength than subunit 1Dx2 using a 2 g mixograph (Bekes et al., 1994, 1995). These results are consistent with a model in which the glutenin polymers interact via inter-chain hydrogen bonds formed between the subunit repetitive domains, with longer subunits forming more stable interaction (Belton, 1999, 2005). The purification and functional characterization of 1Dx HMW subunits of varying lengths are thus necessary in the future to shed light on this question. ACKNOWLEDGEMENTS This work was carried out with the support of the “Research Program for Agricultural Biotechnology (PJ907133, PJ006680)”, National Academy of Agricultural Science, Rural Development Administration, Republic of Korea

Lee et al. (2013)

167

REFERENCES Ahmad M (2000) Molecular marker-assisted

selection of HMW glutenin alleles related to wheat bread quality by PCR-generated DNA markers. Theor. Appl. Genet. 101: 892–896.

Bietz JA, Simpson DG (1992) Electrophoresis and chromatography of wheat proteins: available methods, and procedures for statistical evaluation of the data. J. Chromatogr. 624:53–80.

Belton PS (1999) On the elasticity of wheat gluten. J. Cereal Sci. 29:103–107.

Belton PS (2005) New approaches to study the molecular basis of the mechanical properties of gluten. J. Cereal Sci. 41:203–211.

Bekes F, Anderson OD, Gras PW, Gupta RB, Tam A, Wrigley CW, Appels R (1994) The contribution to mixing properties of 1D HMW glutenin subunits expressed in a bacterial system. In: Henry RJ, Ronalds JA (eds) Improvement of cereal quality by genetic engineering. Plenum Press, New York, pp 97–103.

Bekes F, Gras PW, Gupta RB (1995) The effect of purified cereal polypeptides on the mixing properties of dough. In: Williams YA, Wrigley CW (eds) Proceedings of the 45th Australian Cereal Chemistry Conference. The Royal Australian Chemical Institute, Cereal Chemistry Division, pp 92–98.

Buonocore F, Hickman DR, Caporale C, Porceddu E, Lafiandra D, Tatham AS, Shewry (1996) Characterization of a novel high Mr glutenin subunit encoded by chromosome 1D of bread wheat. J. Cereal Sci. 23:55-60.

D'Ovidio R, Porceddu E, Lafiandra D (1994) PCR analysis of genes encoding allelic variants of high-molecular-weight glutenin subunits at the Glu-D1 locus. Theor. Appl. Genet. 88:175-180.

D'Ovidio R, Lafiandra D, Porceddu E (1996) Identification and molecular characterization of a large insertion within the repetitive domain of a high-molecular-weight glutenin subunit gene from hexaploid wheat. Theor. Appl. Genet. 93:1048-1053.

Fang J, Liu Y, Zhang C, Yang G, Li K, Shewry PR, He G (2009) Cloning and molecular characterization of a novel x-type glutenin gene Glu-Dt1 from Aegilops tauschii. Mol. Biol. Rep. 36:1195–1202.

Liang YS, Bae HJ, Kang SH, Lee T, Kim MG, Kim YM, Hwa SH (2009) The Arabidopsis beta-carotene hydroxylase gene promoter for a

strong constitutive expression of transgene. Plant Biotechnol. Rep. 3:325–331.

Lu XB, Zhang YB, Song QJ, Liu DN, Zhang CL and Zhao HB (2004) Qualitative difference between HMW-GS 5 + 10 and 2 + 12 NILs of four spring wheat cultivars with high quality genetic background. Agric. Sci. China 3:568–574.

Margiotta B, Colaprico G, D'Ovidio R, Lafiandra D (1993) Characterization of high Mr subunits of glutenin by combined chromatographic (RP-HPLC) and electrophoresis separations and restriction fragment length polymorphism analyses of their encoding genes. J. Cereal Sci. 17:221-236.

Odenbach W, Mahgoub ES (1988) Relationships between HMW glutenin subunit composition and the sedimentation value in reciprocal sets of inbred backcross lines derived from two winter wheat crosses. Proceeding of the 7th International Wheat Genetics Symposium, Cambridge, pp987–991.

Park CS, Kim YK, Han OK, Lee MJ, Park JC, Seo JH, Hwang JJ, Kim JG and Kim TW (2005) Characteristics of biochemical markers and whole-wheat flours using small scaled sampling methods in Korean wheats. Kor. J. Crop Sci. 50:346–355.

Park CS, Baik BK, Kang MS, Park JC, Lim JD (2006) Flour characteristics and end-use quality of Korean wheats with 1Dx2.2 + 1Dy12 subunits in high molecular weight glutenin. J. Food Sci. Nutr. 11: 243–252.

Park CS, Kang CS, Jeung JU, Woo SH (2011) Influence of allelic variations in glutenin on the quality of pan bread and white salted noodles made from Korean wheat cultivars. Euphytica 180: 235–250.

Payne PI (1987) Genetics of wheat storage proteins and the effect of allelic variation on bread-making quality. Anun. Rev. Plant Physiol. 38:141–153.

Ren Y, Wang T, Xu ZB, Yang ZJ, Ren ZL (2008) Molecular characterization of a novel HMW-GS 1Dx50 associated with good bread making quality (Triticum aestivum L.) and the study of its unique inheritance. Genet. Resour. Crop Evol. 55:585–592.

Schropp P, Belitz HD, Seilmeier W, Wieser H (1995) Reoxidation of high molecular weight subunits of glutenin. Cereal Chem. 72: 406–410.

Shewry PR, Halford NG, Tatham AS (1992) High molecular weight subunits of wheat glutenin. J. Cereal Sci. 15:105–120.


168

Shewry PR, Tatham AS, Barro P, Lazzeri P (1995) Biotechnology of breadmaking: unraveling and manipulating the multi‐protein gluten complex. Biotechnology 13: 1185‐1190.

Shewry PR, Halford NG, Lafiandra D (2003) The genetics of wheat gluten proteins. In: Hall JC, Dunlap JC, Friedman T (eds) Advances in Genetics, Gol 49. Academic Press, London, pp 111–184.

Shewry PR, Gilbert SMA, Savage WJ, Tatham AS, Wan YF, Belton PS, Wellner N, D’ovidio R, Bekes F and Halford NG (2003) Sequence and properties of HMW subunit 1Bx20 from pasta wheat (Triticum durum) which is associated with poor end use properties. Theor. Appl. Genet. 106:744–750.

Sissons MJ, Bekes F, Skerritt JH (1998) Isolation and functionality testing of low molecular weight glutenin subunits. Cereal Chem. 75: 30–36.

Singh NK, Shepherd KW, Cornish GB (1990) A simplified SDS-PAGE procedure for separating LMW subunits of glutenin. J Cereal Sci. 14:203–208.

Wan YF, Yan ZH, Liu KF, Zheng YL, D’Ovidio R, Shewry PR, Halford NG, Wang DW (2005) Comparative analysis of the D genome-encoded high molecular weight subunits of glutenin. Theor. Appl. Genet. 111: 1183–1190.

Wang YP, Liu LL, Zhao HX, Sun GL, Guo AG (2006) Purification of wheat flour high-molecular-weight glutenin subunits by fast protein liquid chromatography. J. Cereal Sci. 44: 174-181.

Zhang YZ, Li XH, Wang AL, An XL, Zhang Q, Pei YH, Gao LY, Ma WJ, Appels R, Yan YM (2008) Novel x-type highmolecular-weight glutenin genes from Aegilops tauschii and their implications on the wheat origin and evolution mechanism of Glu-D1-1 proteins. Genetics 178:23–33.

Zuo X, Echan L, Hembach P, Tang HY, Speicher KD, Santoli D, Speicher DW (2001) Towards global analysis of mammalian proteomes using sample prefractionation prior to narrow pH range two-dimensional gels and using one-dimensional gels for insoluble large proteins. Electrophoresis 22: 1603–1615.


PERFORMANCE OF PROMISING CLONES OF CASSAVA (Manihot esculenta Crantz.) FOR EARLY MATURITY ON SOME LOCATIONS OVER YEARS IN INDONESIA

SHOLIHIN

Indonesian Legumes and Tuber Crops Research Institute, Malang

Jl. Raya Kendalpayak Corresponding author’s email: [email protected]

SUMMARY

It is predicted that demand for cassava in Indonesia will increase markedly in the future. Therefore cassava production in Indonesia needs to increase. The aim of the experiment reported in this paper was to test the performance of promising clones of cassava for early maturity in different location over years. Mean square of standard deviation of CMM9904-100, SM 2361, MLG 10.312, Malang 1, Adira 1, and Adira 4 were not significantly different from zero, but the others were significantly different. The coefficient of regression (bi) of clones/varieties tested were not significantly different from 1. CMM9904-100, SM 2361, MLG 10.312, Malang 1, Adira 1, and Adira 4 were stable clones across environments, but the others were not stable. The mean of the fresh tuber yield of OMM 02048-6 over locations and years was the highest (32.67 t/ha, 67 % higher than Adira 1, equal to Rp 9.191.000/ha or around US $ 10132 (US$1 = Rp 9000). Keywords: cassava, fresh tuber yield, stability, early maturity. Manuscript received: February 5, 2012; Decision on manuscript: September 22, 2012; Manuscript accepted: September 29, 2012


INTRODUCTION

Demand for cassava tends to increase specially for industry in Indonesia. Therefore increased demand requires increased production, using the best varieties. There are 10 released varieties in Indonesia including Adira 1. This variety has been classified as an early variety. This variety has been cultivated in East Java and West Java, and can be harvested in 7 months. New varieties which are better than Adira 1 are being developed. If a new early variety is available which is higher yielding than Adira 1, the farmers will get benefits. Furthermore, management becomes easier in terms of scheduling planting time and harvesting time

Based on harvest time, cassava varieties can be grouped into three groups: early maturity (7-8 months), medium maturity (9-10 months), and late maturity (> 10 months).

Determination of harvesting time of cassava is different from the other crops like rice, maize, groundnut, soybean, and mungbean because it is a tuber crop. Harvest time of cassava is based on tuber yield and starch content of tuber. If any of the tuber yield and starch content of the tuber is relatively high at 7-8 months, that variety can be classified as early maturity. Usually the characteristic of the early maturity is the development of the tuber is earlier than the late maturity, and the growth rate at the beginning of growth of early maturity is faster than the late maturity.

RESEARCH ARTICLE

mailto:[email protected]�


170

Environmental conditions for cassava planting in Indonesia are variable. In general the condition of land is marginal with levels of fertility from low to medium. Sumatra area is relatively wet with rain distribution evenly spread through the year, West Java has medium rainfall, Central Java and East Java are relatively dry with clear differences between rainy season and dry season. Thus, the young phase of cassava is in rainy season and the mature phase is in dry season. On the contrary, in Sumatera, the rain is consistent through the year.

Basically the environment can be classified as predictable or unpredictable. Irrigation, fertilization, plant population, and planting method can be classified as predictable factors, while rainfall, humidity, and temperature can be classified as unpredictable factors. To respond to predictable environments, variety trials should be done in target environments which are similar to that of the main production areas of cassava. To respond to the unpredictable factors, the variety trial should be done on some environment during some planting season. Regarding variety trials, the important information that can be acquired is information about the stability of any genotype. There are a few techniques of analysis, one of them a technique based on additive models. This technique is focused on analysis of stability. This technique is based on regression of variety performance on environment index as was proposed and modified by Finlay and Wilkinson (1963), Eberhart and Russel (1966), Perkins and Jinks (1968), Freeman and Perkins (1971) and Shukla (1972). Gauch (1992) proposed a technique that is based on the model of AMMI (additive main effects and multiplicative interaction).

The aim of the trial was to test the performance of promising clones of cassava for early maturity in some locations over years.

MATERIALS AND METHODS

The experiments were done during two years (2007/2008 and 2009) in Lumajang (East Java), Pati (Central Java), Banyuwangi (East Java), Lampung Selatan (Lampung), Lampung Timur (Lampung), and dan Lampung Tengah

(Lampung). The experiments were conudcted using RCBD design, three replications. Tuber yield was analyzed using MSTAT (Michigan Statistic), version C software (released by Michigen State University) to obtain the combined analysis of variance. Stability analysis based on the technique of Eberhart and Russel (1966) was used. IRRISTAT (International Rice Research Institute Statistic) software, version 5.0 (released by International Rice Research Institute) was used to analyze the variance based on AMMI model and IPCA (interaction principal component analysis) score. IPCA score is Σnλnγgnδen; λn = the singular value for PCA axis n; γgn = the genotype eigenvector for axis n; δen = the environment eigenvector. The plot size was 5 m x 5 m. Plant distance was 100 cm x 80 cm. Doses of fertilizers were 93 kg N+ 36 kg P2O5 + 60 kg K2O/ha. The clones used were CMM 02048-6, CMM9904-100, SM 2361, MLG 10.333, MLG 10.312, CMM 97001-158 as promising clones, as well as Malang 1, Adira 1, and Adira 4 (released varieties) as checks. Clone CMM 02048-6, CMM9904-100, CMM 97001-158, MLG 10.333, MLG 10.312, CMM 97001-158, Malang 1, Adira 1, and Adira 4 are developed by ILETRI (Indonesian Legumes and Tuber Crops Research Institute), while SM 2361 is developed by CIAT. The variable recorded was fresh tuber yield (t/ha) of six-or seven-month-old plants. RESULTS

Tuber yields of clones at 7 months in Lumajang 2007/2008 ranged from 11.39 to 42.77 t/ha, with mean of 32.06 t/ha (Table 1). Tuber yield of CMM 97001-158 was the highest. Tuber yield of OMM9904-100 and SM 2361 were similar to CMM 97001-158. Tuber yields of clones at 7 months in Pati 2007/2008 ranged from 23.53 to 46.36 t/ha, with a mean of 38.30 t/ha. Tuber yields of SM 2361 was the highest. Tuber yields of clones in 7 months in Lampung Selatan 2007/2008 ranged from 16.54 to 27.61 t/ha (mean of 20.74 t/ha). Tuber yield of OMM9904-100 was the highest. Tuber yield range of clones at 7 months in Lampung Tengah 2007/2008 was 16.35 – 35.09 t/ha, with a mean of 25.87 t/ha.

Sholihin (2013)

171

Tuber yield of CMM 97001-158 was the highest. Tuber yields of OMM9904-100, MLG 10.333 and MLG 10312 were similar to CMM 97001-158. Tuber yields of clones in 6 months in Banyuwangi 2009 ranged from 22.83 to 58.68 t/ha, with mean 34.87 t/ha. Tuber yield of CMM 02048-6 was the highest. Tuber yields of clones in 7 months in Lumajang 2009 ranged from 20.56 to 36.82 t/ha, with a mean of 29.74 t/ha. Tuber yield of CMM 02048-6 was the highest. Tuber yields of MLG 10.312, SM 2361, and Adira 4 were similar to CMM 02048-6. Tuber yields of clones in 7 months in Pati 2009 ranged from 26,35 to 42.57 t/ha, with mean 33.80 t/ha. Tuber yield of CMM 02048-6 was the highest. Tuber yield of OMM9904-100 and MLG 10.333 were similar to CMM 02048-6. Tuber yield of clones in 7 months in Lampung Tengah 2009 ranged from 8.02 to 21.8 t/ha, with a mean of 15.74 t/ha. Tuber yield of Adira 4 was the highest. Tuber yield of Malang 1 was similar to Adira 1. The mean of the fresh tuber yield of OMM 02048-6 over locations and years was the highest (32.67 t/ha) and was significantly different from Adira 1.

The analysis of variance for the 9 clones, four locations, and two years for fresh tuber yield are shown in Table 1. The clone x location interaction were significantly different for fresh tuber yield. Because the clone x location interactions were significant, analysis of stability of clones should be done to assess stability and adaptability of clones. Stability analysis based on the technique of regression The range of fresh tuber yield, mean of the fresh tuber yield, mean square of standard deviation (Sdi

2), and coefficient of regression of cassava promising clones are shown in Table 2. The mean of the fresh tuber yield of OMM 02048-6 over locations and years was the highest (32.67 t/ha, 67% higher than Adira 1. Mean square of standard deviation (Sdi

2) of CMM9904-100, SM 2361, MLG 10.312, Malang 1, Adira 1, and Adira 4 were not significantly different from zero (0), but the others were significantly different from zero (0). A coefficient of regression (bi) of clones/varieties tested were

not significantly different from one (1). So, CMM9904-100, SM 2361, MLG 10.312, Malang 1, Adira 1, and Adira 4 were stable clones, but the others were not stable clones. AMMI analysis Analysis of variance based on AMMI models for fresh tuber yield presented at Table 3. It can be seen from Table 3 that the effect of clone x location interaction differed significantly at 1%, with AMMI models, source of variance of clone x location interaction can be divided into some components, i.e. IPCA 1, IPCA 2 and IPCA 3, and IPCA 1 and IPCA 2 were significantly different, while IPCA 3 was not significantly different. Fifty one percent of interaction sum of squares was contributed by IPCA 1, 34% by IPCA 2, 12% by IPCA 3 and 10% by IPCA4.

Biplot of IPCA 1 and mean of tuber yield was presented in Figure 1. The tuber yield of CMM 02048-6 was the highest, while that of Adira 1 was lowest. Based on this figure, it can be determined that a clone, which was on one point on the horizontal axis, had the same main effect (tuber yield), and a clone which was on one point on the vertical axis had the same interaction effect. The tuber yields of clone 4 (MLG 10.333) and 5 (MLG 10.312) were similar, but the interaction effect with location was different. Clone 5 (MLG 10.312) had positive interaction with F location, while clone 4 (MLG 10.333) was negative interaction with F location.

IPCA scores for four locations and mean of starch yield are given in Table 4. Biplots IPCA 1 and IPCA 2 for environment based on tuber yield are given in Figure 2. Based on this figure, it can be seen that locations used was good enough. Position of A (Lumajang, 07/08) was far from B (Lampung S., 07/08), C (Lampung T., 07/08), D (Pati, 07/08), E (Lumajang, 09), F (Banyuwangi, 09), G (Lampung S.) and H (Lampung T., 09) in Figure 1. This meant the environments were varied.

IPCA scores for 9 clones and tuber yield were presented in Table 5. The average of tuber yield of CMM 02048-6 was the highest, followed by CMM 9904-100, SM 2361, and


172

Adira 4. Clone CMM 02048-6 and Adira 1 had low scores by IPCA 1 (-1.987 and -2.541).

Biplots of IPCA1 and IPCA2 for clones based on tuber yield are given in Figure 3. Based on this figure, it can identify the stability of a clone. Stable clone was the clone that is near point (0, 0). Clones 9 (Adira 4), 2 (CMM 9904-100), 3 (SM 23 61) and 5 (MLG 10312) were more stable than the others.

The map of the best clone over the range of AMM-2 environment scores is given in Figure 4. Table 4 and Figure 4 can be used to determine the best and adaptive clone for a target area. In the environment with score of IPCA 1 less than 0. 91 and score of IPCA 2 more than 0.61, clone CMM 02048-6 was the best and most adaptive clone.

Table 1. Fresh tuber yield (t/ha) of cassava clones/varieties in eight environment, 2007/2008 and 2009. Clone/variety Tuber yield (t/ha)

Lumajang 2007/08

Pati 2007/08

Lampung Selatan 2007/08

Lampung Tengah 2007/08

Banyuwangi 2009

Lumajang 2009

Pati 2009

Lampung Tengah 2009

Rata-rata

CMM 02048-6 OMM9904-100 SM 2361 MLG 10.333 MLG 10.312 CMM 97001-158 Malang 1 Adira 1 Adira 4

31.43 bcd 38.73 ab 38.37 abc 30.80 d 33.37 bcd 42.77 a 31.07 cd 11.39 e 30.63 d

36.07 d 44.16 ab 46.35 a 37.60 cd 37.26 cd 43.93 abc 37.50 cd 23.53 e 39.33 bcd

16.54 d 27.61 a 22.47 b 20.35 bc 22.27 b 19.20 bcd 19.35 bcd 17.41 cd 21.43 b

26.42 bc 29.19 ab 19.38 cd 29.08 ab 28.68 ab 35.09 a 22.40 bcd 16.35 d 26.25 bc

53.68 a 34.72 b 39.53 b 27.13 c 38.74 b 34.78 b 22.83 c 27.78 c 34.65 b

36.82 a 30.95 bc 33.28 ab 24.39 de 31.72 ab 29.81 bc 25.99 cd 20.56 e 34.10 ab

42.57 ab 37.27 ab 34.55 bc 44.53 a 26.77 cd 24.22 d 34.17 bc 26.35 cd 33.75 bcd

17.82 b 17.34 b 13.93 c 17.19 b 13.37 c 8.02 d 19.14 ab 12.93 c 21.88 a

32.67 a 32.50 a 30.98 ab 28.88 b 29.02 b 29.60 b 26.55 c 19.54 d 30.25 b

Mean 32.06 38.30 20.74 25.87 34.87 29.74 33.80 15.74 C.V. (%) 12.21 8.99 9.12 15.32 10.54 9.82 15.12 10.93 Note: Numbers in the same columns with the same letters are not significantly different at the 5% level. Table 2. Combined ANOVA for 9 cassava clones, 4 locations and 2 years for tuber yield. Source Degrees of freedom Mean squares Environment(E) 7 1573.392** Error (a) 16 14.34 Clones (C) 8 380.862**

C x E 56 91.390** Error (b) 128 12.199 Coefficient Variation (%) 12.09 ** = 1 % significantly different.

Sholihin (2013)

173

Table 3. Mean values of the fresh tuber yield, mean square of standard deviation, and coefficient of regression of cassava promising clones.

Clone/ Variety

Mean of fresh tuber yield (t/ha)

Coefficient of regression

Mean square of standard deviation

CMM 02048-6 32.67 1.352 ns 45.092 ** CMM 9904-100 32.50 1.015 ns -4.051 ns SM 2361 30.98 1.391 ns 2.869 ns MLG 10.333 28.88 0.916 ns 23.721 ** MLG 10312 29.02 0.987 ns 2.252 ns CMM 97001-158 29.60 1.285 ns 44.039 ** Malang1 26.55 0.737 ns 6.926 ns Adira1 19.54 0.513 ns 12.591 ns Adira 4 30.25 0.803 ns -7.228 ns ns = not significant; ** = significantly different at 1 %. Table 4. Analysis of variance based on AMMI model for tuber yield. Source of variance Degrees of freedom Mean squares Location (L) Error Clone (C) C x L IPCA1 IPCA2 IPCA3 IPCA4 Combined error

7 16 8 56 14 12 10 8 128

506829848** 2672609 27432162** 4410850 49.5818* 48.0892** 21.0310 20.6131** 12.199

**, * = significantly different at 1% and 5%, respectively Table 5. IPCA score for locations and mean of fresh tuber yield in 7 months. Location Fresh tuber yield

t/ha IPCA1 IPCA2

A. Lumajang,Inceptisol, 110 m above sea level, 2007/2008 B. Pati, 5 m above sea level, 2007/2008 C. Lampung Tengah, 58 m above sea level, 2007/2008 D. Lampung Selatan, 135 m sea level, 2007/2008 E. Banyuwangi,Entisol, 168 above m sea leavel, 2009 F. Lumajang,Inceptisol, 110 m above sea level, 2009 G. Pati, 5 m above sea level, 2009 H. Lampung Tengah, 58 m above sea level, 2009

32.06 38.30 25.87 20.74 34.87 29.74 33.80 15.74

3.140 1.751 -0.119 1.123 -1.019 -0.196 -2.582 -2.098

-0.959 -0.651 -1.305 -0.551 4.043 1.258 1.258 -1.433


174

Table 6. IPCA scores for clones and tuber yield in 6/7 months. Clones Tuber yield (kg/ha) IPCA1 IPCA2

CMM 02048-6 CMM 9904-100 SM 2361 MLG 10.333 MLG 10312 CMM 97001-158 Malang1 Adira1 Adira 4

32,67 32,50 30,98 28,88 29,02 29,60 26,55 19,54 30,25

-1.987 0.823 0.865 -0.887 0.911 3.539 -0.127 -2.541 -0.597

3.229 -0.875 0.850 -2.264 1.142 0.657 -2.266 -0.234 -0.241

Figure 1. Biplot of IPCA 1 and fresh tuber yield in 7 months. 1: CMM 02048-6; 2: CMM 9904-100; 3: SM 2361; 4: MLG 10.333; 5: MLG 10.312; 6: CMM 97001-158; 7: Malang 1; 8: Adira 1; 9: Adira 4; A: Lumajang, 07/08; B: Lampung S., 07/08; C: Lampung T., 07/08; D: Pati, 07/08; E: Lumajang, 09; F: Banyuwangi, 09; G: Lampung S., 09; H: Lampung T., 09.

MEANS3934.229.424.619.815

IPCA

1

3.6

2.36

1.12

-0.12

-1.36

-2.6

01

0203

04

05

06

07

08

09

A

B

C

D

E

F

G

H

Sholihin (2013)

175

Figure 2. Biplot of IPCA 1 and IPCA 2 for environment based on fresh tuber yield in 7 months. A: Lumajang, 07/08; B: Lampung S., 07/08; C: Lampung T., 07/08; D: Pati, 07/08; E: Lumajang, 09; F: Banyuwangi, 09; G: Lampung S., 09; H: Lampung T., 09.

Figure 3. Biplot of IPCA 1 and IPCA 2 for clones based on fresh tuber yield in 7 months. 1: CMM 02048-6; 2: CMM 9904-100; 3: SM 2361; 4: MLG 10.333; 5: MLG 10.312; 6: CMM 97001-158; 7: Malang 1; 8: Adira 1; 9: Adira 4; A: Lumajang, 07/08; B: Lampung S., 07/08; C: Lampung T., 07/08; D: Pati, 07/08; E: Lumajang, 09; F: Banyuwangi, 09; G: Lampung S., 09; H: Lampung T., 09.

3 C O SCO S O O

IPCA13.62.361.12-0.12-1.36-2.6

IPCA

32.4

1.48

0.56

-0.36

-1.28

AB

CD

E

F

G

H

IPCA13.62.361.12-0.12-1.36-2.6

IPCA

2

4.1

2.82

1.54

0.26

-1.02

-2.3

01

02

03

04

05

06

07

08 09 A

B

C

D

E

F

G H


176

IPCA 2 3.95 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010606 3.86 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101060606 3.76 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101060606 3.67 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010106060606 3.58 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010106060606 3.49 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010106060606 3.39 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010606060606 3.30 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010606060606 3.21 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101060606060606 3.11 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101060606060606 3.02 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010106060606060606 2.93 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010106060606060606 2.84 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010606060606060606 2.74 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010606060606060606 2.65 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010606060606060606 2.56 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101060606060606060606 2.47 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101060606060606060606 2.37 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010106060606060606060606 2.28 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010106060606060606060606 2.19 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010606060606060606060606 2.09 010101010101010101010101010101010101010101010101010101010101010101010101010101010101010606060606060606060606 2.00 010101010101010101010101010101010101010101010101010101010101010101010101010101010101060606060606060606060606 1.91 010101010101010101010101010101010101010101010101010101010101010101010101010101010101060606060606060606060606 1.82 010101010101010101010101010101010101010101010101010101010101010101010101010101010101060606060606060606060606 1.72 010101010101010101010101010101010101010101010101010101010101010101010101010101010106060606060606060606060606 1.63 010101010101010101010101010101010101010101010101010101010101010101010101010101010106060606060606060606060606 1.54 010101010101010101010101010101010101010101010101010101010101010101010101010101010606060606060606060606060606 1.44 010101010101010101010101010101010101010101010101010101010101010101010101010101010606060606060606060606060606 1.35 010101010101010101010101010101010101010101010101010101010101010101010101010101060606060606060606060606060606 1.26 010101010101010101010101010101010101010101010101010101010101010101010101010101060606060606060606060606060606 1.17 010101010101010101010101010101010101010101010101010101010101010101010101010106060606060606060606060606060606 1.07 010101010101010101010101010101010101010101010101010101010101010101010101010106060606060606060606060606060606 0.98 010101010101010101010101010101010101010101010101010101010101010101010101010106060606060606060606060606060606 0.89 010101010101010101010101010101010101010101010101010101010101010101010101010606060606060606060606060606060606 0.79 010101010101010101010101010101010101010101010101010101010101010101010101010606060606060606060606060606060606 0.70 010101010101010101010101010101010101010101010101010101010101010101010101060606060606060606060606060606060606 0.61 010101010101010101010101010101010101010101010101010101010101010101010101060606060606060606060606060606060606 0.52 010101010101010101010101010101010101010101010101010101010101010101010106060606060606060606060606060606060606 0.42 010101010101010101010101010101010101010101010101010101010101010101010202060606060606060606060606060606060606 0.33 010101010101010101010101010101010101010101010101010101010101010101020202060606060606060606060606060606060606 0.24 010101010101010101010101010101010101010101010101010101010101010202020202020606060606060606060606060606060606 0.14 010101010101010101010101010101010101010101010101010101010101020202020202020606060606060606060606060606060606 0.05 010101010101010101010101010101010101010101010101010101010102020202020202020206060606060606060606060606060606 -0.04 010101010101010101010101010101010101010101010101010101020202020202020202020206060606060606060606060606060606 -0.13 010101010101010101010101010101010101010101010101010102020202020202020202020202060606060606060606060606060606 -0.23 010101010101010101010101010101010101010101010101020202020202020202020202020202060606060606060606060606060606 -0.32 010101010101010101010101010101010101010101010102020202020202020202020202020202020606060606060606060606060606 -0.41 010101010101010101010101010101010101010101010202020202020202020202020202020202020206060606060606060606060606 -0.51 010101010101010101010101010101010101010102020202020202020202020202020202020202020206060606060606060606060606 -0.60 010101010101010101010101010101010101010202020202020202020202020202020202020202020202060606060606060606060606 -0.69 010101010101010101010101010101010102020202020202020202020202020202020202020202020202060606060606060606060606 -0.78 010101010101010101010101010101010202020202020202020202020202020202020202020202020202020606060606060606060606 -0.88 010101010101010101010101010101020202020202020202020202020202020202020202020202020202020606060606060606060606 -0.97 010101010101010101010101010202020202020202020202020202020202020202020202020202020202020206060606060606060606 -1.06 010101010101010104040404040402020202020202020202020202020202020202020202020202020202020206060606060606060606 -1.15 010101040404040404040404040404020202020202020202020202020202020202020202020202020202020202060606060606060606 -1.25 040404040404040404040404040404040202020202020202020202020202020202020202020202020202020202060606060606060606 -1.34 040404040404040404040404040404040402020202020202020202020202020202020202020202020202020202020606060606060606 -1.43 040404040404040404040404040404040402020202020202020202020202020202020202020202020202020202020606060606060606 ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ -2.58 -2.0 -1.42 -0.836 -0.254 0.328 0.910 1.49 2.07 .2.66

IPCA 1 Figure 4. Map of the best clone over the range of AMM-2 environment scores.

Sholihin (2013)

177

DISCUSSION The results demonstrate that interaction of clones and locations was significantly different for fresh tuber yield at 6 or 7 months. Genotype x environment interactions can produce phenotypes that are highly variable across locations. This phenomenon was also reported by Sholihin (2009) and (2011), Sundari et al. (2010) and Kalkani and Sharma (2010).

The mean of the fresh tuber yield of OMM 02048-6 over locations and years was the highest (32.67 t/ha, 67% higher than Adira 1, equal to Rp 9.191.000/ha or around US$ 10132, at US$1 = Rp 9000). However, based on the technique developed by Eberhart and Russel (1966) and the AMMI model developed by Gauch (1992), CMM 02048-6 clone was unstable based on AMMI analysis. This clone has specific adaptation, and this clone was adaptive to environment with score of IPCA 1 of less than 0 (0.91), and score of IPCA 2 more than 0.61. The environments E (Banyuwangi 2009), F (Lumajang 2009) and G (Pati 2009), are located in Java Island. In East Java and Central Java, cassava is planted during climate types C2, C3 and D3 in which there are 2-6 dry months of <100 mm rainfall with alfisols, entisols, and inceptisols soils.

Information on HCN content is important in determining the appropriate use of cassava. Cassava varieties with HCN content >50 ppm is not suitable for fried or steamed cassava. Sholihin et al. (2011) reported that HCN content of CMM 02048-6 was 27.8 ppm meaning that this variety can be used for fried or steam cassava. In addition, the β-carotene content of this promising clone was 791 ug/100 g dry weight (Anonymous, 2011). In terms of resistance to mites, CMM 02048-6 was resistant to tuber mite (Anonymous, 2011).

Based on Table 3, it can be determined that clones CMM9904-100, SM 2361, MLG 10.312, Malang 1, Adira 1, and Adira 4 were stable clones, but the others were not. There are two possibilities in which cassava can act as a buffer to varying environmental conditions. The first is that a clone is a hybrid. The second is that it has genetic potential to perform well, irrespective of the environment where they are

grown. There is a correlation between tuber yield and starch yield because starch yield is related to fresh tuber yield and starch content. Sholihin (2011) reported that environmental factors which are important in determining stability of cassava clone/variety based on the starch yield in 6 months were bulk density of subsoil, soil pH of topsoil, and maximum air humidity 4 months after planting. REFERENCES Anonymous (2011). Perbaikan proposal usulan

pelepasan varietas ubikayu klon harapan CMM 02048-6, hasil tinggi dan umur genjah. [in Bahasa Indonesia] (Unpublished).

Eberhart SA Russel WA (1966). Stability parameters for comparing varieties. Crop Sci. 6:36-40.

Finlay KW Wilkinson GN (1963). The analysis of adaptation in a plant-breeding programme. Aust. J. Agric. Res. 14:742-754.

Freeman GH Perkins JM (1971). Environmental and genotype-environmental components of variability. VIII. Relations between genotypes grown in different environment and measure of these environments. Heredity 27:15-23.

Gauch HG (1992). Statistical analysis of regional yield trial: AMMI analysis of factorial designs. Elsevier Science Publishers. Amsterdam, Netherlands..

Kakani RK, Sharma Y (2010). Genetic component analysis for yield and yield contributing traits under diverse environments in barley. SABRAO J Breed. Genet. 42(1):9-20.

Perkins JM, Jinks JL (1968). Environmental and genotype environmental components of variability III. Multiple lines and crosses. Heredity. 23:339-356.

Shukla GK (1972). Some statistical aspects of partitioning genotype-environmental components of variability. Heredity 29:237-245.

Sholihin (2011). AMMI model for interpreting clone-environment interaction in starch yield of cassava. HAYATI J. Biosci. 18(1):21-26.

Sholihin (2009). The genotypes x environment interaction for starch yield in nine-month old cassava promising clones. Indonesian J. Agric. Sci. 10(1):12-18.

Sholihin, Sundari T, Ginting E (2011). Keragaan


178

klon-klon harapan ubikayu. In: Adie et al. (Ed.). Inovasi Teknologi Untuk Pengembangan Kedelai Menuju Swasembada, Prosiding Seminar Nasional Hasil Penelitian Tanaman aneka kacang dan Umbi. Pusat Penelitian dan Pengembangan Tanaman Pangan, Bogor. [in Bahasa

Indonesia] pp. 540-549. Sundari TK, Noerwijati IMJ Mejaya (2010).

Hubungan antara komponen hasil dan hasil umbi klon harapan ubikayu. Jurnal Penelitian Pertanian Tanaman Pangan. [in Bahasa Indonesia] 29(1):29-35.


PHOTOPERIOD-INSENSITIVE MUTANTS WITH SHORTER PLANT HEIGHT IDENTIFIED IN THE M1 GENERATION OF RICE IRRADIATED

WITH CARBON ION BEAMS

M.A.K. AZAD1*, M.N.N. MAZUMDAR1, A.K. CHAKI1, M. ALI1, M.L. HAKIM2, A.N.K. MAMUN2

, Y. HASE3, S. NOZWA3, A. TANAKA3, A. KOIKE4, H. ISHIKAWA4

and M.A. AZAM1

1Plant Breeding Division, Bangladesh Institute of Nuclear Agriculture, Bangladesh Agricultural University Campus, Mymensingh 2202, Bangladesh

2Biotechnology and Genetic Engineering Division, Institute of Food and Radiation Biology, Atomic Energy Research Establishment, Bangladesh Atomic Energy Commission, Savar, Dhaka, Bangladesh

3Quantum Beam Science Directorate, Japan Atomic Energy Agency, Takasaki, Gunma, Japan 4Nuclear Safety Research Association, Tokyo, Japan

*Corresponding author’s email: [email protected]

SUMMARY To investigate the possibility of inducing fixed mutants in M1 generation of rice, 150 seeds of the highly photoperiod-sensitive and tall indica type rice cv. Ashfal were irradiated with different doses of carbon ion beams from the Radiation Applied Biology Division, Quantum Beam Science Division, Japan Atomic Energy Agency. M1 was grown during boro (December-June) season and M2 during shorter days in aman season (July–December) in 2009, M3 was grown during the 2010 boro season. Nine M1 hills, 2 from each of 40, 120 and 160 Gy, and 1 from each of 60, 80 and 200 Gy headed, even under long days in boro season. None of the plants of parent Ashfal headed. Of the heading M1 plants that had plant height as the parent and that could not produce any seed, only the shorter plants produced seed. The M2 progenies took 85 days and 110 days for heading and maturity, respectively, during shorter days in aman season. In contrast, the parent Ashfal headed after 7 November and matured in 152 days. In M3 generation, the progenies bred true as in M2 for heading, maturity and plant height despite the fact that the number of Days taken were a little bit higher than M2. Simple sequence repeat analysis confirmed that the photoperiod-insensitive M1 plants that bred true in the M2 and M3 generations were not contamination of another photoperiod-insensitive variety as monomorphic bands were exhibited for the primer RM351 among the parent and the mutants. It was further confirmed by the fact that the parent Ashfal and the mutants had similar 1000-grain weight. Finally, it could be concluded that it is possible to induce fixed photoperiod-insensitive mutants with shorter plant height with 40- 200 Gy doses of carbon ion beams in M1 of indica type rice cv. Ashfal.

Keywords: Rice, indica type, ion beams, mutation, photoperiod sensitivity, fixed M1 mutant, simple Sequence repeat

Manuscript received: March 6, 2012; Decision on manuscript: June 6, 2013; Manuscript accepted: January 8, 2013

© Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Rice (Oryza sativa L.) is a short-day plant (Vergara and Chang, 1985). Almost all rice

cultivars mature in a shorter time when they were grown under a short photoperiod (about 10 h) than under a long photoperiod (about 14 h). The degree of photoperiod sensitivity greatly

RESEARCH ARTICLE



180

varies among cultivars. In Bangladesh, rice is being grown in three seasons in a year; aman season (July to December), boro season (December to June) and aus season (March to June). Ashfal is a local T. aman (July to December) cultivar being cultivated in the coastal southwestern part of Bangladesh since time immemorial. It is a tall highly photoperiod-sensitive, long-duration cultuvar. It is prone to lodging but tolerant of salinity and submergence. Its grains are medium-bold and contain 10.3% protein and 28.6% amylose. Ashfal cannot be grown in boro (long day) season as it is highly photoperiod-sensitive. It often lodges even in the aman season due to its very long culms, thus causing yield loss to a considerable extent.

Recently, ion beam irradiation has been utilized as an effective method of inducing mutations. The biological effects of ion beams have been investigated and observed to have shown high relative biological effectiveness (RBE) in lethality, mutation and so on, compared with low linear energy transfer (LET) radiation like Gamma-rays, X-rays and electrons (Blakely, 1992). It has been also demonstrated that ion beams induce mutations at high frequency and induce novel mutants in Arabidopsis (Hase et al. 2000, Shikazono et al. 2003, Tanaka et al. 1997, 2002). The use of ion beams for inducing mutations in rice (Oryza sativa L.) breeding has also been attempted (Hayashi et al. 2007; Hidema et al. 2003; Rakwal et al. 2008). It is generally accepted that, in mutation breeding, induced traits become fixed during M2-M4 generations (Azad et al., 2010; Hamid et al., 2006; Shamsuzzaman et al., 1998; Azam and Uddin, 1999). But it has been reported that it is possible to isolate fixed mutants, even in the M1 generation of heavy ion-irradiated sweet pepper (Honda et al., 2006).

In this study, we aimed to elucidate whether it is possible to get fixed photoperiod-insensitive M1 plants with shorter height and duration from Ashfal by ion beam irradiation.

MATERIALS AND METHODS Dehusked seeds of Ashfal were exposed to 26.7 MeV/n carbon ions with doses of 0, 10, 20, 40, 60, 80, 100, 120, 160 and 200 Gy in January 2009 at the Japan Atomic Energy Agency (Takasaki, Gunma, Japan). One hundred and fifty seeds were used for each irradiation dose. The irradiated seeds were allowed to germinate in petri dishes. Seedlings were transplanted on February 11, 2009. Single seedlings were transplanted into each hill. Spacing between rows and hills were 20 cm and 15 cm, respectively. All plants were grown under natural conditions at the Bangladesh Institute of Nuclear Agriculture (BINA) headquarters’ farm (Mymensingh, Bangladesh). Daylength at the time of heading in boro (December to June) season is between 12 and 13 h while that in aman season (July-December) is 10 ~ 11 h.

Fertilizers were applied at a rate of 90 kg N, 18 kg P, 63 kg K, 10 kg S and 1.6 kg Zn per hectare in the forms of urea, triple superphosphate, muriate of potash, gypsum and zinc sulphate equally in all plots. All fertilizers except ureas, were applied during final land preparation. Urea was applied at three equal installments at 7, 30 and 55 days after transplanting. Hand weeding was done three times and the plots were kept saturated with irrigation water till maturity.

Heights of 10 randomly chosen M1 plants were measured for each irradiation dose at 3 weeks after transplanting. M2 seeds were harvested individually from three fertile M1 plants at 5 months after transplanting. The M2 seeds were sown on a seedbed on July 16, 2009. Surviving seedlings were transplanted on August 17, 2009 and they were grown during aman season. Days to heading was recorded as the number of days required from sowing to the time when 50% of plants of each line headed. Days to maturity was recorded as the number of days required from sowing to the time when 90% of plants of each line appeared with yellowish grains. These two data were recorded through visual observation by visiting the plots every other day. Plant height, number of effective tillers per plant and panicle length were measured at maturity from 5 randomly selected competitive plants. M3 seeds were harvested

Azad et al (2013)

181

separately from each M2 plant. M3 plants were grown as plant progeny rows during next boro (long-day) season (December 2009 to June 2010). Data recording was the same as that for M2. Grain characteristics like length (L) and width (W) of unhusked and husked grains were recorded in October 2012 from 25 seeds of each of the eight mutants (as shown in SSR analysis) along with the parent Ashfal while grain weight from 1000 unhusked seeds are presented in Table 3.

To confirm that the plants that showed photoperiod insensitivity in M1, M2 , M3 and onward generations were not contamination of other photoperiod-insensitive varieties, simple sequence repeat (SSR) analysis of DNA was performed during September-October 2012 of these eight mutants: RM(1)-160 (C)-1, RM(1)-200 (C)-1-1, RM(1)-200 (C)-1-3, RM(1)-200 (C)-1-9, RM(1)-200 (C)-1-10, RM(1)-200 (C)-1-13, RM(1)-200 (C)-1-17, RM(1)-200 (C)-1-18 together with the parent Ashfal. The mutants derived from 140 Gy dose were excluded previously and could not be included.

DNA isolation Healthy leaves from 21-day-old seedlings of the mutants and parent Ashfal were cut into 2-3 cm pieces. The modified cetyl trimethyl ammonium bromide (CTAB) mini-prep method was used to extract DNA. Before PCR amplification, it is important to optimize the amount of DNA to achieve reproducibility and strong signal in PCR assay. Spectrophotometry was used to quantify of DNA concentration. Amplification of SSR markers by polymerase chain reaction (PCR) The primer RM351 used previously by many workers (Salil et al., 2007; Islam, 2004; Bonilla et al., 2002; Niones, 2004 and Gregorio et al., 2002) for tagging different traits in rice was used in this study. The PCR reaction conducted in a volume of 15.0 µl reaction mixture including 2 µl DNA, 8.25 µl of sterilized ddH2O, 1.5 µl of 10 X buffer, 0.75 µl of dNTPs, 1.0 µl of forward and reverse primers, 0.5 µl of Taq polymerase. Amplification reaction consists of preheating for 5 min at 94 oC and of 35 cycles of 1 min at 94 oC

(denaturation), 1 min at 55 oC (primer annealing) and 2 min at 72 oC (primer extension) followed by 7 min incubation at 72 oC. The products were separated on 8% polyacrylamide gels. The molecular weight markers used were 100 bp ladder. The gels were viewed by the GelDoc software. RESULTS Seeds of Ashfal irradiated with carbon ions were grown during boro (long day) season in 2009. Survival rate and plant height were determined to examine the effect of carbon ion irradiation on plant growth. Survival rate of the irradiated seeds were significantly decreased at 60 Gy and higher doses (Table 1). Seedling height also gradually decreased as the irradiation dose increased. The height of seedlings irradiated with 80 Gy and higher doses were around half that of non-irradiated plants. Since Ashfal is a highly photoperiod-sensitive cultivar, none of the control plants headed during boro season. However, nine M1 plants headed under the same conditions (Table 1).

In order to confirm that photoperiod sensitivity was genetically altered in the fertile M1 plants, M2 seeds harvested from them were grown during next aman (short day) season in 2009. Grown in the seedbed were 301, 650 and 246 M2 seeds, which were derived from 40, 160 and 200 Gy irradiation, respectively. Germination rate was very low probably due to seed dormancy. Of these 9, 126 and 18 M2 seedlings were transplanted. Ashfal headed around 122 days and matured around 152 days after sowing. In contrast, M2 progenies derived from 40 and 200 Gy took around 85 and 110 days to heading and maturity, respectively. The number of effective tillers per plant was 13 in Ashfal, while it ranged from 4 to 13 in M2 progenies. Panicle lengths were similar to that of Ashfal in all M2 progenies. These results suggest that the photoperiod-insensitive phenotype of three fertile M1 plants was inherited by all M2 progenies without segregation.


182

Table 1. Effects of carbon ion irradiation on survival rate, seedling height and heading in highly photo-period-sensitive rice cultivar Ashfal grown in boro season. Dose (Gy)

Number of seeds sown

Number of surviving plants

Survival rate (%)

Seedling height* (cm)

Number of headed plants

Number of fertile plants

0 150 147 98.0 23.5 ± 0.4 0 − 10 150 122 81.3 23.0 ± 0.7 0 − 20 150 141 94.0 21.7 ± 0.4 0 − 40 150 121 80.7 19.2 ± 0.8 2 1 60 150 94 66.0 15.5 ± 0.2 1 0 80 150 15 10.0 13.5 ± 1.5 1 0 100 150 9 6.0 12.5 ± 0.5 0 − 120 150 6 4.0 11.6 ± 1.3 2 0 160 150 11 7.3 12.9 ± 1.1 2 1 200 150 3 2.0 12.3 ± 2.0 1 1

* Mean ± SE of seedling heights at 3 weeks after transplanting

To further confirm the photoperiod sensitivity of mutant lines, M3 plants derived from each M2 plant were grown during the next boro (long day) season in 2010. While none of parental Ashfal headed in this season, all mutant lines headed around 90 days after sowing and matured around 119 days after sowing. Days to heading and maturity were slightly longer than the M2 generation grown during aman season. As was observed in M2 generation, plant heights of mutant lines were markedly shorter than that of Ashfal (Figure 1). The mutant lines from 40 Gy were slightly shorter in plant height than mutant lines from 200 Gy. This was consistent with the plant height of each mutant line observed in M2 generation.

Length (L) and width (W) of unhusked and husked grains and their ratio differed significantly in the mutants from those of the parent Ashfal (Table 3). Length of unhusked and husked grain increased significantly in all the

mutants, except RM(1)-160(C)-1, where it decreased significantly. In contrast, width of unhusked and husked grain decreased significantly in all mutants while ratios of length and width (L/W) of unhusked and husked grain increased significantly. Finally, 1000-grain weight remained unchanged in most of the mutants, except for 3. SSR analysis revealed monomorphic bands for the parent Ashfal and the mutants derived from it (Figure 2).

Azad et al (2013)

183

Figure 1. Photoperiod-sensitive cultivar ‘Ashfal’ (right) and M3 progenies of photoperiod-insensitive mutant (left) grown in boro (long day) season, 2010.

100bp 1 2 3 4 5 6 7 8 9

Figure 2. Amplified DNA of Ashfal and 8 mutants derived from irradiating the seeds of Ashfal with carbon ion beams with primer RM351 after resolution in 8% polyacrylamide gel. Lane 1 = Ashfal, 2 = RM(1)-160 (C)-1, 3 = RM(1)-200 (C)-1-1, 4 = RM(1)-200 (C)-1-3, 5 = RM(1)-200 (C)-1-9, 6 = RM(1)-200 (C)-1-10, 7 = RM(1)-200 (C)-1-13, 8 = RM(1)-200 (C)-1-17, 9 = RM(1)-200 (C)-1-18.


184

Table 2. Characteristics of M2 progenies grown in aman season and M3 progenies grown in boro season.

Mutant line

M2 generation grown in aman season, 2009 M3 generation

grown in boro season, 2010 Plant height (cm)

Effective tillers/plant

(no.)

Panicle length (cm)

Plant

height* (cm)

Effective tillers/plant*

(no.)

Panicle length*

(cm) RM(1)-40(C)-1-1 75 10 21 75.6 ± 1.0 10.0 ± 0.6 22.6 ± 0.8 RM(1)-40(C)-1-2 73 11 23 77.8 ± 1.7 8.2 ± 0.7 22.8 ± 0.5 RM(1)-40(C)-1-3 77 5 23 79.0 ± 0.7 9.0 ± 0.6 22.4 ± 0.5 RM(1)-40(C)-1-4 73 10 19 77.0 ± 1.8 8.4 ± 1.0 22.8 ± 0.6 RM(1)-40(C)-1-5 76 7 24 76.8 ± 2.0 9.2 ± 0.7 22.8 ± 0.2 RM(1)-40(C)-1-6 72 7 21 74.0 ± 1.3 9.0 ± 0.8 22.8 ± 0.5 RM(1)-40(C)-1-7 74 7 21 81.0 ± 1.7 11.8 ± 0.6 23.0 ± 1.7 RM(1)-40(C)-1-8 81 6 22 − − − RM(1)-40(C)-1-9 91 13 28 83.4 ± 5.1 11.6 ± 1.4 26.2 ± 1.6 RM(1)-200(C)-1-1 68 13 26 70.6 ± 2.0 9.2 ± 1.1 23.0 ± 0.4 RM(1)-200(C)-1-2 64 10 22 65.9 ± 0.7 6.0± 0.7 21.6 ± 0.7 RM(1)-200(C)-1-3 68 6 23 67.4 ± 2.4 7.8 ± 0.9 23.2 ± 0.7 RM(1)-200(C)-1-4 67 8 22 69.4 ± 4.8 8.6 ± 0.9 24.6 ± 1.4 RM(1)-200(C)-1-5 66 5 22 66.0 ± 1.4 6.2 ± 0.5 22.6 ± 0.9 RM(1)-200(C)-1-6 69 8 21 67.8 ± 1.2 7.8 ± 0.6 22.4 ± 0.7 RM(1)-200(C)-1-7 61 6 20 69.8 ± 1.6 6.6 ± 0.7 22.4 ± 0.5 RM(1)-200(C)-1-8 62 9 22 67.6 ± 2.0 8.6 ± 0.5 22.6 ± 0.4 RM(1)-200(C)-1-9 71 7 22 70.2 ± 0.7 7.6 ± 0.4 22.0 ± 0.4 RM(1)-200(C)-1-10 56 5 19 69.4 ± 1.6 9.8 ± 0.6 22.6 ± 0.5 RM(1)-200(C)-1-11 58 7 22 69.4 ± 1.1 7.4 ± 0.2 22.2 ± 0.5 RM(1)-200(C)-1-12 64 6 21 69.2 ± 1.2 9.6 ± 1.7 21.8 ± 0.7 RM(1)-200(C)-1-13 75 6 23 69.4 ± 1.0 11.0 ± 0.9 22.6 ± 0.2 RM(1)-200(C)-1-14 66 9 21 69.6 ± 1.0 8.8 ± 0.5 23.2 ± 0.9 RM(1)-200(C)-1-15 63 7 21 67.0 ± 0.8 8.4 ± 0.9 22.6 ± 0.8 RM(1)-200(C)-1-16 56 4 16 70.6 ± 1.9 10.6 ± 1.0 22.6 ± 0.7 RM(1)-200(C)-1-17 61 8 21 65.0 ± 1.5 10.2 ± 0.8 22.8 ± 0.5 RM(1)-200(C)-1-18 48 4 20 70.6 ± 2.0 9.2 ± 1.1 23.0 ± 0.4 Ashfal (parent) 148 13 24 − − −

* Mean ± SE of 5 plants, RM indicates rice mutant; 40 (C) and 200 (C) indicate irradiation with 40 and 200 Gy doses of carbon ion beams.

Table 3. Grain characteristics of eight mutants along with parent Ashfal. Mutant/variety Grain length, (L)

(mm) Grain breadth (W)

(mm) L: W 1000-grain weight (g)

RM(1)-200 (C)-1-1 9.4 (7.0) 2.4 (2.0) 3.92 (3.50) 24.0 RM(1)-200 (C)-1-3 8.9 (7.0) 2.4 (2.0) 3.71(3.50) 23.9 RM(1)-200 (C)-1-9 9.1 (7.0) 2.3 (2.2) 3.96 (3.18) 22.5 RM(1)-200 (C)-1-10 9.4 (7.2) 2.4 (2.0) 3.92 (3.60) 24.0 RM(1)-200 (C)-1-13 8.5 (6.8) 2.3 (2.0) 3.70 (3.40) 23.8 RM(1)-200 (C)-1-17 9.3 (7.0) 2.4 (2.0) 3.88 (3.50) 24.2 RM(1)-200 (C)-1-18 9.4 (7.0) 2.4 (2.0) 3.92 (3.50) 22.6 RM(1)-160 (C)-1 7.5 (5.8) 2.8 (2.2) 2.68 (2.64) 20.5 Ashfal 7.9 (6.2) 3.0 (2.6) 2.63 (2.38) 24.2 SE 0.2 (0.2) 0.1 (0.1) 0.18 (0.15) 0.4 Figures in parentheses indicate grain characteristics from husked grain.

Azad et al (2013)

185

DISCUSSION Heavy ion irradiation has attained much importance inmutation breeding as it induces rare mutations. Keeping this in mind, the seeds of Ashfal were irradiated with different doses of carbon ion beams (Table 1) and were grown in boro season. Three of nine headed plants were fertile and the rest six were sterile. Since Ashfal is a highly photoperiod-sensitive cultivar, none of the control plants headed during boro season. However, we found nine M1 plants headed under the same conditions (Table 1). This raised the question whether the headed plants in M1 generation were contamination of other photoperiod-insensitive varieties or induced by carbon ion beam irradiation. To exclude the possibility of contamination, SSR analysis of DNA was conducted and it was revealed that the primer used produced monomorphic bands for the parent Ashfal and the mutants derived from the M1 plants (Figure 2). These results suggest that photoperiod sensitivity was altered by carbon ion irradiation in a few M1 plants. Plant heights of M2 progenies were significantly shorter than that of Ashfal (Table 2). This is consistent with the fact that the fertile M1 plants had shorter plant height than Ashfal. These results could be due to the fact that the fertile mutant lines shifted from vegetative growth to reproductive growth earlier than parental Ashfal. It has been reported that one dominant and one recessive gene are responsible for photoperiod sensitivity in rice (Yu and Yao, 1968; Yokoo and Fujimaki, 1971; Yokoo et al., 1980). It could be that carbon ion beam irradiation inactivated or down-regulated the activity of the dominant gene in the irradited seeds, making the M1 plants homogygous monogenic recessive for photoperiod sensitivity. As a result, these M1 plants bred true in M2 and M3 generations. The M2 progenies from 160 Gy were omitted from these evaluations because they were prone to lodging; they had weak culms although plants were shorter. Moreover, these progenies had shorter panicle length and smaller grain number. The mutants derived from 40 Gy dose also did not continue after M3 generation because of their inferior performance. Of the grain characteristics (although size and shape of most of the mutants were altered) grain

weight mostly remained unchanged just like the parent (Table 3). This also confirms that the M1 plants that mutated as photoperiod insensitive with shorter plant height were not contamination of other varieties. If it was contamination, then 1000-grain between Ashfal and the contaminants must differ significantly.

It could be concluded that it is possible to induce fixed mutants in the M1 generation of indica rice Ashfal through 40-200 Gy doses of carbon ion beam irradiation. This result confirms the findings of Honda et al. (2006) who reported genetically fixed mutants in M1 generation of sweet pepper for dwarf height and yellow pericarp. All these suggest that heavy ion beam irradiation has unique properties that induce fixed mutation even in M1 generation. In other mutational studies, mutants are usually detected in the M2 and M3 generations, and thus it necessitates of numerous plants screening. If mutants can be screened in M1 generation, the number of plants that must be handled can be dramatically decreased. Therefore, further studies are needed to investigate whether other crop species induce fixed mutants in the M1 generation. REFERENCES Azad MAK, Hamid MA, Yasmine F (2010).

Binachinabadam-4: a high-yielding mutant variety of groundnut with medium pod size. Plant Mutation Rep. 2(2): 45-47.

Azam MA, Uddin I (1999). Binadhan-4, an improved rice variety bred through induced mutation. Bangladesh J. Nuclear. Agric. 15: 59-65.

Blakely EA (1992). Cell inactivation by heavy charged particles. Rad. Environ. Biophys. 31:181-196.

Bonilla PS, Dvorak J, Mackill D, Deal K, Gregorio G (2002). RFLP and SSLP mapping of salinity tolerance genes in chromosome 1 of rice (Oryza sativa L.) using recombinant inbred lines. Philipp. Agric. Sci. 85(1): 64-74.

Gregorio GB, Senadhira DRD, Mendoza RD, Manigbas NL, Roxas JP, Guerta CQ (2002). Progress in breeding for salinity and associated abiotic stresses in rice. Field Crops Res. 76: 91-101.

Hamid MA, Azad MAK, Howelider MAR (2006). Development of three groundnut varieties


186

with improved quantitative and qualitative traits through induced mutation. Plant Mutation Rep. 1(2): 14-16.

Hase Y, Tanaka A, Baba T, Watanabe H (2000). FRL1 is required for petal and sepal development in Arabidopsis. Plant J. 24: 21-32.

Hayashi Y, Takehisa H, Kazama Y, Ichida H, Ryuto H, Fukunishi N, Abe T (2007). Effects of ion beam irradiation on mutation induction in rice. In: Proc. Cyclotrons and their applications 2007, 18th International Conference, pp. 237-239.

Hidema J, Yamoto, M, Kumagai T, Hase Y, Sakamoto A, Tanaka A (2003). Biological effects of carbon ion on rice (Oryza sativa L.). JAERI- Review 2003-033: 85-87.

Honda I., Kikuchi K, Matsuo S, Fukuda M, Saito H, Ryuto H, Fukunishi N, Abe T (2006). Heavy ion induced mutants in sweet pepper isolated by M1 plant selection. Euphytica 152: 61-66.

Islam MM (2004). Mapping salinity tolerance genes in rice (Oryza sativa L.) at reproductive stage. Ph.D. Thesis, University of the Philippines Los Banos, College Laguna, Philippines.

Niones JM (2004). Fine mapping of the salinity tolerance gene on chromosome 1 of rice (Oryza sativa L.) using near isogenic line. M.Sc. thesis, University of the Philippines Los Banos, College Laguna, Philippines.

Rakwal R., Kimura S, Shibato J, Nojima K, Kim YK, Nahm BH, Jwa NS, Endo S, Tanaka K, Iwahashi H (2008). Growth retardation and death of rice plants irradiated with carbon ion beams is preceded by very early dose- and time-dependent gene expression changes. Mol. Cells 25(2): 272-278.

Salil KB, Islam MM, Emon MR, Begum SN, Siddika A, Sultana S (2007). Identification of salt-tolerant rice cultivars via phenotypic and marker assisted procedures. Pakistan J. Biol. Sci. 10(24): 4449-4454.

Shamsuzzaman KM, Hamid MA, Azad MAK, Shaikh MAQ (1998). Development of early maturing and high-yielding cotton genotypes through induced mutations. Bangladesh J. Nuclear Agric. 14: 1-8.

Shikazono N, Yokota Y, Kitamura S, Suzuki C, Watanabe H, Tano S, Tanaka A (2003). Mutation rate and novel tt mutants of Arabidopsis thaliana induced by carbon ions. Genetics 163: 1449-1455.

Tanaka A, Tano S, Chantes T, Yolota Y, Shikazono N, Watanabe H (1997). A new Arabidopsis mutant induced by ion beams affects

favonoid synthesis with spotted pigmentation in testa. Genes Genet. Syst. 72: 141-148.

Tanaka A, Sakamoto A, Ishigaki Y,Nikaido O, Sun G, Ase Y, Shikazono N, Tano S, Watanabe H (2002). An ultra violet-B- resistant mutant with enhanced DNA repair in Arabidopsis. Plant Physiol. 129: 64-71.

Vergara BS, Chang TT (1985). The flowering response of the rice plant to photoperiod. The International Rice Research Institute, Los Baños, Laguna, Philippines.

Yokoo M, Fujimaki H (1971). Tight linkage of blast resistance with late maturity observed in different Indica varieties of rice. Jpn J. Breed. 21: 35-39

Yokoo M, Kikuchi F, Nakane A, Fujimaki H (1980). Genetical analysis of heading data by aid of close linkage with blast resistance in rice. Bull. Natl. Inst. Agric. Sci. D31: 95-126.

Yu CJ, Yao TT (1968). Genetische studien beim Reis. II. Die Koppelung des Langhullspelzengens mit dem Photoperiodizitatsgen. Bot. Bull. Acad. Sin. 9: 34-35.


INHERITANCE OF CELL MEMBANE STABILITY UNDER HEAT AND OSMOTIC STRESSES IN BREAD WHEAT

H. SHAHBAZI1*, E. AALI, A.A. IMANI and M. ZAEEFIZADEH

1Department of Agronomy and Plant Breeding, Ardabil Branch, Islamic Azad University, Ardabil, Iran *Corresponding author’s email: [email protected]

SUMMARY To evaluate the inheritance of cell membrane stability of bread wheat, the F1 seeds of a 7×7 half diallel, along with their parents were grown in greenhouse in a randomized complete block design with three replications. After flowering, leaf blades were used for the evaluation of membrane relative injury (RI) under heat stress and index of injury (ID) under osmotic stress induced by polyethylene glycol. The experiment was carried out at the Islamic Azad University, Ardabil, Iran, in 2010. Results showed that the additive-dominant model was adequate for determining the inheritance of the traits. In spite of high broad-sense heritability of the traits, they had relatively low narrow-sense heritability. Both of the traits were governed by overdominance, while the magnitude of average degree of dominance was greater in ID. High heterotic effects were observed only in ID among crosses. Higher frequency of dominant alleles in the parents was also demonstrated in both traits. Recessive alleles were favorable in ID. However, such relation was not observed in the case of RI. The significant GCA mean square suggests that genetic gain is achievable through selection over the segregant population. However, due to high average degree of dominance, selecting for drought tolerance must be done in advanced generations of wheat breeding programs. Results of combining ability analysis also showed that Pishtaz, Azar2, Fengkang and Sisson had favorable additive genes for ID, while Alvand, Pishtaz and Gaspard had favorable genes for RI. Keywords: Cell membrane stability, diallel, drought, heritability, osmotic stress, bread wheat

Manuscript received: March 12, 2012; Decision on manuscript: November 14, 2012; Manuscript accepted: December 12, 2012

@ Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Drought is a major abiotic stress, limiting crop production in arid and semi-arid climates. Stress resistance in plants is a complex character that depends on many genes and thus is determined by the interactions of many morphological, physiological and biochemical processes. Yield, which is a major selection criterion, has low heritability and shows high genotype-environment interaction, and hence, selection

becomes more difficult in a given environment (Jackson et al., 2001). Over the past few decades, breeding efforts for improved drought tolerance have revolved around the selection of genotypes for morphophysiological characters responsible for drought resistance under field conditions.

Oxidative stress is common feature of abiotic stresses in which formation of reactive oxygen species (ROS) can seriously disrupt normal metabolism, resulting in lipid peroxidation and consequently membrane injury,

RESEARCH ARTICLE


https://www.agronomy.org/publications/cs/articles/47/4/1416#ref-14�


188

protein degradation, enzyme inactivation, pigment bleaching and disruption of DNA strands (Pan et al., 2006; Quiles and López, 2004). Therefore, cell membrane stability in plants under stress is usually regarded as an indicator of the tolerance of genotypes against stress conditions (Sairam et al., 1998; Grzesiak et al., 2003; Dhanda and Munjal, 2006; Sayar et al., 2008). Grzesiak et al (2003) found that leaf injury under osmotic stress induced by manitol and heat stresses were significantly correlated with drought tolerance in Triticale.

According to Dhanda and Munjal (2006) membrane thermostability is significantly correlated with drought tolerance. Sayar et al (2008) reported that under atmospheric osmotic stress, tolerant cultivars had more integrated membrane compared with sensitive cultivars. Based on the results of Yildirim et al (2009), membrane stability of flag leaf at the early milk stage was significantly correlated with grain yield.

According to Blum and Ebercon (1979) and Singh et al. (1992), maximal separation of wheat cultivars in drought tolerance was obtained with 40% (w/v) solution of PEG. Cell membrane stability methods also can be measured easily and are nondestructive to the whole plant, so

potentially stable cell membranes can be considered as selection criteria if they also have high heritability. The objective of this study was to compare the inheritance of cell membrane injury induced by heat and osmotic stresses in bread wheat. MATERIALS AND METHODS Seven bread wheat varieties (Sabalan, Azar2, Fengkang, Pishtaz, Gaspard, Alvand and Sisson) were crossed in half diallel fashion (Table 1). The F1 seeds, along with their parents, were grown in the greenhouse in well=watered condition using randomized complete block design with three replications at the Islamic Azad University, Ardabil, Iran in 2010. Seeds of the genotypes were sown in plastic pots filled with 10 kg of soil composed of a mixture of garden soil, compost and sand (1:1:1, v/v). Two weeks after anthesis, flag leaves were collected for cell membrane injury under heat and PEG=induced osmotic stresses.

Table 1. Wheat cultivars used in the present study.

Cultivar Pedigree/Origin GH PH Irrigation need Sabalan 908/Fn A12//1-32-4382 W T RF Azar2 Kvz/Tr71/3/Maya"s"//Bb/Inia/4/Sefid W T RF Fengkang Fengkank/sefid3… W T RF Pishtaz Alvand//Aldan"S"/Ias58 40072-48 S SD RF Gaspard Introduction Cultivar-France W D IR Alvand CF1770/1-27-6275 F SD RF Sisson Introduction Cultivar-France W D IR

GH = growth habit, W = winter, F = facultative, S = spring, PH = plant height, D = dwarf (<70cm), SD = semidwarf (70 to 85 cm), T = tall (>85cm), RF = rainfed, IR = irrigated. Cell membrane injury under heat-induced stress (RI) Leaf blades were cut into 1 cm-long Sections immediately after sampling. Two 0.5 g leaf samples were placed into vials, one as control and the other for heat treatment. The samples were washed three times with distilled water to remove electrolytes adhering to the leaves or

released from the cut ends of the tissues. Distilled water (10 ml) was added to each vial. Control vials were covered and held in a water bath at 25 °C, and treatment vials were heated at 50 °C for 1 h. Both control and treatment vials were held at 10 °C for 24 h to allow exosmosis and then warmed to room temperature. Electrolytes were measured directly by an electrical conductivity meter for control (C1) and

Shahbazi et al. (2013)

189

heat treated (T1) tubes. The tubes were heated in boiling water for 10 min and cooled. A second conductivity reading of the aqueous phase (C2 and T2) was taken at 25 °C after the samples were cooled. Membrane thermostability was expressed as relative injury (RI) proposed by Martineau et al., (1979).

Cell membrane injury under PEG=induced stress (ID) Fully expanded flag leaf blades were collected, immediately weighed and cut into segments (ca.1 cm), washed for 15 min in sterile deionized water and then exposed either to 0% (control) or to PEG 6000 solution with -1 MPa osmotic potential for 15 h in the dark. Electrolyte leakage was then measured before (ECi) and after (ECf) 4 h of rehydration and ultimately after autoclaving (ECt). Cell membrane injury was expressed as an index of injury (ID) (Flint et al., 1967) calculated as follows:

where Rs and Rc represent

for control or PEG-treated tissues, respectively. Statistical analysis The diallel analysis was done according to the theoretical basis developed by Hayman (1954a), and adapted for the half diallel by Walters and Morton (1978). The goodness of fit of the

additive-dominant model was performed based on the analysis of variance of Wr-Vr and linear regression of Wr on Vr (Hayman, 1954a). The genetic components D, H1, H2, F and h2 were estimated according to Singh and Singh (1984). Standard errors of these components were calculated from expected and observed values of Wr, Vr, rV , nVp and (mL1-mL0)2 over replications (Hayman, 1954a). From the estimates of the genetic components, the genetic parameters presented in Table 4 were estimated. Average degree of dominance, broad=sense heritability and narrow=sense heritability were calculated according to Mather and Jinks (1971). Combining ability analysis was also carried out following Model I and Method II of Griffing (1956). Following Baker (1978), the variance ratio 2S2gca/2S2gca+S2sca was computed from expected components of mean squares assuming a fixed model, to assess the relative importance of additive and non-additive gene effects. Analysis of variance of diallel was performed using the DIAL98 software (Ukai, 1989); genetic components were estimated by electronic spreadsheets in the Excel program (Microsoft® Excel 2003). Combining ability analysis was performed using SAS 9.2 software. RESULTS The results of the goodness of fit of the additive-dominant model are shown in Table 2. Non-significant Wr-Vr mean squares for treatment (crosses) indicated the adequacy of additive dominant model for both of the traits. The slopes of linear regression were also significantly higher than zero and did not show significant differences with 1 (Table 2).

Table 2. Goodness of fit of additive-dominant model for evaluated traits.

Character Heterogeneity of Wr-Vr (mean squares)

t-test of b on the null hypothesis

b=0 b=1 Index of injury (ID) 6895.0ns 0.661**± 0.313 0.661ns± 0.313 Relative injury (RI) 2192.9ns 1.05**± 0.195 1.05ns ± 0.195

ns* and ** non-significant and significant at the 5% and 1% levels, respectively.


190

The analysis of variance of the diallel is shown in Table 3. Additive variance (a compo-nent) was highly significant in both traits, indicating the presence of additive effects in their control. The significance of (A) in Table 3 was in accordance with the significance of additive effects (D component) in table 4. The dominant genetic effects (B source of variation) showed highly significant effects in both traits, indicating the importance of dominant genetic effects in these traits. The “b1” component, which measures the mean deviations of the F1s from the mid-parental values, was highly significant for ID (Table 3). The significance of the “b1” component indicates that dominance was predominantly in one direction and measures average heterosis (Singh and Singh, 1984). The significance of the b1 component was generally in accordance with higher magnitude of dominance ( PF −1 ) where F1s had lower ID than their parents. Since the mean dominance effect of the heterozygote locus (h2) was significant only for ID, high heterotic effect values would be expected for ID among crosses (Table 4). The “b2” component was significant only for ID. The significance of the b2 item indicated that the mean dominance deviations of the Fls from their mid=parental values differed significantly over the F1 arrays; this implies the presence of asymmetry in the distribution of alleles among the parents (Hayman, 1954b). This also means that there was evidence that some parents had a significantly better performance than others (Ramalho et al., 1993). Since b2 is significant for ID, the 'a' item will not measure additive variance unambiguously, but it will be contaminated with non-additive variance also (Singh and Singh, 1984; Chaudhary et al., 1977). The proportion of positive and negative genes was estimated by calculating (H2/4H1) in Table 4. This ratio was lower than 0.25 in both traits, indicating the presence of asymmetry in the distribution of the positive and negative alleles in the parents. This is also substantiated by H1 being greater than H2 in these traits. The “b3” component, which is equivalent to specific combining ability variance, was significant only in RI. b3 estimates residual dominance effects combined from additive × additive, additive × dominance

and dominance × dominance interaction effects that are not attributed to b1 and b2 (Chaudhary et al., 1977). The estimate of the genetic component F was significant in both cases, which is an indication of asymmetry in the distribution of dominant and recessive alleles in the parents. The ratio of the total number of dominant and recessive alleles in the parents (KD/KR) was higher than one in both traits, demonstrating a higher frequency of dominant alleles in the parents. Positive values for F were substantiated by KD/KR being greater than one. The degree of average dominance was higher than one, indicating the presence of overdominance in the control of both traits. Overdominance also was confirmed by the negative intercept of the regression line in both cases (Table 4). Yildirim et al. (2009) also showed that membrane thermal stability was mediated mainly by non-additive gene actions. Despite the high broad=sense heritability ( ) in both cases, narrow-sense heritability ( ) of the traits was relatively low (Table 4). The differences observed between and reflected the presence of the dominant effects. Ibrahim and Quick (2001) in parent-offspring regression using F3 plants and their F4 progeny means found that, narrow sense heritability was relatively low (0.32–0.38) for RI. Correlation coefficients between the parental means and order of dominance “rYr(Wr+Vr)” were positive in ID, indicating that recessive alleles are favorable. However, such relation was not clearly observed in the case of RI.

Combining ability analysis by Grifting's method indicated the significance of both GCA and SCA mean squares, which is in accordance with the results of Dhanda and Munjal (2006). This shows the importance of both additive and dominance effects (Table 5). The higher Baker’s ratio for RI suggests that additive effects are more important than non-additive effects in the case of RI (Table 5). Dhanda and Munjal (2006) also found that the magnitude of GCA effects was considerably higher than SCA effects for RI. Dhanda and Munjal (2009) showed that the mean square for GCA was higher in magnitude than that of SCA, but the components of genetic variance indicated considerable influence of dominance variance in determining inheritance of


191

cell membrane stability. The importance of GCA effects was also

evident from the higher correlation between the parental means and the GCA effects in the case of RI (r = 0.947**) compared with ID (0.798*).

Thus, the combining ability analysis was in good agreement with conclusions from Hayman's method in showing gene action in the inheritance of RI and ID.

Table 3. Analysis of variance of the diallel tables for the evaluated traits. SV df Index of injury(ID) Relative injury (RI)

REP 2 125.15ns 4.25ns A 6 558.38** 717.45** B 21 417.42 ** 244.24 ** b1 1 2584.39 ** 152.83ns b2 6 383.34* 207.19ns b3 14 277.23ns 266.65** Error 54 164.42 104.79

ns,* and ** non-significant and significant at the 5% and 1% levels, respectively. Table 4. Estimates of genetic components and related statistics in half- diallel design.

Index of injury (ID) Relative injury (RI) D= 251.22**±6.22 274.91**±4.67 H1= 713.43**±14.98 446.29**±11.25 H2= 558.07**±13.20 354.80**±9.91 F= 275.57**±14.93 235.50** ±11.21 h2 569.2**±14.93 14.2ns±6.66 E 54.81**±2.20 34.93**±1.65

H2/4H1 0.195 0.198 KD/KR 1.96 2.01

Average d 1.68 1.34 0.252 0.346 0.789 0.815

F1-P(%) -12.81 3.11 rYr (Wr+Vr) 0.829* -0.456 ns

A(intercept) -85.59ns±73.34 -80.99ns ±32.51 ns,* and ** non-significant and significant at the 5% and 1% levels, respectively. Table 5. Analysis of variance of the diallel tables for the evaluated traits.

Source of variation df Mean Squares

Index of injury(ID) Relative injury (RI)

General Combining Ability 6 711.83** 811.54** Specific Combining Ability 21 372.30** 217.36*

Error 54 164.42 104.79 S2GCA 20.27 26.18 S2SCA 69.29 37.52

SCAGCA

GCA

SSS

22

2

22

+ -- 36.9% 58.25%

* and ** significant at the 5% and 1% levels, respectively


192

Results of GCA effects indicated that Pishtaz, Azar2, Fengkang and Sisson had lower GCA for ID (smaller values are favorable), while Alvand, Pishtaz and Gaspard had favorable genes for RI under heat injury (Table 6). Additional information on the mode of inheritance of characters was inferred from Wr,Vr graphs (Figure 1). The distance of each parent from the origin showed that Sabalan, Siosson, Azar2 and Alvand had higher concentrations of dominant

alleles for RI. Among them, Sabalan, Siosson and Azar2 had high GCA. In contrast, Pishtaz had lower concentrations of dominant alleles and had relatively good GCA, suggesting that dominant alleles are more or less unfavorable. In the case of ID all of the genotypes (except for Alvand) had an equal number of dominant and recessive alleles and hence, the relationship between dominance and favorability was not clear (Figure 2).

Table 6. General combining ability estimates of parents used in the experiment. Parent Index of injury(ID) Relative injury (RI)

Sabalan 6.125 b 3.15 b Azar2 -5.040 a 5.99 b

Fengkang -2.32 a 3.23 b Pishtaz -5.478 a -4.93 a Gaspard 1.637ab -4.17 a Alvand 7.246 b -7.83 a Sisson -2.167 a 4.56 b

Standard error of GCA 5.22 3.45

Figure 1. Regression of Wr on Vr for relative injury (RI).


193

Figure 2. Regression of Wr on Vr for injury index (ID).

DISCUSSION The fulfillment of assumption for Hayman's analysis indicated that a relatively simple genetic model was involved in the inheritance of membrane stability under heat and osmotic stresses. Since the degree of average dominance was higher than 1, the presence of overdominance and the greater importance of dominance effects in the control of traits were suggested. The results also showed that, in the case of ID, dominance was predominantly in one direction, indicating the presence of heterosis in the control of this trait. Regarding the high broad-sense heritability of traits under study, it can be concluded that they can be considered candidates for selecting drought tolerance in wheat. However the relatively lower narrow-sense heritability of ID and RI suggests the use of multiple replications during selection to limit environmental effects. Regarding the low narrow-sense heritability of the traits, it can be concluded also that these traits can be considered selection over the segregant population. In the combining ability analysis, additive × additive epistasis forms a specific part of variance due to GCA, while epistasis of additive × dominance and dominance × dominance types are included

in GCA (Griffing, 1956). As additive gene action and additive × additive types of epistatic gene action are exploitable in homozygous genotypes, the estimates of GCA effects of individual lines are a useful predictor for progeny performance in self-fertilizing species (Baker, 1978). Finally, it is obvious that the two methods of diallel analysis together provide more useful information on the mechanism of inheritance than each alone does. ACKNOWLEDGEMENTS We are grateful to the Islamic Azad University, Ardabil Branch for financial support. REFERENCES Baker RJ (1978). Issues in diallel analysis. Crop Sci.

18: 533-536. Blum A, Ebercon A (1979). Cell membrane stability

as a measure of drought and heat tolerance in wheat. Crop Sci. 21(1): 43-47.

Chaudhary BD, Kakar SN, Singh RK (1977). Comparison of diallel and its modification. Silvae Genet. 26: 2-3.


194

Dhanda SS, Munjal R (2006). Inheritance of cellular thermotolerance in bread wheat. Plant Breed. 125: 557-564.

Dhanda SS, Munjal R (2009). Cell membrane stability: Combining ability and gene effects under heat stress conditions. Cereal Res. Commun. 37(3): 409-417.

Flint HL, Boyce BR, Beattie DJ (1967). Index of injury - A useful expression of freezing injury to plant tissues as determined by the electrolytic method. Can. J. Plant Sci. 47: 229–230.

Fronza V, Vello NA, Camargo LEA (2004). Genetic analysis of soybean resistance to Fusarium solani f.sp. glycines. Genet. Mol. Biol. 27(3): 400-408.

Griffing B (1956). Concept of general and specific combining ability in relation to diallel crossing systems. Aust. J. Biol. Sci. 9: 463-493.

Grzesiak S, Grzesiak MT, Feilk W, Stabryla J (2003). Evaluation of physiological screening tests for breeding drought resistant triticale (X Triticosecale Wittmack). Acta Physiol. Plant. 25 (1): 29- 37.

Hayman BI (1954a). The theory and analysis of diallel crosses. Genetics 39: 789-809.

Hayman BI (1954b). The analysis of variance of diallel tables. Biometrics 10: 235-244.

Ibrahim AMH, Quick JS (2001). Heritability of heat tolerance in winter and spring wheat. Crop Sci. 41: 1401—1405.

Jackson PA (2001). Direction of physiological research in breeding: Issues from a breeding perspective. In M.P. Reynolds et al. (ed.) Application of physiology in wheat breeding. CIMMYT, Mexico, DF. p. 11–16.

Martineau JR, Specht JE, Williams JH, Sullivan CY (1979). Temperature tolerance in soybeans. I. Evaluation of a technique for assessing cellular membrane thermostability. Crop Sci. 19: 75-78.

Mather K, Jinks JL (1971). Biometrical genetics. Chapman & Hall: London.

Pan Y, Wu LJ, Yu ZL (2006). Effect of salt and drought stress on antioxidant enzymes activities and SOD isoenzymes of liquorice (Glycyrrhiza uralensis Fisch). Plant Growth Regul. 49: 157–165.

Quiles MJ, López NI (2004). Photoinhibition of photosystems I and II induced by exposure to high light intensity during oat plant grown effects on the chloroplastic NADH dehydrogenase complex. Plant Sci. 166: 815-823.

Ramalho MAP, Santos JB, Zimmermann MJO (1993).

Genética quantitativa em plantas autógamas: aplicações ao melhoramento do feijoeiro. UFG, Goiânia. 271 pp.

Sairam RK, Deshmukh PS, Saxena DC (1998). Role of antioxidant systems in wheat genotypes tolerance to water stress. Biol. Plant. 41 (3): 387-394.

Sayar R, Khemira H, Kameli A, Mosbahi M (2008). Physiological tests as predictive appreciation for drought tolerance in durum wheat (Triticum durum Desf.). Agron. Res. 6(1): 79-90.

Singh M, Srivastava JP, Kumar A (1992). Cell membrane stability in relation to drought tolerance in wheat genotypes. J. Agron. Crop Sci. 168 (3): 186- 190.

Singh M, Singh RK (1984). A comparison of different methods of half-diallel analysis. Theor. Appl. Genet. 67: 323-326.

Ukai Y (1989). A microcomputer program DIALL for diallel analysis of quantitative characters. Jpn. J. Breed. 39: 107–109.

Walters DE, Morton JR (1978). On the analysis of variance of a half diallel table. Biometrics 34: 91-94.

Yildirim M, Bahar B, KOÇ M, Barutcular C (2009). Membrane thermal stability at different developmental stages of spring wheat genotypes and their diallel cross populations. Tarim Bilimlari Dargisi 15 (4): 293-300.

http://www.akademiai.com/content/w10h9574167u/?p=93d3723fd63f4612a3ab8d8999e3552c&pi=0�

195


AMMI AND BI-PLOT ANALYSES TO IDENTIFY STABLE GENOTYPES OF INDIAN MUSTARD (Brassica juncea L.) FOR OIL AND SEED MEAL QUALITY CHARACTERS

J.S. CHAUHAN1,3*, K.H. SINGH1 and D.C. MISHRA2

1Directorate of Rapeseed-Mustard Research, Bharatpur, Rajasthan, India 2Indian Agricultural Statistics Research Institute, Pusa Campus, New Delhi, India

3Present address: Indian Council of Agricultural Research (ICAR), New Delhi, India *Corresponding author’s email: [email protected]

SUMMARY

Stability of oil, protein and glucosinolates, erucic, ecosenoic, oleic, linoleic, linolenic, palmitic and stearic acid along with seed yield was investigated in 25 genotypes (23 for seed yield) of Indian mustard (Brassica juncea L.) using AMMI (additive main effect and multiplicative interaction) and bi-plot analyses based on double centred principal component analysis (PCA). Experiments were conducted during 2003-04 to 2005-06 using randomized complete block design with three replications. Combined analysis of variance showed highly significant (P<0.01) difference between the genotypes, locations and G×E interaction, suggesting differential response of genotypes across growing environments, which could be attributed to differential ranking of genotypes. PCA 1 and PCA2 axes were significant (P<0.01) and captured the largest portion of the variation of the G×E interaction for all the characters, indicating that AMMI 2 model was the best for the data set, therefore, a bi-plot of PCA1 and PCA2 was constructed to identify the most stable genotypes for different quality characters. AMMI’s stability parameter - ASV (AMMI stability value) has been calculated for each genotype. Higher ASV reflected variable response to different environment and lower ASV reflects stability across the environments. AMMI analysis revealed that IPCA1 and IPCA2 captured almost 100% of the interaction sum of squares and were also significant (P < 0.01) for all the characters investigated, suggesting that the AMMI model with two principal components was the best predictive model. Considering results from AMMI and n\bi-plot analyses, genotypes RH 30, RH 8812; Sanjucta Asech, Sej 2 and Varuna were most stable for oil, protein, glucosinolate content, fatty acid profile and seed yield.

Keywords: AMMI, bi-plot, PCA, stability, genotype × environment interaction, Indian mustard, Brassica juncea L. Manuscript received: February 2, 2012; Decision on manuscript: September 16, 2012; Manuscript accepted: December 21, 2012

@ Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION The rapeseed-mustard (Brassica spp.) group of crops represents the second important oilseed-crops in India. Indian mustard (Brassica juncea L. Czernj & Coss.) is a predominant crop among the oilseed Brassicas. Oil content of the seed is very important from the commercial point of view whereas fatty acid composition of oil is its

major quality determinant. It is desirable to have edible oil with high oleic (about 60%), intermediate linoleic (25-30%), low erucic (< 2%) and very low linolenic acid (< 3%). Glucosinolates, belong to a group of plant thioglucosides found principally among members of family Cruciferae (Brassicaceae). They principally determine the quality of seed meal, an important source of high-protein animal

RESEARCH ARTICLE



196

feed. Cleavage products from hydrolysis of glucosinolates like isothiocyanates, oxazolidimenthones, etc. are very toxic to non-ruminants such as pigs, poultry and are detrimental to animal health. They reduce the feed palatability and affect iodine uptake by the thyroid glands, thus reducing feed efficiency and weight gains (Fenwick et al., 1983).

Erucic acid in oil and glucosinolate content in seed meal of the Indian rapeseed-mustard varieties are quite high (Chauhan et al., 2007). Because of known adverse effects of high erucic acid in oil (Ackman et al., 1977) and glucosinolate content in seedmeal (Fenwick et al., 1983), reducing erucic acid up to 2% (low) and glucosinolate content up to 30 μ moles/g defatted seed-meal (low) and combining both the traits to develop double-low varieties has been the focus of Indian breeding programs in recent years.

Genotype, environment and their interaction play a major role in influencing the oil and seed meal quality. Stability in the genotypes for these quality characters in multi-location/ season is foremost for both selection of a variety as well as for breeding. Knowledge of the interaction and stability is important in breeding varieties for wider adaptation in diverse agro-climatic conditions. Genotype × environment interaction effects for seed yield and its contributing characters have been well documented in Indian mustard (Dhillon et al, 1999; Singh et al., 2006; Brar et al., 2009; Sah et al., 2009). However, limited information is available on these aspects for oil and seed-meal quality characters (Chauhan et al., 2010). Further, no report is available on the use of AMMI and bi-plot analyses to assess stability in such characters in Indian mustard. Therefore, the present investigation aimed at estimating the magnitude and nature of G x E interaction and to identify stable genotypes for oil, protein, glucosinolates, fatty acid profile and seed yield in Indian mustard using AMMI and bi-plot analyses.

MATERIALS AND METHODS Field and laboratory experiment The experimental materials consisted of 25 predominantly grown varieties of Indian mustard for quality characters; however, yield was recorded on 23 varieties because of low plant population in two varieties, ‘Basanti’ and ‘RCC 4’. Breeder seed was used to grow the crop in different cropping seasons. These varieties were grown in randomized complete block design from winter (rabi) 2003-04 to 2005-06 with three replications in 5-row plots of 5 m length, keeping 45 cm row-to-row and 15 cm plant-to- plant spacing. The experiment was conducted using the recommended package of practices and two irrigations were given at 35 and 60 days after sowing. Observations were recorded on composite samples from three central rows. The oil and protein content were analysed using pre-calibrated NIR (Dicky John Instalab 600) following Kumar et al. (2003). Fatty acid profile of oil was analysed by gas liquid chromatography (Nucon Model 5765) using SP 2300+2310 SS columns. The detailed method for fatty acid analysis has been described earlier (Chauhan et al., 2002). Individual fatty acids have been expressed as the percentage of total fatty acids present in the oil. Total glucosinolate content in the seed meal was estimated by complex formation between glucosinolates and sodium tetrachloropalladate solution (Kumar et al., 2004). The intensity of the colour produced was measured using ELISA reader at 405 nm. ‘Hyola 401’, a double-low hybrid of gobhi sarson (Brassica napus), and ‘Varuna’, non-canola variety of Indian mustard (Brassica juncea), were used as standard checks for comparing varieties for quality characters. Observations were also recorded on seed yield (kg/ha).

Statistical analysis

Combined analysis of variance across the test environments was performed using SAS software. Following testing of the significance of the genotype x environment interaction (GEI) mean square, means over three replications for

Chauhan et al. (2013)

197

genotype i at location j i.e. ijY were subjected to AMMI stability analysis using SAS software (Burgueno et al., 2002). The AMMI model postulates additive components for the main effects of genotype iα and environments jβ and multiplicative components for the effect of the interaction .ijφ (Gauch, 1992 a,b). Thus, the mean response of genotype i in the environment j was modeled by:

ijijjkikkiY ερδγλβαµ ˆm

1kj +++++= ∑

=

(1)

in which ijφ is represented by ∑=

m

kjkikk

1δγλ

where kλ is the size, ikγ is the normalized genotype vector of the genotype scores or sensitivities, jkδ is the normalized environmental of the scores describing environments, ijρ are the AMMI residuals, and

the ijε is the error term. The ASV is the distance from the

coordinate point to the origin in a two-dimensional scattergram of PCA1 scores against PCA2 scores. Because the PCA1 score contributes more to the G x E sum of squares, a weighted value is needed. This weight is calculated according to the relative contribution of PCA1 to PCA2 to the interaction SS. ASV was calculated using the following:

ASV =

( )[ ] [ ]{ } 2/12GPCA2score

2)1(2/PCA1SS +scoreGPCAPCASS

(2) where 21 / PCAPCA SSSS is the weight given to PCA1 value by dividing the PCA1 sum of squares by the PCA2 sum of squares, GPCA1score is the PCA1 score for that specific genotype, and GPCA2score is the PCA2 score for that specific genotype.

Bi-plot, which graphically represents the two way data in a scatter plot, has also been used in the study for graphical representation of PCA1 score and PCA2 score for the purpose of finding out the most stable genotype (Gabriel, 1971; Kempton, 1984).

RESULTS Pooled analysis of variance showed highly significant differences for environments, genotypes and GEI for oil, protein and oleic, linoleic, linolenic, eicosenoic, erucic, palmitic + stearic acid and seed yield. The mean sum of squares due to genotypes and GEI were not significant, whereas environment effects were highly significant for glucosinolates. Partitioning of variance components for oil content indicated that residual, followed by GEI and genotypes together, accounted for 79% of the total variation. Results from AMMI analysis showed that the first principal component axis (IPCA1) of the interaction captured 63.55% of the interaction sum of squares. The mean squares for IPCA1 and IPCA2 were significant (P < 0.01) and cumulatively contributed 99.9% to the total GEI (Table 1). Genotypes Krishna, CS 52, PBR 97 and RL 1359 were found to be more stable than other genotypes on the basis of ASV ranking (Table 2) and bi-plot analysis (Figure 1a). In case of protein content, 64.32%, 27.14% and 7.2% variation was due to GEI, genotypes and residual, respectively. AMMI analysis indicated that IPCA1 captured 96.54% of the interaction sum of squares and IPCA2 explained a further 3.46% of the GEI sum of squares (Table 1). On the basis of ASV ranking (Table 2), genotypes RH 30, Sej 2, RH 8812, Sanjucta Asech and Kranti were most stable. Bi-plot analysis (Figure 1b) revealed that only RH 30, Sej 2 and RH 8812 were most stable.

Residual followed by GEI and genotypes captured the maximum variation in glucosinolates and environment contributed only 6.96%. IPCA1 and IPCA2 accounted for 88.12% and 11.87% of the GEI sum of squares. ASV values indicated genotypes Kranti, Pusa Jagannath, Rohini, Sanjucta Asech and RH 819 were most stable, whereas bi-plot analysis (Figure 1c) revealed that only Pusa Jagannath, RH 819 and Sanjucta-Asech were most stable genotypes. Partitioning of variance components for linolenic acid showed that environments, residual and GEI accounted for 26.59%, 24.67% and 24.19% of the total variation, respectively. AMMI analysis indicated that the IPCA1 and IPCA2, respectively, captured 58.06% and 41.93% of the interaction sum of squares (Table


198

1). The mean squares for IPCA1 were only significant (P < 0.01). The post-predictive evaluation using an F-test (P < 0.01) suggested that two principal component axes of the interaction were significant for the model with 48 degree of freedom. Genotypes, RL 1359, Sej 2, RH 30 and RH 819 were the most stable on the basis ASV ranking and bi-plot analysis (Figure1 d).

Environments captured 39.67% of the total variation for linoleic acid followed by GEI (24.61%) and genotypes (19.97%) and IPCA1 captured 60.04% of the interaction sum of squares while IPCA2 explained the rest GEI sum of squares (Table 1). On the basis of ASV values and bi-plot analysis (Figure1 e) the five most stable genotypes were Pusa Bold, PCR 7, RL 1359, PBR 91 and PBR 97. Environments (30.19%) followed by GEI (26.59%) and residual (24.41%) captured 81.19% of the total variation in oleic acid. The mean squares for IPCA1 and IPCA2 were significant (P < 0.01) and cumulatively contributed almost 100% to the total GEI (Table 1). ASV ranking (Table 2) and bi-plot analysis (Figure 1f) indicated that genotypes Pusa Bold, Pusa Jagannath, RH 8812 and RH 30 were most stable.

Variation assessment for eicosenoic acid revealed that 28.68%, 17.67%, 28.95% and 22.83% of the total variation was explained by environments, genotypes, GEI and residual, respectively. AMMI analysis indicated that IPCA1 accounted for 79.10% of the interaction sum of squares and the rest was explained by IPCA2 (Table 1) of the GEI sum of squares and both were significant (P < 0.01). ASV ranking (Table 2) suggested that genotypes Bio 902, Rohini, Varuna, Pusa Bahar and PCR 7 were most stable. However, bi-plot analysis (Figure 1g) indicated Bio 902, Pusa Bahar, Pusa Bold, RH 8812 and PCR 7 to be most stable genotypes. Genotypes followed by GEI and residual contributed maximum to the total variation in erucic acid (Table 1). IPCA1 and IPCA2 together contributed 99.99% to the interaction sum of squares and found to be significant (P < 0.01). Genotypes Pusa Jagannath, RH 8812, Varuna and RH 8113 were found to be most stable genotypes on the basis of ASV ranking and bi-plot analysis (Figure 1h). Contribution of genotype, environment and GEI

was 10.16%, 23.81% and 27.4%, respectively, to the total variation in palmitic + stearic acid. Nevertheless, a high proportion was attributed to the residual factors (Table 1). Cumulatively, IPCA1 and IPCA2 captured 99.99% of interaction sum of squares (Table 1). Both ASV ranking (Table 2) and bi-plot analysis (Figure 1i) revealed Pusa Bahar, RH 8113, GM 2, Pusa Bold and Pusa Jagannath to be most stable genotypes.

Variance analysis for seed yield indicated that of the total variation, 31.45% was due to environments, 39.87% due to genotypes, 14.67% was due to GEI and residual contributed 13.78%. IPCA1 and IPCA2 together accounted for 99.99% of the interaction sum of square, which was significant (P < 0.01). ASV ranking (Table 2) and bi-plot (Figure 1j), analysis revealed PBR 97, RH 781, PCR 7 and RL 1359 were most stable genotypes.

DISCUSSION The overall results of the analysis of variance suggested that oil, seed meal quality and seed yield were highly influenced by the environmental conditions and that genotypes performed differentially in 3 cropping seasons indicating the need for stability analysis. In earlier studies, quality characters were observed to be largely influenced by environment and genotype x environment interaction (Chauhan et al., 2010). The contribution of genotype, environment, GEI and residual factors to total variation varied from character to character. Similar results were, by and large, reported for seed yield, its components and oil content in Indian mustard (Singh et al., 2006; Brar et al., 2009; Sah et al., 2009). IPCA1 and IPCA2 together contributed almost 100% to the interaction sum of squares and both principal component axes were significant (P < 0.01) for oil, protein, glucosinolates, fatty acid profile and seed yield, therefore, AMMI with two interaction principal component axes could be considered as the best predictive model.


199

Table 1. Analysis of variance table for AMMI of oil, seed meal quality and yield in Indian mustard. Sources of variation

d.f %SS Oil Protein Gluco-

sinolate Linolenic acid

Linoleic acid

Oleic acid

Eicosenoic acid

Erucic acid

Palmitic + stearic

Yield

Environments 2 16.68** 1.10** 6.96** 26.59** 39.67** 30.19** 28.68** 12.26** 23.81** 31.45** Replication 2 4.32** 0.26 0.30 0.25 1.30** 0.50 1.86** 0.84** 2.25* 0.23 Genotypes 24 25.10** 27.14** 11.5 23.69** 19.97** 18.32** 17.67** 47.29** 10.16* 39.87** Env*Gen 48 26.80** 64.32** 24.38 24.19** 24.61** 26.59** 28.95** 32.63** 27.44** 14.67** IPCA1 25 63.55** 96.54** 88.12** 58.06** 60.04** 61.19** 79.10** 63.66** 67.72** 61.23** IPCA2 23 36.44** 3.46** 11.87 41.94** 39.95** 38.80** 20.89* 36.33** 32.27 38.76** Residuals 148 27.10 7.20 56.85 24.67 14.45 24.41 22.83 6.98 36.34 13.78**

Table 2. AMMI stability value (ASV) for oil, seed meal quality and yield in Indian mustard. Genotype Oil Protein Gluco-

sinolate Linolenic acid

Linoleic acid

Oleic acid

Eicosenoic acid

Erucic acid

Palmitic + stearic

Yield

Basanti 0.36 1.23 1.89 0.20 (1) 2.40 0.69 0.92 1.49 0.52 ----- Bio 902 0.93 1.70 1.25 0.93 0.94 1.00 0.36 (1) 0.86 0.37 0.23 CS 52 0.26 (2) 0.77 1.51 1.37 0.74 0.74 1.23 1.49 0.28 0.46 GM 1 0.92 0.66 2.46 0.98 1.03 1.09 1.34 1.95 0.37 0.43 GM 2 1.06 0.51 1.48 0.65 1.04 1.50 1.86 1.69 0.19 (3) 0.28 Kranti 0.62 0.48 (5) 0.79 (1) 1.63 1.44 2.14 0.77 0.78 0.71 0.14 Krishna 0.14 (1) 0.75 2.01 0.91 1.80 1.91 1.30 2.00 0.45 0.05(2) Pusa Bahar

0.33 2.42 2.41 0.60 0.45 0.82 0.44 (4) 0.87 0.14 (1) 0.41

Pusa Bold

0.53 2.01 24.48 0.46 0.10 (1) 0.12 (1) 0.53 0.90 0.21 (4) 0.40

Pusa Jagannath

0.89 1.74 0.90 (2) 1.26 1.47 0.26 (2) 2.03 0.03 (1) 0.26 (5) 0.26

PBR 91 1.19 0.97 2.23 1.40 0.27 (4) 0.97 (5) 0.59 1.73 1.04 0.25 PBR 97 0.30 (4) 1.57 4.00 0.62 0.29 (5) 0.39 0.69 1.35 0.93 0.01(1) PCR 7 0.75 0.72 1.22 0.48 0.18 (2) 0.52 0.45 (5) 0.83 1.17 0.11(4) RCC 4 0.50 20.91 2.71 1.05 2.26 1.09 2.25 1.53 0.50 ----- RH 781 0.77 1.01 3.27 0.83 0.71 1.10 1.65 1.12 0.81 0.07(3) RH 8113 0.76 0.66 1.24 0.81 0.73 0.69 1.66 0.37 (4) 0.17 (2) 0.19 RH 819 0.63 0.75 1.16 (5) 0.45 (5) 0.31 0.56 0.79 0.88 0.28 0.21 RH 30 0.44 0.15 (1) 1.50 0.34 (4) 0.54 0.33 (4) 1.47 0.61 (5) 1.16 0.28 RL 1359 0.30 (5) 0.83 2.75 0.27 (2) 0.19 (3) 0.89 0.59 1.38 1.02 0.11(5) RH 8812 0.27 (3) 0.35 (3) 1.68 0.60 0.76 0.29 (3) 0.37 0.11 (2) 0.30 0.52 Rohini 1.80 2.26 1.04 (3) 0.71 1.20 0.80 0.85 (2) 1.46 1.29 0.81 Sanjuncta Asech

0.49 0.40 (4) 1.15 (4) 0.82 0.45 1.16 1.17 1.77 0.66 0.46

Sej 2 0.75 0.33 (2) 1.51 0.27 (3) 1.03 0.67 0.80 0.75 0.32 0.16 Vardan 0.61 2.39 3.86 0.90 0.82 0.67 1.51 1.27 0.89 0.16 Varuna 0.75 1.65 2.82 1.35 0.39 1.40 0.42 (3) 0.22 (3) 0.44 0.29 Value in the bracket indicates ASV ranking of the genotypes, 1 indicates most stable, 2 indicates second most stable, and so on.


200

(a) (b)

(c) (d)

(e) (f)

(g) (h)


201

Figure 1. Bi-plot of IPCA1 against IPCA2 for oil content (a), protein (b), glucosinolate (c), linolenic acid (d), linoleic acid (e), oleic acid (f), eicosenoic acid (g), erucic acid (h), palmitic + stearic acid (i) and yield (j). 1=Basanti, 2= Bio 902, 3=CS 52, 4=GM 1, 5=GM 2, 6=Kranti, 7=Krishna, 8=P. Bahar, 9=P. Bold, 10=P. Jagannath, 11=PBR 91, 12=PBR 97, 13=PCR 7, 14=RCC 4, 15=RH 781, 16=RH 8113, 17=RH 819, 18=RH 30, 19=RL 1359, 20=RH 8812, 21=Rohini, 22=S. Asech, 23=Sej -2, 24=Vardan, 25=Varuna. ASV of the varieties showed stable performance for all the nine characters investigated. Except for linoleic, eicosenoic and palmitic + stearic acid, ASV and bi-plot analysis indicated different stable varieties. Nevertheless, of the five genotypes, at least 3 were also similar on the basis of these parameters for rest of the characters.

Three varieties, namely, Pusa Jagannath (glucosinolate, oleic acid, erucic acid and palmitic + stearic acid), RH 30 (protein, linolenic acid, oleic acid and erucic acid) and RH 8812 (oil, protein, oleic acid and erucic acid), showed stability for maximum of 4 characters. Two (Pusa Bold, RL 1359) and eleven varieties (Kranti, Pusa Bahar, PBR 91, PBR 97, PCR 7, RH 8113, RH 819, Rohini, Sanjucta Asech, Sej-2 and Varuna) respectively exhibited stable performance for 3 and 2 of the 9 quality characters studied whereas Basanti, Bio 92, GM 2, Krishna, were stable for only one quality character. Further 5 varieties (PBR 97, Krishna, RH 781, PCR 7 and RL 1359) showed stable performance for seed yield over environments. Using the Eberhart and Russel model of stability analysis, Chauhan et al. (2010) reported oil and protein content to be fairly stable across the environments. The results of the present investigation revealed that G × E interaction largely affected oil, seed meal quality

and seed yield of Indian mustard to varying levels, depending on the character. Therefore, data from this study would be helpful for breeders to adopt the appropriate selection strategy to accommodate the influence of G × E interaction on the expression of quality and seed yield. Stable genotypes for various characters identified could be used in the breeding program to improve seed yield and quality. Selection for target trait should be based on multi-environment/location evaluation. REFERENCES Ackman RG, Eaton CA, Sipos JC, Loew FM,

Hancock D (1977). Comparison of fatty acids from high levels of erucic acid of RSO and partially hydrogenated fish oil in non-human primate species in a short-term exploratory study. Nutr. Diet 25: 170–185.

Brar KS, Mittal VP, Singh Paramjit, Yadav DK (2009). Bi-plot analysis of diallel data set for seed yield of Indian mustard, Brassica juncea L. Czern & Coss. under south-western region of Punjab. J. Oilseeds Res. 26: 11-12.

Burgueno J, Crossa J, Vargas M (2002). SAS Programs for graphing GE and GGE bi plots. CIMMYT, D.F. CIMMYT. www.cimmyt.org/english/wps/biometrics.

(i) (j)


202

Chauhan JS, Tyagi Poonam, Tyagi MK (2002). Inheritance of erucic acid in two crosses of Indian mustard (Brassica juncea L.). SABRAO J. Breed. Genet. 34: 19-26. Chauhan J S, Bhadauria V P S, Singh M, Singh K H and Kumar A (2007). Quality characteristics and their interrelationships in Indian rapeseed-mustard (Brassica sp.) varieties. Indian J. Agric. Sci. 77: 616–620.

Chauhan JS, Meena SS, Singh Maharaj, Singh KH (2010). Estimating genotype × environment interaction and stability parameters for oil and seed meal quality, seed yield and its contributing characters in Indian mustard (Brassica juncea). Indian J. Agric. Sci. 80: 110-115.

Dhillon SS, Singh K, Brar KS (1999). Stability analysis of elite strains in Indian mustard. (in) New horizons for an old crop. Proceedings of the 10th International Rapeseed Congress. Canberra, Australia, pp. 1-5.

Fenewick GR, Heaney RK, Mullin J (1983). Glucosinolates and their breakdown products in food and food plants.CRC Crit. Rev. Food Sci. Nutr. 18: 123–201.

Gabriel KR (1971). The biplot-graphical display of matrices with applications to principal component analysis. Biometrika 58: 453-467.

Gauch HG (1992a). Introduction. In: H. G. Gauch, ed. Statistical analysis of regional yield trials: AMMI analysis of factorial designs. Elsevier Science Publisher B.V., Amsterdam, pp. 1-14.

Gauch HG (1992b). AMMI and related models. In: H. G. Gauch, ed. Statistical analysis of regional yield trials: AMMI analysis of factorial designs. Elsevier Science Publisher B.V., Amsterdam, pp. 53-110.

Kempton RA (1984). The use of bi-plots in interpreting variety by environment interactions. J. Agric. Sci. 103: 123-135.

Kumar Satyanshu, Singh AK, Kumar M, Yadav SK, Chauhan JS, Kumar PR (2003). Standardization of near-infrared reflectance spectroscopy (NIRS) for determination of seed oil and protein contents in rapeseed-mustard. J. Food Sci. Technol. 40: 306-309.

Kumar Satyanshu, Yadav SK, Chauhan JS, Singh AK, Khan NA, Kumar PR (2004). Total glucosinolate estimate by complex formation between glucosinolates and tetrachloropalladate (II) using ELISA reader. J. Food Sci. Technol. 41:63-65.

Sah RP, Kumar Arun, Haider ZA, Ghosh J, Verma N, Kumar Rahul (2009). Stability study in Indian mustard (Brassica juncea L.). In: Book of abstracts. National Seminar on Designing Crops for Changing Climate. Indian Society of Genetics and Plant Breeding, BAU,Ranchi, Jharkhand, India. p. 204.

Singh KH, Srivastava KK, Chauhan JS, Kumar A (2006). Genetic divergence and stability analysis in Indian mustard Brassica juncea L. Czern. & Coss. J. Oilseeds Res. 23: 151-155.

Singh et al. (2013)


INTERRELATIONSHIPS AMONG MORPHOLOGICAL AND SEEDLING CHARACTERS IN F5 PROGENIES OF INDIAN MUSTARD (Brassica juncea L.)

V.V. SINGH1*, A. RAJORIA2, J.S. CHAUHAN1, M.L. MEENA1 and S. KUMAR1

1Directorate of Rapeseed-Mustard Research, Sewar, Bharatpur (Raj)-321303, India

2 Institutes of Agricultural Sciences, Bundelkhand University, Jhansi, India *Corresponding author’s email: [email protected]

SUMMARY

One hundred and eighty progenies (F5) of Indian mustard were evaluated under rainfed conditions for 18 morphological and seedling characters. The estimates of heritability were high (>75%) for plant height, fresh seedling weight, seedling elongation, vigor index and root-shoot ratio. Genetic advance was moderate to low for most of the characters studied. Seed yield per plant had positive and significant correlation with plant height, primary branches per plant, main shoot length, fruiting zone length, siliquae per plant, siliquae on main shoot, siliqua length, seeds per siliqua, germination percentage and root-shoot ratio. On the basis of significantly superior mean values over best check, progenies BPR-1195-B-27-19, BPR-1187-92-31, BPR-1195-B-22-14, BPR-1153-33-6 and BPR-1153-1-1 were selected for seed yield and characterized for morphological and seedling traits. These progenies will be tested in multilocational trials after stabilization and will also be used in hybridization programs as parents to develop high-yielding varieties. Keywords: F5 progenies, morphological and seedling characters, augmented block design, genetic advance, correlations, vigor index, Indian mustard

Manuscript received: March 21, 2012; Decision on manuscript: October 17, 2012; Manuscript accepted: February 19, 2013 © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Indian mustard [Brassica juncea (L) Czern & coss, 2n=4x=36] is an annual, rabi oilseed crop and possesses a unique position by virtue of its high oil content (37-42%). This species is a natural amphidiploid between Brassica campestris (2n=20) and Brassica nigra (2n=16). Central Asia-Himalayas is a primary center of diversity for this species with migration to China, India and Caucasus (Hemingway, 1976). In India, B. juncea is a predominant species, which accounts for nearly 80% of the production area of the oilseed Brassica. The crop is economically important for its oil, which is largely used for cooking. The oil is also used in

the manufacture of salad dressing and vegetable oils, confectionary fats and shortenings. The fatty acids and derivatives are widely used for industrial purposes. Oil may not be used directly in the preparation of paints, linoleum or inks due to its semi-drying nature. However, its derivatives and fatty acids can be used for this purpose. It is also used in the production of rubber, in tanning industries, as lubricants and in the manufacture of soaps and detergents. Oilseed cake or meal, which is a by-product during the extraction of oil from the seeds, is an important source of protein feed for animals while the leaves of the plant are used as fodder for cattle.

Most oilseeds have lower yield potential than cereal crops and Indian mustard (B. juncea)

RESEARCH ARTICLE



204

is no exception. There is lack of adequate genetic variability for morphological and seedling traits and a high degree of genotype x environment interaction. Concerted breeding efforts are required to bring out favorable change in yielding ability of this crop. About 25% of the total rapeseed-mustard area is rainfed (Chauhan et al., 2011), which suffers from moisture stress at one or more stages of crop growth, thus lowering productivity. Besides high yield, the varieties developed and recommended for cultivation should have the traits that ensure stability under a diverse set of conditions. Kant and Tomar (1995) reported positive correlation of siliquae characteristics, seed size, seed weight with seed germination and seedling vigor. High germination percentage and vigor resulted in an excellent crop having adequate plant population. Gontia and Awasthi (1999) observed profound effects of these traits on ultimate yield. Seed yield is a complex character governed by polygenes and is the ultimate result of a complex relationship among different yield components, along with morphological and seedling characteristics. Thus, improvement in seed yield/plant is very difficult to achieve by direct selection. Therefore, it is imperative to determine major morphological and seedling components affecting seed yield and also to know the relationship among these components. MATERIALS AND METHODS The material for the present investigation consisted of 180 advanced F5 progenies of Indian mustard. These progenies were derived from five different crosses involving exotic (Australian and Chinese germplasm) and Indian genotypes. Thirty-seven genotypes (BPR 1195) were derived from EC 564641 x (NUDHYJ 3L88 x PCR 7) cross, 23 progenies (BPR 1191) were obtained from the cross EC 564647 x (NUDHYJ 3L188x PCR 7), 85 progenies (identity BPR 1153) were derivatives of JNO 33 x BPR 897-4-8 cross, 22 progenies (identity BPR 1187) were selected from EC 564649 x (NUDHYJ 3L83 x Varuna) and 13 progenies (identity BPR 1165) were from EC 552583 x BPR 897-4-3. During 2007-08, the F1s obtained from these five crosses were advanced to F2,

from where individual plants were selected on the basis of oil content (%), 1000 seed weight and phenotypic expression of the plant under rainfed conditions. Individual plants selected from F2 generation were grown as single plant progeny rows in the F3 generation. Generation advancement to F4 was made through individual plants and evaluation was done under rainfed conditions along with standard checks. Total number of progenies evaluated during 2009-10 in F4 generation was >300. On the basis of oil content (%), 1000-seed weight and yield superiority, 180 progenies were selected and advanced to F5 generation through individual plants. F5 progenies thus selected were evaluated during 2010-2011 rabi in an augmented block design (Federer, 1956). The material was divided into five blocks consisting of 36 progenies and 4 check varieties, namely, Bio 902, PBR 97, RB 50 and RH 819. In each block, progenies and check varieties were sown in 2- row plots of 5m length, spaced 30 cm apart with plant-to-plant spacing of 10 cm achieved by thinning at 15-20 days after sowing. The recommended package of practices was followed to raise a healthy crop. The experiment was conducted under rainfed conditions on conserved moisture. Initial moisture content of experimental area at sowing time was 9.9% (0-15 cm depth) and 14.9% (15-30 cm depth). After 50 days, moisture content reduced to 7.5% (0-15 depth) and 9.1% (15-30 cm depth). However, 61.4 mm rainfall during the crop growth period from October 2010 to March 2011 was received.

Ten plants were randomly taken from each plot to record plant height, primary branches per plant, main shoot length, fruiting zone length, siliquae per plant, siliquae on main shoot, siliqua length, seeds per siliqua, seed yield per plant, 1000-seed weight and oil content. The laboratory experiment to study germination percentage, root length, shoot length, fresh seedling weight, seedling elongation, vigour index and root-shoot ratio was carried out in petri dishes (9 cm diameter). Fifty seeds of each genotype were placed in petri dish lined with two layers of Whatman filter paper moistened with distilled water four replications. The petri dishes were kept in B.O.D. incubator (25±1oC and 75% RH). The mean data were subjected to analysis of variance

Singh et al. (2013)

205

(Federer, 1956; Sharma, 1998) using SPAD (Abhishek et al., 2004) software. Genetic parameters and simple correlations in all possible combinations were worked out as per standard procedure (Burton, 1952; Johnson et al., 1955) using Microsoft Office Excel 2007. Vigour index was calculated according to Abdul-Baki and Anderson (1973) using the following formula:

Vigor index = germination (%) x seedling elongation

RESULTS AND DISCUSSION The analysis of variance revealed significant differences among progenies for all the characters. except germination percentage. This indicates the scope of selection in breeding programs for seed yield improvement. Seed yield per plant, followed by siliquae per plant, and fresh seedling weight, showed higher estimates of genotypic as well as phenotypic coefficients of variation as compared with other characters (Table 1) suggesting that selection may be exercised directly for these traits. Earlier workers (Kumar et al., 2001, Singh et al., 2009) made similar reports.

The estimates of heritability in the present investigation were of higher magnitude (>75%) for plant height, fresh seedling weight, seedling elongation, vigor index and root-shoot ratio, which are in conformity with earlier reports (Singh et al., 2009, Singh et al., 2011). Contrary to these results, Meena et al., (2008) observed low heritability for seed yield per plant. Genetic advance was highest for seed yield per plant, siliquae per plant, fresh seedling weight, root length and vigor index. High genetic advance for these traits was also reported by (Patel et al., 2006, Singh et al., 2010). These findings indicate a good scope for developing genotypes having deep root system, siliquae per plant, seed yield per plant, which would be vigorous under water stress conditions. High heritability and high genetic advance were recorded for root length, fresh seedling weight and vigor index. Similar results indicating effectiveness of direct selection for these

characters have been reported by Singh et al. (2009).

Seed yield per plant showed positive and significant correlation with plant height, primary branches per plant, main shoot length, fruiting zone length, siliquae per plant, siliquae on main shoot, siliqua length, seeds per siliqua, germination percentage and root-shoot ratio (Table 2). These results were in agreement with earlier reports of Kardam and Singh (2005), Meena et al. (2008) and Singh et al. (2011). Positive and significant association of 1000 seed weight with plant height, siliquae length and fresh seedling weight supported earlier findings (Singh et al., 2011). This indicated that bold seed size had profound effect on vigour of the crop under rainfed conditions.

Oil content exhibited negative association with most of the traits of economic importance. Similar results were also reported earlier by (Singh et al., 2009, Singh et al., 2010). Germination percentage exhibited positive and significant correlation with primary branches per plant, main shoot length, fruiting zone length, siliquae per plant, siliquae on main shoot, root length, seedling elongation, vigor index and root-shoot ratio. Ozer and Dogru (1999) also reported a similar trend of association. Root length exhibited positive and significant correlation with main shoot length, oil content, seedling elongation, vigour index and root-shoot ratio. Shoot length exhibited positive and significant correlation with root length, seedling elongation, vigour index and root-shoot ratio, suggesting that more growth depends on a deep root system. Seedling elongation exhibited positive and significant correlation with main shoot length, fruiting zone length, siliquae on main shoot, seeds per siliqua, root length and shoot length. Vigor index showed positive and significant correlation with main shoot length, siliquae per plant, siliquae on main shoot, fruiting zone length, seeds per siliqua and seedling elongation. The results revealed that seedling vigour under rainfed conditions would be the result of high germination, rapid growth of root and shoot and increased seedling elongation.

Root-shoot ratio exhibited positive and significant correlation with main shoot length, siliquae on main shoot, fruiting zone length,


206

seeds per siliqua, seed yield, germination percentage, root length and shoot length. The results of the present investigation are consistent

with earlier reports (Meena et al., 2008, Singh et al., 2009).

Table 1. Range, mean, genotypic and phenotypic coefficients of variation, broad-sense heritability and genetic advance as percentage of mean for morphological and seedling traits in Indian mustard.

Character Mean Range GCV PCV h2 GA (%)

Plant height (cm) 188.9 134.5 - 221.5 6.2 6.9 79.3 11.3

Primary branches per plant (no.)

5.4 3.5 - 8.4 11.2 13.6 67.3 18.9

Main shoot length (cm) 62.7 47.4 - 82.5 7.6 10.2 55.9 11.7

Fruiting zone length (cm) 67.3 37.6 - 92.6 8.8 11.8 55.6 13.6

Siliquae per plant (no.) 180.2 41.9 - 398.5 23.3 29.8 60.9 37.5

Siliquae on main shoot (no.) 43.8 33.1 - 58.7 8.6 11.0 61.3 13.9

Siliquae length (cm) 4.2 2.7 - 5.9 9.4 11.5 66.6 15.8

Seeds per siliqua (no.) 13.7 10.0 - 18.1 6.6 8.0 68.0 11.2

Seed yield per plant (g) 10.1 2.9 - 21.5 25.4 31.3 65.8 42.4

1000-seed weight (g) 4.6 3.3 - 6.4 8.9 11.8 56.6 13.8

Oil content (%) 42.6 38.8 - 44.2 1.4 1.8 59.6 2.2

Germination (%) 91.4 83.2 - 98.2 * * * *

Root length (cm) 7.7 4.3 - 10.3 13.9 14.1 95.9 27.9

Shoot length (cm) 3.8 3.0 - 5.2 6.8 9.3 53.8 10.3

Fresh seedling weight (mg) 51.9 31.1 - 77.6 16.6 17.2 93.3 33.1

Seedling elongation 11.5 7.7 - 14.4 10.4 10.6 95.9 20.9

Vigor index 10.5 6.9 - 13.3 11.2 11.6 93.3 22.3

Root-shoot ratio 1.9 1.1 - 2.6 10.2 11.4 80 18.8

GCV= genotypic coefficient of variation, PCV= phenotypic coefficient of variation, h2= heritability in broad sense, GA= genetic advance expressed as percentage of mean.

Singh et al. (2013)

Table 2. Correlation coefficients between different morphological and seedling characters in India mustard

PH PB MSL FZL SPP SMS Sl L SPS SY SW OC GP RL SL FSW SE VI PB 0.118

MSL 0.205** 0.093

FZL 0.298** 0.62** 0.592**

SPP 0.071 0.63** 0.493** 0.787**

SMS 0.075 0.263** 0.615** 0.563** 0.59**

Sl L 0.423** 0.265** -0.03 0.154** 0.071 -0.12

SPS -0.02 0.291** 0.445** 0.463** 0.452** 0.297** 0.012

SY 0.151* 0.597** 0.513** 0.656** 0.75** 0.58** 0.154* 0.398**

SW 0.236** 0.08 0 0.044 -0.02 0.029 0.429** -0.18 0.073

OC 0.137 -0.34** -0.11 -0.353** -0.45** -0.25** 0.107 -0.35** -0.372** 0.079

GP 0.06 0.196** 0.27** 0.271** -0.229** 0.16** 0.051 0.132 0.208** 0.061 -0.23**

RL 0.023 -0.03 0.039** 0.08 0.049 0.146 0.074 0.141 0.093 0.053 0.031 0.217**

SL -0.12 0.055 -0.01 0.071 0.086 0.038 -0.1 0.052 -0.084 -0.06 -0.09 0.018 0.271**

FSW -0.09 0.048 -0.01 0.033 0.46 -0.03 -0.01 0.006 -0.067 0.181* -0.01 -0.01 0.069 0.065

SE -0.05 0.07 0.29** 0.17** 0.135 0.215** -0.01 0.207** 0.07 0.031 -0.04 0.213** 0.84** 0.608** 0.103

VI -0.02 0.131 0.338** 0.228** 0.192* 0.234** 0.019 0.222** 0.14 0.046 -0.1 0.483** 0.826** 0.535** 0.078 0.954**

RSR 0.044 0.041 0.377** 0.156** 0.097 0.239** 0.062 0.224** 0.171* 0.095 0.029 0.239** 0.738** -0.25** 0.083 0.608** 0.622**

PH=plant height, PB=number of primary branches, MSL=main shoot length, FZL=fruiting zone length, SPP= siliquae per plant, SMS= siliquae on main shoot, SlL=siliqua length, SPS=seeds per siliqua, SY=seed yield per plant, SW=1000-seed weight, OC=oil content (%), GP=germination percentage, RL=root length, SL=shoot length, FSW=fresh seedling weight ,SE=seedling elongation, VI=vigor index,.RSR=root-shoot ratio.

Singh et al. (2013)

Among other morphological characters, plant height exhibited positive and significant correlation with main shoot length, fruiting zone length, siliqua length and 1000-seed weight, this conforms the findings of Meena et al., (2008) and Singh et al., (2011). Main shoot length exhibited positive and significant correlation with siliquae per plant, siliquae on main shoot, fruiting zone length, seeds per siliqua, seed yield per plant, germination percentage, root length, seedling elongation, vigor index and root-shoot ratio. Similar results have been made by Yadav et al., (2000) and Mishra et al., (2004). Primary branches per plant showed positive and significant correlation with siliquae per plant, siliquae on main shoot length, fruiting zone length, seeds per siliqua, seed yield per plant and germination percentage. These results supported the report of Kardam and Singh (2005). Siliquae per plant exhibited positive and significant correlation with main shoot length, primary branches, siliquae on main shoot, fruiting zone length, seeds per siliqua, seed yield per plant and germination percentage. Similar results were earlier reported by Kardam and Singh (2005) and Singh et al., (2010). Siliquae on the main shoot showed positive and significant correlation with main shoot length, primary branches, siliquae per plant, fruiting zone length, seed yield per plant, germination percentage, seedling elongation, vigor index and root-shoot ratio. The findings of the present investigation are in accordance with earlier studies (Mishra et al., 2004). Fruiting zone length exhibited positive and significant correlation with plant height, main shoot length, primary branches, siliquae per plant, main shoot length, siliqua length,

seeds per siliqua, seed yield per plant, germination percentage, seedling elongation, vigour index and root-shoot ratio. Earlier workers (Singh et al., 2010) obtained similar results. Siliqua length showed positive and significant correlation with plant height, primary branches, fruiting zone length, seed yield per plant and 1000-seed weight. Gupta and Thakur (2002) reported a similar pattern of association with siliqua length. Seeds per siliqua exhibited positive and significant correlation with main shoot length, primary branches, siliquae per plant, main shoot length, fruiting zone length, seed yield, seedling elongation, vigor index and root-shoot ratio. These results are in conformity with the earlier reports of Kardam and Singh (2005), and Singh et al. (2011).

In conclusion, the study revealed adequate variability for 18 morphological and seedling traits. All the significant characters had high heritability but some had low to moderate genetic variation. Further, plant height, primary branches per plant, main shoot length, fruiting zone length, siliquae per plant, siliquae on main shoot, siliqua length, seeds per siliqua, germination percentage and root-shoot ratio exhibited significant association with seed yield per plant. On the basis of significantly superior mean values over the best check for seed yield per plant, progenies, BPR-1195-B-27-19, BPR-1187-92-31, BPR-1195-B-22-14, BPR-1153-33-6 and BPR-1153-1-1 were selected for seed yield and characterized for morphological and seedling traits (Table-3). These progenies will be tested in multilocational trials after stabilization and will also be used in hybridization programs as parents to develop high-yielding varieties.

Singh et al. (2013)

Table 3. Characterization of significantly superior F5 progenies for morphological and seedling characters in Indian mustard.

Progeny Seed yield

per plant (g)

Plant height (cm)

Primary branches per plant (no.)

Main shoot length (cm)

Fruiting zone length (cm)

Siliquae per plant (no.)

Siliquae on main shoot (no.)

Siliqua length (cm)

Seeds per silaqua (no.)

BPR-1195-B-27-19 21.5 181.5 6.1 72.1 87 342.9 57.5 3.4 14.9 BPR-1187-92-31 19.2 203.2 6.4 71.1 74.5 242.8 53.5 4.4 15 BPR-1195-B-22-14 18.6 204.1 6.6 72.5 73 245.1 49.7 4.5 14.5 BPR-1153-33-6 18.5 190.8 7.4 63.1 75.9 209.6 44.9 4.9 15.2 BPR-1153-1-1 18.4 186.1 7.2 68.7 88.7 376.2 55.9 5.3 15.6 BPR-1187-107-33 18.1 187.6 5.6 66.1 72.7 200.4 48.5 3.8 15.5 Bio 902 (best check) 11.02 195 5.3 71.9 74.5 210.1 47.2 3.6 13.3

CD (5%) 6.69 21.6 1.5 15.4 19.2 121.6 10.8 1.0 2.2 Progeny 1000-seed

weight (g)

Oil content (%)

Germination (%)

Root length (cm)

Shoot length (cm)

Fresh seedling weight (mg)

Seedling elongation

Vigor index

Root-shoot ratio

BPR-1195-B-27-19 3.5 41.1 85.9 7.1 3.7 57.3 10.8 9.3 1.9 BPR-1187-92-31 4.1 42.8 90.9 8.8 3.9 43.4 12.7 11.5 2.2 BPR-1195-B-22-14 4.5 43.3 90.7 7.8 3.6 54.8 11.4 10.5 2.1 BPR-1153-33-6 5.9 42.3 96.9 8.2 3.9 43.4 12.2 11.8 2.1 BPR-1153-1-1 5.5 42.4 89.9 8.4 3.7 77.1 12.1 10.8 2.3 BPR-1187-107-33 4.4 43.2 89.9 7.4 3.5 31.2 10.9 9.8 2.1 Best Check 4.3 42.3 91.5 7.8 3.8 45.7 11.6 10.6 2 CD (5%) 1.3 1.8 9.5 0.1 0.9 8.3 0.9 1.1 0.5

Singh et al. (2013)

ACKNOWLEDGEMENTS We express our sincere thanks to the director of DRMR, Bharatpur for providing us the necessary facilities. REFERENCES Abdul-Baki AA, Anderson JD (1973). Vigour

determination in soybeans seed by multiple criteria. Crop Sci. 13:360-363.

Abhishek R, Rajendar Prasad, Gupta VK (2004). Computer aided construction and analysis of augmented designs. J. Indian Soc. Agric. Stat. 57 (special volume): 320-344.

Burton GW (1952). Quantitative inheritance of grasses. Proc. 6th Int. Grassland Cong. p. 227-283.

Chauhan JS, Singh KH, Singh VV, Kumar S (2011). Hundred years of rapeseed-mustard breeding in India: accomplishments and future strategies. Indian J. Agric. Sci. 81: 1093-1109.

Federer WT (1956). Augmented design. Hawain Planters Record. 20: 191 – 207.

Gontia AS, Awasthi MK (1999). Effect of seed grading and size on various seed vigour attributes, morphological characters and seed yield in soybean genotypes. Seed Res. 27(1): 25-30.

Gupta SK, Thakur HL (2002). Variability, correlation and path analysis of seed yield in Brassica carinata. J. Oilseed Res. 19 (2): 231.

Hemingway JS (1976). Mustard: Brassica spp. and Sinapis alba (Cruciferae). In NW Simmonds, ed. Evolution of crop plants. Longmans, London. pp 56-59.

Johnson HW, Robinson HF, Comstock RE (1955). Estimate of genetic and environmental variability in soybean. Agron. J. 47: 314-318.

Kant K, Tomar SRS (1995). Effect of seed size on germination, vigour and field emergence in mustard (Brassica juncea L. Czern and Coss.) cv. Pusa bold. Seed Res. 23 (1): 40-42.

Kardam DK, Singh VV (2005). Correlation and path coefficient analysis in Indian mustard (Brassica juncea L.) grown under rainfed condition. J. Spic. Arom. Crops. 14 (1):56-60.

Kumar A, Chauhan Y S, Kumar K, Maurya KN (2001). Genetic variability for seed yield and component traits in newly collected germplasm of yellow sarson [Brassica rapa

(L.)] syn. Brassica campestris (L.) spp. Yellow Sarson. Indian .J. Plant Genet. Resour. 14 (3): 396-397.

Meena SS, Yadav R, Singh VV (2008). Genetic variability for seed and seedling traits in the advance breeding lines of Indian mustard [Brassica juncea (L.) Czern & Coss]. Seed Res. 36 (2): 152-156

Mishra AK, Ratan Shiv, Kumar A. (2004). Germplasm evaluation of Indian mustard. J Oilseeds Res. 21 (2): 248-251.

Ozer H, Dogru O (1999). Relationship between yield and components on currently improved spring rapeseed [Brassica napus spp. olifera (L.)] cultivars. Turkey J. Agric. For.. (23): 603-607.

Patel JM, Patel KM, Patel CJ, Prajapati KP (2006). Genetic parameter and inter relationship analysis in Indian mustard [Brassica juncea (L.) Czern and Coss.] J. Oil Seed Res. 23 (2):159-160.

Sharma JR (1998). Statistical and biometrical techniques in plant breeding. New Age International Publishers, New Delhi.

Singh VV, Meena SS, Yadav Rajbir, Shrivastav Nidhi, Kumar A. (2010). Genetic variation and character association in F6 progenies of interspecific crosses between Brassica juncea and Brassica carinata. Indian J. Plant Genet. Resour.. 23 (2): 168-171.

Singh VV, Singh M, Chauhan JS, Kumar S (2011). Development and evaluation of full sib progenies in Indian mustard (Brassica juncea L.) for moisture stress conditions. Indian J. Genet. 71:78-81.

Singh VV, Singh Sudheer, Verma Vandna, Meena SS, Kumar A. (2009). Genetic variability for seedling traits in Indian mustard under moisture stress condition. Indian J. Plant Genet. Resour. 22 (1): 46-49.

Yadav SK, Chauhan JS, Kumar PR, Pareek, Tyagi MK (2000). Evaluation and analysis for variability in two ecotypes of Indian mustard germplasm under rainfed condition. Indian J. Plant Genet. Resour. 13 (2):177-182.


UNDERSTANDING AND OVERCOMING SEED DORMANCY IN EGGPLANT (Solanum melongena L.) BREEDING LINES

J. ZDRAVKOVIC1*, N. RISTIC1, Z. GIREK1, S. PAVLOVIC1, N. PAVLOVIC1, R.

PAVLOVIC2 and M. ZDRAVKOVIC1

1Institute for Vegetable Crops, Serbia

2University of Kragujevac, Faculty of Agronomy, Cacak, Serbia *Corresponding author’s email: [email protected]

SUMMARY The seed dormancy of six eggplant (Solanum melongena L.) genotypes were studied. Two genotypes (33 and 34) were breeding lines while the other four (1-00261, 2-02619, 7-00568, 12–00823) belonged to the exotic germplasm of the Institute for Vegetable Crops in Smederevska Palanka. In order to break the dormancy, the seed was treated with: (1) low temperature, at 4 0C for 96 hours (4 days), 72 hours (3 days) and 48 hours (2 days), continually; (2) hormones, gibberellic acid (GA3) in three concentrations: 5 ml/100 ml, 15 ml/100 ml, and 25 ml/100 ml for 24 hours and after that seeds were allowed to germinate; (3) chemicals, with potassium nitrate KNO3 solution at concentrations 0.5, 1 and 1.5% for 24 hours. Non-treated seed was the control. The trial was set according to two-factorial ANOVA, with four repetitions. The AMMI model was applied for analyzing the interaction of genotypes and treatments for increasing seed germination (break of dormancy). Genotype 12 reacted most favorably to treatments KNO3 1%, KNO3 1.5%, GA3 15 mg/100 ml and GA3 25mg/100 ml. Three genotypes (33, 7 and 12) represent the least stable genotypes, since they reacted positively to all treatments, and most intensively to treatment with GA3 25 ml/100 ml. The low temperature treatment increased germination but not in the expected range, since the highest impact had the low temperature regime for 48 hours. Keywords: Solanum melongena, seed dormancy, gibberellic acid, potassium nitrate, low temperature, germination

Manuscript received: April 15, 2012; Decision on manuscript November 30; Manuscript accepted: December 30 @ Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Dormancy or inaction of seed is the impossibility of expressing maximum germination in a certain period of time. This was studied previously by many researchers (Bawley, 1997). Seed dormancy in eggplant (Solanum melongena L.) is a rare trait for bred genotypes (Yogeesha et al., 2006) and seems to be positively correlated to the content of ABA. Seed dormancy after collecting is a big problem

in studying this vegetable. It is considered that the seed maturity during seed collection (the day after anthesis [DAA]) is an important factor for resolving dormancy of eggplant (Demir et al., 2002, Passam et al., 2010; Agbo and Nwosu 2009). In this study, seed dormancy was observed in eggplant fruits, which were picked in different maturity phases. Fully mature fruits had seeds with 90% of germination, while half mature fruits had 70%, and 90% 3 months after shelving.

RESEARCH ARTICLE



212

Seed dormancy in eggplant is under genetic control (Padmini et al., 2008). This trait has cytoplasm (maternal effects) in the F1 generation, and monogenic and two gene complementary and recessive gene action can be studied in F2 generations of some cross combinations. Seed dormancy is a complex trait and there are types of dormancy based on their mode of action (Finch-Savage and Leubner-Matzger, 2006): physiological (embryo dormancy, dormancy of endosperm and seedcoat); morphological; morpho-physiological, physical and combined (complex) dormancy.

Breaking the dormancy of seed with exogenous hormones and low temperatures increases germination energy (the first count) and total seed germination (the final count). Treatment of fresh eggplant seed with acetone solution of gibberellic acid (GA3) increased germination from 81% to 100%, and germination energy (the first count) from 68 to 98% (Zhinzhang et al., 1993). By applying treatments with exogenous hormones, chemical agents and low temperatures (Yogeesha et al., 2006), partially proved the mechanism of overcoming dormancy in eggplant seed in two breeding lines. Vujaković et al. (2000) used, besides hormone and chemical treatment, a temperature treatment (low and high temperatures and the combination of both) and scarification in the hilum area. Germination after this treatment was the lowest. Similar to findings in sunflower (Bajaj et al., 2009), germination could depend on different external conditions under which the eggplant fruits were formed. In order to overcome seed dormancy in S. melongena, producers used various and alternative preparations in order to initiate germination (Sharma and Sharma, 2010).

This paper is attempts to answer two questions: (1) are germination energy (the first count) and germination (the final count) stable in bred eggplant genotypes; and (2) is seed dormancy in bred genotypes stable and is it possible that certain treatments do not give any effect?

The variety Domaci srednje dugi (DSD) is widespread both in the open field and glasshouse production of eggplant in Serbia. This variety has dormant seed. New lines with

DSD genes in its pedigree were compared with eggplant lines from the Institute for Vegetable Crops. Treatments using hormones, chemicals and low temperatures were done to investigate the level of dormancy in the new lines and assess the quality of seed. MATERIALS AND METHODS Genotypes Six genotypes were studied: 2 lines and 4 genotypes. Line 33 and 34 were derived from breeding combination DSD x Junior. Both lines have large fruits, colored intensively. Line 34 is not dormant, while line 33 has dormancy at DSD level. Four genotypes originated from the eggplant germplasm collection of the Institute for Vegetable Crops: 1 - 00261, 2 - 02619, 7 - 00568, and 12 - 00823. Genotype 2 does not have seed dormancy, while others are dormant. The seed of genotype 1 is strongly dormant. Seed extraction Seed was extracted by hand from mature fruits 73 days after flowering. Manual seed extraction means cutting fruits and washing them in water. The seed was dried until use in the trial, and stored in paper bags. Low temperature treatments began 5 days later. Treatments Low temperature treatment was conducted at 4 oC for 96 hours (4 days), 72 hours (3 days) and 48 hours (2 days) continually. Hormone treatment with gibberellic acid (GA3) was conducted using three concentrations: 5 ml/100 ml, 15 ml/100 ml and 25 ml/100 ml for 24 hours after which seeds were allowed to germinate. Chemical treatment was done using KNO3 at concentrations 0.5, 1, and 1.5% solution for 24 hours. The untreated seed was the control. Germination The germination test was conducted using the standard ISTA (1985) method: 100 seeds were sown on filter paper in Petri dishes with 4

Zdravkovic et al. (2013)

213

replicates for all treatments and control. The germination energy (first count) was determined 7 days and total germination (final count) 14 days after sowing. Statistics The two-factorial ANOVA with four replicates based on LSD (the lowest significant difference) for comparison of differences among genotypes and treatments was used. In order to analyze the interaction of genotype and treatment for increased germination energy and germination (breaking of dormancy), the AMMI model (additive mean effects and multiactive interaction) (Zobel et al. 1988) was used. For analyzing the residual from the interaction, we used the PCA model (Gauch, 1988). The AMMI’s stability value (ASV) was calculated in order to rank genotypes in terms of stability using the formula suggested by Purchase (1997) as shown below: AMMI stability value (ASV) =

2] [IPCA2 + 2] (IPCA1 SIPCA2[SSIPCA1/S scorescore

Where SS= sum of squares IPCA1= interaction principal component axis 1 IPCA 2 = interaction principal component axis 2 The R software (version 2.14.0, 2011) was used for data analysis. RESULTS

The results of the study showed that genotypes can be classified into three groups:

1) genotypes without dormancy: genotypes 2 and 34 with average germination values of 96.75 and 81%, respectively;

2) genotypes with deep dormancy: genotype 1 (0% germination, except after treatment with GA3 25 ml/100 ml, when germination was 7.25%); and

3) treated dormant genotypes: genotypes 12, 33 and 7 with germination of 59%, 18.5% and 21.5%, respectively (Table 2). The situation was similar with

respect to the average germination energy values of the genotypes studied.

As for treatment with KNO3, only genotype was significant, whereas the treatments and genotype x treatment interaction did not cause any change in germination energy of the eggplant seed. In treatments with GA3, all genotypes and treatments and their interactions were significant. The treatment of low temperature had a significant effect for genotype, treatment and their interaction for germination energy (Table 1).

For total seed germination of different eggplant genotypes, genotype, the KNO3 treatment and their interaction were significant. GA3, the genotype and their interaction significantly influenced germination. Low temperature treatment was highly significant (99%) for the treatment and the genotype, while the interaction of these two factors was significant (95%). Stability analysis The ANOVA showed that the average values of genotype treatments and their interaction were highly significant (P < 0.01: Table 3). The ratio of SS treatment of germination energy of 6 lines tested with 10 treatments showed that 82% from the total SS can be influenced by genotypes, 10.03% by treatment and 6.03% by interaction genotype x treatments. The results clearly showed a significant difference in reaction of genotypes to treatments. The highest variability for the observed traits was caused by a high SS.

The results have also proved that 81.96% of the total SS of the seed germination was influenced by genotype, 9.45% by treatment and 7.50% by interaction.

The results of the variance analysis showed 78.08% SS of the genotype x treatment interaction. Furthermore, the second main component explains 9.63% of the SS of interaction; other components included 12.29% SS genotype x treatment.

The results of our analysis have also revealed that the first main component for germination (IPCA1) has 89.5% SS of interaction G x T. Similarly, the second main component explains 7.8% SS of interaction in


214

the third 0.8%. Other components were not significant.

As for the energy of germination, 4 IPCA, and for germination, 3 IPCA values were highly significant (P < 0.01). For both traits, IPCA1 was higher than IPCA2 and other interaction components (Table 3). AMMI biplot analysis of germination energy and germination of seed The analysis of the main components for germination energy have proved that genotype 33 reacts best to treatments GA3 15 ml/100 ml and GA3 25 ml/100 ml. Genotype 7 reacted best to 1% KNO3 treatment. Genotype 2 did not react to treatments and belongs to constant genotypes (Figure 1a).

The AMMI 1 biplot proved the stability of genotypes 1, 2 and 34 for germination energy (first count). Genotype 1 did not react to treatments while genotypes 2 and 34 did not react to treatments since they have good germination energy (higher than average) and they are not genotypes with dormant seed.

The analysis of the main components for germination according to biplots showed genotypes with the best reaction to treatments. Genotype 12 interacted with KNO3 1.5% and KNO3 1% to increase germination, while genotypes 33 and 7 had the best interaction with GA3 15 ml/100 ml and GA3 25 ml/100 ml. Genotypes 34 and 2, near the x-axis were the most stable genotypes that did not react to the treatments (Figure 2a).

The genotypes in the positive part of the biplot, farthest from the coordinate beginning and x-axis, were the genotypes not shown that the first main component influence by treatment, and these are the genotypes with the lowest and highest germination: 1, 2 and 34. Genotypes 33, 7 and 12 (dormant compared with control) were

in the negative part of the biplot. Genotype 12 had the best reaction to treatments KNO3 1%, KNO3 1.5%, GA3 15 ml/100 ml, GA3 25 ml/100 ml. Three genotypes, 33, 7 and 12, were the least stable genotypes, since the most intensive treatment with GA3 0.25% had no influence on them. The treatments above the X-axis were those with low impact on seed dormancy of dormant genotypes. The farthest were the control and the low temperature treatment (Figure 2b). AMMI parameter of stability for germination energy (the first count) and germination (the final count) Treatments proved highly variable for both factors and treatment x genotype interaction The highest values of ASV coefficient were seen in treatments with highest impact to break dormancy. The ASV was the highest for GA3 25 ml/100 ml (range 10) and GA3 15 ml/100 ml (range 9) for germination energy. These were the treatments with the highest variability, i.e. they caused the breakage of seed dormancy. The lowest impact on seed dormancy had treatment KNO3 1.5 (range 1) for germination energy.

The ASV for seed germination was highest for GA3 25 ml/100 ml, GA3 15 ml/100 ml (the same as that for germination energy). These were the treatments with the highest variability and caused the breakage of seed dormancy. The lowest impact to germination was the KNO3 0.5% treatment (Table 4.)

The highest variability for germination energy per treatment had genotype 12. The treatments of this genotype had the highest impact on breaking dormancy. Treatments were least effective on genotype 33. This is not ideal and necessitates research to find other ways to break dormancy (Table 5).


215

Table 1. Average value of germination energy (first count after 7 days).

Treatment Genotypes 34 12 7 33 2 1

KNO3 0.5% 97.50 50.75 39.75 35.00 81.75 0 KNO3 1% 97.75 56.75 52.00 48.75 69.75 0 KNO3 1.5% 95.50 51.50 42.00 49.00 83.75 0 LSD 0.05 13.13 LSD T 0.05 9.28ns LSD 0.05 G x T 22.73ns 0.01 16.15 0.01 11.42ns 0.01 27.96ns

GA3 5 ml/100 ml 92.25 41.00 37.50 37.00 84.00 0

GA3 15 ml/100 ml 96.50 70.00 69.50 74.50 84.50 2.25

GA3 25 ml/100 ml 99.25 73.75 69.25 85.50 90.75 7.25 LSD 0.05 5.38 LSD T 0.05 3.82 LSD 0.05 G x T 9.35

0.01 6.64 0.01 4.69 0.01 11.51

Low temperatures 96 h 89.50 43.25 27.75 26.25 83.50 0 Low temperatures 72 h 90.25 26.50 34.50 25.50 84.75 0 Low temperatures 48 h 92.25 48.75 42.75 35.75 86.75 0 LSD 0.05 5.62 LSD T 0.05 3.98 LSD 0.05 G x T 9.74

0.01 6.93 0.01 4.90 0.01 11.99

Control 66.00 3.00 17.75 16.00 66.75 0

Table 2. Average value of germination (final count after 14 days). Treatment Genotypes

34 12 7 33 2 1 KNO3 0.5% 98.50 84.75 54.00 48.00 99.00 0 KNO3 1% 99.25 91.50 62.50 59.75 98.75 0 KNO3 1.5% 98.75 92.00 58.80 67.25 99.25 0 LSD G 0.05 4.68 LSD T 0.05 3.31 LSD 0.05 G x T 8.12

0.01 5.77 0.01 4.08 0.01 9.99

GA3 5 ml/100 ml 93.50 75.00 41.50 41.75 100 0

GA3 15 ml/100 ml 98.75 97.50 78.50 83.00 99.75 2.50

GA3 25 ml/100 ml 99.25 98.00 94.25 98.00 99.75 7.15 LSD G 0.05 5.23 LSD T 0.05 3.70 LSD 0.05 G x T 9.07

0.01 6.44 0.01 4.56 0.01 11.16

Low temperatures 96 h 92.25 58.75 31.50 27.75 100 0 Low temperatures 72 h 91.50 50.00 36.50 28.25 99.25 0 Low temperatures 48 h 94.00 63.00 45.00 36.50 99.25 0 LSD G0.05 5.14 LSD T 0.05 3.64 LSD 0.05 G x T 8.91*

0.01 6.33 0.01 4.47 0.01 10.96*

Control 81.00 59.25 21.50 18.50 96.75 0


216

Table 3. Analysis of variance of main components for germination energy (first count) and germination (final count).

Source of variation

Germination energy Germination df SS MS F SS MS F

Treatments 9 25520 2836 88.23** 30483 3387 125.9*** Repetitions 30 964 32 1.2141ns 807 27 1.5ns Genotypes 5 208629 41726 1576.21** 264312 52862 2945.3***

G x T 45 15343 341 12.88** 24191 538 29.9** IPCA1 13 11979.05 921.47 34.81** 21640.30 1664.60 92.75** IPCA2 11 1429.85 129.97 4.91** 1898.8 172.60 9.62** IPCA3 9 1128.62 125.4 4.74** 448.10 49.80 2.77** IPCA4 7 565.69 80.81 3.05** 197.90 28.30 1.57ns. Residue 8 239.46 47.89 1.81ns 5.90 0.74 0.04ns Error 150 3971 26 2692 18 Environment 59 249492 4228.68 318986 5406.54 Total 239 254427 1064.55 322485 1349.31 ** indicate significant at P<0.01, df - degrees of freedom, SS - sum of squares, MS – mean of squares, IPCA= interaction principal component axis.

Table 4. Value of AMMI parameters stability (ASV) treatment for germination energy (first count) and germination (final count) of eggplant seed. Treatment Germination energy PC1 PC2 ASV Germination PC1 PC2 ASV

Mean Rank Value Rank Mean Rank Mean Rank Control 32.96 10 2.05 -3.71 17.58 6 46.17 10 3.35 0.88 38.20 8 GA3 5 48.63 7 1.14 0.71 9.53 4 58.63 6 1.11 1.02 12.74 2 hml/100

ml 15 66.21 2 -3.92 -0.07 32.83 9 76.67 2 -3.60 0.03 41.03 9

25 70.96 1 -4.17 -0.69 34.92 10 82.83 1 -5.10 -2.20 57.93 10 hours 48 51.04 5 0.94 0.46 7.93 2 56.29 7 1.56 -1.30 17.80 4

72 43.58 9 2.65 1.57 22.25 8 50.92 9 2.72 -2.60 31.16 6 96 45.04 8 2.24 -0.26 18.77 7 51.71 8 2.85 -0.50 32.51 7

KNO3 0.5 50.79 6 0.95 0.88 8.04 3 64.04 5 0.01 1.47 1.47 1 % 1 54.17 3 -1.60 0.66 13.42 5 68.54 4 -1.30 1.48 14.78 3 1.5 53.63 4 -0.28 0.44 2.42 1 69.29 3 -16.00 1.60 18.66 5

Table 5. Value of AMMI parameters of stability (ASV) of genotypes for germination energy (first count) and germination (final count) of eggplant seed.

Genotype Germination energy PC1 PC2 ASV Germination PC1 PC2 ASV

Mean Rank Value Rank Mean Rank Value Rank 1 0.95 6 3.33 -2.57 28.05 5 1.03 6 3.49 -0.92 39.76 4

2 81.63 2 3.29 0.37 27.60 4 76.98 3 -1.75 4.01 20.32 1

7 43.28 5 -2.27 1.04 19.01 3 99.18 1 4.01 -0.34 45.73 5

12 49.35 3 -1.71 -1.48 14.42 1 50.83 5 -4.85 -0.89 55.32 6

33 43.33 4 -4.58 -0.32 38.37 6 94.68 2 2.59 0.13 29.56 2

34 91.68 1 1.93 2.97 16.43 2 52.38 4 -3.47 -1.99 39.64 3


217

Figure 1. (a) AMMI2 biplot (IPCA1-IPCA2) for energy of germination (first count) of seed with vectors of genotypes; (b) AMMI1 (germination–IPCA1) biplot for germination energy (final count).

(a)

(a)

(b)


218

Figure 2. a) AMMI 2 biplot (IPCA1-IPCA2) with germination energy (first count) with vectors of genotypes b) AMMI1 (germination –IPCA1) biplot for germination (final count) of eggplant seed. DISCUSSION

The aim of this research was to answer two questions: whether the germination energy and germination are stable in selected eggplant genotypes and whether seed dormancy for bred genotypes is stable and if it is possible to break dormancy by certain treatments.

Seed dormancy of eggplant was investigated including 4 exotic germplasm acessions. Although seed dormancy is considered to be a rare trait in selected genotypes (Yogeesha et al., 2006), in our study dormancy was found in one genotype, while another had high germination energy and germination just after the seed was collected. The analysis indicated that genotype, treatment and their interactions were highly significant. The highest significant interaction indicated the IPCA1 value of 89.5% of the total variability for seed germination.

Seed dormancy is determined by genetic control (Padmini et al., 2008), although understanding of regulation and breakage of dormancy should be consider the physiological mechanism of seed maturity in the days after anthesis (DAA) period. Finch-Savage and Leubner-Matzger (2006) found that seed dormancy happens due to the ratio of ABA and gibberellic acid concentrations, so the exogenous intake of gibberellins acid affected the deeply dormant seed of genotype 1.

Fruits were picked 73 days DAA in order to avoid the effect of non-mature seeds, since the maturity of seed at the moment of collection is a relevant factor for seed dormancy in eggplant, Demir et al. (2002), Agbo and Nwosu (2009). The best time for extraction was 55 DAA (Passam et al., 2010).

The treatment with KNO3 is significant only for the energy of germination in eggplant, while for the total germination of seed: genotype, KNO3 treatment and their interaction, were significant. Van Pijlen et al. (1995) and Geetharani and Ponnuswamy (2002) found that KNO3 increases the energy of germination and total germination in eggplant seed.

Many researchers found that potassium nitrate increases germination in eggplant (Jagadish 1993; Yogananda et al. 2004; Geetharani and Ponnuswamy 2002) while pre-treatment of seed with potassium nitrate increases the energy of germination, germination and vigour of plants of eggplant (Barlow and Haigh 1987; Geetharani and Ponnuswamy 2002; Van Pijlen et al. 1995 - tomato; Jagadish 1993 - onion; Bradford et al. 1990 – pepper; Kumar 2005).

ASV (AMMI stability value) was highest for GA3 25 ml/100 ml and GA3 15 ml/100ml for the energy of germination and total germination. These were the treatments that increased variability and broke the dormancy of seed.

(b)


219

Treatments with gibberellic acid significantly influenced all genotypes, energy of germination and total germination. Gibberellic acid causes the softness of seed endosperm in Solanaceae (Leubner-Metzger 2003). Only the highest concentration of GA3 (25ml/100ml) caused 7.15% of germination in genotype 1. Other treatments in this study did not break the dormancy in genotype 1. This genotype needs to be treated in another way in order to break dormancy. The fact that the gibberellic acid is a strong hormone that increases germination and breaks the dormancy in some species is consistent with the findings of Solanki and Joshi (1985). They significantly increased the germination in cauliflower and tomato by applying gibberellic acid for 12 hours on seed of these species. Arun et al. (1997) sprayed 200 ppm of gibberellic acid on eggplant seeds and increased germination, while Yogananda et al. (2004) obtained the same results for pepper seed. Pre-treatment of eggplant seed with 200 ppm of GA3 for 6 hours increased the seed quality (Kumar 2005).

Our results indicated (based on the analysis of the main components (IPCA)), that the energy of germination of genotype 33 responds best with GA3 15 ml/100 ml and GA3 25 ml/100ml treatments. Genotype 7 interacts best with KNO3 1% treatment which is consistent with results of Van Pijlen et al. (1995), Geetharani and Ponnuswamy (2002).

AMMI biplot for seed germination indicated which treatments were the best for each genotype. Genotype 12 interacted with KNO3 1.5% best and KNO3 1% in increasing germination, and genotypes 33 and 7 reacted best with GA3 15 ml/100 ml and GA3 25 ml/100 ml.

Low temperature treatment was effective, but did not produce germination level in the ideal range. In this study, the low temperature treatment gave the best results when applied for 48 hours, which is also the shortest period. Low temperature treatments had significant effects for genotypes, seed germination and their interaction. The AMMI parameter of stability (ASV) for germination energy and germination for treatments with low temperatures were weaker comparing to treatments with gibberellic acid. Low

temperature treatment had an effect on genotypes regarding breaking dormancy and increasing of seed germination. This was assumed to be due to divergence of eggplant genotypes. Breakage dormancy with low temperature treatment of humid seed of Solanum species from Solanaceae took 1 to 5 days (Hayati et al., 2005). Demir et al. (2004) found positive and significant effects on emergence after a few days of heating seed at 20% humidity for 5 eggplant genotypes. Baskin and Baskin (2004) found that breakage of dormancy in Solanaceae can be performed by using endogenous hormones for Angiospermae.

ACKNOWLEDGEMENTS

This research was supported by the Ministry of Science and Technology, Republic of Serbia, through Project TR-31059:“Integrating Biotechnology Approach in Breeding Vegetable Crops for Sustainable Agricultural Systems” REFERENCES

Agbo CU, Nwosu P U (2009). The influence of seed

processing and drying techniques at varying maturity stages of Solanum melongena fruits on their germination and dormancy. Afr. J. Biotech. 8(18): 4529-4538.

Arun KG, Hariwallabh, Jandial KC (1997). Effect of GA3 on growth and yield of brinjal. Haryana J. Hort. Science 26: 143-145.

Bajaj RK, Kaur S, Dhillon SK, Charma SR (2009). Variability and association studies and plant characters affecting germinability under different environments in sunflowers (Helianthus annuus L.). SABRAO J. Breed. Genet. 41(2): 137-148.

Barlow EWR, Haigh AM (1987). Effect of seed priming on the emergence, growth and yield of UC 82B tomatoes in the field. Acta Horticult., 200: 153-164.

Baskin JM, Baskin CC (2004). A classification system for seed dormancy. Seed Sci. Res. 14: 1–16.

Bawley DJ (1997). Seed germination and dormancy. Plant Cell 9: 1055-1066.

Bradford KJ, Steiner JJ, Travatha SE (1990). Seed priming influence on germination and


220

emergence of pepper seed lots. Crop Sci. 30 (3): 718-721.

Demir I, Ermis S, Okcu G, Matthews S (2004). Vigour tests for predicting seedling emergence of aubergine (Solanum melongena L.) seed lots. 27th ISTA Congress Seed Symposium, Budapest, Hungary, May 17-19. pp: 13

Demir I, Mavi K, Seermanli T, Ozkoban M (2002). Seed development and maturation in aburgine (Solanum melongena L.). Gartenbauwissenschaft 67 (4), S: 148-154.

Finch-Savage W.E., and Leubner-Matzger G. (2006). Seed dormancy and control of germination. New Phytol. 171: 501-523.

Gauch HG (1988). Model selection and validation for yield trials with interaction. Biometrics 46: 1488-1500.

Geetharani, P. Ponnuswamy, A.S., (2002). Effect of seed hardening in tomato to increase the germination and vigour. M.Sc. (Agri.) Thesis, Tamil Nadu Agricultural University, Coimbatore

Hayati NE, Sukprakarn S, Juntakool S (2005). Seed Germination Enhancement in Solanum stramonifolium and Solanum torvum. Kasetsart J. (Nat. Sci.) 39: 368 - 376

International Seed Testing Association (ISTA) (1985). Rules for seed testing. Seed Sci. Technol. 13: 300-520.

Jagadish GV (1993). Seed storability, ageing and effect of pre-sowing treatment on the performance of some vegetable crops. M.Sc. (Agri.) Thesis, University of Agricultural Sciences, Dharwad.

Kumar S (2005). Influence of pre-sowing seed treatment and seed pelleting on storability in brinjal (Solanum melongena L.). M.Sc. (Agri.) Thesis. University of Agricultural Sciences, Dharwad.

Leubner-Metzger G (2003). Hormonal and molecular events during seed dormancy release and germination, In Nicolás G., Bradford K.J., Comê D. and Pritchard H.W. The Biology of Seeds: Recent Research Advances, CABI publishing. pp. 101–112.

Padmini K, Yogeesha HS, Naik LB (2008). Genetics of fresh seed dormancy in brinjal (Solanum melongena L.). Indian J. Agri. Sci. 78, (4): 304-308.

Passam H.C. Teodoropoluolu S., Karanossa T.m Karapanos IC (2010). Influence of harvest time and after-ripening on the

seed quality of eggplant. Sci. Hortic. 125, 3: 518-520.

Purchase JL (1997). Parametric analysis to describe G x E interaction and yield stability in winter wheat. PhD Thesis. Department of Agronomy, Faculty of Agriculture. University of the Orange Free State Bloemfontein, South Africa.

R Development Core Team (2011). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org/

Sharma S, Sharma K (2010). Study on germination and growth behavior of brinjal Solanum melongena var BR 112 in admiration to effect of C.M.L. (Country Made Liquor). Nature and Science 8(5): 163- 165.

Solanki, SS, Joshi RP (1985). Studies on invigoration of vegetable seeds of tomato (Lycopersicon esculentum L.) and cauliflower (Brassica oleraceae L.). Progressive Hortic. 17 (3): 267-269.

Vanpijlen, JG, Kraak HL, Bino RJ, De Vos CHR (1995). Effects of ageing and osmo-priming on germination characteristics and chromosome aberrations of tomato (Lycopersicon esculentum Mill) seeds. Seed Sci. Technol. 23: 823-830.

Vujaković M, Milošević M, Zlokoloca M, Balešević-Tubić S, Nikolić Z (2000). Primena različitih treatmenta za prekid mirovanja kod semana plavog patlidžana (Solanum melongena L.) III JUSEM Zlatibor 28. Maj-1. Jun, Srbija: 98 (sr)

Yogananda DK, Vyakaranahal BS, Sekhargouda M (2004). Effect of seed invigoration with growth regulators and micronutrients on germination and seedling vigour of bell pepper cv. California wonder. Karnataka J. Agr. Sci. 17 (4): 811-813.

Yogeesha HS, Upreti KK, Padmini K, Bhanuprakash K, Murti GSR (2006). Mechanism of seed dormancy in eggplant (Solanum melongena L.). Seed Sci. Technol. 34: 319-325.

Zhinzhang Y, Enrang Z, Binkui Z (1993). Research on optimum pretreatment methods of eggplant seeds with exotic hormones. J. Shanghai Agr. College 11(4): 291-296.

Zobel RW, Wright MS, Gauch HG (1988). Statistical analysis of a yield trial. Agron J. 80: 388-393.

http://www.r-project.org/�

http://www.r-project.org/�


EVALUATING THE TOLERANCE OF BREAD WHEAT GENOTYPES FOR POST-ANTHESIS WATER STRESS: WATER USE EFFICIENCY

AND STRESS TOLERANCE INDICES

A. BAHRANI1, A. MADANI2٭ and H. MADANI3

1Islamic Azad University, Ramhormoz Branch, Ramhormoz, Iran 2Islamic Azad University, Gonabad Branch, Gonabad, Iran

3Islamic Azad University, Arak Branch, Arak, Iran *Corresponding author’s email: [email protected]

SUMMARY

Bread wheat (Triticum aestivum L.) is one of the most important cereals being grown in many areas of the world. However, there is limited information about water use efficiency and stress drought tolerance to determine the tolerance of wheat cultivars to water stress. A 2-year field study was conducted with the objective of determining the effect of different rates of nitrogen (N) (control, 80 kg N ha-1, and 160 kg N ha-1) and two irrigation treatments (nonstress and post-anthesis water stress) on three bread wheat cultivars. Water deficiency markedly affected grain yield, straw yield and harvest index. Under water deficit (WD) grain yield was reduced by 58%, showing that wheat cultivars were sensitive to drought stress. Tolerance index (TOL) was lower under WD than under well-watered (WW) conditions in both years. Stress tolerance index (STI), mean productivity (MP), and TOL indices were efficient in evaluating sensitivity to drought stress. Grain yield under stress condition showed positive correlation with STI, MP, and TOL indices. Water use efficiency and irrigation water use efficiency increased with increasing N rates in the WW treatment and decreased in WD treatment. In conclusion, if irrigation is for maximum yields, fertilization should be too; but if irrigation is limited, fertilizer supply should be adjusted accordingly to prevent N loss and pollution of the environment. Keywords: Wheat, nitrogen, post-anthesis water deficit, stress tolerance indices, water use efficiency

Manuscript received: March 12, 2012; Decision on manuscript: July 2, 2012; Manuscript accepted: April 4, 2013 © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Drought limits agricultural production of many crops around the world. Wheat is particularly susceptible to water stress and it can be affected at different growing stages with the most important occurring during grain-filling stage (Blum, 1998). In many Mediterranean-type climates such as the ones that exist in many areas of Iran, rainfall decreases and evaporation and temperature increase in the spring, when

wheat enters the grain-filling stage (Ahmadi and Baker, 2001; Fathi, 2005). Consequently, wheat plants often experience severe water deficit during grain filling, causing significant decrease in grain yield (Spiertz et al., 2006; Barnabas et al., 2008; Blum, 1998).

The responses of different wheat genotypes in stressed and non-stressed environments were described by Fernandez (1992), who described four types of responses: first, genotypes with similar response under

RESEARCH ARTICLE



222

stress and nonstress conditions (Class A); second, genotypes with suitable response in nonstress conditions (Class B); third, genotypes with high yield in stress condition (Class C) and fourth, genotypes with poor response in both conditions (Class D).

A number of different indices were also developed to determine the stress sensitivity of the different genotypes. Fischer and Maurer (1978) proposed the stress susceptibility index (SSI), which indicates higher tolerance for drought when SSI <1. Relative tolerance or susceptibility of cultivars can be determined by comparing their SSI (Andarzian, 2000). On the basis of SSI, A and B class genotypes can generally be distinguished more than the other groups. This index cannot separate genotypes of B and C groups and includes them in the same group. Another index which was introduced by Rosielle and Hamblin (1981) was the tolerance index (TOL) and the mean productivity (MP) index, which involve yield difference and mean productivity in stress and nonstress conditions, respectively. High values of TOL indicate higher sensitivity to drought; therefore, lower rates are favorable. In addition, more tolerant cultivars have higher MP rates and it is possible to separate B and C genotype groups using MP and TOL indices. However, separation of C from A is impossible with these indices. Therefore, Fernandez (1992) suggested the stress tolerance index (STI) to distinguish genotypes with high yield either under stress or nonstress conditions. Higher STI values show higher tolerance for drought and more yield potential (Fernandez 1992). In addition, STI can be used to distinguish group A genotypes from B and C and STI is considered a more relevant index than TOL and MP. Fernandez (1992) also introduced the geometrical mean productivity index (GMP). Indices showing high correlation with grain yield were introduced as the best indices because of their ability to separate genotypes with high yield, both in stress and nonstress conditions.

In Mediterranean environments, crop canopy development in winter is slow and rain occurs as frequent and small events. Soil water evaporation may account for 30 to 60% of seasonal evapotranspiration (Cooper et al., 1983; French and Schultz, 1984; Siddique et al., 1990; Zhang et al., 1998). Thus, agronomic practices

that reduce soil water evaporation via a larger plant canopy and early ground cover and at the same time increase the crop's ability to extract soil water may increase water use efficiency (WUE). Nitrogen (N) deficiency is another major constraint to canopy development in the Mediterranean region (Anderson, 1985). Crop responses to N fertilization depend on the level of water availability (Pala et al., 1996). Application of fertilizers not only increases plant shoot and root growth (Brown et al., 1987) but also increases evapotranspiration through a larger root system and greater extraction of stored water (Cooper et al., 1987). In addition, a large and an earlier canopy cover resulting from the application of N can reduce soil water evaporation and increase crop WUE (Zhang et al. 1998). In a study done in Kentucky, Corak et al. (1991) found that WUE increased with the addition of N fertilizer and the use of hairy vetch (Visia villosa Roth) residues.

Wheat grown under the Mediterranean climate is generally affected during the grain-filling period when there is soil water deficit that restricts N uptake from the soil. The objectives of this study were (i) to compare responses of three wheat cultivars to different levels of N and post-anthesis water deficit in terms of grain yield, water use, and WUE and (ii) to compare the different tolerance indices under water stress using the same three wheat cultivars. MATERIALS AND METHODS Experimental set up The experiment was conducted at Shiraz Agricultural Research Station, Iran (52o 36' E, 29o 33' N) during the growing seasons of 2006-2007 and 2007-2008. The cultivars used in the study to determine their response to fertilization and water stress were Shiraz (early maturity), Marvdasht (medium maturity), and Chamran (medium maturity). Soil was sampled pre-planting at a depth of 30 cm and before fertilizer application. The soil characteristics were determined at the Shiraz Soil Testing Laboratory. The soil sample contained 3.2 g organic matter kg-1, 0.3 mg N kg-1, 11 mg P (Olsen) kg-1, and 270 mg exchangeable K kg-1.

Bahrani et al. (2013)

223

The pH was 7.9 (1:2 water). Weather data (rainfall and average temperature) were recorded daily in the experimental site and reported as mean monthly data for the 2 years of the study together with the 30-year averages for temperature and rainfall (data not shown). Crop management and experimental design The experimental design was a split-split plot with irrigation treatments as main plots, fertilizer treatments as split plot and the cultivars as split-split plots with four replications. The experimental plots were 8 by 1.5 m with 6 rows 0.2 m apart. The irrigation treatments were I1 (nonstressed) and I2 (post-anthesis water- stressed plots with 65% field capacity). The fertilization treatments were 0, 80, and 180 kg N ha-1 applied pre-planting in the form of (NH4)2SO4 (NPK, 20.5-0-0). The N rates were according to soil available N. In addition, P and K were applied at a rate of 60 kg ha-1 and 100 kg ha-1 (pre-planting) in the form of superphosphate and K2SO4, respectively. Wheat cultivars were sown on 11 November 2006 and 14 November 2007. Irrigation of main plot was measured volumetrically using a field-calibrated gypsum block. Six gypsum blocks were installed in each replication randomly. Measurements were made post anthesis, prior to the irrigation, during the growing period. The change in soil moisture was measured weekly to a depth of 30 cm. A drip irrigation system was used to avoid runoff losses.

Measurements The total aboveground biomass at heading was determined by cutting at ground level in 0.3 m2 quadrants per plot (Zadoks growth stage 60). At maturity (Zadoks growth stage 95), the central four rows from each plot were harvested for grain yield and the total biomass was weighed. Harvest index (HI) was determined by dividing grain yield by the total biomass. All samples were dried in an air-forced oven at 70 0C for 48 stress indices.

A number of different drought stress indices were determined to evaluate their use in the selection of wheat cultivars:

Stress susceptibility index (SSI) was calculated with the equation described by Fisher and Maurer (1978):

SIYpYsSSI )/(1−

= ,

−=

YpYsSI 1 .

where SI is stress intensity, Yp is yield of each genotype in nonstress conditions, Ys is potential yield of each genotype in stress condition, pY is mean yield of all genotypes in nonstress

condition, and sY is mean yield of all genotypes in stress condition. Tolerance index (TOL) and mean productivity (MP) index, which are the differences in yield and mean productivity under stress and nonstress conditions, respectively were calculated using the formula of Rosielle and Hambelin (1981):

YsYpTOL −= ,

2YsYpMP +

= .

Stress tolerance index (STI) was calculated according to Fernandez (1992):

2)())((

pYYSYp

pYYs

YsYs

YpYpSTI =

=

P and S stands for nonstress and stress

conditions, respectively. Geometrical mean productivity (GMP) was calculated according to Fernandez (1992):

))(( YpYsGMP =

Water use efficiency (WUE) was

calculated according to Aase and Pikul (1995):

ETYWUE /= where Y is the yield of the crop, either in total harvestable biomass or marketed yield, and ET is evapotranspiration of water from the soil surface, plant leaves, and through the stomata (transpiration).


224

Evapotranspiration (ETP) was determined according to the equation: ETPcrop = ETPpan × Kc where ETPcrop, ETPpan and Kc are crop evapotranspiration, pan evaporation and crop coefficient, respectively. Statistical analysis The data were analyzed using ANOVA following a factorial design with four replications per treatment combination. More specifically, the experiment was set up as a randomized complete block design for the irrigation treatments (main plots), with N levels as split plots and cultivars as split-split plots. A combined analysis over the growing season was carried out. Duncan’s multiple range test (DMRT) was used for testing the differences between treatment means. The significance level of all hypotheses testing was preset at P<0.05. All statistical analyses were performed using SAS version 9.2 (SAS Institute 1995). RESULTS Water deficit affected grain yield, straw yield and harvest index (Table 1). Under water deficit, grain yield was reduced by 58%, showing the sensitivity of wheat to drought stress conditions (Table 2). The reduction was mainly due to reduction in grain weight rather than in grain number (data not shown). The cultivars showed differences in grain yield in the different treatments (Table 2), indicating that yield-potential of cultivars differed in well-watered and stress conditions. TOL index was lower in Chamran than in the other cultivars in both years (Table 2), indicating that Chamran was more tolerant of drought stress than Marvdasht and Shiraz. Between cultivars, MP index of Shiraz was higher than that of other cultivars. The highest SSI value belonged to WD treatment (Table 2). Therefore, SSI like MP and TOL, were suitable in evaluating the sensitivity to drought stress. Chamran showed the lowest SSI index compared with the other cultivars and had the lowest sensitivity to drought stress in

both years. Therefore, it can be deduced that this index was not efficient in the identification of cultivars tolerant of drought stress. Since there was significant interaction between cultivar and water regime (Table 1) it was expected that the different cultivars would show different sensitivities to water stress, which was not observed in other indices. In addition, the use of STI showed different responses among different cultivars. TOL and SSI indices showed that Chamran had lower sensitivity and the STI and MP indices showed that this cultivar was more sensitive to drought stress. Therefore, STI was more appropriate for the selection of cultivars with drought tolerance. The results of GMP were significantly similar to those of STI. Therefore, it can be concluded that STI and GMP indices are more accurate for the selection of tolerant cultivars under drought condition. Grain yield in stress condition showed positive correlation with STI, MP and TOL indices (Table 3). The highest correlation was between grain yield in stress condition and STI index. In addition, there was a high correlation between TOL and MP with grain yield under stress conditions, indicating that tolerant cultivars can be indentified using this indices. STI was more effective than the GMP and STI in separating the stable and more productive cultivars. Water use efficiency Water use efficiency increased with increasing N rates in WW treatments but decreased in WD treatments (Table 4). On WW treatments, the 160 kg N ha -1 rate was sufficient for maximum WUE and the 80 kg N ha -1 rate was sufficient for WD treatment. On WD, the N fertilizer treatment did not affect WUE in 2007-2008. In 2006-2007, there was a trend towards increasing grain yield up to 160 kg N ha-1, more N resulted in significantly higher WUE. In 2006-2007, on adequately fertilized treatments, WUE was highest on WW and lowest on WD treatment. In 2007-2008, however, efficiencies were highest on WD treatment and lowest on WW treatment. The differences in WUEs in WD treatment between the two growing seasons resulted from less water use on this treatment in 2007-2008


225

than in 2006-2007. The higher WUE in 2006-2007 may have been at least partially due to differences in temperature and rainfall distribution during the two seasons. Temperatures during grain filling and maturity periods were above normal in 2007-2008 and below normal in 2006-2007. The cooler, moist conditions in 2006-2007 allowed normal grain filling and maturity on WD treatment and reduced the effects of stress during that period. Drought conditions hastened maturity in this treatment in 2007-2008, resulting in reduced grain and test weights and, consequently, lower WUE. WUE for Chamran increased when soil moisture and/or N were in limited supply during the grain-filling period. When N and soil moisture were probably abundantly available, no significant difference in WUE were observed between Shiraz and Marvdasht.

Irrigation water use efficiency Irrigation water use efficiency increased with increasing N rates up to 160 kg N ha -1 on WW treatment in 2006-2007 (Table 4). In WD treatment, applied N had little effect on yield because the soil contained enough N for yields as high as the water supply would allow. Without N application, N was a limiting factor when irrigation was moderate. Both applied N and water were required for substantial yield increases from either variable. For the most efficient use of both N and water, the supply of one should be adjusted to that of the other. If moderate irrigation is possible for maximum yields, fertilization should be too; but if irrigation is limited, fertilizer supply should be limited accordingly.

Table 1. Results of analysis of variance combined across years, Irrigation, fertilizer and wheat cultivars to both years.

Source df Grain yield Straw yield Harvest index (%) kg ha-1 kg ha-1 Year (Y) 1 ** ** ns Water stress (S) 1 * * * Y×S 1 ns ns ns Cultivar (C) 2 * ** * Y×C 2 ns ns ns S×C 2 * * ns Y×S×C 2 ns ns ns CV, % 13 16 17

*,** Significant at 0.05 and 0.01 probability levels, respectively. ns=nonsignificant at P> 0.05.


226

Table 2. Stress susceptibility indices of the different cultivars used in this study and water deficit treatment for grain yield during the 2006-2007 and 2007-2008 growing seasons.

Table 3. Correlation coefficients of stress tolerance indices with grain yield under normal and stress conditions. STI SSI MP TOL Ys Yp

SSI 0.39 * MP 0.95** 0.55*

TOL -0.44* 0.52* -0.19 ns Ys 0.90** 0.09 ns 0.78** -0.76**

Yp 0.39* 0.89** 0.61** 0.59* 0.08ns GMP 0.98** 0.34 ns 0.95** 0.49* 0.90** 0.32 ns

* , ** Significant at 0.05 and 0.01 probability levels, respectively. Table 4. Water use, water use efficiency, and irrigation water use efficiency of three wheat cultivars as affected by N fertilizer and irrigation treatment in 2006-2008. Treatment Cultivars Water use Water use efficiency Irrigation water use efficiency kg N ha -1

WW WD WW WD Ave WW

WD Ave

(mm) (mm) kg grain m -3 kg grain m -3 (mm) (mm) 2006-2007 0 Shiraz 0.56 0.51 0.53 0.2 0.12 0.16 Marvdasht 0.58 0.57 0.57 0.24 0.15 0.2

Chamran 0.65 0.61 0.63 0.28 0.21 0.25 Ave 562 364 0.59 0.56 0.24 0.16 80 Shiraz 0.7 0.54 0.62 0.7 0.5 0.6 Marvdasht 0.76 0.6 0.68 0.74 0.52 0.63

Chamran 0.85 0.64 0.74 0.84 0.61 0.73

Cultivars Grain yield (kg ha-1) Stress tolerance indices 2006-2007 Yp Ys TOL MP SSI STI GMP Shiraz 8857 3253 5503 6053 1.07 0.50 5380 Marvdasht 7482 3170 4211 5326 0.97 0.40 4879 Chamran 6649 3066 3482 4858 0.91 0.34 4523 Mean 7662 3163 4398 5412 0.98 0.41 4927 LSD 1021 316 439 636 0.08 0.04 418 2007-2008 Shiraz 7066 2582 4383 4824 1.06 0.49 4284 Marvdasht 5966 2515 3349 4241 0.97 0.40 3884 Chamran 5300 2432 2766 3866 0.91 0.34 3597 Mean 6110 2509 3499 4310 0.98 0.41 3921 LSD 678 334 507 763 0.09 0.04 522


227

Ave 631 400 0.77 0.59 0.76 0.54 160 Shiraz 0.81 0.49 0.65 0.9 0.7 0.8 Marvdasht 0.9 0.55 0.72 0.95 0.75 0.85

Chamran 0.99 0.59 0.79 1.18 0.8 0.99 Ave 701 490 0.9 0.54 1.01 0.75 LSD (0.5) 0.16 0.1 0.13 0.11 2007-2008 0 Shiraz

0.45 0.4 0.48 -0.03 -0.13 -0.06

Marvdasht

0.48 0.44 0.46 0.1 0.02 0.05

Chamran

0.55 0.5 0.53 0.18 0.1 0.14

Ave 620 320 0.49 0.45

0.1 0.01

80 Shiraz

0.67 0.59 0.63 0.25 0.09 0.12

Marvdasht

0.7 0.63 0.67 0.29 0.13 0.21

Chamran

0.8 0.71 0.76 0.35 0.2 0.28

Ave 634 394 0.72 0.64

0.3 0.11

160 Shiraz

0.74 0.6 0.67 0.59 0.39 0.49

Marvdasht

0.8 0.65 0.73 0.67 0.56 0.61

Chamran

0.92 0.72 0.82 0.98 0.71 0.85

Ave 719 400 0.82 0.66

0.75 0.55

LSD (0.5) 0.15 0.12 0.07 0.09 *: WW=well-watered treatment, and WD=post anthesis water deficit. * , ** Significant at 0.05 and 0.01 probability levels, respectively. DISCUSSION Post-anthesis water stress affected grain yield, straw yield and harvest index, which correlates with several other studies (Siahpoosh et al., 2011; Nouri et al., 2011). One of the objectives of the present study was to determine the effect of post-anthesis water stress on the different stress tolerance indices that have been proposed to describe the behavior of a given genotype under stressed and nonstressed conditions, while most of the previous studies focused on water stress during the growth period of the crop (Mohammadi et al., 2010; Nouri et al., 2011). More tolerant cultivars show low TOL values (Fischer and Wood, 1979). TOL index was lower in Chamran than the other cultivars in both years, indicating that Chamran was more tolerant of water stress. Consequently, MP index showed that Shiraz had lower sensitivity to water deficit compared with Marvdasht and Chamran. Therefore, the MP index is better than TOL for evaluating tolerant cultivars to water stress as suggested by Sanjari et al. (2005). SSI

index was lower in Charman compared with the other two cultivars. Also, SSI was not an efficient index that can be used to distinguish cultivars tolerant of drought stress. In agreement with these findings, the result found by Moghadam and Hadizadeh (2002) showed that SSI index did not show great efficiency for selecting tolerant cultivars and it must be used only when a severe stress occurs. In other words, the SSI index must be used for determining the sensitive cultivars and not for selecting tolerant cultivars. The index GMP showed response similar to that of STI which indicates that STI and GMP indices are more important for the selection of tolerant cultivars under drought condition. These results are similar to findings from other studies (Moghadam and Hadizadeh 2002; Sanjari et al., 2005; Samizadeh, 1997; Mozafari, 1996; Nouri et al., 2011). NikhKhah (1999) also introduced GMP, MP and STI as the best indices for the selection of tolerant genotypes under stress condition, which was confirmed by other authors (Nouri et al., 2011; Sio-Se Mardeh et al., 2006).


228

Grain yield during stress conditions showed positive correlation with STI, MP and TOL indices. In addition, it was observed that there was a relationship among GMP, STI and MP. Similarly, Fernandez (1992) found a correlation between STI and GMP. Since, there was a significant correlation among GMP, STI, and MP, these indices can be used for the selection of cultivars tolerant of drought and STI and MP are the more efficient ones (Mohamadi et al., 2005, 2010; Sanjari et al., 2005; Nouri et al., 2011). Moghadam and Hadizadeh (2002) reported that STI was correlated with grain yield under both stress and optimum conditions (0.96 in stress and 0.5 in nonstress condition). Tari Nezhad (1999) also found no correlation between TOL and GMP and the use of MP is more effective than TOL in selecting tolerant genotypes. However, STI is more effective than the other indices in the separation of stable and more productive cultivars. NikhKhah (1999) also introduced GMP, MP and STI as the best indices for finding tolerant genotypes under drought stress condition and also reported that these indices have positive and significant correlation with grain yield in stress and nonstress conditions. However, Nouri et al. (2011) proposed that MP cannot be used to select high-yielding genotypes in both stressed and nonstressed environments, if the correlation of grain yield in contrasting environments is highly negative. Also, MP is related to yield under drought stress if it is not too severe and the difference between YR and YI is not too large.

Water use efficiency and irrigation water use efficiency Water use efficiency increased with increasing N rates in WW treatments but decreased in WD treatments. According to several researchers (Van Herwaarden et al., 1988; Angus and Van Herwaarden, 2001; Halvorson and Reule, 2004), N fertilization can improve WUE; however, high N levels can reduce yields through ''haying off'' due to excess WU in the pre-anthesis period, leaving insufficient water for the post-anthesis grain filling period. It is therefore important to balance N fertilization with the available seasonal water supply. Van den Boogaard et al.

(1996) and El Hafid et al. (1998) concluded that drought did not increase WUE.

In conclusion, GMP, STI, and MP were better indices to distinguish tolerance among the different cultivars for drought. Also, grain yield under stress condition showed positive correlation with STI, MP, and TOL indices. Thus, the different tolerance indices can be classified into distinct groups considering different concepts of drought tolerance, resistance and susceptibility under mild drought stress. In general, for the most efficient use of both N and water, the supply of one should be adjusted to that of the other. If irrigation is for maximum yields, fertilization should be too; but if irrigation is limited, the fertilizer supply should be adjusted accordingly to prevent N loss and pollution of the environment.

REFRENCES Aase JK, Pikul JL (1995). Crop and soil response to

long-term tillage practices in the Northern Great Plains. Agron. J. 87(4): 652–656.

Ahmadi A, Baker DA (2001). Effect of drought stress and drought tolerance heredity on nitrogen efficiency of winter wheat. J. Agric. Sci. 136(3): 257-269.

Andarzian B (2002). Study and comparison of wheat and barley under limiting irrigation in Ahwaz region. MSc thesis Shahid Chamran University. Ahvaz-Iran.

Anderson K (1985). Differences in response of winter cereal varieties to applied nitrogen in the field. I. Some factors affecting the variability of response between sites and seasons. Field Crops Res. 11: 353-367.

Angus JF, van Herwaarden AF (2001). Increasing water use and water use efficiency in dryland wheat. Agron. J. 93(2): 290–298.

Barnabas B, Jager K, Feher A (2008). The effect of drought and heat stress on reproductive processes in cereals. Plant Cell Environ. 31: 11-38.

Blum A. (1998). Improving wheat grain filling under stress by stem reserves mobilisation. Euphytic. 100: 77-83

Brown SC, Keatinge JDH, Gregory PJ, Cooper PJM (1987). Effects of fertilizer, variety and location on barley production under rainfed conditions in northern Syria. I. Root and shoot growth. Field Crops Res. 16(1): 53–66.


229

Cooper PJM, Gregory PJ, Keatinge JDH, Brown SC (1987). Effects of fertilizer, variety and location on barley production under rainfed conditions in Northern Syria. II. Soil water dynamics and crop water use. Field Crops Res. 16(1): 67–84.

Cooper PJM, Keatinge JDH, Hughes G (1983). Crop evapo-transpiration—a technique for calculation of its components by field measurements. Field Crops Res. 7: 299–312.

Corak S J, Frye WW, Smith MS (1991). Legume mulch and nitrogen fertilizer effects on soil water and corn production. Soil Sci. Soc. Am. J. 55(5): 1395–1400.

El Hafid R, Smith DH, Karrou M, Samir K (1998). Root and shoot growth, water use and water use efficiency of spring durum wheat under early-season drought. Agronomie 18(3): 181–195.

Fathi GH (2005). Effect of drought and nitrogen on remobilization of dry matter and nitrogen in six wheat cultivars. Iranian J. Agric. Sci. 36(5): 1093-1101.

Fernandez GCJ (1992). Effective selection criteria for assessing plant stress tolerance. In: Kuo. C.G (Ed). Proceedings of the International Symposium on Adaptation of Vegetables and Other Food Crops to Temperature Water Stress. Taiwan. pp. 257-270

Fischer RA, Maurer R (1978). Drought resistance in spring wheat cultivars. I. Grain yield responses. Aust. J. Agric Res. 29(5): 897-912.

Fischer RA, Wood JT (1979). Drought resistance in spring wheat cultivars. III. Yield association with morpho-physiological traits. Aust. J. Agric. Res. 30(6): 1001-1020.

French RJ, Schultz JE (1984). Water use efficiency of wheat in a Mediterranean-type environment. I. The relation between yield, water use and climate. Aust. J. Agric. Res. 35(6): 743–764.

Halvorson AD, Reule CA (1994). Nitrogen fertilizer requirements in an annual dry land cropping system. Agron. J. 86(2): 315-318.

Moghadam A, Hadizadeh H (2002). Responses of corn hybrids and their parental lines to drought using stress tolerance indices. Iranian J. Seed Seedling 18: 255-272.

Mohamadi A, Ahmadi J, Habibi D (2005). Selection indices for drought tolerance in bread wheat (Triticum aestivum). Iranian J. Agron. Plant Breed. 1(1): 47-62.

Mohammadi R, Armion M, Kahrizi D, Amri A (2010). Efficiency of screening techniques for evaluating durum wheat genotypes under mild drought conditions. Int. J. Plant Prod. 4(1): 11-24.

Mozafari K (1996). Path analysis in sunflower under drought stress and normal condition of irrigation. MSc. Thesis, Tehran University, Tehran-Iran.

NikhKhah H (1999). Evaluation of heritability of drought tolerance in bread cultivars. IN: Proceedings of the 5th Crop Sci. Congress. Karaj, Iran. pp. 345.

Nouri A, Etminan A, Teixeira da Silva JA, Mohammadi R (2011). Assessment of yield, yield-related traits and drought tolerance of durum wheat genotypes (Triticum turjidum var. durum Desf). Aust. J. Crop Sci. 5(1): 8-16.

Pala M, Matar A, Mazid A (1996). Assessment of the effects of environmental factors on the response of wheat to fertilizer in on-farm trials in a Mediterranean type environment. Exp. Agric. 32(3): 339–349.

Rosielle AT, Hamblin J (1981). Theoretical aspects of selection for yield in stress and non stress environment. Crop Sci. 21(6): 943-946.

Samizadeh L (1997). Study of phenotypic and genotypic diversity of quantitative traits and their relationships with chickpea yield. MSc thesis, Islamic Azad University, Karaj–Iran.

Sanjari AGH, Valizadeh M, Majidi A, Shiri MR (2005). Evaluation of modern wheat genotype of bread wheat to drought condition on grain yield and some agronomic and physiological traits. Agric. Sci. 16: 96-122.

Sio-Se Mardeh A, Ahmadi A, Poustini K, Mohammadi V (2006). Evaluation of drought resistance indices under various environmental conditions Field Crops Res. 98: 222–229.

Siddique KHM, Tennant D, Perry MW, Belford RK (1990). Water use and water use efficiency of old and modern wheat cultivars in a Mediterranean-type environment. Aust. J. Agric. Res. 41: 431–447.

Spiertz JHJ, Hamer RJ, Xu HY, Primo-Martin C, Don C, Van de Putten PEL (2006). Heat stress in wheat; effects on grain weight and quality within genotypes. Eur. J. Agric. 25(1): 89-95.

Tari Nezhad, A. (1999). Evaluating of responses of wheat lines to irrigation and drought stress. MSc thesis, Tabriz University, Iran. pp.163.

Van den Boogaard R, Veneklaas EJ, Peacock J M, Lambers H (1996). Yield and water use of wheat (Triticum aestivum) in a Mediterranean environment: Cultivar differences and sowing density effects. Plant Soil 181: 251–262.

Van Herwaarden AF, Farquhar GD, Angus JF, Richards RA, Howe GN (1998). ‘Haying-off’, the negative grain yield response of dryland wheat to nitrogen fertilizer. I. Biomass, grain


230

yield, and water use. Aust. J. Agric. Res. 49: 1067–1081.

Zhang H, Oweis TY, Garabet S, Pala M (1998). Water-use efficiency and transpiration efficiency of wheat under rain-fed and irrigation conditions in a Mediterranean environment. Plant Soil 201: 295–305.


EVALUATION OF GENETIC VARIATION AMONG FINGER MILLET (Eleusine coracana L. Gaertn) ACCESSIONS USING RAPD MARKERS

G. MALAMBANE1, P. JAISIL1,2*, J. SANITCHON1,2, B. SURIHARN1,2

and D. JOTHITYANGKOON1

1 Department of Plant Science and Agricultural Resources

Khon Kaen University, Khon Kaen, 40002Thailand 2Plant Breeding Research Center for Sustainable Agriculture, Khon Kaen University,

Khon Kaen 40002, Thailand *Corresponding author’s email: [email protected]

SUMMARY

A study was conducted to evaluate the genetic variation in finger millet (Eleusine coracana) using random amplified polymorphic DNA (RAPD) markers. Forty-four primers showing reliable polymorphism were used to produce a total of 295 bands of which 184 (62.4%) were polymorphic. The size range of bands was 400-2000 bp, with a number of bands per individual primer ranging from 4 to 14. The dendrogram constructed using the UPGMA method separated Eleusine coracana subsp. Africana, a wild relative (IE 4709) from the cultivated finger millet (Eleusine coracana subsp. coracana). The cultivated finger millet was further separated into five groups showing similar morphological traits. The results obtained from this study show high genetic variation among finger millet accessions. The study also demonstrates high reliability, ease of applicability and importance of RAPD markers in evaluating genetic variation among finger millet landraces. Keywords: Eleusine coracana, genetic diversity, random amplified polymorphic DNA, molecular markers, cluster analysis

Manuscript received: May 21, 2012; Decision on manuscript: November 24, 2012; Manuscript accepted: December 26, 2013 © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Finger millet (Eleusine coracana L. Gaertn) is considered an important food source among the world’s poorest people mostly found in Asia and Africa (Upadhyaya et al., 2007; 2010). Finger millet is a tetraploid crop with a genome composition of AABB and a basic chromosome number of 9. It belongs to the family Poaceae and subfamily Chloridoideae (Dida et al., 2008; Panwar et al., 2010). It is commonly called bird foot millet, Coracana, African millet and ragi (Panwar et al., 2010). Finger millet was

domesticated approximately 5000 years ago from its wild progenitor E. coracana Africana (Dida et al., 2006) and is believed to have originated from East Africa, mainly Ethiopia.

The economic importance of finger millet to the drought-prone arid and semi-Arid areas of Africa and Asia comes from its inherent capacity to tolerate several stresses including severe water deficit, and adaptability to marginal soils with very low fertility. It is rich in protein (6-13%) and calcium (0.3-0.4%), fat and minerals. The nutritional quality of finger millet grain makes it an ideal food for infants and

RESEARCH ARTICLE


SABRAO J. Breed. Genet. 45(2) 231-239

patients (Vavidoo et al., 1998). Finger millet has high levels of amino acids, especially methionine (Bisht and Mukai, 2000) and iron, which reduces malnutrition among the poor (Upadhyaya et al., 2007). In spite of the great value associated with this crop, its average yield has remained low (Tenywa et al., 1999). While finger millet production figures consistently show an increase in area planted, yield per hectare has indicated a parallel decline (Tenywa et al., 1999). The precise global area under finger millet production is not known because production data are often combined with other millets (Dida et al., 2008). Recently, increasing attention has been directed towards improving finger millet because of the crop’s inherent ability to grow in areas with low rainfall and infertile soils (Babu et al., 2007).

There is therefore a need to conduct research that focuses on crop varieties with improved traits to increase finger millet production and consequently food security and to resolve the world energy crisis. The basic information on the existence of genetic variability, population diversity and the relationships between different traits is essential for successful plant breeding programs. However, such research is limited on finger millet production (Dida et al., 2006). It is therefore appropriate to carry out baseline studies on genetic variability of finger millet germplasm for improvement of the crop. To fully exploit genetic potential of finger millet, it is important to know the extent of the already existing genetic variability in available landraces. Such research justifies the continued characterization and evaluation of different landraces stored in germplasm banks in many research centers.

Molecular marker techniques have been used successfully in DNA fingerprinting of plant genomes (Hongtrakul et al., 1997, Cervera et al., 1998). These techniques have proved to be valuable tools in the characterization and evaluation of genetic diversity within and between species and populations (Karim et al., 2010). Among the molecular markers widely used is random amplified polymorphic DNA (RAPD) marker, which comparatively shows sensitiveness, convenience and rapidness in the detection of polymorphism. Among the uses of RAPD is analysis of genetic relationship (diversity), cultivar identification and gene localization in a genome

(Lima et al., 2007). RAPDs have been used for measuring genetic diversity in several plant species (Fernandez et al., 2002). RAPDs are fairly cheap, can be automated and produce good polymorphisms. Also, it produces results quicker as compared to other techniques. Kumari and Pande (2010) found RAPD markers to be very informative and useful in monitoring the genetic diversity present in a sample of 11 finger millet genotypes germplasm. The objective of this study was to evaluate the genetic variation in finger millet (Eleusine coracana) using RAPD markers. MATERIAL AND METHODS Plant material A total of 83 finger millet (Eleusine coracana) accessions originating from four different continents, namely, Africa (Malawi and Zimbabwe), Asia (India, Nepal, and Maldives), America (USA), Europe (Germany) and two of unknown origin were acquired from (ICRISAT) India (Table 1). The landraces were planted in a field with treatments (landraces) replicated three times and laid out in a randomized complete block design. The experiment was carried out at the Field Crops Experiment Station of Khon Kaen University, Khon Kaen, Thailand (latitude 16° 30´ N, longitude 102° 47 ́E, 204 m above sea level). Planting and Morphological Classification A field experiment was conducted for two consecutive seasons from November 2010 to May 2011 and from June 2011 to October 2011 at the Field Crops Experiment Station of the Faculty of Agriculture, Khon Kaen University. A total of 83 accessions of finger millet were arranged in a randomized complete block design (RCBD) with three replications. Data were recorded for quantitative and qualitative agronomic traits at different growth stages of the crop from five competitive plants in the middle of every plot. Harinarayana and Seethaaram (1985) plant descriptor for finger millet was followed when collecting data.

Malambane et al. (2013)

233

Table 1. Origin of finger millet varieties used in the study.

DNA extraction Fresh leaf samples weighing 3 g were collected 7 days after emergency and snap-frozen in liquid nitrogen before DNA extraction. DNA extraction was done using the method described by Dellaporta et al. (1983), with minor modifications. Aliquots of the extracted DNA were run on 1% agarose gels to assess quality and quantity against the λ-DNA standard. DNA quality and quantity DNA quality and quantity were assessed using gel electrophoresis in 1% agarose gels and concentrations were estimated by visual assessment using the λ DNA of known quantity and quality (50 and 100 ng/ul). DNA aliquots of 50 ng were taken from large samples and used as working concentration. When the DNA quantity was higher than 50 ng the aliquots were diluted with double distilled H2O (ddH2O). Primer screening and selection Initially, 120 Operon (OP) primers were acquired from Biodesign Co., LTD and screened to select the one that gave good banding. Five finger millet accessions were randomly selected for primer screening. The primer showing good banding was then selected to be further screened for polymorphism. From the initial 120 primers, 60 showed a good banding pattern. Five new

randomly selected finger millet accessions were then selected to screen polymorphic primers. Primers that showed four or more bands and are polymorphic were selected for DNA fingerprinting of 83 finger millet accessions. A total of 44 polymorphic primers were selected based on polymorphism. PCR optimization and amplification Ten RAPD primers were randomly selected to optimize the polymerase chain reaction (PCR) amplification. The PCR assay was carried out in 20µl reaction volume containing 50 ng of DNA, 5X buffer (4 µl), 25 mM MgCl2 (2.4 µl), 0.2mM dNTP (0.4 µl), 0.2 mM random primer (0.8 µl), 0.5 U Taq polymerase and topped up with dH2O. The standardized amplification was initially denatured at 94 ºC for 5 min, followed by 44 cycles of denaturation at 94 ºC for 0.5 min. Primer annealing was carried out at 38 ºC for 1 min; primer extension at 72 ºC for 2 min and final extension at 72 ºC for 5 min and then cooled at 10 ºC for 2 min. The amplified products were separated on 2% agarose gel run for 2 ½ hours in 1X TBE and stained with Ethidium bromide. The gels where visualized with a UV trans-illuminator. Gel photographs were printed using the Gene Flash (Sony Syngene Bio Flash).

Continent Country No. of accessions Africa Burundi

Kenya Malawi Nigeria Senegal Uganda Zimbabwe Zambia

1 8 4 1 1 10 21 3

America USA 1 Asia India

Nepal Maldives

20 9 1

Europe Germany 1 Unknown Unknown 2


Data scoring For statistical analysis, the amplification products were scored across the lanes by comparing their molecular weight. The presence of a band was scored (1) and absence was (0). The genetic variations between accessions were evaluated by calculating Jaccard’s similarity coefficient for pairwise comparisons based on the proportion of shared band produced by primers. Data analysis was performed using the numerical taxonomy system (NTSYS-PC Version 2.11). The dendrogram was constructed using unweighted pair group method for arithmetic mean (UPGMA). RESULTS AND DISCUSSION Dendrogram A dendrogram (UPGMA) with cophenetic correlation of 0.85 (Fig. 2) managed to separate E. coracana subsp africana (IE 4709) from 82 accessions of E. coracana subsp. coracana at 50% similarity level. Dendrogram grouped 83 accessions of finger millet into two major clusters. Furthermore, the two main clusters had three (A) and two (B) distinct subcluster respectively, with each group having similar morphological traits. Group 1 had varieties with a high number of basal tillers, all having more than 10 tillers per plant. Group 2 was made up of late-maturing plants, whereas accessions in Group 3 had the same color (ragi brown). Group 4 was made up of dwarf plants all growing to a height of less than 100 cm, while Group 5 was made up of plants with high leaf number. The maximum likelihood or similarity of 0.91 was observed between IE 2043 and IE 2217 which showed very close morphological traits (days to flowering, dwarfness, maturity, and days to maturity). The landrace IE 2043 and IE 2217 apparently originated from the same country (India). Maximum dis-similarity was observed between accession 2872 and 5066 at 0.74 similarity level (Fig. 2).

RAPD markers proved to be very informative and useful in DNA fingerprinting of 83 accessions of finger millet. The current results confirm previous studies which showed that RAPD was increasingly being employed in genetic research owing to its speedy process and simplicity (Williams et al., 1990). The use of RAPD has led to the identification of genetic markers in many finger millet germplasm (Fakrudin et al., 2004). The current study has proved that RAPD markers can be used extensively in fingerprinting of finger millet accessions and germplasm. RAPD primers have shown a reproducible manner as most of them were able to reproduce the same banding pattern when reproduced; there is therefore potential of reproducibility (Kumari and Pande, 2010).

One importance of fingerprinting is to identify which primers are informative or are likely to be efficient in fingerprinting accessions of finger millet and thus in this research primers which showed higher polymorphism were recorded so that they can be further used to check diversity in other finger millet accessions. The primers that showed higher or better polymorphic information contact (PIC) can then be recommended for evaluation of more finger millet accessions, especially OPY 2 that had a high PIC value.

With a total of 44 of 60 (73%) primers being polymorphic and more than 4 major bands used, these proved very informative. A total of 255 scorable bands had been obtained which translated into 5.8 amplicons per primer showing polymorphism of 63% (160 polymorphic bands). The mean number of RAPD bands per primer in this study is higher than that reported by Kumari and Pande (2010), but lower than that reported by Fakrudin et al. (2004). The level of polymorphism in this study is higher than that reported by Panwar et al. (2010) and Gupta et al. al. (2010) but lower than that reported by Babu et al. (2007).


235

Table 2. Number of total bands, number of polymorphic bands and polymorphic information content (PIC) generated by each RAPD primer Primer

Total DNA bands (no.)

Polymorphic DNA bands (no.)

Polymorhism ratio (%)

PIC

OPA 1 13 9 69.23 0.555 OPA 2 7 4 57.14 0.419 OPB 10 8 6 75.00 0.559 OPB 18 4 2 50.00 0.294 OPC 5 11 7 63.64 0.440 OPC 7 8 5 62.50 0.497 OPC 15 7 4 57.14 0.323 OPC 20 13 8 61.54 0.666 OPF 2 10 7 70.00 0.618 OPG 3 9 7 77.78 0.527 OPG 4 13 9 69.23 0.508 OPG 5 14 10 71.43 0.580 OPG 6 7 6 85.71 0.605 OPP 2 11 6 54.55 0.555 OPP 4 7 4 57.14 0.603 OPP 5 12 8 66.67 0.542 OPP 6 14 9 64.29 0.542 OPP 9 10 4 40.00 0.413 OPP 10 7 4 57.14 0.242 OPQ 2 10 7 70.00 0.209 OPQ 3 6 3 50.00 0.158 OPQ 4 8 5 62.50 0.417 OPQ 5 7 4 57.14 0.531 OPN 2 8 4 50.00 0.167 OPN 3 10 7 70.00 0.503 OPN 4 5 3 60.00 0.135 OPN 5 10 5 50.00 0.444 OPN 9 10 4 40.00 0.305 OPS 1 6 4 66.67 0.539 OPS 3 4 3 75.00 0.348 OPS 7 4 3 75.00 0.324 OPS 8 8 5 62.50 0.368 OPS 9 4 2 50.00 0.324 OPY 2 10 6 60.00 0.707 Total 295 184 62.37 14.966 Average 9 5 0.440


(a) M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

(b) M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Figure 1. Polymorphic banding pattern of RAPD primers OPF-4 (a) and OPN-3 (b).

The cluster analysis was able to separate landraces into major groupings confirming morphological differences and place of origin, this separation increases the understanding of genetic relatedness and genetic diversity of the finger millet landraces used in the current study. The information is critical in future breeding programs for finger millet production because selection of parents is mostly based on genetic variation of genotypes. A breeding program for finger millet will use genetically and morphologically diverse parents to transfer genes from one genotype to another to improve the qualities of the selected genotypes (Karim et al., 2010).

In this study the geographical origin of finger millet accession did not contribute significantly to classification or grouping of finger millet accessions. Previous studies on classification of accessions also showed insignificant contribution of geographical origin to genetic diversity, possibly because of genetic drift and selection of different environments (Murthy and Arunanchalam, 1996). The relationship between genetic diversity and geographical diversity has been a point of debate in

literature. Lang et al. (2009) reported that varieties from different places grouped together because of close similarities of qualitative traits. In contrast Rohman et al. (2004) reported that genetic diversity is generally associated with geographical diversity. Agrama and Tuinstra (2003) and Dalkilic et al. (2011) however, observed that the genotypes collected from the same regions showed dissimilarity when subjected to cluster analysis. Fahima et al., (1999) showed no association with geographical distance between wheat population sites of the same origin.

The results obtained from this study showed that there is high genetic variation among finger millet accessions. The study also demonstrates the high reliability, ease of applicability and importance of RAPD markers in evaluating genetic variation among finger millet accessions. This information will help in the breeding of finger millet varieties and characterization of landraces that are selected by farmers and those stored in national and international gene banks.


237

Figure 2. Dendrogram showing genetic similarity of 83 finger millet accessions revealed by UPGMA clusters analysis based on RAPD fingerprinting. A and B are main groups while 1-5 represent the subgroups.

1

2

3

4

5

B

A


ACKNOWLEDGEMENTS The authors gratefully acknowledge the financial assistance provided in the form of a scholarship partnership between Khon Kaen University (KKU), Thailand International Development Cooperation Agency (TICA) and Botswana College of Agriculture (BCA). They also thank the International Crops Research Institute for the Semi-arid Tropics (ICRISAT), India, for providing the research materials. REFERENCES Agrama HA, Tuinstra MR (2003). Phylogenetic

diversity and relationships among sorghum accessions using ISSRs and RAPDs. Afr. J. Biotechnol. 10: 334-340.

Babu KB, Babu RC, Biji KR, Gomez MS, Kumar SS, Rajendraprasad NS, Senthil N (2007). Assessment of genetic diversity among finger millet (Eleusine coracana (L.) Gaertn) accessions using molecular markers. Genet. Res. and Crop Evol. 54: 399-404.

Bisht MS, Mukai Y (2000). Genome in situ hybridization identifies genome donor of finger millet (Eleusine coracana). Theor. Appl. Genet. 402: 825-832.

Cervera MT, Cabezas JA, Sancha JC, Martίnez de Toda F, Martίnez-Zapater JM (1998). Application of AFLPs to the characterization of grapevine Vitis vinifera L. genetic resources. A case study with accessions from Roja (Spain). Theor. Appl. Genet. 97: 154-162.

Dalkiliç Z, Mestav HO, Gǜnver -Dalkiliç G, Kocataş H (2011). Genetic diversity of male fig (Ficus carica caprificus L.) genotypes with random amplified polymorphic DNA (RAPD) markers. Afr. J. Biotechnol. 10: 519-526.

Dellarporta SL, Wood VP, Hicks JB (1983). A plant DNA mini-preparation: version II. Plant Mol. Biol. 1:19-21.

Dida MM, Srinivasacheary S, Ramakrishnan S, Bennetzen JL, Gale MD, Devos KM (2006). The genetic map of finger millet, Eleusine coracana. Theor. Appl. Genet. 114: 321-332.

Dida MM, Wanyera N, Ramakrishnan S, Harrison-Dunn ML, Bennetzen JL, Devos KM (2008). Population structure and diversity in finger millet (Eleusine coracana) germplasm. Trop. Plant Biol. 1: 131-141.

Fahima T, Sun GL, Beharav A, Krugman T, Beiles A, Nevo E (1999). RAPD polymorphism of

wild emmer wheat populations, Triticum dicoccoides, in Israel. Theor. Appl. Genet. 98: 434- 447.

Fakrudin B, Kalkarni RS, Shashidhar HE, Hittalmani S (2004). Genetic diversity assessment of finger millet, Eleusine coracana (Gaertn), germplasm through RAPD analysis. Plant Genet. Resour. Newsl. 138: 50-54.

Fernandez ME, Figueirus AM, Benito C (2002). The use of ISSR and RAPD markers for detecting DNA polymorphisms, genotype identification and genetic diversity among barley cultivars with known origin. Theor. Appl. Genet. 104: 845-851.

Gupta R, Verma K, Joshi DC, Yadav D, Singh M (2010). Assessment of genetic relatedness among three varieties of finger millet with variable seed coat color using RAPD and ISSR markers. Genet. Eng. Biotech J. 2: 1-9.

Harinarayana G, A. Seethaaram A (1985). Descriptors for finger millet. AGPG. IBPGR/85/106.

Hongtrakul V, Huestis GM, Knapp SJ (1997) Amplified fragment length polymorphism as a tool for DNA fingerprinting sunflower: Genetic diversity among oilseed inbred lines. Theor. Appl. Genet. 95: 400-407.

Karim K, Rawda A, Hatem C, Mbarek B, Mokhtar T (2010). Analysis of genetic diversity and relationships in local Tunisian barley by RAPD and SSR analysis. Afr. J. Biotechnol. 9: 7429-7436.

Kumari K, Pande A (2010). Study of genetic diversity in finger millet (Eleusine coracana L. Gaertn) using RAPD markers. Afr. J. Biotechnol. 29: 4542-4549.

Lang MT, Tu PTB, Thanh NC, Buu BC, Ismail A (2009). Genetic diversity of salt tolerant rice landraces in Vietnam. J. Plant Breed. Crop Sci. 1: 230-243

Lima JM, Dass A, Sahu SC, Behera L, Chauhan DK (2007). A RAPD marker marker identified a susceptible specific locus for gall midge resistance gene in rice cultivar ARC5984. Crop Prot. 26: 1431-1435.

Murty BR, Arunachalam V (1996). The nature of divergence in relation to breeding system in some crop plants. Indian J. Genet. Plant Breed. 26: 188-198

Panwar P, Riki Saini, Sharma N, Yadav D, Kumar A (2010). Efficiency of RAPD, SSR and cytochrome P450 gene-based markers in accessing genetic variability amongst finger millet (Eleusine coracana) accessions. Mol. Biol. and Genet. Eng. 37: 4075-4082.


239

Rohman MM, Hakim MA, Sultana NA, Kabir ME, Hasanuzzan M, Ali M (2004). Genetic divergence analysis in sorghum (Sorghum bicolor L.). Asian J. Plant Sci. 3: 211-214

Tenywa JS, Nyende P, Kidoido M, Kasenge V, Oryokot S, Mbowa S (1999). Prospects and constraints of finger millet production in Uganda. Afr. Crop Sci. 4: 569-583.

Upadhyaya HD, Gowda CLL, Reddy VG (2007). Morphological diversity in finger millet germplasm introduced from Southern and Eastern Africa. J. SAT Agric. Res. 3: 1-3.

Upadhyaya HD, Sarma NDRK, Ravishankar CR, Albrecht T, Narasimhudu Y, Singh SK, Varshney SK, Reddy VG, Singh S, Dwivedi SL, Wanyera N, Oduori COA, Mgonja MA,

Kisandu DB, Parzies HK, Gowda CLL (2010). Developing a mini-core collection in finger millet using multilocation data. Crop Sci. 50: 1924-1931.

Vavidoo AS, Joseph R, Ganesan NM (1998). Genetic variability and diversity for protein and calcium contents in finger millet (Eleusine coracana (L.) Gaertn) in relation to grain color. Plant Foods Hum. Nutr. 52: 353-364.

Wiliiams JGK, Kubelik AR, Livak KJ, Rafalski AJA, Tingey SV (1990). DNA polymorphism amplified by arbitrary primers are useful l as genetic markers. Nucleic Acids Res. 22: 6531-6535.


GENETIC DIVERGENCE IN FRENCH BEAN FOR POD YIELD-RELATED TRAITS

A. SHARMA* and J. DEVI

C.S.K. Himachal Pradesh Agricultural University, Palampur 176062, India 1Department of Vegetable Science and Floriculture, College of Agriculture,

Corresponding author’s email: [email protected], [email protected]

SUMMARY

The characterization of genetic resources allows the identification of sources of genetic variability, which is important to select potential genotypes for use in future breeding programs. An attempt was made to determine the degree of divergence by utilizing 33 genotypes of bush-type French bean, which were evaluated in randomized complete block design at Vegetable Research Farm, CSK HPKV, Palampur for two consecutive seasons during summer 2008 and 2009. Based on Mahalanobis D2 statistics, these genotypes were classified into 6 groups each during 2008 and pooled over the years and 4 during 2009. Cluster I was found to be the largest. Intracluster distances were found to be highest for cluster II for both years and for those pooled over the years. The clustering pattern indicated the absence of any relationship between genetic divergence and geographical distribution. Critical study of cluster means for different characters indicated that genotypes Palam Mridula, DWDFB-53, HAFB-1, Aparna, JFB-97-1, Chandini, Surya, Falguni and IVRFB-1 showed greater potential for fresh pod yield and seed yield based on maximum intercluster distances for horticulturally desirable traits. It was concluded that the line JFB-97-1 was the most divergent, followed by KPV-2 and Palam Mridula. Superior lines may be obtained from segregating populations by using these in hybridization—e.g., Contender × JFB-97-1/Palam Mridula, JFB-97-1 × Palam Mridula, Palam Mridula × KPV-2 and JFB-97-1 × KPV-2. Keywords: Genetic divergence, Phaseolus vulgaris L.

Manuscript received: June 7, 2012; Decision on manuscript: November 24, 2012; Manuscript accepted: February 19, 2013 @ Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO)

Communicating Editor: Arbind K. Choudhary

INTRODUCTION French bean (Phaseolus vulgaris L.), a member of family Fabaceae, is one of the most ancient and popular warm-season vegetable crops grown worldwide for its green pods. It is an important source of protein, calcium, iron and vitamins in the human diet. It enjoys a much coveted position in the hill states of India for its off-season cultivation and quality pods, which in turn results in high remuneration to vegetable growers. In recent years, tremendous genetic

improvement efforts have been made by breeders with the basic objective of high yield, quality of fresh produce, sloughing free cultivars for canning industry, short pod for whole processing and resistance to pest and diseases. Thus, estimation of genetic diversity is important as this crop is used for diverse purposes.

Genetic diversity analysis enables identification of genetically diverse genotypes to be used in the breeding program (Coelho, 2007). Hybridization of genetically diverse genotypes

RESEARCH ARTICLE



241

results in the release of a great amount of genetic variability due to recombination of genes. This has been observed in several crop species. Multivariate analysis is a powerful tool for determining the degree of divergence among genotypes in the population and also for determining the nature of forces operating at different levels. Evaluation of genotypes in a single environment does not give a clear picture about genotypic value and the degree of divergence as genotype x environment affects character expression. Thus, evaluation over environments is important to get relevant results (Teixeira, 2004). Genetic diversity analysis also reveals the redundancy of accessions with respect to a particular trait or combination of traits, which avoids wastage of resources. The progenies derived from diverse parents are expected to show a broad spectrum of genetic variability and provide better scope to isolate superior recombinants. Therefore, genetically diverse genotypes should be used in a hybridization program to get superior recombinants. Thus, in the present study, 33 genotypes of French bean were evaluated over environments for morphological traits in order to identify diverse genotypes.


The experiment was undertaken at the Vegetable Farm, CSK HPKV, Palampur (1, 290.8 m asl, with 320 6' N latitude, 760 3' E longitude). The 33 genotypes (Table 1) were evaluated in randomized complete block design with 3 replications for two consecutive seasons (March-June 2008 and 2009). These genotypes were assigned to 2.7m length plots with spacing of 0.45 m between rows and 0.15 m between plants within rows. Data were recorded on 10 randomly selected competitive plants in each entry in each replication for 15 yield and other related traits—pod width (cm), green pod length (cm), average pod weight (g), branches per plant, pods per plant, fresh pod yield per plant (g), days to seed maturity, mature pod length (cm), seeds per pod, biological yield per plant (g), plant height (cm), 100-seed weight (g) and seed yield per plant (g). Data on days to 50 per cent flowering and days to first marketable picking were recorded on a

plot basis. The recommended/standard agronomic practices were followed to raise a good crop. Statistical analysis The data collected were subjected to multivariate analysis utilizing Mahalanobis D2 statistic as suggested by Mahalanobis (1936) and Rao (1952) using statistical software WINDOSTAT 8.0 developed by Indostat Services. Genotypes were grouped into various clusters following Tocher’s method as suggested by Rao (1952). RESULTS

The French bean genotypes were significantly different for all characters but a few during both years and pooled over the years. Significant genotype × environment interactions were also observed for all the traits (Table 2).

D2 analysis grouped 33 genotypes into 6 clusters showing reasonable variability (Table 3). Cluster I was the largest, comprising 73% genotypes. Clusters III (‘HAFB-1’), IV (‘IVRFB-1’), V (‘KPV-2’) and VI (‘JFB-97-1’) were monogenotypic - i.e., they contain only one genotype. Intracluster distance was high in Cluster II, followed by Cluster I (Table 4). The clusters with single genotype had zero intracluster divergence. On the other hand, the intercluster distance ranged from 2.42 to 5.27. The maximum intercluster distance was observed between cluster II and VI. The intercluster proximity was maximum between clusters I and III, which showed the lowest variability.

Substantial differences in clusters means (Figure 1) were observed for each character. Based on cluster means, it was observed that the most economic characters (branches per plant, pods per plant and fresh pod yield per plant with respect to fresh green crop) were contained in clusters II and III. For seed crop, clusters II, VI and III were observed to be the most important, having maximum cluster means for plant height, 100-seed weight and seed yield per plant, respectively.

The contribution of individual characters to divergence has been worked out in terms of

Sharma &Devi (2013)

242

number of times it appeared first (Figure 2). Percentage of the character contribution toward

total divergence among the French bean genotypes was the most for 100-seed weight.

Table 1. Genotypes used in the experiment. Genotype Source Genotype Source

Arka Suvidha IIHR, Hasarghata KPV-2 CSKHPKV, Palampur Arka Anoop IIHR, Hasarghata MFB-1 NEH, ICAR, Barapani Palam Mridula CSKHPKV, Palampur MFB-2 NEH, ICAR, Barapani DPDFB-1(M) CSKHPKV, Palampur MFB-3 NEH, ICAR, Barapani DPDFB-2(M) CSKHPKV, Palampur MFB-4 NEH, ICAR, Barapani DWDFB-I UAS, Dharwad MFB-5 NEH, ICAR, Barapani DWDFB-53 UAS, Dharwad VLB-8 VPKAS, Almora DWDFB-57 UAS, Dharwad VLB-9 VPKAS, Almora HAFB-1 HARP, Ranchi VLB-2003 VPKAS, Almora HAFB-2 HARP, Ranchi VLFB-130 VPKAS, Almora HAFB-3 HARP, Ranchi Aparna Prabhakar Hybrid Seed

Company, Hyderabad HAFB-4 HARP, Ranchi Chandini Sutton and Sons (India) Pvt.

Ltd. IVRFB-1 IIVR, Varanasi Falguni Seminis Vegetable Seeds (India)

Ltd. IVFB-1 IIVR, Varanasi Surya Solar Seeds IVFB-2 IIVR, Varanasi Arka Komal IIHR, Hasarghata IVFB-3 IIVR, Varanasi Contender CSKHPKV, Palampur JFB-97-1 GAU, Junagadh


243

Table 2. Analysis of variance for different characters in French bean during 2008 and 2009. Mean sum of squares Pooled over years Character Replication Genotype (G) Error Genotype

(G) Environment (E)

G x E Error F test (homogeneity test) 2008 2009 2008 2009 2008 2009

df 2 32 64 32 1 32 128 Days to 50 percent flowering

0.58 5.18 16.73* 6.84* 2.22 1.4 16.65* 8.91* 6.91* 1.81 2.51*

Days to first marketable picking

12.16 1.98 17.89* 4.41* 4.13 1.57 15.83* 2.44* 6.47* 2.85 6.93*

Pods width (cm) 0.02 0.02 0.03* 0.02* 0.002 0.004 0.46* 0.34 0.59* 0.26 1.98*

Green pod length (cm)

6.58 0.71 3.62* 5.19* 0.98 0.65 6.52* 0.23 2.27* 0.82 2.24*

Average pod weight (g)

0.36 2.77 3.61* 1.29 0.4 0.37 3.28* 213.72* 1.62* 0.38 1.18

Branches per plant (no.)

5.69 0.32 0.67* 0.69* 0.23 0.06 0.99* 0.21 0.99* 0.15 8.20*

Pods per plant (no.) 5.48 1.23 52.51* 98.01* 2.24 4.65 112.21* 0.703 38.31* 3.45 4.33*

Fresh pod yield per plant (g)

36.13 247.16 2917.45* 2219.4* 28.73 22.18 3806.07* 50345.38* 1330.8* 25.45 1.68

Days to seed maturity 0.07 0.07 20.54* 20.54* 1.01 1.01 15.69* 861.5* 9.33* 1.03 1.04

Mature pod length (cm)

1.44 1.44 8.54* 8.54* 0.41 0.41 16.89* 0.39 1.40* 0.72 6.30*

Seeds per pod (no.) 0.05 0.05 0.85* 0.84* 0.11 0.11 1.30* 0.633 0.23* 0.14 2.90*

Plant height (cm) 176.50 2.00 30.74* 25.68* 3.93 5.95 40.12* 145.19* 16.30* 4.94 2.29*

Biological yield per plant (g)

4.06 4.06 383.67* 383.66* 3.99 3.99 362.23* 5083.97* 203.55* 5.049 2.34*

Seed yield per plant (g)

24.87 24.87 228.40* 228.4 15.08 15.08 399.87* 477.71* 126.65* 17.47 1.73

100-seed weight(g) 24.17 24.17 283.51* 283.52 7.29 7.29 389.31* 57.89* 35.26* 7.04 1.16

*Significant at P ≤ 0.05.

Sharma &Devi (2013)

244

Table 3. Cluster compositions in French bean following multivariate analysis of data.

Table 4. Intra and intercluster distances in French bean germplasm.

Cluster 1 II III IV V VI

1 2.07 3.6 2.42 3.03 2.57 2.97

(1.44) 1.99 1.56 1.74 1.60 1.72

II 2.36 3.58 4.50 3.56 5.27

(1.54) 1.89 2.12 1.89 2.30

III 0.00 3.02 2.81 3.35

(0.00) 1.74 1.68 1.83

IV 0.00 3.75 4.03

(0.00) 1.94 2.01

V 0.00 3.09

(0.00) 1.76

VI (0.00)

Values in bold figures are intracluster distances; values in parentheses are √D2 = D values.

Cluster Number of genotypes

Name of genotypes

I 24 VLFB-130, Arka Komal, MFB-5, MFB-4, IVFB-3, VLB-9, IVFB-2, HAFB-4, VLB-2003, Contender, Arka Suvidha, DPDFB-2(M), VLB-8, IVFB-1, MFB-1, HAFB-3, Arka Anoop, DWDFB-57, MFB-2, HAFB-2, MFB-3, DWDFB-53, DPDFB-1(M), DWDFB-1

II 5 Chandini, Surya, Falguni, Aparna, Palam Mridula

III 1 HAFB-1

IV 1 IVRFB-1

V 1 KPV-2

VI 1 JFB-97-1


245

Phonological characters Plant growth characters Yield Figure 1. Cluster means of different yield-contributing, yield and quality traits in French bean germplasm.

Figure 2. Contribution (%) towards genetic divergence of different characters. DISCUSSION

Quantification of genetic diversity is very important in a breeding program because it reveals the genetic structure of the population. Significant differences were observed for almost all characters, which indicated the existence of genetic variability among these materials (Singh, 2006; Sofi et al., 2011). Significant genotype × environment interaction further indicated the need for multilocation trials to identify the superior genotypes. The composition of clusters

on the basis of D2 statistics revealed that cluster I was the largest, containing majority of the genotypes and showing high homogeneity among them (Ceolin et al., 2007) or the least genetic variation (Galvan et al., 2003). This is due in part to the fact that these cultivars within each cluster might have been developed from a relatively narrow genetic base. Singh (2006) and Sharma et al. (2009) also observed majority of genotypes with similar genetic background in a single cluster. The clustering of the genotypes indicated no parallelism between genetic

Sharma &Devi (2013)

246

diversity and geographical diversity. Similar views were also expressed by Singh (2006), Lavanya et al. (2008) and Mishra et al. (2010). Since the intracluster distance was low, the likelihood of developing good segregants by hybridization among parents within a cluster would be low. Therefore, crosses should be attempted between genotypes falling in diverse clusters to get superior recombinants.

It is imperative to identify clusters with the highest genetic diversity. This should be followed by identification of the genotypes in each cluster, which have higher expression for the traits and are also complementing the important traits with the genotypes of the diverse cluster. Clusters I and III displayed the lowest degree of divergence, suggesting close genetic makeup of the strains included in these groups. Hence, they cannot be used for recombination breeding to evolve desirable new genotypes. Since a wider genetic base may provide more useful genes to be used by breeders, research programs aimed at transferring desirable traits between gene pools should be encouraged. The clusters with single genotypes (HAFB-1, IVRFB-1, KPV-2 and JFB-97-1) indicated their potential to exploit heterosis and transgressive segregation (Yadav et al., 2009). A high mean value for important traits is fundamental for the selection of superior material, since it is essential to account not only for performance but also for genetic divergence (Palomino et al., 2005). Based on cluster means, cluster II and III clearly indicate that genotypes included in these clusters had sufficient genetic diversity. The results obtained indicate that, in order to get segregating populations with possibilities of superiority upon the parents from bi-parental crosses, the genotypes more productive of each clustering should be intercrossed for improving green and seed yield (Ceolin et al., 2007).

The contribution of individual character toward diversity showed that biological yield per plant, seed yield per plant and 100-seeds weight was the most important characteristic for divergence (Figure 2). Earlier workers who evaluated genetic divergence among French bean genotypes also verified that the characters that most contributed to genetic divergence were 100-seedweight (Fonseca and Silva, 1999;

Coimbra et al., 1999) and seed yield (Coelho, 2007; Durga, 2012). In relation to characteristics of lower importance, such as pod width, average pod weight, mature pod length, seed per pod and plant height in the mean of the environments, no consistency was observed. Similar results were obtained for plant height by Coimbra and Carvalho (1999) when studying genetic divergence in common bean using data from two environments. Accordingly, genotypes Arka Komal, MFB-3, Contender, Arka Suvidha and Arka Anoop from Cluster I along with Palam Mridula, Aparna and Falguni from Cluster II showed higher expression for fresh pod and seed yield along with desirable yield characteristics. These genotypes can be used to complement the important traits with the genotypes of the diverse cluster with maximum intercluster distance -e.g., DWDFB-53, HAFB-1, JFB-97-1 and IVRFB-1. These genotypes, therefore, show greater potential as a breeding stock and could be used as parents in hybridization to get transgressive segregants for further exploitation in French bean improvement programs.

It can be concluded that superior lines may be obtained in segregating populations using the combinations Contender × JFB-97-1/Palam Mridula, JFB-97-1 × Palam Mridula, Palam Mridula/Falguni × HAFB-1 and JFB-97-1 × MFB-3/Arka Suvidha.

REFERENCES Ceolin ACG, Goncalves-Vidigal MC, Vidigal Filho

PS, Kvitschal MV, Gonela A, Scapim CA (2007). Genetic divergence of Phaseolus vulgaris L. using morpho-agronomic traits by multivariate analysis. Hereditas. 144: 1-9.

Coelho CMM, Coimbra JLM, Souza C, Bogo A, Guidolin AF (2007). Genetic diversity in common bean accessions. Ciencia-Rural 37(5): 1241-1247.

Coimbra JLM, Carvalho FIF (1999). Divergencia genetica em linhagens de feijao preto (Phaseolus vulgaris L.) preditas atraves de variaveis quantitativas. Rev. Cient. Rural 4: 47-53.

Coimbra JLM, Carvalho FIF, Hemp S (1999). Divergencia genetica em feijao preto. Ciencia Rural 29: 427-431.


247

Durga KK (2012).Variability and divergence in horsegram (Dolichos uniflorus). J. Arid Land 49(1):71-76.

Fonseca JR, Silva HT (1999). Identificacao de duplicidades de acessos de feijao por meio de tecnicas multivariadas. Pesquisa Agropecua´ria Brasileira 34: 409-414.

Galvan MZ, Bornet B, Balatti PA, Branchard M (2003). Inter simple sequence repeat (ISSR) markers as a tool for the assessment of both genetic diversity and gene pool origin in common bean (Phaseolus vulgaris L.). Euphytica 132: 297–301.

Lavanya GR, Srivastava J, Ranade SA (2008). Molecular assessment of genetic diversity in mungbean germplasm. J. Genet. 87(1): 65-74.

Mahalanobis PC (1936). On the generalized distance in statistics. Proc.Natl. Inst. Sci. 12: 49-55.

Mishra S, Sharma MK, Singh M, Yadav SK (2010). Genetic diversity of French bean (bush type) genotypes in north-west Himalayas. Indian J. Plant Genet. Resour. 23(3): 285- 287.

Palomino EC, Mori ES, Zimback L (2005). Genetic diversity of common bean genotypes Carioca commercial group using RAPD markers. Crop Breed. Appl. Biotechnol. 5: 80-85.

Rao CR (1952). Advanced statistical methods in biometrical research. John Wiley and Sons, New York pp. 357-64.

Sharma MK, Mishra S, Rana NS (2009). Genetic divergence in French bean (Phaseolus vulgaris L.) pole type cultivars. Legume Res. 32(3): 220-223.

Singh AK (2006). Genetic divergence in French bean (Phaseolus vulgaris L.). Veg. Sci. 33(1): 103-105.

Sofi PA, Zargar MY, Debouck D, Graner A (2011). Evaluation of common bean (Phaseolus vulgaris) germplasm under temperate conditions of Kashmir valley. J. Phytol. 3(8): 47-52.

Teixeira AB, Amaral-Junior AT, Rodrigues R, Pereira TNS, Bressan-Smith RE (2004). Genetic divergence in snap-bean (Phaseolus vulgaris L.) evaluated by different methodologies. Crop Breed. & Appl. Biotecnol.4(1): 57-62.

Yadav R, Srivastava RK, Kant R, Singh R (2009). Studies on genetic divergence in field pea [Pisum sativum (L.) Poir]. Legume Res. 32: 121-124.

1


INHERITANCE OF PUNGENCY IN THAI HOT PEPPER (Capsicum annuum L.)

VEERAYA TEMPEETIKUL1, SUCHILA TECHAWONGSTIEN1*, KAMOL LERTRAT1 and SUNGCOM TECHAWONGSTIEN1

1Department of Plant Science and Agricultural Resources, Faculty of Agriculture, Khon Kaen University,

Khon Kaen 40002, Thailand. *Corresponding author’s email: [email protected], [email protected]

SUMMARY

To develop new improved varieties of hot pepper (Capsicum annuum) L.) with the desired pungency, suitable parental lines and mode of inheritance for pungency is prerequisite. Two experiments were conducted at the experimental farm, Faculty of Agriculture, Khon Kaen University, Thailand. In the first experiment, 5 hot pepper lines with different pungency levels were used in a half diallel mating design to develop 10 F1 crosses. All the F1s and parents were evaluated to determine combining ability and heterosis. Based on the first experiment, variety Yodson (YS) and Num Chiangmai (NM) were selected in the second experiment due to their good combining ability for pungency. Both these lines were used to develop a set of F1, F2 and their backcross (BC1P1 and BC1P2) generations to study the inheritance of pungency. The results from the first experiment showed that the crosses derived from ‘YS’ had maximum mean values for all the capsaicinoids. They were also superior to their parents in terms of capsaicinoid quantity. ‘YS’ was considered a good parent because of its high actual mean value and high general combining ability (GCA) for all capsaicinoid traits. The F1 hybrid ‘YS/NM’ had high specific combining ability (SCA) for all capsaicinoids. In addition, narrow-sense heritability for capsaicinoids in YS/NM was moderate to high for total capsaicinoids (0.62). Generation mean analysis indicated a large magnitude of dominance, and dominance by dominance gene effects for all capsaicinoids. Keywords: Capsaicinoids, combining ability, gene effects, heterosis, narrow-sense heritability, Thai chilli

Manuscript received: June 25, 2012; Decision on manuscript: March 30, 2013; Manuscript accepted: April 27, 2013. © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION

Hot pepper (Capsicum spp.) is well recognized as one of the most important spices and condiments in many countries, especially in Asia and Africa (FAOSTAT, 2009). In Thailand hot pepper is cultivated on about 80,000 ha with 600,000 metric tons of production (Department of Agriculture Extension, 2009). However, Thailand annually imports hot pepper worth more than US$10 million, mainly for industrial purposes (Customs Department, 2009). Hot pepper production in Thailand is characterized by low

productivity and quality. There are many problems related to hot pepper production in Thailand. The pests (insects and diseases) problem is commonly spread all year-round in the cultivation areas. The lack of good varieties and their production packages are the main reasons for the low yield and poor quality of Thai hot peppers. Currently, northeastern Thailand is the major region of rainfed hot pepper production. The local varieties are popular among the growers due to their better adaptation and low seed price. The small-fruited varieties (‘Hauysithon’, ‘Yodson’ or ‘Jinda’) with high pungency are preferred in

RESEARCH ARTICLE


249

the northeast region, while the long cayenne with mild pungency (Num-Chiangmai) is popular in the north. However, the demand of highly pungent varieties for food and pharmaceutical industries has been continuously increasing. Hence, a breeding program for developing hot pepper varieties with high pungency has been developed by many institutes and private companies. The information on inheritance of the pungency required for developing improved hot pepper varieties with high pungency. The high broad-sense heritability (h2

b) and narrow-sense heritability (h2

n) for capsaicinoids (pungency compounds) have been reported (Doshi, 2003 and Techawongstien et al., 2008). Now we know that synthesis of capsaicinoids in the placenta glands of the fruit is influenced by the environment and many QTLs associated with capsaicinoids (Ben-Chaim et al., 2006), although it is genetically inherited as a dominant trait and controlled under C locus or Pun1 locus (Blum et al., 2002). Ahmed et al. (1982) reported additive effect, dominance effect, and dominance by dominance effects for pungency inheritance. Zewdie and Bosland (2000) found that the additive, dominance and interaction effects were significant for capsaicin, dihydrocapsaicin and isomer of dihydrocapsaicin in an interspecific hybridization of C. annuum x C. chinense. General combining ability (GCA) and specific combining ability (SCA) effects were significant for capsaicinoids in C. pubescens, thereby indicating non-additive gene effect (Zewdie and Bosland, 2001). Furthermore, additive and epistasis gene effects for capsaicin and dihydrocapsaicin were also reported within C. annuum L. study (Garcés -Claver et al., 2007).

In order to identify suitable parental lines and to study the inheritance of capsaicinoids (pungency) under Thai environments, we conducted a set of two experiments and the results are described.


Description of experiments There were 2 experiments in this study: (1) study on combining ability of Thai hot pepper varieties and (2) generation mean analysis (GMA) of a cross for capsaicinoids. In the first

experiment, 5 hot pepper varieties (C. annuum L.) with different levels of pungency and genetic background were used - Yodson: ‘YS’ small-fruited type with high pungency, Huaysithon: ‘HS’ a small fruited highly pungent local variety, Num-Chiangmai: ‘NM’ long-cayenne fruit type with mild pungent local variety, Yuyi: ‘YY’, small and shrink-fruited low pungent variety from China and California Wonder: ‘CW’ non-pungent commercial sweet pepper variety. These five were used as parents to produce 10 F1 crosses hybrids (by half-diallel). All the 10 F1s were tested for their combining ability compared to their parents during two consecutive seasons, rainy season (May-October, 2004) and dry season (November, 2004-April, 2005) at the experimental field of Khon Kaen University, Thailand. A randomized complete block (RCB) was designed using 15 treatments (10 F1s and 5 parents) with 3 replications. Each replication had 20 plants.

Based on the results of this experiment, two parents (‘YS’ and ‘NM’) with high general combining ability for capsaicinoids were selected and used to produce F1, F2, BC1P1 and BC1P2 populations. Both parents (P1, P2) and 4 generations (F1, F2, BC1P1 and BC1P2) were tested for narrow sense heritability and GMA by using RCB design with 4 replications during rainy season (May and October, 2005). Drip-based irrigation and fertigation was applied to the plants in both seasons to ensure normal crop stand. Standard crop management practices from nursery to harvest were applied for both seasons. HPLC analysis of capsaicinoids The ripe fruits from all the treatments in both the experiments were harvested and dried in a hot-air oven at 80 ºC until a stable weight is attained. The dry fruits were grinded, extracted and quantified for capsaisinoids using high performance liquid chromatography (HPLC) according to Collins et al. (1995). For each analysis, 1 g of fruit powder was extracted by shaking in 10 ml of acetonitrile at 80 °C for 4 h. The extract was syringe-filtered using a 0.45 µm polyamide (Sartorius, # 0440-293) and 10-µl of the filtered extract was injected for HPLC analysis, using a Shimadzu-Model 10AT-VP series. The mobile phase was CH3OH:H2O=80:20 at a flow rate of 1.5 ml min-1 with the ODS C-18

Tempeetikul et al. (2013)

column. Detector was set at 284. Concerning the high components of capsaicin (69%) and dihydrocapsaicin (22%) in hot pepper (Bennett and Kirby, 1968), external standards for capsaicin and dihydrocapsaicin (Fluka # 76390 and #37274, respectively) were prepared in 100% methanol by dilution of 200 ppm stock solution. Capsaicin (Cap) and dihydrocapsaicin (Di) were determined and pooled as sum of Cap and Di (CAPS).

Data analysis Capsaicinoid data in the first experiment were calculated by Microsoft Excel for GCA and SCA following Griffing (1956) method II, model I. The parent Vs cross comparisons were tested for heterobeltiosis (%) or high-parent heterosis = {[(F1-HP)/HP]*100} (Falconer, 1981). The selected cross ‘YS/NM’ was used for estimating gene effects in capsaicinoids, and narrow-sense heritability (h2

n) following Falconer (1981), due to their high mean Cap, Di and CAPS, and SCA. This cross was also tested for generation mean analysis by weighted least squares following Mather and Jinks (1977) using their coefficients for different generations based on F-infinity metric. The notation of Gamble (1962) was used, i.e. mean (m), additive effect

(a), dominance effect (d), additive x additive (aa), additive x dominance (ad) and dominance x dominance (dd). The joint scaling test (Cavalli, 1952) was also performed to provide the best estimates of the genetic parameters. Only significant parameters were fitted using the weighted least squares method as described by Rowe and Alexander (1980). RESULTS

Combining ability and heterosis for capsaicinoids Interaction between seasons and crosses were highly significant for all capsaicinoid components (Table 1). The GCA analysis showed highly significant differences for all capsaicinoids, while the SCA for capsaicin (Cap) and the sum of Cap and Di (CAPS) were significant and highly significant, respectively (Table 2). However, SCA for dihydrocapsaicin (Di) was not significant. In addition, the GCA for capsaicinoids was greater than SCA by 3 to 4 times. This suggests that highly pungent inbred lines or true varieties may be bred by selection from the highly pungent crosses.

Table 1. Analysis of variance of capsaicin (Cap), dihydrocapsaicin (Di) and sum of Cap and Di (CAPS) from 5 parental lines and 10 F1 crosses of Thai hot pepper. Df 1 14 14 56

CV (%) Characteristic Mean squares Error Season Cross Season x Cross Cap 17.20** 22.26** 1.5** 0.15 13.41 Di 6.16** 5.37** 0.41** 0.05 14.11 CAPS 43.26** 49.2** 3.36** 0.36 13.38 ** significant at P ≤ 0.01 level. Table 2. Analysis of variances for combining abilities for capsaicin (Cap), dihydrocapsaicin (Di) and sum of Cap and Di (CAPS) of 5 parents and their 10 crosses. Source of variation df Mean square

Cap Di CAPS GCA 4 40.51** 5.80** 22.85** SCA 9 2.54* 2.13ns 5.74** Error 28 0.06 0.02 0.15 GCA/SCA 4.14 : 1 2.72 : 1 3.98 : 1 ns: non-significant, *, ** significant at P ≤ 0.05 and 0.01 levels, respectively.


251

Among parental varieties, Yodson (YS) had maximum values for all capsaicinoids, followed by Hauysithon (HS), Num Chiangmai (NM) and Yuyi (YY). As expected California wonder (CW) had the lowest values for all capsaicinoids (Table 3). The GCA effects for capsaicinoids in highly-pungent ‘YS’ were high with positive value, while it was low and negative for low-pungency ‘CW’. The mean values of capsaicinoids significantly differed among all 10 crosses under both seasons (Table 4). In general, mean values for capsaicinoids in all crosses under rainy season were higher than those under dry season. It was interesting to observe that the crosses derived from the highly pungent parent ‘YS’ gave the highest means of all capsaicinoids. They were also superior to those of their parents in all capsaicinoid quantity.

Among the crosses produced from ‘YS’, ‘YS/NM’ and ‘YS/YY’ showed the highest values for all mean of capsaicinoids traits, SCA and their HP%. Although, the cross ‘NM/YY’ showed the highest HP % for all capsaicinoid traits in both seasons, their SCA were not as high as those of the crosses ‘YS/NM’ and ‘YS/YY’. Heritability and gene effects for capsaicinoids Variation in F2 generation showed high values than those in BCP1 and BCP2 (Table 5). The values of narrow-sense heritability (h2

n) varied among capsaicinoids. These values were low for Cap (0.32), very high for Di (0.92) and medium to high for CAPS (0.62). Additive (a), dominant (d) and dominant x dominant (dd) gene effects for capsaicinoid were significantly observed, with d and dd having the large effects for all capsaicinoid traits (Table 6). In addition, the interaction for additive x additive (aa) and additive x dominant (ad) gene effects were not significant. DISCUSSION The interaction between seasons and crosses for all capsaicinoids studied indicated the different responses of different varieties to different seasons. Although, significant differences in SCA for capsaicin (CAP) and

total capsaicinoids (CAPS) were observed, GCA for capsaicinoids was greater than SCA by 3 to 4 times. This indicated that the additive gene effects were more important than the non-additive gene effects for this characteristic (Griffing, 1956).

High capsaicinoid expression in all varieties during the rainy season as compared to dry season may be due to season effect as observed by Gurung et al. (2011), where capsaicinoid production was found to increase at a high-rainfall site. They also explained that although drip irrigation was appropriately applied during both seasons, the plants grown in the rainy season may suffer from excess water supply and increased capsaicinoid accumulation. Considering the highest values for all capsaicinoids in ‘YS’ with positive GCA effect, this variety was considered to be a good parent, which can pass on its capsaicinoid content to the offspring (Lippert, 1975).

Although the cross ‘YS/HS’ was the cross between both highly pungent parents ‘YS’ and “HS’, its capsaicinoid content, SCA and HP% were not as high as those in the crosses ‘YS/NM’ and ‘YS/YY’, which were derived from high-medium and high-low pungent parents, respectively. This could be due to the diverse genetic background for capsaicinoids between the parents of ‘YS/NM’ and ‘YS/YY’, compared to ‘YS/HT’, resulting in high heterosis for capsaicinoids in ‘YS/NM’ and ‘YS/YY’. In contrast to the results from ‘YS/NM’ and ‘YS/YY’, the SCA for capsaicinoids in ‘NM/YY’ was low, even though heterosis was very high. This result could be explained in the light of the low value of the actual capsaicinoids in the ‘NM/YY’ parents, especially in ‘YY’. Therefore, the better parent heterosis, even though important for hybrid development, must somehow involve the actual mean value of target characteristics. Due to the high SCA, better parent heterosis and high actual mean capsaicinoids in YS/NM and YS/YY, these 2 hybrids might be good for developing high pungency varieties.


Table 3. Mean and general combining ability (GCA) effects of the five hot pepper genotypes for capsaicin (Cap), dihydrocapsaicin (Di) and sum of Cap and Di (CAPS) in 5 Thai hot pepper varieties grown under 2 seasons.

. Table 4. Mean, specific combining ability (SCA) and high parent heterosis (HP) for capsaicin (Cap), dihydrocapsaicin (Di) and sum of Cap and Di (CAPS)

in 10 crosses grown under 2 seasons.


253

Table 5. Variations in F2, BCP1, BCP2, additive and narrow sense heritability (h2n) for capsaicin (Cap),

dihydrocapsaicin (Di) and sum of Cap and Di (CAPS) of the cross between Yodson and Num Chiangmai, and their relative generations.

Parameters VF2 VBC1P1 VBC1P2 VA h2n

Cap 66266 80002 31466 21064 0.32 Di 42972 22906 23502 39535 0.92

CAPS 209670 181346 107325 130669 0.62 Table 6. Joint scaling test and estimates (±SE) with six parameters model using six families mean for capsaicin (Cap), dihydrocapsaicin (Di) and sum of Cap and Di (CAPS) of the cross between Yodson and Num Chiangmai, and their relative generations.

Parameters1/ Cap Di CAPS m 3.17 ± 0.18** 1.89 ± 0.04** 5.07 ± 0.18** a 1.16 ± 0.10** 1.61 ± 0.02** 1.77 ± 0.10** d 3.74 ± 0.25** 2.09 ± 0.05** 5.83 ± 0.24** aa ns ns ns ad ns ns ns dd 4.61 ± 0.88* 2.13 ± 0.18** 6.69 ± 0.84*

1/ m = mid-parent, a = additive, d = dominance, aa = additive x additive, ad = additive x dominance, dd = dominance x dominance ns, *, ** mean non-significant, significant at 0.05 and 0.01 probability levels, respectively.

The high heritability normally implies the high possibility of obtaining the high pungent variety in the early generation, particularly by using simple breeding method (Falconer, 1981). In addition, the high broad-sense heritability (h2

b) and h2n of these traits

were also reported by Gill et al. (1973), Doshi (2003) and Techawongstien et al. (2008). From our results with medium to high narrow-sense heritability (h2

n) for capsaicinoid traits, especially for Di in the cross ‘YS/NM’, indicated that improvement of hot pepper with high capsaicinoid might be potentially expected within early or F2 generations. Due to the larger effects of dominance and dominance by dominance for all capsaicinoids parameters studied, it might be possible to produce F1-hybrids with high pungency and yield between YS and NM (Deshpande, 1965; Garcés-Claver et al., 2007). CONCLUSION ‘YS’ variety was considered to be a good parent because of its high actual mean value and high general combining ability for all capsaicinoids. ‘YS/NM’ had the maximum amount of a capsaicinoids during both seasons. The F1 cross ‘YS/NM’ had high specific combining ability for all capsaicinoids. ‘YS/NM’ and ‘YS/YY’ crosses were found to

be potential for high pungency, due to their high SCA, better parent heterosis and also high actual mean capsaicinoids. Medium to high heritability for total capsaicinoids in ‘YS/NM’ indicated the high possibility for obtaining the high pungency segregants in the early generation of breeding program. In addition, due to the large magnitude of dominance and dominance by dominance gene effects for capsaicinoids, F1 crosses between YS and NM were suggested for breeders to produce. ACKNOWLEDGEMENTS The authors are grateful to the National Science of Thailand and Development Association, and Plant Breeding Research Center of Sustainable Agriculture, Khon Kaen University, Thailand for providing financial support for this research. Acknowledgement is extended to Khon Kaen University and the Faculty of Agriculture for providing financial support for manuscript preparation activities. REFERENCES Ahmed N, Singh J, Bajaj KL (1982). Genetics of

capsaicin content in chilli pepper (Capsicum annuum L.). Capsicum and Eggplant Newsletter 1: 33.


Bennett DJ, Kirby GW (1968). Constitution and biosynthesis of capsaicin. J. Chem. Soc. (C): 442-446.

Blum E, Liu K, Mazourek M, Yoo EY, Jahn M, Paran I (2002). Molecular mapping of the C locus for presence of pungency in Capsicum. Genome 45: 702-705.

Ben-Chaim A, Borosky Y, Falise M, Mazourek M, Kang BC, Paran I, Jahn M (2006). QTL analysis for Capsaicinoid content in Capsicum. Theor. Appli. Genet. 113: 1481-1490.

Cavalli LL (1952). An analysis of Linkage in Quantitative Inheritance. In: E.C.R. Reever, and C.H. Waddington, eds., Quantitative Inheritance. HMSO, London, pp. 35-144.

Collins MD, Wasmund LM and Bosland PW (1995). Improved Method for quantifying capsaicinoids in Capsicum using high performance liquid chromatography. HortScience 30: 137-139.

Customs Department, Thailand (2009). Import/Export Statistics, (cited 1 September 2009). Available from: URL: http//www.customs.go.th/wps/wcm/connect/Library

Department of Agricultural Extension, Thailand (2009). Agricultural Statistics. (cited 1 September 2009). Available from URL: http://www.agriinfo.doae.th/.

Deshpande RB (1935). Studies in Indian chillis: 4. Inheritance of pungency in Capsicum annuum L. Indian J. Agric. Sci. 5: 513-516.

Doshi KM (2003). Genetic architecture of chilli (Capsicum annuum L.). Capsicum and Eggplant Newsletter 22: 33-36.

Falconer DS (1981). Introduction to quantitative genetics. (2nd ed.) Longman. New York.

FAOSTAT. 2009. Production crops. FAOSTAT Agricultural production database. (cited 1 September 2009). Available from URL: http://faostat.fao.org.

Gamble EE (1962). Gene effects in corn (Zeamays L.) I. Separation and relative importance of gene effects for yield. Can. J. Plant Sci. 42: 339-348.

Garcés-Claver A, Gil-Ortega R, Álvarez-Fernández A, Arnedo-Andrés MS (2007). Inheritance of capsaicin and dihydrocapsaicin, determined by HPLC-ESI/MS, in an intraspecific cross of Capsicum annuum L. J. Agric. Food Chem. 55: 6951-6957.

Gill K, Ghai S, Singh JR (1973). Inheritance of amount of capsaicin in chilli (Capsicum frutescens L. and Capsicum annuum L.). Indian J. Agric. Sci. 43(9): 839-841.

Griffing B (1956). Concept of general and specific combining ability in relation to diallel crossing systems. Australian J. Biol.Sci. 9: 463-493.

Gurung T, Techawongstien S, Suriharn B, Techawongstien S (2011). Impact of environments on the accumulation of capsaicinoids in Capsicum spp. HortScience 46(12):1-6.

Techawongstien S, Khemnak S, Theerakulpisut P (2008). Determination of the pungency inheritance in Capsicum by molecular technique. Acta Hort. 765 (ISHS): 109-116.

Lippert LF (1975). Heterosis and combining ability in chilli peppers by diallel analysis. Crop Sci. 15: 323-325.

Mather K, Jinks JL (1977). Introduction to biometrical genetics. Cornell University Press. New York.

Ohta Y (1962). Physiological and genetical studies on the pungency of capsicum: Inheritance of pungency. Jpn. J. Genet. 37: 169-175.

Rowe KE, Alexander WL (1980). Computations for estimating the genetic parameters in joint-scaling tests. Crop Sci. 20: 109-110.

Webber HJ (1912). Preliminary notes on pepper hybrids. Am. Breed. Assoc. Annu. Rep.7: 188-189.

Zewdie Y, Bosland PW (2000). Capsaicinoid inheritance in an interspecific hybridization of Capsicum annuumx C. chinense. J. Am. Soc. Hortic. Sci. 125 (4): 448-453.

Zewdie Y, Bosland PW (2001). Combining ability and heterosis for capsaicinoids in Capsicum pubescens. HortScience 36(7): 1315-1317.


SCREENING OF WHEAT GENOTYPES FOR DROUGHT TOLERANCE BASED ON DROUGHT RELATED INDICES

YAHYA RAUF1, ABID SUBHANI1, MUHAMMAD SHAHID IQBAL1, 3*, MUHAMMAD TARIQ1, ABID MAHMOOD1, 2and MUHAMMAD KAUSAR

NAWAZ SHAH1, 3

1Barani Agricultural Research Institute, Chakwal, Pakistan

2Cotton Research Institute, Ayub Agricultural Research Institute, Faisalabad, Pakistan 3PMAS-Arid Agriculture University, Rawalpindi, Pakistan *Corresponding author’s email: [email protected]

SUMMARY

Germplasm comprising of 61 entries was evaluated at Barani Agricultural Research Institute, Chakwal, under drought and irrigated conditions during 2008-09 and 2009-10. Under irrigated field conditions, 3 irrigations (75 mm each) were applied in addition to seasonal rainfall (252.6 and 104.1) at three critical crop growth stages (40 days after planting, booting and grain-filling stage). No irrigation was applied under drought field conditions during both crop seasons; it was also protected from seasonal rains by using a rain shelter for 70 days from crop emergence. Morphological and physiological parameters were recorded along with drought-related indices such as drought tolerance efficiency (DTE) and drought susceptibility index (DSI) to evaluate the best performing lines under water stress conditions. The drought tolerant genotypes with the minimum yield reduction and highest DTE and lowest DSI were TE173/4/LEE/KVZ/3/CC/RON/CHA, DH-7, WC-15, ML-177 and DH-4 in 2008-09 while LLR-30, DH-4, LLR-42, 5C003, and DH-6 in 2009-10. Inqilab-91, DH-11, DH-12, DH-8, DH-9 and WC-9 (in 2008-09) and DH-15, Inqilab-91, DH-20, LLR-8, DH-26 and WC-18 (in 2009-10) were the most drought-susceptible genotypes with maximum yield losses, lowest DTE, and highest DSI. Keywords: Drought tolerance, wheat physiology, drought indices

Manuscript received: July 14, 2012; Decision on manuscript: March 28, 2013; Manuscript accepted: April 28, 2013. © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION

In the arid and semi-arid regions of the world, precipitation is low and erratic, and drought is frequent. Water shortage is a major constraint to agricultural production (Corbeels et al., 1998; Zhang and Oweis, 1999; Oweis et al., 2000).Yield of cereal crops in rainfed areas are generally low at 0.6-1.5 t/ha. Application of nitrogen (N) and

phosphorus (P) fertilizers and other management practices can increase dry matter production and grain yield (Cooper & Gregory, 1987; Ludlow & Muchow, 1990) provided ample soil moisture is available for efficient nutrient uptake. Several studies indicate that these practices could also have negative effects during seasons when water is severely limiting (Copper et al., 1987; Van den Boogaard et al., 1996; El Mejahed and

RESEARCH ARTICLE


256

Aouragh, 2005; Rusan et al., 2005). The unpredictability of rainfall makes it difficult to determine the level and timing of fertilizer needed to attain optimum yields, as it might result in over- or under-fertilization of N (Rusan et al., 2005). As N fertilizer is an expensive input, to get the considerable economic yield, it is necessary to develop wheat varieties with high nutrient use efficiency.

A water crisis is a severe threat to sustainable agriculture, particularly in most of the Asian countries where irrigated agriculture accounts for 90% of total diverted fresh water (Huaqi et al. 2002). Plants can resist high temperatures by losing water as a result of transpiration through stoma closure (Mujdeci et al., 2007).

Achieving genetic increases in yield under rainfed conditions has always been a difficult challenge for plant breeders. A likely reason is that dry environments have unpredictable and highly variable seasonal rainfall and, hence, highly variable yields. In addition to this, farmers in rainfed areas do not apply the recommended fertilizer doses due to less moisture availability and unpredictable rains. This is directly related to nutrient uptake and constitutes a major constraint to yield.

Yield and drought resistance are controlled at separate genetic loci (Morgan, 1984). Breeding should involve the identification of physiological traits responsible for drought resistance. The utility of a particular measurement depends on the rapid assessment of the plant at a critical stage using large quantities of plant material. Some physiological screening tests are suitable for testing a large number of plant genotypes (Sayar et al., 2005).

Water loss can lower leaf water potential, leading to reduced turgor, stomatal conductance, photosynthesis and ultimately, reduced growth and yield. Several physiological characters, which can contribute to continued growth under water stress, have been identified (Garcia del Moral et al., 2003).

This study was conducted to identify drought-tolerant wheat genotypes for use in recombination breeding to develop varieties

with high yield potential under low fertilizer input for rainfed areas of Pakistan. MATERIALS AND METHODS An experiment was conducted to screen wheat genotypes for drought tolerance at Barani Agricultural Research Institute, Chakwal, during 2008-09 and 2009-10. Wheat genotypes comprising 61 entries were planted under irrigated and drought conditions by using a RCBD split-plot design having three replications and three fertilizer levels (recommended dose, half of the recommended dose, and no fertilizer). Plot size for each entry was 0.45m2 and all the fertilizers were applied before planting at the recommended dose of 90-60-60 kg NPK/ha. In the irrigated field, three irrigations (75 mm each) were applied to supplement seasonal rainfall (Table 2) at three different critical crop growth stages - 40 days after planting, booting and grain fill stage. No irrigation was applied to the drought field in both seasons; it was also protected from seasonal rains for 70 days ater crop emergence

Soil analysis Soil analysis at pre-sowing was done to check the soil fertility status in both irrigated and drought fields by taking samples from two soil depths - 0-15 and 16-30 cm. (Table-1). The representative soil samples were collected at two depths with a soil augur from different sites and composite samples were made. Then the samples were air-dried, ground and passed through a 2 mm sieve and preserved for further analysis. The soil texture (particle size distribution) was determined by hydrometer method of Day (1965). Gravimetric soil water contents, electrical conductivity of the soil extract (ECe), pH and extractable potassium were determined according to the methods described by U. S. Salinity Laboratory staff (U. S. Salinity Lab. 1954). Organic matter and available phosphorus were determined by Walkely’s and Olsen’s methods, respectively as described by Jackson (1962). Total N was

Rauf et al. (2013)

257

determined by micro-Kjeldahl method as described by Anderson and Ingram (1993). Drought susceptibility index calculations Drought intensity index (DII) was used to compare the stress between two or more experiments, the higher the value the greater the drought stress (Ramírez-Vallejo and Kelly, 1998). DII = (1-Xs / Xi) where: Xs is the mean experiment yield of all genotypes grown under stress Xi is the mean experiment yield of all genotypes grown under non-stress conditions.

DSI of a line was calculated as follows: DSI = (1-Ys / Yi)/ DII (Fischer and Maurer, 1978) where Ys is yield of particular line in stress Yi is yield of that line in non-stress condition.

The morphological parameters were recorded for days to heading, plant height, spike length, grains per spike, 1000 grain weight, grain yield and total biomass. The analysis of variance was carried out using MSTAT-C statistical package (Nissen, 1991).

Table 1. Physical and chemical characteristics of field soil before sowing. Characteristic Drought field Irrigated field

0-15cm 15-30cm 0-15cm 15-30cm Textural class Sandy loam Sandy loam Sandy loam Sandy loam Soil water content (%) 8.2 8.9 8.3 8.8 pH 8.1 8.0 8.1 8.1 ECe (d S/m) 0.70 0.68 0.71 0.66 Organic matter 0.65 0.47 0.62 0.49 Available P (mg/kg) 6.20 5.10 5.70 5.30 Extractable K (mg/kg) 115 85 119 80 Total N (%) 0.033 0.024 0.036 0.025 Table 2: Total rainfall and irrigation (mm) in irrigated and drought fields during two crop seasons (2008-09 and 2009-10). ______________________________________________________________________________________

Month Irrigated Drought (Rain shelter) 2008-09 2009-10 2008-9 2009-10 Rain Irr. Rain Irr. ________________________________________________________________________________________November 2.0 - 7.1 - - - December 68.5 75.0 0.0 75.0 - - January 19.7 - 19.7 - - - February 39.0 75.0 55.0 75.0 39.0 55.0 March 38.8 75.0 16.3 75.0 38.8 16.3 April 84.6 6.0 - 84.6 6.0 ________________________________________________________________________________________ Total 252.6 225.0 104.1 225.0 162.4 77.3 ________________________________________________________________________________________


258

Table 3. Mean squares for morphological and physiological parameters of 61 wheat genotypes under irrigated and drought conditions during 2008-09.

IRRIGATED

df Grain yield 1000 grain Grains/spike Pn E C weight

Replications 2 112280.649 16.298 1636.036 26.03 3.68 11988.02 Genotypes 60 19136.016* 98.635* 398.253* 17.06 1.94* 2385.68 Error 360 6082.944 34.382 141.846 12.95 1.31 1904.60 DROUGHT

Replications 2 2257.772 64.240 222.551 3.505 8.218 385.98 Genotypes 60 3827.681* 101.015* 293.446* 26.698 1.502* 1989.286* Error 360 2689.270 68.154 108.378 20.261 0.945 1288.216 ________________________________________________________________________________________ Pn-= net photosynthetic rate (µmol/m2/s), E= transpiration rate (mmol/m2/s), C= Stomatal conductance (mmol/m2/s). *P < 0.05 Table 4. Mean squares for morphological and physiological parameters of 61 wheat genotypes under irrigated and drought conditions during 2009-10. ________________________________________________________________________________________ IRRIGATED

df Grain yield 1000 grain Grains/spike Pn E C weight

Replications 2 31786.739 9.611 1622.253 1686 46.327 28663.8 Genotypes 60 8035.470* 120.602* 99.881* 289.9* 8.7* 27784.052* Error 360 1692.074 1.860 57.277 87.5 0.538 1035.479 DROUGHT

Replications 2 22772.610 3.145 27.492 100.8 12.543 9766.878 Genotypes 60 3291.369* 50.268* 69.074 144.6* 4.306* 3278.071* Error 360 954.857 0.720 53.685 19.24 0.075 271.102 Pn-= net photosynthetic rate (µmol/m2/s), E= transpiration rate (mmol/m2/s), C= Stomatal conductance (mmol/m2/s). *P < 0.05 RESULTS AND DISCUSSION The analysis of variance (Tables 3 and 4) for 2008-09 and 2009-10 respectively, revealed highly significant differences between genotypes for most of the studied traits under irrigated and drought conditions. These results indicate that the genotypes responded differently to different water regimes, suggesting the importance of assessing genotypes under different environments in order to identify the best genetic makeup for drought areas.

The drought tolerance efficiency (DTE) value, which was one of the drought

resistance parameters, ranged from 31.6-99% in 2008-09 to 41.5-97.8% in 2009-10. Thus, Inqilab-90 (31.6%), DH-11 (33.2%), DH-12 DH-2 (37.1%), DH-8 (46.8%), DH-9 (47.5%) and WC-9 (47.6%) showed the lowest DTE in 2008-09. DH-15 (41.5%), Inqilab-91(47.9%), DH-20 (49.4%), LLR-8 (50.7%), DH-26 (57.1%) and WC-18 (57.3%) showed the lowest DTE values in 2009-10.

Values for drought susceptibility index (DSI) ranged between 0.03-2.39 and 0.08-2.05 during 2008-09 and 2009-10, respectively. The DSI values of the above-mentioned genotypes were 2.39, 2.33, 2.20, 1.88, 1.83 and 1.83 in 2008-09 while, 2.05,

Rauf et al. (2013)

259

1.83, 1.78, 1.73, 1.50 and 1.50 in 2009-10, respectively.

On the other hand, E173/4/LEE/KVZ/3/CC/RON/CHA (99.0%), DH-7 (96.5%), WC-15 (91.1%), ML-177 (89.6%) and DH-4 (90%) in 2008-09. In the following year (2009-10), LLR-30 (97.8%), DH-4 (97.7%), LLR-42 (97.1%), 5C003 (89.8%), and DH-6 (88.6%) had the highest DTEs. The lowest DSI values in 2008-09 were 0.03, 0.12, 0.31, 0.31 and 0.35, and in 2009-10, these were 0.08, 0.08, 0.1, 0.36 and 0.4. This study was consistent with the findings of Parameshwarappa et al. (2008). They reported that minimum yield reduction was realized in genotypes with the highest DTE and lowest DSI. Actually, TE173/4/LEE/KVZ/3/CC/RON/CHA, DH-7, WC-15, ML-177 and DH-4 in 2008-09 and LLR-30, DH-4, LLR-42, 5C003 and DH-6 in 2009-10 were found to be the most drought-resistant genotypes with minimum yield reduction coupled with the highest DTE and lowest DSI; Inqilab-91, DH-11, DH-12, DH-8, DH-9 and WC-9 in 2008-09. In 2009-10, the most drought susceptible genotypes with maximum yield losses, lowest DTE and highest DSI were DH-15, Inqilab-91, DH-20, LLR-8, DH-26 and WC-18.

Most findings (Ozkan et al., 1998; Ozturk, 1999; Golabadi et al., 2006; Sio-Se Mandeh et al., 2006) showed that genotypes with the lowest DSI values were more tolerant than those with the highest DSI. Ahmad et al. (2003) have also reported that drought-susceptible varieties had higher values (DSI > 1), while resistant varieties had lower values (DSI < 1). In the crop season of 2009-10, rainfall was less and ultimately total water availability (329 mm) in the irrigated field (Table 2) was less as compared with the 1st year`s (2008-09) total water availability (477.6mm). These caused yield losses and restricted the yield in the range of 144.4 to 272.6 g/plot (Table 6) and 90.0 to 293.0 g/plot (Table 5) in 2009-10 and 2008-09, respectively. The water availability in the field exposed to drought (rain shelter) was 77.3 mm during 2009-10 by rains after 70 days of drought and yield ranged from 98.8 to 190.7 g/plot (Table 6), while it was

162.4mm and yield range was 70.0-162.0 g/plot (Table 5) in 2008-09.

Morphological parameters such as 1000-grain weight, number of grains per spike and harvest index gave highly significant results, except for the number of grains per spike, which showed non-significant results under drought conditions during 2009-10. Physiological characteristics such as net photosynthetic rate, stomatal conductance and transpiration rate showed significant results during the two consecutive crop cycles under drought and irrigated conditions, except for net photosynthetic rate under drought conditions during 2008-09.

CONCLUSION

In this experiment, genotypes, TE173/4/LEE/KVZ/3/CC/RON/CHA, DH-7, WC-15, ML-177 and DH-4 (2008-09) and LLR-30, DH-4, LLR-42, 5C003, and DH-6 (2009-10) were found to be the most drought-resistant with minimum yield reduction and also the highest DTE and lowest DSI. Inqilab-91, DH-11, DH-12, DH-8, DH-9 and WC-9 (2008-09) and DH-15, Inqilab-91, DH-20, LLR-8, DH-26 and WC-18 (2009-10) were the most drought susceptible genotypes with maximum yield losses lowest DTE and highest DSI values.


260

Table 5. Means of morphological and drought-related parameters of wheat genotypes under irrigated (I) and drought (D) conditions during 2008-09.

Genotype

Grain yield (g)

1000 grain Wt (g)

Grains/spike (no.)

*HI DTE (%)

DSI

I D I D I D I D Blue Silver 132 105 42.6 42.6 65.9 40.5 35.4 28.5 79.3 0.72 LU-26 130 116 43.5 40.4 60.9 44.9 33.7 23.1 89.3 0.37 Shalimar-88 200 112 39.9 40.8 57.5 43.4 36.6 27.2 56.0 1.54 Inqilab-90 244 77 44.0 40.4 51.9 31.5 36.3 25.4 31.6 2.39 Kohistan-97 108 90 43.9 40.8 55.8 40.3 29.1 20.4 82.9 0.60 GA-2002 183 117 44.0 37.9 55.8 37.3 34.6 25.5 64.0 1.26 Chakwal-50 120 70 41.3 39.7 52.1 35.3 33.5 24.6 58.6 1.45 DH-1 104 92 42.9 39.5 59.5 40.3 31.3 25.5 88.1 0.42 DH-2 148 96 39.8 39.2 64.4 35.4 31.4 23.8 65.0 1.22 DH-3 124 88 40.6 36.0 55.0 36.5 33.3 21.6 71.1 1.01 DH-4 137 123 41.9 39.4 52.1 44.9 33.5 27.2 90.0 0.35 DH-6 156 110 43.9 37.4 62.3 36.3 36.1 21.8 70.4 1.03 DH-7 132 128 43.2 37.1 58.0 42.4 34.3 27.4 96.5 0.12 DH-8 265 124 42.8 39.8 63.5 43.8 37.0 25.9 46.8 1.86 DH-9 293 139 45.7 47.9 63.8 35.4 35.6 28.8 47.5 1.83 DH-11 250 83 48.3 43.9 58.7 29.3 32.7 23.9 33.2 2.33 DH-12 190 70 43.3 38.5 56.3 32.7 32.8 17.9 37.1 2.20 DH-13 165 99 42.4 39.5 56.8 31.3 32.6 23.3 60.4 1.38 DH-15 178 113 37.2 38.8 58.4 39.4 32.2 17.4 63.4 1.28 DH-20 134 82 40.1 35.9 60.4 30.9 28.7 29.3 61.7 1.34 DH-24 90 81 38.4 38.0 46.9 37.1 30.1 22.5 89.6 0.36 DH-26 120 95 38.3 37.6 40.7 36.3 26.3 20.4 79.4 0.72 DH-28 156 104 46.4 31.3 45.6 38.3 34.5 26.1 66.4 1.17 DH-29 111 139 42.2 41.6 55.4 42.8 34.6 21.6 60.2 0.75 DH-30 150 107 41.2 35.9 60.1 45.9 36.3 29.0 71.4 1.00 DH-31 161 115 38.3 44.2 54.1 42.4 33.3 26.7 71.2 1.01 LLR-8 257 144 48.8 41.8 45.6 35.8 34.5 27.2 56.0 1.54 LLR-15 170 87 39.7 39.0 46.1 38.0 31.8 24.6 51.4 1.70 LLR-18 142 103 34.0 40.0 48.2 37.7 32.4 20.4 72.7 0.95 LLR-19 146 122 39.5 39.5 48.1 35.8 31.8 20.4 83.8 0.57 LLR-30 173 110 42.7 42.9 42.1 41.1 33.3 21.5 63.6 1.27 LLR-34 127 102 40.3 40.5 41.4 36.7 32.6 22.0 80.0 0.70 LLR-42 187 111 39.8 39.0 43.4 34.8 32.5 24.4 59.6 1.41 LLR-44 103 119 41.0 44.9 46.7 35.9 32.4 21.1 58.3 0.83 WC-2 169 120 46.0 41.3 57.7 50.7 38.0 30.2 71.0 1.01 WC-4 131 107 44.4 40.3 47.0 46.0 35.4 28.1 81.7 0.64 WC-5 138 121 38.9 40.3 54.0 52.7 34.8 24.6 87.7 0.43 WC-9 216 103 52.8 47.9 47.5 42.1 35.6 23.5 47.6 1.83 WC-11 184 115 39.4 37.7 52.3 43.1 37.9 21.4 62.8 1.30 WC-13 129 93 36.7 36.1 53.4 41.1 35.6 29.1 72.2 0.97 WC-15 151 138 45.7 46.0 46.7 40.6 37.9 28.7 91.1 0.31 WC-16 168 114 47.1 47.6 45.4 36.2 37.5 26.7 68.1 1.11 WC-18 141 154 47.8 46.2 54.7 50.3 37.1 30.5 59.8 0.75 ERA 265 134 43.5 44.2 55.5 44.9 42.5 30.3 50.8 1.72 MILAN 125 153 38.2 45.2 52.0 44.7 32.8 32.6 62.7 0.75 PF70402/ALDS/PAT72/160/ALDS/3.PEWS

144 128 43.7 44.1 46.9 39.1 32.2 29.9 89.1 0.38

CM079*2/PRLS 108 109 45.2 39.0 41.4 48.9 33.8 32.2 65.3 0.69 TE173/4/LEE/KVZ/3/CC/RON/CHA 134 133 43.1 41.7 45.6 43.4 31.1 26.8 99.0 0.03 VEES/3/TAST/MO//NAC 99 124 42.0 42.1 51.2 46.3 33.6 28.8 64.3 0.64 HD2329 223 138 41.5 40.8 56.9 52.6 39.3 29.3 61.7 1.34 KUFRA-1 96 126 39.2 41.6 57.8 48.2 33.4 20.8 57.5 0.74 BOWS/NAC/VEES/3/BYJS/COC 151 116 42.4 39.7 57.7 45.9 35.9 28.7 76.7 0.81 CUMHURIYET/NE/10443 149 121 37.9 37.1 46.9 47.8 33.4 26.1 81.2 0.66 00FJ03 211 115 42.0 45.5 50.0 44.6 39.4 26.1 54.4 1.59 HD2169 202 162 41.3 39.2 59.1 54.6 35.4 34.3 80.4 0.68 CMH76A.912/CMH76A.769 132 105 43.2 43.9 59.9 40.2 32.8 35.7 79.4 0.72 BOWS/GHK’S// PRL’S 122 131 38.9 39.0 39.4 45.6 32.6 25.4 63.4 0.68 Milan/Kauz//Babax/3/Babax/FRET2 122 79 44.8 40.7 44.8 35.5 30.8 25.5 64.8 1.23 Sitta 125 100 41.4 35.0 48.9 43.4 31.0 21.8 80.3 0.69 ML-177 147 134 43.1 44.6 49.9 45.4 34.3 27.2 91.1 0.31 5C003 163 120 46.5 39.5 55.2 42.0 38.2 25.7 73.6 0.92 LSD (0.05) 72.30 48.08 5.44 7.65 11.04 9.65 6.07 7.43 - - *HI = harvest Index, DTE= drought tolerance efficiency, DSI= drought susceptibility index

Rauf et al. (2013)

261

Table 6. Means of morphological and drought-related parameters of wheat genotypes under irrigated and drought conditions during 2009-10.

HI = harvest index, DTE= drought tolerance efficiency, DSI= drought susceptibility index

Genotypes

Grain yield (g)

1000-grain wt (g)

Grains/spike (no.)

HI* DTE (%)

DSI

I Dt I D I D I D Blue Silver 219.2 130.4 37.5 29.1 34.0 31.6 33.1 34.8 59.5 1.42 LU-26 157.9 137.5 35.3 30.8 32.1 34.3 30.9 36.6 87.1 0.45 Shalimar-88 221.9 134.7 36.1 30.9 38.4 38.3 37.8 39.7 60.7 1.38 Inqilab-91 272.6 130.5 34.8 31.5 35.3 37.7 35.2 40.6 47.9 1.83 Kohistan-97 179.0 140.8 27.5 29.1 36.4 36.3 33.6 39.1 78.6 0.75 GA-2002 224.2 129.7 36.0 29.6 39.0 34.1 38.2 38.5 57.9 1.48 Chakwal-50 194.6 169.4 36.9 31.1 38.6 38.4 38.8 40.5 87.0 0.45 DH-1 152.0 125.0 39.5 31.1 32.3 33.4 30.6 33.5 82.2 0.62 DH-2 187.3 153.7 36.1 31.0 35.8 36.1 31.4 35.4 82.1 0.63 DH-3 191.7 143.0 37.1 30.4 41.1 34.8 32.5 37.6 74.6 0.89 DH-4 161.8 158.1 35.8 34.0 36.2 33.0 28.6 34.8 97.7 0.08 DH-6 161.3 142.9 34.9 30.1 34.7 31.4 32.4 37.9 88.6 0.40 DH-7 179.8 150.3 33.8 31.2 36.4 32.3 35.6 37.0 83.6 0.58 DH-8 183.6 152.9 30.9 30.5 35.4 32.8 35.3 38.2 83.3 0.59 DH-9 194.1 126.7 37.3 31.2 38.7 33.8 34.6 34.4 65.3 1.22 DH-11 176.3 128.7 41.0 34.2 31.7 34.1 30.7 32.7 73.0 0.95 DH-12 170.0 127.7 38.0 34.7 34.5 36.1 29.4 36.1 75.2 0.87 DH-13 177.4 137.5 36.6 36.0 33.7 33.0 27.5 34.1 77.5 0.79 DH-15 257.9 107.0 32.2 29.1 33.0 36.1 38.5 38.1 41.5 2.05 DH-20 200.2 98.8 36.1 28.0 34.7 35.8 37.2 38.6 49.4 1.78 DH-24 144.4 107.8 37.2 30.0 30.5 36.8 29.3 34.6 74.6 0.89 DH-26 203.9 116.4 33.2 26.6 32.7 34.9 37.4 36.9 57.1 1.50 DH-28 199.6 145.2 32.6 28.6 38.6 33.3 38.5 41.3 72.8 0.95 DH-29 219.9 136.8 32.2 32.4 35.6 33.6 38.1 39.0 62.2 1.33 DH-30 220.8 143.6 31.4 25.8 37.6 36.1 36.6 37.1 65.0 1.23 DH-31 190.9 137.5 35.3 33.4 36.1 35.2 34.5 36.7 72.0 0.98 LLR-8 243.6 123.5 31.8 29.3 32.0 33.8 28.1 35.0 50.7 1.73 LLR-15 180.1 134.3 28.8 30.5 36.0 34.6 27.1 33.7 74.5 0.89 LLR-18 177.3 119.6 27.8 28.3 28.9 36.3 29.7 30.9 67.4 1.14 LLR-19 175.6 152.0 30.0 30.0 38.6 40.8 30.8 34.4 86.6 0.47 LLR-30 162.8 159.2 29.7 31.5 34.6 41.6 29.9 35.7 97.8 0.08 LLR-34 198.1 155.8 30.6 31.5 32.4 37.8 32.1 37.6 78.7 0.75 LLR-42 196.3 190.7 31.4 31.5 34.3 36.8 30.3 36.6 97.1 0.10 LLR-44 146.9 112.9 34.4 31.7 32.4 36.7 34.4 37.5 76.8 0.81 WC-2 154.8 122.9 34.6 28.3 31.1 33.9 34.9 36.3 79.4 0.72 WC-4 159.8 119.2 37.0 28.5 28.9 36.8 37.6 39.0 74.6 0.89 WC-5 147.5 109.4 32.7 27.6 29.0 35.9 29.2 31.3 74.2 0.91 WC-9 147.2 110.7 33.3 33.8 30.2 34.9 30.1 31.4 75.2 0.87 WC-11 181.9 107.4 32.6 27.4 39.8 35.1 34.1 35.9 59.1 1.43 WC-13 191.8 122.9 35.0 27.6 37.2 38.2 39.2 39.3 64.1 1.26 WC-15 204.2 142.4 39.3 34.2 35.9 35.3 36.8 37.6 69.8 1.06 WC-16 153.8 112.3 42.2 35.3 33.9 37.6 34.1 37.4 73.0 0.95 WC-18 252.3 144.6 41.8 33.8 39.7 36.8 36.2 38.7 57.3 1.50 ERA 247.4 152.5 38.2 31.2 40.2 40.3 37.2 40.0 61.7 1.34 MILAN 249.9 167.0 41.0 30.2 35.6 37.6 37.8 39.5 66.8 1.16 PF70402/ALDS/PAT72/160/ALDS/3.PEWS 191.8 142.3 43.0 32.8 34.6 39.6 35.8 40.0 74.2 0.90 CM079*2/PRLS 183.9 131.0 42.0 31.6 36.8 36.4 35.7 37.1 71.2 1.01 TE173/4/LEE/KVZ/3/CC/RON/CHA 190.2 125.4 37.5 29.8 36.1 39.0 33.9 35.7 65.9 1.19 VEES/3/TAST/MO//NAC 221.8 136.6 35.3 28.6 40.7 41.0 35.4 35.4 61.6 1.35 HD2329 174.6 134.7 34.1 27.7 32.9 37.6 33.9 36.8 77.1 0.80 KUFRA-1 207.3 139.0 36.5 31.0 39.0 36.6 36.9 38.4 67.1 1.15 BOWS/NAC/VEES/3/BYJS/COC 183.1 124.0 35.7 29.0 35.9 38.1 34.4 34.8 67.7 1.13 CUMHURIYET/NE/10443 199.1 136.3 31.1 26.6 32.5 37.3 34.9 35.2 68.5 1.11 00FJ03 183.5 154.9 37.0 26.7 33.8 40.0 37.6 40.0 84.4 0.55 HD2169 201.3 168.4 34.5 29.7 36.5 39.8 33.8 39.0 83.6 0.57 CMH76A.912/CMH76A.769 196.3 153.6 36.1 34.9 34.5 38.2 35.6 39.2 78.3 0.76 BOWS/GHK’S//PRL’S 177.4 152.7 29.3 28.7 38.7 39.1 30.3 34.7 86.1 0.49 Milan/Kauz//Babax/3/Babax/FRET2 214.8 154.4 39.0 31.8 43.5 43.3 40.2 41.1 71.9 0.98 Sitta 196.4 133.4 29.0 28.4 42.9 43.2 35.8 37.1 67.9 1.12 ML-177 245.9 184.7 36.1 33.3 36.8 41.67 40.3 44.4 75.1 0.87 5C003S 181.5 163.0 36.1 30.8 39.7 39.1 33.4 46.1 89.8 0.36 LSD (0.05) 38.13 28.65 1.264 0.7866 7.016 6.793 3.287 3.299 - -


262

REFERENCES Ahmad R, Quadir S, Ahmad N, Shah KH (2003).

Yield potential and stability of nine wheat varieties under water stress conditions, Int. J.Agric.Biol. 5 (1): 7-9.

Anderson JM, Ingram JSI (1993). Tropical soil biology and fertility. p. 47-104. In: A handbook of methods. CAB International, U.K.

Chowdhry MA, Rasool I, Khaliq I, Mahmood T, Gilani MM (1999). Genetics of some metric traits in spring wheat under normal and drought environment. Rachis Newsl. 18: 34-39.

Cooper PJM, Gregory PJ (1987). Soil water management in the rain fed farming systems of the Mediterranean region. Soil Use Manage. 3: 57-62.

Cooper PJM, Gregory PJ, Keatings JDH, Brown SC (1987). Effect of fertilizer, variety and location on barley production under rainfed conditions in Northern Syria. 2. Soil water dynamics and crop water use. Field Crops Res. 16: 67-84.

Corbeels M, Hofman G, Van-Cleemput O (1998). Analysis of water use by wheat grown on a cracking clay soil in a semi arid Mediterranean environment: weather and nitrogen effects. Agric. Water Management 38: 147-148.

Day PR (1965). Particle fractionation and particle size analysis. p. 545-567. In: C.A. Black (Ed.). Methods of soil analysis., Part-1. Agronomy Monograph, Academic Press, New York.

El-Mejahed K, Aouragh L (2005). Green manure and N fertilizer in soil quality and profitability of wheat-based system in semi arid Morocco using nuclear techniques. In: Management of nutrients, water in rainfed arid, semi-arid areas for increasing crop production, IAEA-TECDOC-1468 p. 89-106.

Fischer KS, Wood G (1981). Breeding and selection for drought tolerance in tropical maize. In: Proc. Symp. on Principles and methods in crop improvement for drought resistance with emphasis on rice, IRRI, Philippines.

Fischer RA, Maurer R (1978). Drought resistance in spring wheat cultivars. I. Grain yield responses in spring wheat. Aust. J. Agric. Sci. 29: 892-912.

Garcia-del Moral, Rharrabti LF, Villegas Y Royomm C (2003). Evaluation of grain yield and its components in durum wheat

under Mediterranean conditions: An ontogenic approach. Agron. J. 95: 266-274.

Golabadi M, Arzani A, Maibody SAM (2006). Assessment of drought tolerance in segregating populations in durum wheat, Afr. J. Agric. Res. 1 (5): 162-171.

Huaqi W, Bouman BAM, Zhao D, Changgui W, Moya PF (2002). Aerobic rice in northern China: opportunities and challenges. Workshop on water wise rice production, 8-11 April IRRI, Los Banos, Philippines.

Jackson ML (1962). Soil chemical analysis, Edition-1, Constable and Company Limited, London.

Ludlow MM, Muchow RC (1990). A critical evaluation of traits for improving crop yields in water-limited environments. Adv. Agron. 43, 107-153.

Mujdeci M, Sarıyev A, ÇelikI, Tulun Y, Karaca H (2007). Investigation of the relationships between stoma resistance and soil water matric potential with other environmental factors in respect of mathematical modeling of plant water consumption. Asian J. Chem. 19 (3): 2045-2052.

Morgan JM (1984). Osmoregulation and water stress in higher plants. Annu. Rev. Plant Physiol. 35: 299-319.

Nissen O (1990). MSTAT-C: A microcomputer program for the design management and analysis of agronomic research experiments. Michigan State University.

Oweis T, Zhang H, PalabM (2000). Water use efficiency of rainfed and irrigated bread wheat in a Mediterranean environment. Agron. J. 92: 231-238.

Ozkan H, Yagbasanlar T, Genc I (1998). Tolerance and stability studies on durum wheat under drought and heat stress conditions, Cereal. Res. Commun. 26: 405-412.

Ozturk A (1999). Drought resistance in bread wheat genotypes. Trop. J. Agric. For. 23 (5): 1237-1247

Parameshwarapp SG, Salimath PM (2008). Field screening of chickpea genotypes for drought resistance, Karnataka. J. Agric. Sci. 21(1): 113-114.

Rusan MM, Battikhi A, Zuraiqi S (2005). Enhancement of nitrogen and water use efficiency by optimizing the combination of soil, crop and nitrogen management. In: Management of nutrient and water in rainfed arid and semi-arid areas of

Rauf et al. (2013)

263

increasing crop production, IAEA-TECDOC-1468 p. 155-177.

Sayar R, Khemira H, Bensalem M, Kameli A (2005). Drought tolerance evaluation tests for durum wheat, Poster presented at the International conference on integrated approaches to sustain and improve plant production under drought stress. Inter Drought-11: p. 3-61.

Sio-Se MA, Ahmadi A, Poustini K, Mohammadi V (2006). Evaluation of drought resistance indices under various environmental conditions. Field Crops. Res. 98: 222-229.

United States Department of Agriculture Salinity Lab (1954). Diagnosis and Improvement of Saline and Alkaline Soils. USDA. Handbook No. 60, Washington, D.C.

Van-den Boogaard R, Goubits S, Veneklaas EJ, Lambers H (1996). Carbon and nitrogen economy of four Triticum aestivum cultivars differing in relative growth rate and water use efficiency. Plant Cell Environ. 19: 998-1004.

Zhang H, Oweis T (1999).Water-yield relations and optimal irrigation scheduling of wheat in the Mediterranean region. Agric. Water Manage. 38: 195-211.

z SABRAO Journal of Breeding and Genetics 45 (2) 264-275, 2013

RELATIONSHIP BETWEEN ELEVATION LEVELS AND YIELDS OF TROPICAL WAXY CORN GENOTYPES

BHALANG SURIHARN1,2* and KAMOL LERTRAT1,2


Khon Kaen, 40002 Thailand 2Plant Breeding Research Center for Sustainable Agriculture, Faculty of Agriculture, Khon Kaen University Thailand 40002


SUMMARY

An understanding of the responses of waxy corn (Zea mays L. var. ceratina) to elevation is important for identification of the environments for cultivar evaluation. The objectives of this study were to determine the effects of different elevation levels on yields, yield components and agronomic traits of tropical-adapted waxy corn genotypes and to study the relationship between elevation levels and studied traits. Eleven waxy corn genotypes were grown in six environments in Thailand in the rainy season during 2009 and the dry season 2009/2010. The elevation levels in these locations ranged from 5 to 388 m above sea level. Combined analysis of variance showed that the interactions between genotype and environment were significant for all traits. Whole ear yield, husked yield, ear diameter and ear length in the dry season were higher than in the rainy season. Waxy corn grown at low elevation levels (5 to 34 m asl) had higher yields than that grown at moderate elevations (134 to 388 m asl). The relationships between elevation level with yield, yield components and agronomic characters were negative and significant for yield, ear length, plant height, and ear height, whereas the correlations between elevation level and days to silking and days to tasseling were positive and significant. The results suggested that the test sites for waxy corn in the tropics could be clustered into similar elevation levels. Keywords: Correlation, elevation, Zea mays L. var. ceratina, test site

Manuscript received: September 25, 2012; Decision on manuscript: March 28, 2013; Manuscript accepted: April 3, 2013 © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Waxy or glutinous corn (Zea mays L. var. ceratina) is an important vegetable crop in Asia. It has been grown for green corn as a cash crop and an additional source of income for large scale and small household farmers in Asian countries for century (Lertrat and Thongnarin, 2008; Kesornkeaw et al., 2009). Waxy corn in the tropics is grown in a wide range of growing conditions and altitudes above sea level. Altitude generally affects crop yield because it alters

solar radiation, temperature and photosynthesis of crop plants (Gale, 2004). The extent to which the altitude affects yield in vegetable waxy corn has not been clearly understood. This information is important for selection of suitable test sites for waxy corn yield trials and breeding of vegetable waxy corn for wide adaptation and specific adaptation to specific altitudes. Furthermore, waxy corn genotypes may respond differently to elevations, and genotype by elevation interaction could be the important factor for yield variation for waxy corn grown

RESEARCH ARTICLE


265

under different altitudes. This information is also important for waxy corn breeding. Unfortunately, this information is still lacking for waxy corn.

Waxy corn is a specialty crop that is grown under optimum, controlled irrigation, and well-managed conditions. As some environmental conditions can be partially controlled, the effect of variation in altitudes and genotype × elevation interaction can be studied in details. This information can help plant breeders classify test locations for multi-environment trials based on elevation and identify waxy corn production zones in tropical areas. Therefore, the objectives of this research were to determine the effects of different elevation levels on yield, yield components and agronomic characteristics of tropical-adapted waxy corn genotypes and to study the relationships between elevation levels with yields, yield components and agronomic characteristics. MATERIALS AND METHODS Plant material and experiment sites The 11 genotypes of waxy corn used in this study were obtained from waxy corn breeding program at Khon Kaen University, participatory institutions and seed companies of the testing network of waxy corn in Thailand (Table 1). These genotypes were evaluated in six locations in Thailand in the dry season 2009/10 and the rainy season 2010 (Table 2). The experimental sites included Nakhonratchasima (SW), Khon Kaen (KK), Ubon (UB), Saraburi (PAC), Chinat (CH) and Supanburi (EWS). The elevation level was highest in SW (388 m above sea level; m asl) followed by KK, UB, PAC, CH and EWS (190, 124, 34, 17 and 5 m asl, respectively) (Table 2). These locations represent the important areas of waxy corn production in Thailand. The planting dates were May 2009 for the rainy season and December 2009 for the dry season in all sites.

Soil chemical properties were similar among the test sites (data not reported), and the crop was planted under optimum management

for each site. Therefore, soil chemical property is not a limiting factor for yield.

Minimum daily air temperature was 24.2oC in the rainy season 2009 and 20.2oC in the dry season 2009/10, while the maximum daily air temperature was 33.0oC in the rainy season 2009 and 32.2oC in the dry season 2009/10 (Table 3). Solar radiations were 17.5 MJ m-2 d-1 in the rainy season 2009 and 16.9 MJ m-2 d-1in the dry season 2009/10. Annual rainfall was 325.3 mm in the rainy season 2009 and 33.7 in the dry season 2009/10, and irrigation was available for all test sites if necessary.

Table 1. Waxy corn varieties used in this study. Genotype Type of Variety

1116 F1 hybrid 1129 F1 hybrid 101LVN F1 hybrid H9VN F1 hybrid Check 1 F1 hybrid Check 2 F1 hybrid Check 3 F1 hybrid Check 4 F1 hybrid Check 5 F1 hybrid Check 6 Open-pollinated Check 7 Open-pollinated Field experiment

A randomized complete block design with four replications was used in each location. The experimental plots were six-row plots 5 m in length and spacing of 80 cm x 25 cm; plot size was 16 m2. Crop management practices were uniform across the locations. Conventional tillage was practiced for soil preparation, and 15-15-15 fertilizer as basal dose (at the rate of 171 kg ha-1) was incorporated into the soil during soil preparation.

The seeds were over-planted and later thinned to obtain desired stands at seedling stage. Two splits of 15-15-15 fertilizer at the rate of 93.75 kg ha-1 plus urea (46-0-0) at the rate of 93.75 kg ha-1 for first split and 15-15-15 fertilizer at the rate of 125 kg ha-1 plus urea at the rate of 62.5 kg ha-1 for second split were applied to the crop at 14 days after planting (DAP) and 30 DAP, respectively. At flowering

Suriharn & Lertrat (2013)

266

stage, 13-13-21 fertilizer was applied at the rate of 156.25 kg ha-1. Therefore, the total doses of fertilizers were 150.65 kg ha-1 nitrogen, 78.78 kg ha-1 phosphorus and 91.27 kg ha-1 potassium. Irrigation was supplied regularly to avoid drought stress. Insect pests, diseases and weed were properly managed to obtain optimum growth and yield of the crop for all sites in both rainy and dry seasons.

Data collection Data from each plot were recorded for un-husked ear weight (kg ha-1), husked ear weight (kg ha-1), ear diameter (cm), ear length (cm), plant height(cm), ear height (cm), days to tasseling and days to silking. Days to 50% tasseling and silking were recorded from total number of plants in each plot. After tasseling, plant height and ear height were also recorded from 10 randomly chosen plants in each plot. Plant height was measured from soil level to leaf collar of flag leaf, whereas ear height was measured from soil level to the node of the top ear. Therefore, days to harvest were calculated as days to 50% silking plus 18 days.

Harvest time for waxy corn is somewhat later than that for sweet corn. It is equivalent to late R3 to early R4 growth stage with kernel moisture of about 70%. At harvest, total number of ears were weighted and husked. Fresh kernels were separated from cops, and un-husked weight

and husked weight ear on per harvest area of 7.5 (2 rows 40 plants) m2 was converted to per hectare.

Data analysis Individual analysis of variance was performed for each character, and a combined three-factor analysis of variance was conducted after variance homogeneity was tested (Gomez and Gomez, 1984). The yield response Yijkr of the genotype i in the location j, season k and block r is:

Yijkr = m + Gi + Lj + Sk + Br (Lj Yk) + GLij + GSik + LSjk + GLSijk + eijkr

where: G = genotype effects, L = location effects, S = season effects, and B = block effects.

When main effects were significant (P <

0.05), mean separation was performed using an LSD test at the 0.05 probability level by using MSTAT-C software (Russel, 1994). The correlations between elevation levels vs. yields, yield components and agronomic characteristics were determined by Pearson’s correlation analysis.

Table 2. Six test locations used in this study.

Location Code Geographical coordinates Elevation (m) Soil type

Nachonratchasima SW 14°39'05.24"N 101°18'41.31"E 388 Sandy KhonKaen KK 16°28'10.78"N 102°48'47.21"E 190 Clayey loam Ubon UB 15°07'09.19"N 104°54'20.89"E 124 Loamy Saraburi PAC 14°43'46.55"N 100°47'09.30"E 34 Clay Chinat CH 15°09'11.50"N 100°11'00.22"E 17 Clay Supanburi EWS 14°32'16.50"N 99°59'28.41"E 5 Loamy sand


267

Table 3. Average minimum and maximum temperature, average solar radiation and total rainfall in six locations during rainy and dry seasons in Thailand, 2009 to 2010. Location Minimum

temperature (oC) Maximum

temperature (oC) Solar radiation

(MJ m-2 d-1) Total rainfall

(mm) Rainy Dry Rainy Dry Rainy Dry Rainy Dry

Nachonratchasima 23.9 18.6 31.3 30.3 17.8 16.9 224.3 56.5 Khon Kaen 24.9 19.9 33.5 31.4 17.4 16.3 260.2 86.3 Ubon 23.1 19.6 33.1 33.1 19.4 17.6 729.2 22.5 Saraburi 23.6 19.1 33 33.8 15.9 17.9 327.9 15.5 Chinat 25.4 22.1 32.9 32.1 15.7 16.1 217.7 20.7 Supanburi 24.1 21.9 33.9 32.5 19 16.6 192.4 0.8 Mean 24.2 20.2 33 32.2 17.5 16.9 325.3 33.7

RESULTS Sources of variation Combined analysis of variance for yield, yield components and agronomic traits of 11 waxy corn genotypes evaluated in two seasons and six locations in 2009 and 2010 showed significant differences between seasons for whole ear yield, husked yield, ear height, days to silking and days to tasseling, ear diameter and ear length (Table 4). Season contributed to 29.7% of total variation for days to silking and 28.9% of total variation days to flowering. Season gave small contributions to other traits, ranging from 0.2% to 8.4%.

Differences among locations were also significant for all traits. In general, location gave large contributions to all traits, ranging from 7.0% for ear height to 38.9% for ear length. Other traits in which location gave large contributions to total variations were ear diameter (29.4%), plant height (25.4%), days to silking (29.0%), days to flowering (31.1%), whole ear yield (17.4%) and husked yield (17.9%).

Genotype was also the large source of variations for most traits, and contributed to total variations ranging from 13.8% for ear diameter to 49.9% for ear height. Other traits in which genotype contributed to large variations were husked yield (22.0%), ear length (24.8%), plant height (43.7%), whole ear yield (14.4%) days to silking (18.9%) and days to tasseling (16.2%).

The interactions between season and location (S×L) were also significant for all traits under investigation. Large contributions of season x location interactions were observed for whole ear yield (34.3%) and husked yield (32.1%). The interactions between genotype and season (G×S) were also significant for whole ear yield, husk yield, ear length, days to silking and days to tasseling. However, the G×S interactions were not significant for ear diameter, plant height and ear height. The interactions between genotype and location (G×L) were significant for all characters. The results indicated that waxy corn genotypes responded differently to the locations.

Effects of season and location Waxy corn grown in the dry season had significantly higher whole ear yield, husked yield, ear diameter and ear length than did waxy corn grown in the rainy season (Table 5). However, waxy corn grown in the dry season had significantly lower ear placement than did waxy corn grown in the rainy season, whereas there was no significant difference in plant height.

The crop grown in Chinat had the highest whole ear yield (12,505 kg ha-1), followed by Saraburi (11,736 kg ha1 and Supanburi (11,458 kg ha-1) (Table 5). The crop grown in Chinat also had the highest husked yield (8,384 kg ha-1) followed by Supanburi (7,738 kg ha-1) and Saraburi (7,600 kg ha-1).


268

The crop grown in Saraburi had the highest ear diameter (4.83 cm) followed by that grown in Khon Kaen (4.62 cm) and Chinat (4.42 cm. Long ears length were observed in the crop grown at the low elevation levels in Saraburi (18.65 cm), Chinat (18.38 cm) and Supanburi (18.00 cm). Taller plants were observed in Suphanburi (197.2 cm) and Chinat (206.4 cm), whereas shorter plants were found in Nakhonrachasima (168.1 cm) and Khon Kaen (171.7 cm). Similarly, high ear placements were recorded for the crop grown in Chinat (100.9 cm) and Suphanburi (95.8 cm), whereas low ear placements were recorded for the crop grown in Nakhonrachasima (87.1 cm) and Khon Kaen (90.7cm). The crop grown in Nakhonrachasima and Khon Kaen took the longest days to silking and days to tasseling (50.0 and 47.1 days for days to silking and 49.5 and 46.4 days for days to tasseling, respectively). Genotypic variation Waxy corn genotypes were significantly different for whole ear yield and ranged from 8,975 to 12,499 kg ha-1(Table 6). Check 7, an open-pollinated variety, had the lowest ear yield of 8,975 kg ha-1, whereas 101LVN and H9VN, F1 hybrids, had the highest whole ear yield. When whole ears were husked, Check 4, H9VN and Check 1 became the highest husked yielding genotypes (8,352; 7,974; and 7,959 kg ha-1, respectively). 101LVN had high un-husked ear yield but its husked ear yield was moderate, showing high husk weight. Ear diameters ranging from 4.1 to 4.7 cm were recorded, and Check 5, H9VN and Check 2 were the best genotypes for ear diameter (4.7, 4.6 and 4.5 cm, respectively). Ear lengths ranged from 15.9 to 19.0 cm, and Check 4 had the longest ears (19.0 cm), followed by Check 1 (18.2 cm) and H9VN (18.0 cm). The tallest plants were observed in Check 1 (210 cm), Check 6 (213 cm) and Check 4 (196 cm). Check 6 and Check 1 also had the tallest ear placement (101 and 106 cm, respectively). Days to silking (43.4-48.6 days) and days to tasseling (43.1-47.8 days) showed good synchronization. The characters were directly

related to maturity and could divide the waxy corn varieties into two maturity classes. Check 5, 1116 and 1129 had early maturity, whereas Check 1, Check 2, Check 3, Check 4, Ratrchata1, Check 7, 101LVN and H9VN had late maturity. Correlations Means of genotypes that were averaged from two seasons were used to calculate correlation coefficients (R2). Elevation had negative and significant correlation coefficients with whole ear yield (R2 = -0.79) and husked yield (R2 = -0.73) (Figure 1a and 1b). Negative and significant correlation coefficient (R2 = -0.88) was also observed between elevation and ear length (Figure 1d), but the correlation coefficient between elevation and ear diameter was not significant (Figure 1e). The results indicated that reductions in yield and ear length were negatively associated with high elevations, but growing waxy corn at high elevations (not higher than 34 m asl) did not significantly affect ear diameter.

Elevation was negatively and significantly correlated with plant height (R2 = -0.59) and ear height (R2 = -0.76) (Figure 2a and 2b). The results indicated that the crop grown at higher altitude had shorter plants and lower ear placements. These effects have not been reported for tropical vegetable waxy corn.

Elevation was significantly and positively correlated with days to silking (R2 = 0.87) and days to tasseling (R2 = 0.80) (Figure 2c and 2d). The results indicated that the crop took more days to tasseling and silking when grown at higher altitudes because of low temperature at higher altitudes.

Responses of waxy corn genotypes Environment index and un-husked ear yield were used for classification of waxy corn genotypes in response to elevation. Based on these criteria, the responses would be either positive or negative direction. The genotypes with b values for un-husked ear yield ranging from 1.14 to 1.43 (1.24 on average) were considered highly sensitive; the genotypes with b values for un-husked ear yield ranging from


269

0.83 to 1.09 (0.06 on average) were considered moderately sensitive and the genotypes with b values for un-husked ear yield ranging from 0.79

to 0.82 (0.82 on average) were considered least sensitive (Figure 3a, 3b and 3c).

Table 4. Mean squares for whole ear yield, husked yield, ear diameter, ear length, plant height, ear height, days to silking and days to tasseling of 11 waxy corn varieties grown in seven different environments in two seasons of Thailand.

Source of variance df

Mean square (% Sum of square) Whole ear yield (kg ha-1)

Husked yield (kg ha-1)

Ear diameter (cm)

Ear length (cm)

Plant Height (cm)

Ear height (cm)

Days to silking

Days to tasseling

Season (S) 1 233,900,000** 73,830,000** 0.43* 4.25* 530 1,269** 1,624.3** 1,616.2** (8.4) (5.5) (0.4) (0.3) (0.2) (1.2) (29.7) (28.9)

Location (L) 5 96,510,000** 47,780,000** 5.75** 113.14** 14,954** 1,444** 317.2** 348.8** (17.4) (17.9) (29.4) (38.9) (25.4) (7.0) (29.0) (31.1)

S x L 5 190,500,000** 85,800,000** 1.83** 25.80** 4,088** 1,483** 137.8** 168.4** (34.3) (32.1) (9.3) (8.9) (7.0) (7.2) (12.6) (15.0)

Genotype (G) 10 40,130,000** 29,480,000** 1.35** 36.08** 12,858** 5,138** 103.3** 90.8** (14.4) (22.0) (13.8) (24.8) (43.7) (49.9) (18.9) (16.2)

G x S 10 4,787,064** 1,453,638** 0.13 2.06** 190 152 7.5** 7.1** (1.7) (1.1) (1.3) (1.4) (0.6) (1.5) (1.4) (1.3) G x L 50 3,625,849** 1,690,521** 0.20** 1.90** 250* 124* 3.1** 2.5** (6.5) (6.3) (10.4) (6.5) (4.3) (6.0) (2.8) (2.3) G x S x L 50 2,089,588** 953,640** 0.16* 1.78** 231* 101 1.7** 1.3** (3.8) (3.6) (8.2) (6.1) (3.9) (4.9) (1.5) (1.1) Error 240 1,279,876 537,386 0.10 0.72 157 79 0.6 0.7 C.V. (%) 10.34 14.11 7.32 4.89 6.8 9.52 1.73 1.77 *, ** Significant at P<0.05 and P<0.01 probability levels, respectively. Number in parenthesis is the percentage of the sum of square that contributes to total variation for the trait. Table 5. Yield, yield components and agronomic characteristics in each season and each location. Husked yield

(kg ha-1) Husk yield (kg ha-1)

Ear diameter (cm)

Ear length (cm)

Plant height (cm)

Ear height (cm)

Days to silking

Days to tasseling

Seasons Dry 11,714 7,652 4.4 17.4 185.6 91.9 48.3 47.6 Rainy 10,177 6,788 4.4 17.2 183.3 95.5 44.3 43.6 F-test ** ** ** ** ns ** ** ** Locations1/ SWAN 9,411 6,274 4.2 15.2 168.1 87.1 50.0 49.5 KK 9,667 6,254 4.6 17.2 171.7 90.7 47.1 46.4 UB 10,899 7,071 4.1 16.5 186.1 93.7 45.8 44.7 PAC 11,736 7,600 4.8 18.0 177.0 93.9 46.5 46.0 CH 12,505 8,384 4.4 18.4 197.2 100.9 44.1 42.9 EWS 11,458 7,738 4.1 18.7 206.4 95.8 44.2 44.1 LSD (0.05) 610 366 0.1 0.3 5.7 4.6 0.6 0.6 ns, ** Not significant and significant at 0.01 probability level. 1/SW (Nachonratchasima) = 388 m asl, KK (Khon Kaen) = 190 m asl, UB (Ubon) = 124 m asl, PAC (Saraburi) = 34 m asl, CH (Chinat) = 17 m asl, EWS (Supanburi) = 5 m asl.


270

Table 6. Yield, yield components and agronomic characteristics of 11 waxy corn genotypes.

Genotype Husked yield (kg ha-1)

Husk yield (kg ha-1)

Ear diameter (cm)

Ear length (cm)

Plant height (cm)

Ear height (cm)

Days to silking

Days to tasseling

1116 10,104 6,430 4.14 17.50 158.5 82 43.8 43.3 1129 10,275 6,290 4.08 16.75 165.8 93 45.1 44.7 101LVN 12,499 7,783 4.47 17.58 184.9 97 47.1 46.4 H9VN 12,274 7,974 4.58 18.03 189.1 94 47.1 46.1 Check 1 11,746 7,959 4.42 18.61 213.3 106 47.1 46.7 Check 2 10,996 7,538 4.50 17.42 154.7 67 47.6 46.9 Check 3 10,537 7,658 4.47 17.11 182.4 92 46.8 46.4 Check 4 11,828 8,352 4.44 19.03 196.1 101 48.6 47.8 Check 5 10,786 7,534 4.69 15.92 188.1 101 43.4 43.1 Check 6 10,383 6,387 4.25 16.50 210.0 110 47.8 46.4 Check 7 8,975 5,517 4.19 15.97 185.7 87 45.0 44.0 LSD (0.05) 525 340 0.15 0.39 5.8 4.1 0.4 0.4

The genotypes (group) had high sensitivity for un-husked yield and also had high sensitivity for most characters as indicated by high b values ranging from 1.13 to 1.30, except for ear length (b=1.04) (Figure 3d, 3g, 4a, 4d, 4g and 4j). Similarly, the genotypes (group) with moderate b values for un-husked ear yield also had moderate b values for other traits (b=0.87 to 1.05) except for Check 2 in this group (Figures 3, 4). Check 2 had low b values for both ear height (Figure 4b) and plant height (Figure 4e), indicating that TSG 0902 is stable for ear placement and plant height. The genotypes (group) with low sensitivity for un-husked ear yield also had low sensitivity for most traits (b=0.70-0.97), except for plant height (b=0.96) (Figure 4f). DISCUSSION Sources of variation The purpose of this study was to determine the effect of elevation levels on yield and agronomic traits of vegetable waxy corn and to group the test sites of waxy corn based on elevation level. Elevation levels in this study ranged from 5 to

388 m asl. The assumption underlying the study is that difference in yield performance and agronomic traits of vegetable waxy corn for the same season is dependent largely on difference in elevation levels because other factors such as irrigation, fertilizer and crop management can be controlled uniformly for some extent for all test sites.

In this study, season was a main source of variation in days to silking and days to flowering. The main cause of variation in season would be due to differences in temperature between the rainy and the dry seasons.

Location was also a main cause of variation in ear diameter, plant height, days to silking, days to flowering, whole ear yield and husked yield. Variety is another important source of variation in many characters such as ear diameter, ear height, husked yield, ear length, plant height, whole ear yield, days to silking, and days to tasseling.

For interactions between main effects, season × location interactions were the main source of variations in whole ear yield and husked yield. There were high genotype × season interactions for most characters under study and, therefore, seasons could not be reduced for multi-location trials.


271

Figure 1. Relationship between elevation levels and whole ear yield (a), husked yield (b), ear diameter (c) and ear length (d) of 11 waxy corn genotypes in 6 locations across rainy 2009 and dry season 2009/10.

Figure 2. Relationship between elevation levels and ear height (a), plant height (b), days to silk (c) and days to tasseling (d) of 11 waxy corn genotypes in 6 locations across 2009 rainy and 2009/10 dry season.

a) Whole ear yieldkg

ha-1

8000

9000

10000

11000

12000

13000 b) Husked yield

kg ha

-1

5500

6000

6500

7000

7500

8000

8500

9000

c) Ear diameter

Elevation (m)

0 100 200 300 400 500

cm

4.0

4.2

4.4

4.6

4.8

5.0d) Ear length

Elevation (m)

0 100 200 300 400 500cm

14

15

16

17

18

19

20

y = -7.318x + 11870R² = 0.79

y = -4.957x + 7846

R² = 0.73

y = -0.000x + 4.434R² = 0.04

y = -0.008x + 18.36

R² = 0.88

a) Ear height

cm

84

86

88

90

92

94

96

98

100

102 b) Plant height

cm

160

170

180

190

200

210

c) Days to silk

Elevation (m)

0 100 200 300 400 500

Days

43

44

45

46

47

48

49

50

51d) Days to tasseling

Elevation (m)

0 100 200 300 400 500

Days

42

43

44

45

46

47

48

49

50

y = -0.027x + 97.17

R² = 0.76

y = -0.078x + 194.3

R² = 0.59

y = 0.013x + 44.53

R² = 0.87

y = 0.014x + 43.83R² = 0.80


272

Figure 3. Comparison between elevation levels and whole ear yield (a-c), husked yield (d-f) and ear length (g-i) with regression lines of waxy corn genotypes (highly sensitive group; HS, moderately sensitive group; MS and low sensitive group; LS) with average b value in each group.

f) husked yield; LSe) husked yield; MSd) husked yield; HS

kg h

a-1

4000

5000

6000

7000

8000

9000

10000

11000

a) Whole ear yield; HS

kg h

a-1

6000

8000

10000

12000

14000

16000

b) Whole ear yield; MS c) Whole ear yield; LS

g) ear length; HS

Elevation levels

0 100 200 300 400 500

cm

12

14

16

18

20

22

Check1Check51129

h) ear length; MS

Elevation levels

0 100 200 300 400 500

Check2Check3Check4

i) ear length; LS

Elevation levels

0 100 200 300 400 500

Check6Check71116101LVN H9VN

b = 1.24 b = 0.96 b = 0.82

b = 1.30 b = 1.00 b = 0.70

b = 1.04 b = 1.05 b = 0.87


273

Figure 4. Comparison between elevation levels and ear height (a-c), plant height (d-f), days to silk (g-i) and days to tasseling (j-l) with regression lines of waxy corn genotypes (high-sensitive group; HS, moderate- sensitive group; MS and low- sensitive group; LS) with average b value in each group.

a) Ear height; HS

cm

60

70

80

90

100

110

120

130

b) Ear height; MS c) Ear height; LS

d) Plant height; HS

cm

120

140

160

180

200

220

240

260e) Plant height; MS f) Plant height; LS

g) Days to silk; HS

Day

s

38

40

42

44

46

48

50

52

54

h) Days to silk; MS i) Days to silk; LS

j) Days to tasseling; HS

Elevation levels

0 100 200 300 400 500

Day

s

38

40

42

44

46

48

50

52

54

Check1Check51129

k) Days to tasseling; MS

Elevation levels

0 100 200 300 400 500

Check2Check3Check4

l) Days to tasseling; LS

Elevation levels

0 100 200 300 400 500

Check6Check7 1116101LVN H9VN

b = 1.26 b = 0.87 b = 0.96

b = 1.18 b = 0.97 b = 0.87

b = 1.13 b = 0.99 b = 0.89

b = 1.20 b = 1.04 b = 0.73


274

Effects of season and location The higher yields observed in the dry season would be possibly due to low temperature in the dry season 2009/10 (Table 3). Lower temperature prolonged crop duration by increasing days to silking and days to tasseling and seed-filling duration. Unhusked ear yield, husked ear yield, ear diameter and ear length were high in the dry season but low in the rainy season, indicating that growing waxy corn in the dry season (with irrigation) was more productive than in the rainy season. However, hotter growing season could reduce days to silking and days and grain-filling period. The difference in yield between different altitudes is similar to the difference in yield between different temperatures.

The results clearly indicated that the crop grown at low elevation levels (5 to 34 m asl) had better yields than those at moderate elevation levels (124 to 388 m asl). The higher yield at low elevation was due largely to higher minimum temperature in both dry and rainy seasons. The results indicated that elevation has significant effect on ear length rather than on ear diameter. The results showed that elevation had significant effects on ear height and plant height of waxy corn, and the crop grown at higher elevation levels had shorter plants and lower ear placement. These results also showed that elevation levels had significant effects on days to silking and days to tasseling. Based on the results and genotypes evaluated, growing waxy corn at low elevations (5 to 34 m asl) was more productive than at higher elevations. Genotypic variation The correlations between elevation level and yield, yield components and agronomic characters were negative and significant for yield, ear length, plant height, ear height, but the correlations between elevation level and days to silking and days to tasseling were positive and significant. These results mean that testing sites of waxy corn in the tropics could be clustered into similar elevation levels.

Responses of waxy corn genotypes The results indicated that yields and other traits of moderate group were more stable and consistent than other groups. It could be recommended that stable waxy corn genotypes should be selected based on yield, yield components and agronomic traits.

Chen et al. (2011) reported that the daily minimum temperature was the major climatic factor for corn production in China and the correlation between corn yield and the daily minimum temperature was positive and significant.

However, high temperature could reduce stem length of field maize (Hunter et al., 1974) and sweet corn (Williams and Lindquist, 2007) in temperate regions. The contrasting results would be due to the differences in plant adaptation to temperate and tropical regions.

Grouping of test sites for waxy corn evaluation using similar altitudes should reduce test sites for multi-location trials. In this study, test sites may be reduced by selecting sites with similar altitudes because elevation had significant effects on unhusked ear yield, husked ear yield, plant height, ear height, ear length, days to silking, days to tasseling and maturity. ACKNOWLEDGEMENTS We are grateful for the financial support provided by the National Center for Genetic Engineering and Biotechnology, Bangkok, Thailand and the Plant Breeding Research Center for Sustainable Agriculture, Khon Kaen University, Thailand. Acknowledgement is extended to Khon Kaen University and the Faculty of Agriculture for providing financial support for manuscript preparation activities. REFERENCES Chen C, Lei C, Deng A, Qian C, Hoogmoed W,

Zhang W (2011). Will higher minimum temperatures increase corn production in Northeast China? An analysis of historical data over 1965-2008. Agric For. Meteorol. 151: 1580-1588.

Gale J (2004). Plants and Altitude — Revisited. Ann. Bot. 94: 199.


275

Gomez KA, Gomez AA (1984). Statistical procedures for agricultural research 2nd ed., John Wiley & Sons.

Kesornkeaw P, Lertrat K, Suriharn B (2009). Response to four cycles of mass selection for prolificacy at low and high population densities in small ear waxy corn. Asian J. Plant Sci. 8: 425-432.

Lertrat K, Thongnarin N (2008). Novel approach to eating quality improvement in local waxy corn: improvement of sweet taste in local waxy corn variety with mixed kernels from super sweet corn. Acta. Hortic. 769: 145-150.

Russel OF, (1994). MSTAT—C v.2.1 (a computer based data analysis software). Crop and Soil Science Department, Michigan State University, USA

Williams MM, Lindquist JL (2007). Influence of planting date and weed interference on sweet corn. Agron. J. 99: 1066-1072.

Nielsen RL, Thomison PR, Brown GA, Halter AL, Wells Wuethrich JKL (2002). Delayed planting effects on flowering and grain maturation of dent corn. Agron. J. 94: 549-558.


COMBINING ABILITY ANALYSIS FOR SEED YIELD AND COMPONENT TRAITS IN PEA UNDER FOOT HILLS OF NORTHEAST INDIA

THIYAM REBICA1, MUKUL KUMAR2*, P. RANJIT SHARMA1, K. S. NOREN3

and SHIV DATT4

1Department of Plant Breeding and Genetics, College of Agriculture, Central Agricultural University, Imphal-795004, Manipur, India

2Department of Plant Breeding and Genetics, Tree Improvement , College of Horticulture and Forestry, Central Agricultural University, Pasighat, East Siang 791102, Arunachal Pradesh, India

3Directorate of Research, Central Agricultural University, Imphal-795004, Manipur, India 4IP &TM Unit, Krishi Anusandhan Bhavan-I, ICAR, Pusa, New Delhi 110012, India


SUMMARY A half-diallel experiment involving 11 parents was carried out using randomized block design with three replications to assess the combining ability and gene action for yield and its component traits in pea under agroclimatic conditions prevailing in the foot hills of northeast region of India. This region is characterized by high rainfall with acidic soil (pH < 5.5). The study revealed the importance of both additive and non-additive gene effects for all the traits with preponderance of non-additive gene effects for days to first picking, pods per plant, pod length, seeds per pod and seed yield per plant, but additive gene effects were found important in controlling for days to 50% flowering, plant height and 100-seed weight. Parents KPMR-728, Arkel, Rachna and CAUP-1 were good general combiners for seed yield and other component traits, which could be exploited to develop prolific pureline varieties of pea. The crosses HUDP-15 x KPMR-728, E-6 x Rachna, Pant P- 25 x KPMR-728 and Arkel x CAUP-1 were the best specific combiners involving either both or one of the parents. A high general combiner for seed yield may be included in recombination breeding to produce desirable segregants for developing high-yielding, acidic soil-tolerant genetic stock/varieties of pea. Keywords: diallel, combining ability, gene action, seed yield, yield components, pea

Manuscript received: September 25, 2012; Decision on manuscript: February 28, 2013; Manuscript accepted: May 25, 2013 © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Pea (Pisum sativum L.) is an important legume grown as a garden and field crop throughout the temperate region of the world. A good source of protein (7.2 g per 100 g) and minerals (Hajra and Som, 1999), pea is grown over wider agro-ecologies, especially in low to mid-altitudes (1300-1700 m asl) of the northeastern region of India. High rainfall, excessive weathering, leaching of bases and acidification contribute to

making low fertility as the major constraint to crop growth (Raichaudhari, 2005). Poor crop productivity and soil fertility in acidic soils are mainly due to a combination of aluminum and manganese toxicities and deficiencies in phosphorus, nitrogen, calcium, magnesium and potassium. For legumes, acid soils pose an additional challenge because their symbiotic rhizobia are acid-sensitive (Hartel and Bouton, 1989). Under such situation, development of varieties adapted to the acid soil complex is the

RESEARCH ARTICLE


277

best solution to exploit the potential of pea cultivation in the northeastern areas of the country. The development of high-yielding varieties through hybridization requires knowledge of combining ability of the parents, particularly information on the nature and magnitude of gene effects controlling quantitative traits. Combining ability analysis on the basis of diallel mating system is one of the most appropriate methods to identify the best combiners. It also gives information about the nature and magnitude of genetic variations. A large number of studies have been carried out in pea but information on combining ability and genetic architecture of quantitative traits under acid soils environments is scanty. Therefore, the present study was undertaken to determine the combining ability of important genotypes and obtain information on the genetic mechanism that controls quantitative traits in pea in acid soils of the foothills of Manipur in northeastern region of India. MATERIALS AND METHODS The eleven genetically diverse varieties of pea - E-6, Arkel, DDR-23, DMR-7, HUDP-15, Rachna, IPFDI-10, CAUP-1, TRCP-8, Pant P-25 and KPMR-728— were crossed in all possible combinations without reciprocals during the winter of 2008. The 55F1

’s and 11 parents were evaluated in an experiment using randomized block design with three replications during the winter season of 2009 at the farm land of College of Agriculture, Imphal in Manipur (India). The site is located between 24051΄N latitude, 93056΄E longitude and at an altitude of 790 m above mean sea level. The total annual rainfall was 1212 mm during the five monsoon months from June to October. During the experimental period, the average maximum temperature was 25.2 0C and minimum temperature was 9.6 oC. Soil at the experimental site is clay loam, with pH 5.2 and nitrogen, phosphorus and potassium contents of 283, 18.6 and 124 kg/ha, respectively. Each cross and parent was grown in two rows of 3.0 m length with inter and intra-row spacing of 45 cm and 15 cm, respectively. The recommended agronomic practices and plant protection measures were

followed during the crop season. Data were recorded on five randomly selected plants for nine quantitative characters: days to 50% flowering, days to first picking, plant height (cm), pods per plant, pod length (cm), seeds per pod, seed yield per plant (g), 100-seed weight (g) and harvest index (%). The mean values of each genotype were subjected to analysis of variance. After finding significant variation among genotypes, the combining ability analysis was carried out using Griffing’s (1956) method-2 and model-1 as elaborated by Singh and Chaudhary (1989) using the SPAR 2.0 (Statistical Package for Agricultural Research) software developed by the Indian Agricultural Statistics Research Institute, New Delhi, India. The ratio of 2σ2

g to 2σ2

g + σ2s was used to estimate the relative

importance of additive or dominance effects (Baker, 1978). RESULTS AND DISCUSSION Analysis of variance for combining ability The analysis of variance (Table 1) for diallel crosses revealed highly significant differences between parents and F1 progenies. The variances due to GCA and SCA were highly significant for all the characters studied. The significance of the variances due to GCA as well as SCA implies that both the additive and non-additive components of heritable variance are responsible for variation observed for all nine characters studied. Similar observations on the involvement of both additive and non-additive components of heritable variation have also been reported by Bhardwaj and Kohli (1998), Ceyhan and Avci (2005) and Sharma et al. (2007). Gene action The ratio of GCA variance (σ2GCA) to SCA variance (σ2SCA) was lower than unity (<1) for all the traits (Table 2), indicating the importance of non-additive gene action for all these characters. This shows that the material used in the present study was heterozygous and that this heterozygosity contributes towards the non-additive components.

Rebica et al. (2013)

278

Table 1. Mean squares from analysis of variance for genotypes and combining ability for various characters by using 11 x 11 half diallel analysis according to Griffing’s method II. Character Source of variation

Replication (2)

Genotypes (65)

Error (130)

GCA (10)

SCA (55)

Error (130)

Days to 50% flowering 4.95** 40.09** 0.68 51.78** 6.38** 0.23 Days to first picking 7.41* 44.87** 2.26 46.14** 9.29** 0.75 Plant height (cm) 19.2 336.87** 6.31 392.34** 61.40** 2.10 Pods per plant (number) 3.25 4.87** 0.92 0.89** 1.72** 0.31 Pod length (cm) 0.02 2.02** 0.18 1.33** 0.55** 0.06 Seeds per pod (number) 0.67 2.18** 0.34 1.24** 0.63** 0.11 Seed yield per plant (g) 0.19 10.18** 1.32 7.74** 2.60** 0.44 100-seed weight (g) 0.10 20.01** 0.42 24.10** 3.60** 0.14 Harvest index (%) 19.73 40.68** 8.20 25.71** 11.35** 2.73 *,** Significant at P < 0.05 and P < 0.01 levels, respectively Table 2. Estimates of components of variance along with their ratio and relative importance for various characters Character Variance estimates and their ratio Relative importance (%) of

σ2GCA σ2SCA σ2GCA/σ2SCA σ2GCA σ2SCA Days to 50% flowering 3.96 6.15 0.65 56.29 43.71 Days to first picking 3.49 8.54 0.41 44.98 55.02 Plant height (cm) 30.02 59.3 0.51 50.31 49.68 Pods per plant (no.) 0.04 1.41 0.03 6.56 93.44 Pod length (cm) 0.10 0.49 0.20 28.98 71.02 Seeds per pod (no.) 0.09 0.52 0.17 25.71 74.29 Seed yield per plant (g) 0.56 2.16 0.26 34.15 65.85 100-seed weight (g) 1.84 3.46 0.53 51.54 48.46 Harvest index (%) 1.77 8.62 0.21 29.11 70.89

Further, according to a method developed by Baker (1978), the relative importance of SCA was higher than 50% for days to first picking (55.02%), pods per plant (93.44%), pod length (71.02%), seeds per pod (74.29%), seed yield per plant (65.85%) and harvest index (70.89%). This shows that these characters were predominantly under the control of non-additive gene effects. Therefore, heterosis and use of hybrid vigor could be applied for improving these traits. The occurrence of both additive and non-additive gene effects with preponderance of non-additive gene effects for yield and most of its important components has also been reported by Singh and Singh (1987), Kumar et al.(1996), Ceyhan and Avci (2005) and Ceyhan et al. (2008). On the other hand the relative

importance of GCA was higher than or equal to 50% only for days to 50% flowering (56.29%), plant height (50.31%) and 100-seed weight (51.54%). It means that additive gene effects predominated in the inheritance of these three characters, hence genetic improvement is easy through selection for such traits. Involvement of additive component in pea was also observed by Dhillon et al. (2006) for days to first flowering, Kumar et al. (1996) for plant height and Seema et al. (2005) for 100-seed weight.


279

General combining ability effects The magnitude and direction of combining ability effects provide guidelines for selecting parents and their utilization. The positive and significant GCA effects of parents indicate that a target genotype with a high proportion of the target trait is able to transfer the character into a new genotype, while negative significant GCA effects indicate a target genotype containing a low proportion of the target trait, but still having the ability to transfer this character into a new genotype (Vejdani and Sepahvand, 1993). Estimates GCA effects of all the parents for 9 characters are shown in Table 3. The parents that proved to be good general combiners on the basis of their significant desirable effects were KPMR-728 for plant height, pods per plant, pod length, 100 seed weight, harvest index and seed yield per plant; CAUP-1 for plant height, pod length, 100 seed weight, harvest index and seed yield per plant; and Rachna for plant height, pods per plant, 100 seed weight and seed yield per plant. The negative combining ability effect is highly desirable for earliness attributes like days to 50% flowering and days to first picking.

The parent Arkel was recorded as a good general combiner for pod length, seeds per pod, harvest index and seed yield per plant along with early flowering and first picking, indicating its potential for exploiting its yield and earliness traits in pea. It is evident that the significant GCA effects for seed yield in either direction resulted from similar GCA effects of some yield components, indicating that the combining ability of seed yield was influenced by the combining ability of its components. In this study, plant height, pods per plant, pod length and 100-seed weight were the common associated traits with seed yield per plant. Krarup and Davies (1970) also suggested that number of pods per plant, plant height, hundred seed weight, pod yield and number of seeds per pod are primarily yield components. Therefore, simultaneous improvement in important yield-associated traits with the ultimate objective to improve overall GCA for seed yield may be a better approach to raise yield potential in pea. Specific combining ability effects

The magnitude of SCA effects is of prime importance in selecting the cross combinations

with higher probability of obtaining transgressive segregants. It may be desirable to consider both SCA effect and mean performance when selection is made. In general, maximum crosses showing significant SCA effects were invariably associated with better mean performance for seed yield and its respective traits. The results were in agreement with those obtained by Seema et al. (2005) who concluded that mean performance of the crosses was closely associated with SCA effects. Therefore, selection of crosses on the basis of heterotic response should prove effective.

Out of 55 cross combinations, the 13 crosses which exhibited highly significant SCA effects for seed yield per plant were also observed to be the good specific combiners for other related traits in the desirable direction (Table 4). The cross combinations possessing significant SCA effects for seed yield were Arkel x CAUP-1with 5 yield components; E-6 x Rachna, Arkel x Pant P-25, HUDP-15 x KPMR-728, Pant P-25 x KPMR-728 each with 4 yield components and Arkel x KPMR-728 with 4 yield components as well as earliness.

All the 13 crosses showing significant positive SCA effects involved parents of high x high, low x high, average x high and low x low general combining ability (Table 4). Thus, the high GCA effects of the parents may not be a reliable criterion for the prediction of high SCA effects. The high performance of these crosses may be attributed to additive x additive (high x high), dominance x additive (low x high and average x high) and dominance x dominance (low x low) type of gene (epistatic) interactions. In the present study, significant SCA effects for seed yield were observed in two crosses (Arkel x CAUP-1 and Arkel x KPMR-728) with both parents as good general combiners. Subhash et al. (2006) also reported that the most promising cross is the one that involves parents with high GCA and shows high SCA effects. Therefore, the presence of additive x additive gene action suggests that a major part of such variance would be fixed in subsequent generations to facilitate further selection.


280

Table 3. Estimates of general combining ability effects of 11 parental lines for various characters in pea. Parent Days to 50%

flowering Days to first picking

Plant height (cm)

Pods per plant (no.)

Pod length (cm)

Seeds per pod (no.)

Seed yield per plant

100-seed weight (g)

Harvest index (%)

(g)

E-6 -4.91**H -4.65**H -7.0**L 0.12A 0.21**H 0.46**H 0.25A -0.25L 0.55*H Arkel -2.52**H -2.24**H -2.48**L 0.12A 0.50**H 0.49**H 0.39**H -0.92**L 1.18**H DDR-23 -0.57**H -0.65**H -4.77**L 0.12A -0.68**L -0.51**L -1.02**L -1.61**L -1.61**L DMR-7 0.84**L 1.81**L 3.16**H 0.07A 0.04A 0.13A -0.15L -0.37L -0.20L HUDP-15 0.58**L 0.78**L -9.45**L -0.67**L 0.01A 0.01A -0.92**L -1.66**L -1.33**L Rachna 0.89**L 0.78**L 5.59**H 0.42**H -0.09L -0.15L 0.52**H 0.72**H 0.55A IPFDI-10 1.14**L 1.19**L -0.48L -0.34*L -0.13*L -0.12L -0.61**L -0.51**L -1.10L CAUP-1 0.31L 0.54*L 7.64**H 0.12A 0.34**H 0.08A 1.08**H 2.47**H 1.83**H TRCP-8 1.43**L 0.44L 4.18A -0.28L 0.20**H -0.08L 0.06A 0.88**H 0.21A Pant P-25 1.47**L 0.93**L -0.44L -0.39L -0.25**L -0.41**L -0.79**L -0.52**L -2.0**L KPMR-728 1.35**L 1.07**L 4.05**H 0.41**H 0.19**H 0.10A 1.19**H 1.78**H 1.92**H SE(gi) 0.18 0.23 0.38 0.15 0.06 0.09 0.18 0.1 0.44 *,** Significant at P<0.05 and P<0.01 levels, respectively H denotes significant GCA effects in favorable direction (high or good general combiner). L denotes significant and non- significant GCA effects in unfavorable direction (Low or poor general combiner). A denotes non-significant GCA effects in favorable direction (average general combiner).


281

Table 4. Crosses showing highly significant SCA effects for seed yield per plant and their performance for other characters in pea. Cross SCA

effects Mean seed yield per plant (g)

GCA effects of its parents

Significant response in other characters for SCA effects in desirable direction

Arkel x CAUP-1 2.84** 9.33 H x H PH, PP, PL, SP, 100-SW and HI HUDP-15 x KPMR-728 2.71** 10.50 L x H PH, PP, PL, SP, and HI DMR-7 x Pant P-25 2.25** 9.33 L x L DF, PH, PP, 100-SW and HI E-6 x Rachna 2.12** 10.13 A x H PH, PP, SP, 100-SW and HI Arkel x KPMR-728 2.04** 5.67 H x H DFP, PP, PL, SP,100-SW and HI Rachna x TRCP-8 2.03** 10.13 H x A DF, PP,100-SW and HI DDR-23 x HUDP-15 1.91** 5.33 L x L PH, PP, and HI Pant P-25 x KPMR-728 1.85** 10.00 L x H PH, PL, SP, 100-SW and HI Arkel x Pant P-25 1.71** 6.00 H x L PH, PP, PL, SP, and HI DMR-7 x HUDP-15 1.71** 8.67 L x L PH and SP DMR-7 x Rachna 1.61** 10.00 L x H DF, PL, and SP DDR-23 x CAUP-1 1.58** 8.33 L x H PH, PP,100-SW and HI IPFDI-10 x TRCP-8 1.53** 9.00 H x A PH, and PL ** Significant at P < 0.01 level. DF= days to 50% flowering; DFP= days to first picking; PH= plant height; PP= pods per plant; PL= pod length; SP= seeds per pod; 100-SW= 100-seed weight; HI= harvest index

However, the majority of the crosses had significant SCA effects for seed yield involving one good and one low or average combining parent. These crosses may give desirable transgressive segregants if the additive effect of one parent and complementary epistatic effects (if present in the cross) act in the same direction and maximize plant character (Sharma, 1999).

These results indicate the importance of both additive and non-additive gene effects with preponderance of non-additive effects of the studied traits. Simple pedigree selection will be ineffective for improving such characters. However, using a population improvement program like reciprocal recurrent selection allows the accumulation of variability and heterozygosity for exploiting non-fixable gene effects and this will prove to be more effective method (Joshi, 1979). CONCLUSION This study revealed the importance of non-additive gene effects for seed yield and most of the yield

attributes. However, additive gene effects are involved for days to 50% flowering, plant height and 100-seed weight. On the basis of overall performance across nine characters, the KPMR-728, Rachna and CAUP-1 were identified as the most promising parents because of their good GCA for seed yield with three or more other major yield components. The parent Arkel was a good general combiner for seed yield with three of its components, along with early flowering and picking. These parents may be utilized in hybridization programs or multiple crossing programs to accelerate the pace of genetic improvement of seed yield in pea. The crosses HUDP-15 x KPMR-728, E-6 x Rachna, Pant P- 25 x KPMR-728 and Arkel x CAUP-1were identified as the most promising hybrid combinations due to their high mean performance, positive SCA effects for yield and majority of the component traits involving both or at least one parent as good general combiner and these could be utilized to produce desirable segregants for developing high yielding acidic soil-tolerant genetic stock/varieties in pea.


282

ACKNOWLEDGEMENTS The authors are grateful to the Directorate of Research, Central Agricultural University, Imphal, Manipur, India for providing the genetic materials for the study.

REFERENCES Baker RJ (1978). Issues in diallel analysis. Crop Sci.

18: 533-536. Bhardwaj RK, Kohli UK (1998). Combining ability

analysis for some important yield traits in garden pea (Pisum sativum L.). Crop Res. Hisar 15: 245-249.

Ceyhan E, Avci MA (2005). Combining ability and heterosis for grain yield and some yield components in pea (Pisum sativum L.). Pak. J. Biol. Sci. 8: 1447-1452.

Ceyhan E, Avci MA, Karadas S (2008). Line x tester analysis in pea (Pisum sativum L.): Identification of superior parents for seed yield and its components. Afr. J. Biotechnol. 7: 2810-2817.

Dhillon TS, Singh M, Brar PS (2007). Assessment of combining ability for some quantitative characters in pea. Crop Improv. 34: 106-109.

Griffings B (1956). Concept of general and specific combining ability in relation to diallel crossing systems. Aust. J. Biol. Sci. 9: 463-493.

Hajra P, Som P (1999). Technology for vegetable production and improvement. Naya Prokash, Kolkata.

Hartel PG, Bouton JH (1989). Rhizobium meliloti inoculation of alfalfa selected for tolerance to acid, aluminum-rich soils. Plant Soil 116: 283–285.

Joshi AB (1979). Breeding methodology for autogamous crops. Indian J. Genet. 10: 567-578.

Krarup A, Davies DW (1970). Inheritance of seed yield and its component in a six parent diallel cross in peas. J. Am. Soc. Hortic. Sci. 95: 795-797.

Kumar S, Singh KP, Panda PK (1996). Combining ability analysis for green pod yield and its components in garden pea. Orissa J. Hortic. 24: 69-70.

Raichaudhari M (2005). Nutrient management of acidic soils of the north-east India. Indian Farming 56: 22-25.

Seema R, Kumar M, Pandey SS (2005). Diallel analysis for yield-contributing characters in pea. Legume Res. 38: 223-225.

Sharma A, Singh G, Sharma S, Sood S (2007). Combining ability and heterosis for pod yield and its related horticulture traits in garden pea (Pisum sativum L.) under mid-hill sub-temperate and high-hill dry-temperate conditions of Himachal Pradesh. Indian J. Genet. 67: 47-50.

Sharma TR (1999). Combining ability and heterosis in garden pea (Pisum sativum var. arvense) in cold desert Himalayan region. Indian J. Agric. Sci. 69: 386-388.

Singh RK, Chaudhary BD (1989). Biometrical methods in quantitative genetic analysis. Kalyani Publishers, New Delhi.

Singh SP, Singh RP (1987). Diallel analysis for combining ability in pea. Acta Agron. Hungarica 36: 89-95.

Subhash K, Srivastava RK, Singh R (2006). Combining ability for yield and its component traits in field pea. Indian J. Pulses Res. 19: 173-175.

Vejdani P, Sepahvand NA (1993). Consideration of general and specific combining abilities of wheat cultivars using diallel cross technique. Plant Seed Agric. Sci. J. 9: 18-22.


STUDY OF INHERITANCE FOR GRAIN YIELD AND RELATED TRAITS IN BREAD WHEAT (Triticum aestivum L.)

RAMEEZ IFTIKHAR1*, SYED BILAL HUSSAIN2, IHSAN KHALIQ3 and SMIULLAH4

1Department of Agricultural, Food and Nutritional Science (AFNS), University of Alberta Edmonton, AB T6G 2P5, Canada 2Faculty of Agricultural Sciences and Technology, Bahauddin Zakariya University, Multan, Pakistan

3Department of Plant Breeding and Genetics, University of Agriculture, Faisalabad, Pakistan 4Ayub Agricultural Research Institute (AARI), Faisalabad, Pakistan.

*Corresponding author’s email: [email protected], [email protected]

SUMMARY The study was conducted to investigate the inheritance pattern of grain yield and component traits related to grain yield in wheat (Triticum aestivum L.) using five varieties/lines –Pasban-90, Sehar-06, NR 356, DN-62 and ZAS-70 –, and their six derived F2 hybrids at the University of Agriculture, Faisalabad, Pakistan. Heritability and genetic advance were determined for number of tillers per plant, flag leaf area, spike density, number of grains per spike, 1000-grain weight and grain yield per plant. The results revealed moderate to high heritability estimates coupled with high genetic advance for tillers per plant, grains per spike, 1000-grain weight and grain yield per plant. These estimates for flag leaf area were relatively low, whereas spike density exhibited moderate heritability along with low genetic advance in all cross combinations. Prospects of quick genetic improvement through selection are evident for most of the traits studied due to presence of higher heritability and genetic advance values. The most promising cross combinations were Pasban-90×DN-62 and Sehar-06×NR-356, suggesting that these crosses deserve more attention in further breeding programs to develop high-yielding wheat varieties. Furthermore, correlation studies showed that flag leaf area, grains per spike and 1000-grain weight had significant positive association with grain yield, pointing out their utility as direct selection criteria for the isolation of superior genotypes/parents from genetically mixed populations. Keywords: Heritability, genetic advance, correlation, selection, grain yield

Manuscript received: March 24, 2012; Decision on manuscript: October 24, 2012; Manuscript accepted: March 28, 2013. © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Wheat (Triticum aestivum L.) is a crop of global importance that is grown all over the world under diversified environmental conditions. Approximately one-sixth of the total cultivated land in the world is devoted to wheat cultivation, making it the third largest cereal crop, after

maize and rice (http://www.cimmyt.org). Due to its importance, breeding programs aim to develop new genotypes with high grain yield potential and desirable quality traits. Grain yield in wheat is a complex trait that is controlled by many genes, creating difficulties in combining genes conferring desirable traits into a single genotype. Therefore, the success of such

RESEARCH ARTICLE


284

breeding programs, which are designed to develop high-yielding genotypes, depends largely on selection of best plants/parents for hybridization. For this purpose, heritability is important to identify suitable parents with desirable traits because heritability estimates provide information about the extent of transmissibility of the quantitative characters of economic importance from parents to progeny. The behavior of succeeding generations is predicted by devising appropriate selection criteria and assessing the level of genetic improvement. Knowledge of heritability for different morphological traits plays an important role in the planning and execution of a successful breeding program.

Similarly, genetic advance gives a clear and precise view of segregating generations for possible selection. Therefore, heritability estimates would not be pragmatic in selection if based only on phenotypic appearance and not on genetic advance. Genetic advance must be considered along with heritability in coherent selection breeding programs. Many researchers (Arshad and Yildirim, 1996; Sattar et al., 2003; Singh et al., 2006; Ahmed et al., 2007; Haq et al., 2008; Akcura, 2009) reported high heritability and genetic advance in most of the traits under study. Likewise, the determination of correlation coefficients among various traits helps to develop superior combination of yield attributes because selection for one trait can reduce the chances of successful selection for some other traits.

However, the combination of traits that in various ways contribute towards yield can result in a maximum gain of each trait individually (Quarrie et al. 1999). Therefore, correlation studies among traits can be of a great use, pointing out the traits that selection should be directed in order to increase the yield per unit area. The work of previous workers like Kashif and Khaliq (2004); Bhuttah (2006); Ali et al., (2008) and Nafouzi et al., (2008) showed that grain yield in wheat had positive association with flag leaf area, tillers per plant, grains per spike and 1000-grain weight. Keeping these in mind, the current study was conducted to investigate the extent of heritability, genetic advance and character association among some quantitative traits in order to establish suitable

selection criteria for developing high-yielding wheat cultivars. MATERIALS AND METHODS The genetic materials used for this research were the two most commonly cultivated wheat cultivars in Pakistan: Pasban-90 (well known for bread quality, it is a commonly used parent in wheat breeding in Pakistan) and Sehar-06 (one of the recently released high-yielding cultivars) along with three superior inbred lines of wheat, NR-356, DN-62 and ZAS-70. During Feb-Mar, crossing was performed keeping commercial cultivars (Pasban-90 and Sehar-06) as female parent and inbred lines (NR-356, DN-62 and ZAS-70) as male parents in order to develop F1 progenies. The crossing patterns were as follows: (1) Sehar-06 × DN-62 (2) Sehar-06 × NR-356 (3) Sehar-96 × ZAS-70 (4) Pasban-90 × DN-62 (5) Pasban-90 × NR-356 (6) Pasban-90 × ZAS-70

In the following cropping season (2009-10), the F1 progenies were selfed to raise F2 generations that, later on, were planted along with their parents in the experimental area of the Department of Plant Breeding and Genetics, University of Agriculture, Faisalabad, Pakistan (longitude 73.74º E, latitude 30.31º N and 184 m asl) during year 2010 - 2011 under normal field conditions in a triplicate randomized complete block design. The field was irrigated prior to sowing as “rauni” and after attaining the proper soil moisture, sowing of experimental material was done by dibbling, keeping rows 30 cm and plants 22.5 cm apart. The first irrigation was applied to each experimental unit after 35 days and subsequent irrigations were done at the time of flowering, anthesis and grain-filling stage. To maintain the identity, each plant was tagged and numbered. At maturity, 200 competitive plants from F2 population of each cross and 20 plants from each parent were selected and data were recorded on individual plant basis for the following traits; number of tillers per plant, flag leaf area (as described by Muller, 1991), spike density (number of spikelets per spike divided

Iftikhar et al. (2013)

285

by spike length), number of grains per spike, 1000-grain weight and grain yield per plant. Means, standard deviations, variances and coefficients of variability for parents and F2 populations were computed by using SPSS version 17.0.1. Broad-sense heritability was calculated by using the formula proposed by Mahmud and Kramer (1951) as follows:

𝐻 = (𝜎2𝐹2) − �(𝜎2𝑃1 × 𝜎2𝑃2)

𝜎2𝐹2

where H = broad-sense heritability, 𝜎2𝐹2 = variance of F2, 𝜎2𝑃1 = variance of P1, and 𝜎2𝑃2 = variance of P2 Genetic advance (GA) was calculated at 10% selection intensity according to formula given by Burton and Devane (1953) as follows: 𝐺𝐴 = 𝑖 × 𝐻 × 𝜎 where i = selection intensity, H = broad-sense heritability, and 𝜎 = standard deviation The value of “i” in this study at 10% selection intensity was 1.7. Genotypic and phenotypic correlation coefficients were calculated using the procedure developed by Kwon and Torrie (1964). RESULTS Heritability and genetic advance Means, standard deviations, coefficients of variability, heritability estimates and genetic advance values for various yield-related traits in six wheat crosses are presented in Tables 1 and 2. It is evident from Table 1 that there existed considerable variability for most traits that can be exploited by adapting proper breeding strategies.

The highest heritability estimate (85.3) for tillers per plant was shown by the cross Pasban-90×DN-62,h followed by the cross Sehar-06×NR-356 with value of 77.3, while the lowest heritability estimate (63.2) was observed

from the cross Sehar-06×ZAS-70 (Table 2). The highest value of genetic advance (16.5) was observed in Sehar-06×NR-356 and the lowest (8.1) in Pasban-90×ZAS-70.

In this study, the highest heritability estimate for flag leaf area was recorded from Sehar-06×NR-356 (63.8), whereas the lowest heritability estimate (38.2) was seen in Sehar-06×DN-62. Similarly, the highest value of genetic advance (11.4) was observed in Sehar-06×ZAS-70 while the lowest value (5.4) was noted in Pasban-90×ZS-70.

It is evident from Table 2 that Pasban-90×DN-62 had the highest heritability for spike density (76.9) and the lowest (59.2) inSehar-06×NR-356. Genetic advance values ranged from 1.9 in cross Pasban-90×NR-356 to 4.9 in cross Sehar-06×NR-356.

Pasban-90×NR-356 exhibited the highest heritability estimate (85.1) for grains per spike, while the lowest value (67.6) was seen in cross Sehar-06×DN-62. The highest genetic advance (17.9) was observed in cross Pasban-90×DN-62; cross Sehar-06×ZAS-70 had the lowest (8.2) (Table 2).

For 1000-grain weight, the highest estimates of heritability (94.0) and genetic advance (13.6) were recorded in Pasban-90×DN-62, while the lowest estimates of these parameters were observed in Sehar-06×ZAS-70 and Pasban-90×ZAS-70 with computed values of 78.3 and 9.6 for heritability and genetic advance, respectively. It is evident from Table 2 that heritability estimates for grain yield per plant were high, ranging from 76.2 in cross Sehar-06×DN-62 to 90.1 in cross Sehar-06×NR-356, along with high genetic advance values ranging from 9.0 in cross Pasban-90×NR-356 to 17.8 in cross Pasban-90×DN-62. Correlation studies The genotypic and phenotypic correlation coefficients for all possible combinations of traits under study are presented in Table 3. In almost all the cases, genotypic correlations for all traits under study were higher than phenotypic ones. This may be due to a predominance of genetic factors over environmental ones in the development of character association. The observations


286

regarding the association of various traits are explained below.

Positive but non-significant correlations of tillers per plant with flag leaf area and grains per spike were recorded at both genotypic and phenotypic levels, whereas tillers per plant was observed to be associated positively and significantly with spike density at the genotypic level but non-significantly at the phenotypic level. There was significant negative correlation between tillers per plant and 1000-grain weight at both levels, whereas negative but nonsignificant correlation between tillers per plant and grain yield was recorded at both genotypic and phenotypic levels.

Flag leaf area was correlated positively but non-significantly with spike density and 1000-grain weight at both genotypic and phenotypic levels, while significant correlations

was noted with grains per spike and grain yield at both levels. There existed positive but non-significant correlation between spike density and grains per spike, while highly significant negative correlation between spike density and 1000-grain weight was recorded at both genotypic and phenotypic levels. Similarly, correlation between spike density and grain yield was negative and significant at genotypic level but non-significant at phenotypic level.

The correlation between grains per spike and 1000-grain weight was recorded non-significantly negative. Significantly positive association between grains per spike and grain yield per plant was observed at both genotypic and phenotypic levels. There was a significant positive correlation between 1000-grain weight and grain yield per plant at both genotypic and phenotypic levels.

Table 1. Means and standard deviations for various yield related traits in six wheat crosses.

Cross

Tillers per plant

Flag leaf area

Spike density Grains per spike

1000-grain weight

Grain yield per plant

X SD X SD X SD X SD X SD X SD

Sehar-06 × DN-62 11.5 9.3 37.6 9.4 1.7 3.6 58.4 10.0 45.9 7.8 26.5 7.3 Sehar-06 × NR-356 12.7 12.6 41.1 6.5 1.6 4.9 63.1 10.4 42.1 12.7 29.8 10.3 Sehar-96 × ZAS-70 11.9 10.9 35.4 16.1 1.7 3.6 59.9 6.6 43.8 7.2 24.6 8.5 Pasban-90 × DN-62 14.9 8.7 40.2 11.3 1.8 1.8 65.4 13.0 46.3 8.5 27.9 12.1 Pasban-90 × NR-356 12.0 8.8 37.7 8.1 1.7 1.9 69.2 7.1 44.7 7.8 24.8 6.4 Pasban-90 × ZAS-70 11.8 7.5 39.0 6.6 1.6 3.5 60.8 9.8 41.5 6.8 26.9 8.0 X = means, SD = standard deviations.


287

Table 2. Coefficients of variability estimates and genetic advance values for various biometric traits in six wheat crosses. Cross Tillers/

plant Flag leaf area

Spike density

Grains/ spike

1000-grain weight

Grain yield

Sehar-06 × DN-62 h2% 69.6 38.2 64.8 67.6 87.6 76.2 GA 11.0 6.1 3.9 11.5 11.7 9.5 CV 7.4 15.2 3.7 9.5 8.0 16.3 Sehar-06 × NR-356 h2% 77.3 63.8 59.1 79.3 81.6 90.1 GA 16.5 7.0 4.9 14.0 17.6 15.7 CV 9.9 12.9 10.7 11.3 13.8 11.9 Sehar-06× ZAS-70 h2% 63.2 41.8 68.5 73.0 78.3 85.0 GA 11.7 11.4 4.2 8.2 9.6 12.3 CV 9.3 17.4 7.8 18.6 7.4 18.5 Pasban-90 × DN-62 h2% 85.3 51.0 76.9 81.2 94.0 87.0 GA 12.6 9.8 2.3 17.9 13.6 17.8 CV 6.7 10.7 6.0 13.6 9.8 10.5 Pasban-90 × NR-356 h2% 69.5 56.7 61.4 85.1 89.7 82.7 GA 11.9 7.8 2.0 10.3 11.9 9.0 CV 11.7 9.8 9.1 13.1 5.9 15.8 Pasban-90 × ZAS-70 h2% 64.1 48.3 66.0 78.4 82.7 85.0 GA 8.1 5.4 3.9 13.1 9.6 11.6 CV 9.6 15.0 7.6 8.0 14.4 12.9 h2 = heritability, GA = genetic advance, CV = co-efficient of variability. Table 3. Genotypic and phenotypic correlation coefficients among yield and its components. Character Flag leaf

area Spike density

Grains/ spike 1000-grain weight

Grain yield

Tillers/plant G 0.58 0.24* 0.20 -0.39* -0.17 P 0.55 0.18 0.19 -0.38* -0.14 Flag leaf area G 0.17 0.18* 0.07 0.56* P 0.08 0.14* 0.06 0.52* Spike density G 0.28 -0.75** -0.30* P 0.25 -0.72** -0.14 Grains/spike G -0.27 0.53* P -0.24 0.50* 1000-grain weight G 0.39* P 0.35* * Significant at 5% level, ** Highly Significant at 1%


288

DISCUSSION Heritability plays a predictive role in breeding, expressing the reliability of phenotype as a guide in giving authentic information about genetic variation and thus reflecting the extent to which a given trait will be transmitted to the next generation. Higher estimates of heritability, coupled with better genetic advance, confirm the scope of selection in developing new genotypes with desirable traits. Moderate to high heritability estimates coupled with high genetic advance were recorded for tillers per plant, revealing transmission of additive-type genetic variability from parents to offspring for this trait. This suggests that tillers per plant would be fixed in genotypes through proper selection, which was consistent with the findings of Ali et al. (2008).

Correlation studies revealed significant positive association of tillers per plant with spike density, indicating that increase in number of tillers per plant may cause increase in spike density but, at the same time, this increase in number of tillers per plant would also cause a reduction in 1000-grain weight due to negative interactions. Therefore, one must consider both positives and negatives while making selection for tillers per plant. Negative but non-significant correlation between tillers per plant and grain yield contradicted the findings of Anwar et al. (2009) who reported significant positive correlation between tillers per plant and grain yield.

Low to moderate heritability estimates along with high genetic advance recorded for flag leaf area in approximately all crosses were possibly due to some variances constituting the environment variance revealing greater influence of environmental factors compared with genetic ones in the expression of this trait. Therefore, careful selection should be practiced in the case of flag leaf area because this trait plays a pivotal role in proper grain filling and development. About 75% of grain composition is composed of carbohydrates produced by green leaf, therefore, proper development of flag leaf is crucial for efficient photosynthetic activity that is directly related with grain yield per plant. These results partially contradicted the work of Kashif and Khaliq (2004) who found very high heritability estimates for flag leaf area in their experiment. This difference in finding may be due to a variety of reasons such as

background of genotypes involved and methodology. From correlation studies, it becomes evident that flag leaf area has significant positive attachment with grains per spike and grain yield per plant (Bhuttah, 2006) such that unfolded flag leaf area is an important direct selection criterion for developing high-yielding wheat genotypes for any target environment.

Heritability estimates for spike density were moderate in general along with low genetic advance (Firouzian, 2003) in all crossing combinations, indicating predominance of non-additive gene action for this trait that may be exploited through heterosis breeding with careful selection of parents in appropriate crosses. Correlation studies reveal that the increase in spike density would exert no effect on grains per spike but may cause reduction in 1000-grain weight considerably along with exerting deleterious effect on grain yield per plant significantly. Therefore, restrictions must be imposed on spike density while making selection for improved grain yield per plant of a genotype.

Moderate to high heritability coupled with high genetic advance for grains per spike was recorded in all crossing combinations (Singh et al., 2006; Haq et al., 2008),showing this trait to be under the control of additive genes exploitable through proper selection strategy in order to evolve genotypes with larger number of grains per spike. Grains per spike constitutes a vital part of grain yield in wheat but depends on other traits like spike length, its compactness and proper grain development in florets. Thus, selection focused on grains per spike along with other traits can bring about fairly rapid improvement in yield potential of a genotype. Furthermore, correlation studies point out significant association of grains per spike with yield, suggesting that any increase in number of grains per spike would have a direct and proportionate impact on grain yield per plant. Efforts must be made in future to develop genotypes with higher number of grains per spike in order to increase economic yield per unit area. These results were consistent with the findings of Ali et al. (2008).

Very high heritability estimates accompanied by very high genetic advance were observed for 1000-grain weight (Sattar et al., 2003; Eid et al., 2009) in all crosses, pointing out additive type gene interaction among alleles


289

governing this trait.This implies the possibility of fast genetic improvement in this parameter through repeated cycles of selection, increasing grain yield in the long run. Grain weight is an important contributor to yield and higher heritability values for this trait obtained in present study indicate the presence of considerable variability that may be exploited through efficient selection in appropriate cross combinations. Moreover, correlation results depicted significant positive association of 1000-grain weight with yield, suggesting that grain yield per plant can be improved by increasing1000-grain weight of a genotype through selection. Therefore, it seems logical to give this trait due attention in future breeding programs for sustained improvement in wheat. This result finds support from earlier works of Lei et al., (2006) and Akram et al.(2008).

Both heritability estimates and genetic advance values were of high magnitude for grain yield per plant, predicting the feasibility of selection for identification and isolation of high-yielding inbred lines from segregating populations at early stages of hybridization. Although yield has complex inheritance mechanism, the presence of high heritability and genetic advance obtained in all crossing combinations reveals the suitability of selection for fast genetic improvement of breeding material. These results were in close agreement with those of Ahmed et al. (2007) and Akcura (2009). CONCLUSION Based on high heritability and high genetic advance values shown by most of the traits under study, especially tillers per plant, grains per spike, 1000-grain weight and grain yield per plant, we conclude that the determinant genetic effects for the phenotypic expression of these parameters are fundamentally of the additive type. For this reason, a high response would be achievable after each selection cycle. Moreover, higher genetic correlations as compared with phenotypic ones obtained in the present study suggest a strong inherent association among these biometric traits at the genetic level and traits like flag leaf area, grains per spike and 1000-grain weight may be used as direct criteria for efficient selection in any breeding program due to their positive association with grain

yield. In addition, the most promising cross combinations were Pasban-90×DN-62 and Sehar-06×NR-356, suggesting that varieties/lines involved in these crosses deserve greater attention in future breeding programs to develop high-yielding wheat varieties on a sustainable basis.

However, these findings are based on a small number of crosses among the elite genetic germplasm of Pakistan., There is therefore a need to conduct further research in this direction to develop efficient selection criteria for improving grain yield in bread wheat. ACKNOWLEDGEMENTS The author is grateful to the Department of Plant Breeding and Genetics, University of Agriculture, Faisalabad for providing inputs for this research and also to Atif Kamran and Katie Satchwell of AFNS Department, University of Alberta, for their careful review of the manuscript. REFERENCES Ahmed N, Chowdhry MA, Khaliq I, Maekawa M

(2007). The inheritance of yield and yield components of five wheat hybrid populations under drought conditions. Indonesian J. Agric. Sci. 8: 53-59.

Akcura M (2009). Genetic variability and interrelationship among grain yield and some quality traits in Turkish winter durum wheat landraces. Turk. J. Agric. For. 33: 547-556.

Akram Z, Ajmal S, Munir M (2008). Estimation of correlation coefficients among some yield parameters of wheat under rainfed conditions. Pak. J. Bot. 40: 1777-1781.

Ali Y, Atta BM, Akhter J, Monneveux P, Latee Z (2008). Genetic variability, association and diversity studies in wheat germplasm. Pak. J. Bot. 40: 2087-2097.

Anwar J, Ali MA, Hussain M, Sabir W, Khan MA, Zulkiffal M, Abdullah M (2009). Assessment of yield criteria in bread wheat through correlation and path analysis. J. Ani. Plant Sci. 19: 185-188.

Arshad Y, Yildirim MB (1996). Studies on determination of some physiological and Agronomic traits used in the simulation of wheat growth. Tur J. Agric. For. 20: 219-224.

Bhutta WM (2006). Role of some agronomic traits for grain yield in wheat (Triticum aestivum L.)


290

genotypes under drought conditions. Rev. UDO Agric. 6: 11-19.

Burton GW, Devane FW (1953). Estimating heritability in tall fescue (Festuc arandinaca) from replicated clonal material. Agron. J. 45: 478-481.

Eid MH (2009). Estimation of heritability and genetic advance of yield traits in wheat (Triticum aestivum L.) under drought condition. Int. J. Genet. Mol. Biol. 1:115-120.

Firouzian A (2003). Heritability and genetic advance of grain yield and its related traits in wheat. Pak. J. Biol. Sci. 4: 2022-2023.

Haq W, Malik MF, Rashid M, Munir M, Akram Z (2008). Evaluation and estimation of heritability and genetic advancement for yield-related attributes in wheat lines. Pak. J. Bot. 40: 1699-1702.

Kashif M, Khaliq I (2004). Heritability, correlation and path coefficient analysis for some metric traits in wheat. Int. J. Agric. Biol. 6: 138-142.

Kwon SH, Torrie JH (1964). Heritability and inter-relationship among traits of two soybean populations. Crop Sci. 4: 196–8.

Lei WD, Mahmood Q, Qureshi AS, Khan MR, Hayat Y, Jilani G, Shamsi IH, Tajammal MA, Khan MD (2006). Heterosis, correlation and path analysis of morphological and biometrical characters in wheat (Triticum

aestivum L. Emp. Thell). Agric. J. 1: 180-185.

Mahmud I, Kramer HH (1951). Segregation for yield, height and maturity following a soybean cross. Agron. J. 43: 605-609.

Muller J (1991). Determining leaf surface area by means of linear measurements in wheat and triticale (a brief report). Arch. fuchtungs frschung. 21: 121-123.

Nofouzi F, Rashidi V, Tarinejad AR (2008). Path analysis of grain yield with its components in durum wheat under drought stress. International meeting on soil fertility land management and agroclimatology, Turkey. 3: 681-686.

Quarrie SA, Stojanovic J, Pekic S (1999). Improving drought resistance in small grained cereals: a case study, progress and prospects. Plant Growth Reg. 29: 1-21.

Sattar A, Chowdhry MA, Kashif M (2003). Estimation of heritability and genetic gain of some metric traits in six hybrid populations of spring wheat. Asian J. Plant Sci. 2: 495-497.

Singh GP, Chaudhary HB (2006). Selection parameters and yield enhancement of wheat (Triticum aestivum L.) under different moisture stress conditions. Asian J. Plant Sci. 5: 894-898.


POPULATION IMPROVEMENT FOR SEED YIELD AND OIL CONTENT BY USING WORKING GERMPLASM IN SUNFLOWER (Helianthus annuus L.)

M. RAMESH1*, J. ARUNAKUMARI2, Y. PRASHANTH1, A.R.G. RANGANATHA3

and M.Y. DUDHE3

1Department of Genetics and Plant Breeding, College of Agriculture, ANGRAU,

Rajendranagar, Hyderabad 500030, A.P. India. 2Department of Biochemistry College of Agriculture, ANGRAU,

Rajendranagar, Hyderabad 500030, A.P. India. 3Directorate of Oilseeds Research, Rajendranagar, Hyderabad 500030. AP, India.


SUMMARY

The study aimed at sunflower population improvement through recurrent selection. The present investigation was undertaken with the third cycle base population (C3) by adopting simple recurrent selection. 20,000 plants from the third cycle, were raised during rainy season 2010 and 2,000 phenotypically superior plants were selfed. Based on self-fertility, the best 200 plants were selected for progeny testing during spring 2011 and evaluated against two national checks (DRSF-108 and 113). In the C3 base population, a wide range of variability was exhibited in days to maturity, plant height, head diameter, seed yield/plant, and oil content. Positive selection differential was observed for plant height, head diameter, seed yield and oil yield, whereas days to maturity and oil content showed negative values. PCV and GCV revealed high values for seed yield/plant, oil yield/plant and moderate values for plant height, head diameter, 100-seed weight and oil content. High heritability associated with high genetic advance as percentage of mean was recorded for plant height, head diameter, oil yield/plant and oil content, indicating less environmental influence on these characters and a role of additive gene action. High heritability estimates associated with moderate genetic advance as percent of mean was recorded for days to 50% flowering, days to maturity, seed yield/plant, suggesting that these characters were less influenced by environment but governed by both additive and non-additive gene action. Superior genotypes may serve to develop inbreds for use in heterosis breeding. Keywords: Recurrent selection, sunflower, heritability, seed yield, oil content

Manuscript received: September 9, 2012; Decision on manuscript: March 13, 2013; Manuscript accepted: April 1, 2013; Revised

online August 25, 2013. © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION

Sunflower (Helianthus annuus) is one of the important edible oil crops of India. This crop has shown distinct superiority over other edible seed crops owing to its wider adoptability, short duration and higher seed yield, oil production

per unit area (Dudhe et al., 2009). Sunflower seeds are rich in protein, a high-quality vegetable oil used in making margarine. Sunflower oil has a light taste, is suitable for frying and has some health benefits. Sunflower

RESEARCH ARTICLE


292

was domesticated in North America and is a crop in temperate regions of the world, including many parts of the U.S and Canada. In temperate regions, sunflower can produce vast quantities of oil per acre, as there are genotypes which contain 40-50% oil by weight. Most of the procured materials from European countries to India show declining oil content in the range of 35-40 % - it may be due to the fact that the Indian climate is tropical (Dudhe 2004). In India, sunflower area and production continued to show a declining trend during 2010-11: area 89 lakh ha; production 62.5 lakh tons; productivity 696 kg/ha (Anonymous, 2012) compared to the earlier years. Hence, there is immediate need to develop hybrids with high oil content along with yield in sunflower.

The extent of gain realized in a quantitative trait in a population in response to selection depends both on the selection intensity employed and the heritability estimate of the trait under selection (Rani and Kumar, 2007). Systematic population improvement programs helps sunflower breeders to generate superior populations with improved performance by directional selection followed by intermating of the reselected lines. This allows the breeder to access favorable recombinations and stabilize traits within the sunflower's gene pool. It gives impetus for the development of elite inbred lines with superior agronomic characteristics and enables the development of superior hybrids (Seneviratne et al. 2004). Also, these inbreds act as a potential genetic resource. Miller et al. (1977) obtained an increment of 12.2% of oil content after 3 cycles of simple recurrent selection. Seneviratne et al. (2004) reported an increase in high seed yield, number of filled seeds and oil content after 3 cycles of simple recurrent selection. Hence, the present study was undertaken to improve the base population by adopting simple recurrent selection with a progeny testing for oil and yield attributes of sunflower with three selection cycles. The concerned parameters evaluated are response to selection, genetic parameters, i.e., variability, heritability (broad-sense), genetic advance as percentage of mean and character association.

MATERIALS AND METHODS The material used in the present study was supplied by the Directorate of Oilseeds Research, from Germplasm Multiplication Unit (GMU), Rajendranagar, Hyderabad. The present investigation was initiated during 2008 to improve sunflower population by using simple recurrent selection as suggested by Seneviratne et al. (2004). The present study aimed at the improvement of the sunflower population in the third selection cycle (C3) which was raised during rainy season 2010. The base gene pool was generated by crossing four promising lines for yield and oil content selected from working germplasm (C0). 2000 composited intercross seeds were planted during rainy season of 2009. Based on morphological observations 300 superior plants were selected for head diameter, plant height and maturity duration. Harvested seeds were evaluated for high oil content and high seed yield and 100 individual plant progenies planted in isolation. Bulk pollens were collected from all plants and pollinated. An equal amount of 10,000 crossed seeds were harvested and composited (C0). The same procedure was followed to raise the second selection cycle. Using seeds of second selection cycle (C2) a population of about 20,000 plants were grown in bulk and about 2,000 plants were selected and selfed. Out of these 2,000 individuals, 200 best performing plants were selected for progeny testing based on high autogamy, high oil content and high seed yield. The progeny test was performed during spring 2011. The selected plants were evaluated against two national checks (DRSF 108 and 113) in randomized block design, replicated twice. Observations were recorded for days to 50% flowering, days to maturity, plant height, head diameter, 100-seed weight, seed yield/ plant, oil content, and oil yield/plant. The genotypic (GCV) and phenotypic (PCV) coefficients of variation were calculated by a formula given by Burton (1952). Heritability in broad sense was calculated by the formula given by Lush (1940) as suggested by Johnson et al. (1955). From the heritability estimates, genetic advance (GA) was estimated by the following formula given by Johnson et al. (1955). The mean performance of the 2,000 individuals and mean, range, (PCV),

Ramesh et al (2013)

293

(GCV), heritability in broad sense, GA, genetic advance as percent of mean and selection differential for 200 progenies were evaluated in this study. RESULTS The genotypes in the third base population exhibited considerable variation in seed yield and yield-contributing characters (Table 1). Plant height ranged from 52 cm to 185 cm, head diameter from as low as 4 cm to as high as 20 cm, seed yield/plant from 1 g to very high 62.0 g, days to maturity from 78 days to 110 days and oil content from 37.0 to 43.0%. Positive selection differential values were observed for plant height (2.4 cm), head diameter (2.1 cm), seed yield/plant (2.6 g) and oil yield/plant (1.8 g), whereas days to maturity (-9.9 days) in

desirable direction and oil content (-4.4%) showed negative values (Table 2). The values of PCV and GCV were high for seed yield/plant (26.6 and 25.0) and oil yield/ plant (31.4 and 27.9) (Table 3). Plant height, head diameter, 100-seed weight and oil content showed moderate PCV and GCV values. The maximum heritability (broad sense) values was exhibited by head diameter (96.0%), plant height (96.0%), followed by oil content (94.0). Progress in shifting the genotypic mean and gene frequencies of a trait in the population toward the desired direction as a result of selection is calculated as genetic advance. The highest genetic advance was recorded for plant height (19.3). High genetic advance as percent of mean were recorded for oil yield/plant (42.4), 100-seed weight (20.1), head diameter (19.1.), oil content (15.1) and plant height (14.6).

Table 1. Variability parameters of the third cycle (C3) base population in sunflower.

Character Mean Range Days to 50% flowering 51.0 45-57 Days to maturity 94.0 78-110 Plant height (cm) 118.5 52-185 Head diameter (cm) 11.8 4-20 100-seed weight (g) 4.7 4.0-5.5 Seed yield per plant (g) 12.4 1-62 Oil content (%) 40.2 37.2-43.4 Oil yield/plant (g) 10.0 8-12

Table 2. Range, mean and selection differential for seed yield and attributes in sunflower.

Character Range Mean Selection Days to 50% flowering BP

S1 45-57 43-60

51.0 51.5

-0.5

Days to maturity BP S1

78-110 76-97

94.0 84.9

-9.9

Plant height (cm) BP S1

52-185 73.1-150.1

118.5 120.9

2.4

Head diameter (cm) BP S1

4-20 7.5-15.0

11.8 13.0

2.1

100-seed weight (g) BP S1

4.0-5.5 3.9-6.0

4.7 4.9

-0.2

Seed yield/plant (g) BP S1

1-62 15.2-49.2

12.4 15.3

2.9

Oil content (%) BP S1

38.8-46.4 27.0-41.9

40.2 35.8

-4.4

Oil yield/plant (g) BP S1

37.2-43.4 4.9-17.8

5.6 7.4

1.8


294

Table 3. Estimates of variability and genetic parameters for seed yield and its attributes for 200 lines in sunflower. Character Mean Coefficient of variation

(%) Phenotypic Genotypic

Heritability (%) (Broad sense)

Genetic advance

Genetic advance as % of mean

Days to 50% flowering 49.5 6.1 5.60 79.2 4.43 8.1 Days to maturity 78.4 4.0 3.6 78.9 4.6 5.6 Plant height (cm) 118.9 12.4 9.4 96.0 19.3 14.6 Head diameter (cm) 11.0 13.7 12.8 96.0 2.5 19.1 100-seed weight (g) 7.0 17.3 15.9 68.2 0.9 20.1 Seed yield/plant (g) 17.0 26.6 25.0 77.3 8.6 4.2 Oil content (%) 34.8 10.1 10.0 94.0 2.7 15.1 Oil yield/plant (g) 9.46 31.4 27.9 80.9 4.9 42.4 DISCUSSION In the C3 base population, a wide range of variability was exhibited for days to maturity, plant height, head diameter, seed yield/plant, and oil content. The selection imposed has narrowed down the variability of days to maturity, plant height, head diameter, seed yield/plant, oil content and oil yield/plant. In our study, positive selection differential values were observed for plant height, head diameter, seed yield/plant and oil yield/plant. Virupakshappa and Sindagi (1987) and Salera and Detti (1990) observed high range of variation for plant height, 100-seed weight, seed yield and oil content. Teklewold et al. (1999) recorded high range of values for percent autogamy, seed yield per plant and percent seed set. The values of PCV and GCV were high for seed yield/plant and oil yield per plant. These findings were consistent with the findings by Teklewold et al. (1999). In our study, four characters showed moderate PCV and GCV values. Moderate PCV and GCV values were reported for oil content by Teklewold et al. (1999). Seneviratne et al. (2004) observed moderate PCV and GCV values for plant height, head diameter along with oil content, which is in agreement with present findings. Among the characters, days to 50% flowering and days to maturity exhibited low PCV and GCV values. These results were in harmony with the results of Teklewold et al. (1999), Ashok et al. (2000) and Seneviratne et al. (2004). The heritability estimates indicate the expressivity of trait with which a genotype can

be assessed by its phenotype and its effective utilization in judging the phenotypic selection. High heritability (h2b) estimates were observed for head diameter, plant height, oil content/plant, oil yield/plant, days to 50% flowering, days to maturity, seed yield /plant and 100-seed weight (Table 3). Similar results were also reported by Teklewold et al. (1999), Ashok et al. (2000) and Khan (2001). Seneviratne et al. (2004) reported high heritability estimates for plant height, days to 50% flowering, head diameter, days to maturity, 100-seed weight, oil content, seed yield per plant and oil yield, which are in agreement with present investigation.

Heritability values alone cannot provide any indication of the amount of progress that would result from selection because heritability in broad sense includes both additive and non-additive gene action. Therefore, high heritability estimates in broad sense would be a reliable tool for selection if accompanied by high genetic advance as percent of mean. High heritability associated with high genetic advance as percent of mean were recorded for plant height, head diameter, oil yield/ plant and oil content, indicating lesser environmental influence on these characters and a role of additive gene action. Similar observations for head diameter and oil yield were also reported by Teklewold et al. (1999) and Ashok et al. (2000). High heritability estimates were associated with days to 50% flowering, days to maturity, and seed yield/plant, suggesting that these characters were less influenced by environment but governed by both additive and non-additive gene action.

Ramesh et al (2013)

295

These findings were in true agreement with the results obtained by Muhammad et al. (1992). Seneviratne et al. (2004) recorded a similar type of observations for plant height, 100-seed weight, and oil content, suggesting that these characters were governed by both additive and non-additive gene action. High heritability estimates associated with low genetic advance recorded for 100-seed weight is influenced by the environment due to its non-additive gene action.

In the present investigation, enormous variability was detected in the sunflower population studied with three selection cycles. This made the cyclic selection more effective and recorded substantial amount of genetic gain for the traits. The elite genotypes with high seed yield, oil content and oil yield can be utilized to develop superior inbred lines for the development of superior synthetics and hybrids. Systemic population improvement programs by using few working germplasm for yield and oil content help breeders generate superior populations with improved performance of the concerned characters by imposing directional selection and intermating them for the development of elite inbred lines. Hence, a population improvement program is essential to sunflower improvement. It will give a new gene pool, which is useful to breeders. REFERENCES Anonymous (2012). Annual report of All India

Coordinated Research Project (AICRP) on Sunflower. pp.1-2.

Ashok S, Narayanan SL, Kumaresan D (2000). Variability studies for yield and its components. J. Oilseeds Res. 17 (2): 239-241.

Burton GW, Devane EH (1952). Estimating heritability in tall fescue (Festuca arundinaceae) from replicated clonal material. Agron. J. 45: 478-481.

Dudhe MY (2004). Diallel analysis of restorer lines in sunflower. M.Sc. Thesis, Dr. PDKV. Akola (Maharashtra).

Dudhe MY, Moon MK, Lande SS (2009). Evaluation of restorer lines for heterosis studies on sunflower (Helianthus annuus L.). J. Oilseeds Res. 26 (Special Issue): 140-142.

Johnson HW, Robinson HF, Comstock RE (1955). Estimation of genetic and environmental variability in soybean. Agron. J. 47: 314-318.

Khan A (2001). Yield performance heritability and interrelationship in some quantitative traits in sunflower. HELIA 24 (34): 35-40.

Lush JL (1940). Intra-sire correlation and regression of offspring in rams as a method of estimating heritability of characters. Proc. of Am. Soc. Animal Prod., 33: 292-301.

Miller JF, Fick GN, Cedeno JR (1977). Improvement of oil content and quality in sunflower. Agron. Abstracts pp. 64.

Muhammad T. Idrees G, Tahir A (1992). Genetic variability and correlation studies in sunflower. Sarhad J. Agri. 8(6): 659-663.

Salera E, Detti GM (1990). Adaptation and yield potential of sunflower (Helianthus annuus L.). Floral Agric. 46: 53-57.

Seneviratne KGS, Ganesh M, Ranganatha ARG, Nagaraj G, Rukmini Devi K (2004). Population improvement for seed yield and oil content in sunflower. HELIA 27 (41): 123-128.

Shobha Rani T, Ravikumar RL (2007). Genetic enhancement of resistance to alternaria leaf blight in sunflower through cyclic gametophytic and sporophytic selections. Crop Sci. 47: 529–536.

Teklewold A, Jayaramaiah H, Ramesh S (1999). Genetic variability studies in sunflower. Crop Improve. 26: 236-240.

Virupakshappa K, Sindagi SS (1987). Characterization, evaluation and utilization of sunflower germplasm accessions, a catalogue. Unit of the Project Coordinator (Sunflower), GKVK, Bangalore.


RESPONSES OF PEANUT (Arachis hypogaea L.) GENOTYPES TO N2-FIXATION UNDER TERMINAL DROUGHT AND

THEIR CONTRIBUTIONS TO PEANUT YIELD

W. HTOON1, W. KAEWPRADIT1*, S. JOGLOY1, N. VORASOOT1, B. TOOMSAN1, C. AKKASAENG1, N. PUPPALA2 and A. PATANOTHAI1

1 Department of Plant Science and Agricultural Resources, Faculty of Agriculture, Khon Kaen University,

Muang, Khon Kaen 40002, Thailand 2New Mexico State University, Agricultural Science Center, Clovis, New Mexico 88101, USA


SUMMARY Five peanut genotypes were tested under well-watered and terminal drought conditions to characterize the effect of terminal drought on N2-fixation (NF), to examine whether there were major differences among peanut genotypes, and to investigate the contribution of NF to yield under terminal drought. A split-plot design with four replications was used. Plants were grown under two main plot treatments: well-watered and terminal drought (1/3 available water [AW]) conditions at R7 growth stage until final harvest. Data on meteorological conditions, soil properties, soil moisture content, plant water status, biomass production (BM), pod yield (PY), nodule dry weight (NDW) and NF were recorded. Terminal drought greatly reduced the NF and NDW but peanut genotypes did differ for such traits. ICGV 98324 was the best genotype as could maintain high NF under terminal drought and NF reduction was relatively low (13 %). ICGV 98308 reduced 58% of NF under terminal drought and ICGV 98348 also showed high reduction of 74%, indicating that terminal drought seriously reduced NF from 0.94 g N plant-1 into 0.24 g N plant-1. When irrigation was withheld at R5, five genotypes in this study continued their nodule growth until R7 but NF might have been decreased soon after soil water depletion due to higher sensitivity of NF to water deficit. NF highly contributed to BM production and PY under well-watered condition. NF contributed to biological yields consistently but there was no contribution to economic yields under terminal drought condition. Keywords: Arachis hypogaea L., biological nitrogen fixation, development, water stress, economic yield

Manuscript received: October 31, 2012; Decision on manuscript: December 26, 2012; Manuscript accepted: January 29, 2013. © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013

Communicating Editor: Sivananda V. Tirumalaraju

INTRODUCTION Drought, a period of dryness, is one of the main causes of reduced growth and yield components in peanut (Arachis hypogaea L.). Since 80% of the world’s peanut production is under rainfed agriculture system with limited inputs, the increasing frequency of drought constitutes a major abiotic stress in peanut (Johensen et al., 1994; Nageswara Rao and Wright, 1994 and FAOSTAT, 2008). Duration and intensity of

drought and the growth stage at which the stress occurs have large effects on growth and productivity. Drought occurring during pre-flowering stage has a small effect on yield or, in some cases was found to increase yield (Nageswara Rao et al., 1988; Nautiyal et al., 1999 and Puangbut et al., 2009). Drought at the time of reproductive phase significantly reduced pod yield (Wright and Nageswara Rao, 1994 and Reddy et al., 2003) and drought at pod setting can reduce 15 to 88 % of yield (Ravindra et al.,

RESEARCH ARTICLE


297

1990; Nautiyal et al., 1999 and Vorasoot et al., 2003). Since peanut is grown as a rainfed and post-rainy season crop, terminal drought, which occurs during the pod-filling phase of peanut, is common and has been observed to cause the greatest reduction in pod yield (Ravindra et al., 1990; Wright et al., 1991 and Girdthai et al., 2010). In fact, yield depressions in response to late-season drought in peanut have also been observed elsewhere (Pallas et al,. 1977 and Wright et al. 1991). The partitioning coefficient might be a more reliable selection criterion for identifying genotypes tolerant of end-of-season drought than yield. But the underlying mechanisms that might be contributing to yield reduction when peanut was subjected to terminal drought need to be investigated further to ensure an effective and efficient peanut breeding program under drought condition.

Drought resistance is a complex trait, the expression of which depends on action and interaction of different morphological and physiological traits. Wright et al. (1996) suggested that the direct use of pod yield as drought resistance trait had been limited by high resource investments and poor repeatability of the results due to large G x E (genotype x environment) interactions resulting in slow breeding progress (Araus et al., 2002). Information on physiological characters contributing to high yield under drought might reveal the underlying mechanisms.

Like other legumes, peanut has an ability to fix nitrogen with the help of Rhizobium bacteria through the process of symbiosis N2-fixation (NF). Nitrogen is an essential element for crop growth, as it is used to produce amino acids, proteins, nucleic acids and other N containing components necessary for life such as chlorophyll. Giller et al. (1987) found that the amount of plant-available soil N was small, so that 86-92% of peanut plant N was derived from NF. Improvement of NF under drought conditions is a promising strategy that would lead to yield improvement (Pimratch et al., 2008). But NF and its related traits such as biomass, nodule number, and nodule dry weight in peanut were also affected by drought (Sinclair and Serraj, 1995 and Pimtrach et al., 2008). The reduction of NF might vary according to the degree of stress, the period of stress and the

stage of crop development (Guerin et al., 1990 and Giller, 2001). Under prolonged drought condition, nitrogenase activity, which is often used as an indicator of NF, decreases and reduction at 60 days after emergence (DAE) is higher than at 90 DAE. But nodule number and nodule dry weight at 90 DAE are higher than those at 60 DAE (Pimratch et al., 2008). In contrast, pre-flowering drought had no effect on NF; the peanut plant could recover after water stress condition (Htoon et al., 2009). Under drought stress conditions, NF is greatly reduced, leading to low N accumulation, dry matter production, and yield (DeVries et al., 1989 and DeSilva et al., 1996). Peanut genotypes were also a source of variation for such NF traits. Unfortunately, information is still lacking on the responses of peanut genotypes to NF under terminal drought.

In general, NF attains a peak during the pod-filling stage and declines at maturity under normal condition (Nambiar and Dart, 1983). But when peanut is exposed to drought at such pod-filling stage until maturity, the response and contribution of NF might play an important role in peanut productivity under terminal drought. Such information on the response of NF should provide insights into the mechanisms underlying the adaptation of peanut under terminal drought condition as well.

The objectives of this study were to characterize the effect of terminal drought on NF, to examine whether there were any major differences among peanut genotypes, and to investigate the contribution of NF to yield under terminal drought.

MATERIALS AND METHODS Experimental design and plant materials Five peanut genotypes have been used in these experiments. Peanut genotypes selected were ICGV 98308, ICGV 98324, ICGV 98348, Tainan 9, and Tifton 8. The ICGV lines are from the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) and they were identified as drought-resistant with different drought tolerance index (DTI). Girdthai et al. (2010) reported that genotype ICGV 98324 had

Htoon et al. (2013)

298

high DTI for total biomass in both years whereas ICGV 98348 also had high DTI for pod yield across years. ICGV 98308 had low DTI for pod yield (Girdthai et al., 2010). Tainan 9 performed poorly for total biomass and pod yield under terminal drought and had the highest reduction in total biomass. Tifton-8 exhibited the highest biomass production under terminal drought but reductions in pod yield of Tifton-8 was also high (low DTI) (Vorasoot et al., 2003 and Girdthai et al., 2010). Moreover, a non-nodulating line, which is used as a check variety for N2-fixation, was also grown in each replication. This genotype was only used to estimate the amount of fixed N2 of each genotype at each treatment but it was not included when the data were subjected to analysis of variance.

These five peanut genotypes were grown under two soil moisture conditions. They are well-watered condition and terminal drought (1/3 available water) at R7 growth stage (Boote, 1982) to final harvest. The water stressed period in this experiment was defined as terminal drought (Girdthai et al., 2010). A split-plot in a randomized complete block design with four replicates was used for both years at the Field Crop Research Station of Khon Kaen University, Thailand for two years from October 2010 to January 2011 and repeated from October 2011 to January 2012. Soil type at the field site has been classified as Yasothon soil series (loamy-sand, Oxic Paleustults).

The soil was disk-plowed thrice, leveled and harrowed. During land preparation, the land was incorporated with lime (CaCO3) at a rate of 625 kg ha-1. Basal doses of P and K fertilizers were also incorporated into the topsoil at the rate of 24.7 kg ha-1 and 31.1 kg ha-1, respectively, shortly prior to planting. Seeds were treated with captan (3a,4,7,7a-tetrahydro-2-[(trichloromethyl)thio]-1H-isoindole-1,3(2H)-dione) at the rate of 5 g kg-1 seed to protect from fungal diseases before planting. Plot size was 5 x 5 m with 50 cm spacing between rows and 20 cm between plants. Three to four seeds were sown per hill by hand and after germination, they were thinned to one plant per hill. In order to increase the Rhizobium population in the soil, inoculation of Rhizobium was accomplished by applying a water-diluted commercial peat-based inoculum of Bradyrhizobium (mixture of strains

THA 201 and THA 205; Department of Agriculture, Ministry of Agriculture and Cooperatives, Bangkok, Thailand) at the rate of 13 kg seed-1 on the rows of peanut seeds soon after planting. Weeds were controlled by the application of alachlor (2-chloro-2’, 6’-diethyl-N-(methoxymethyl) acetanilide 48%, w/v, emulsifiable concentrate) at the rate of 3 L ha-1 at planting and hand weeding was done until pegging stage. Gypsum (CaSO4) at the rate of 312 kg ha-1 was applied at 40 days after emergence (DAE).

Irrigation Subsurface drip irrigation system was installed with a spacing of 50 cm mid-way between peanut plant rows to supply water. Drip lines with a distance of 20 cm between emitters were installed at 10 cm below the soil surface and fitted with a pressure valve and water meter to make sure that water supply is efficient and uniform across each plot. Sub-valves were set up monitor the water supply to respective genotypes in order to get the predetermined water-stressed level (1/3 AW). For the well-watered treatment, water was applied daily from planting to harvest. But growth stage of plants was checked regularly in water-stressed plots in order to determine the right time to stop irrigation. In general, the drought treatment was initiated at 15-20 days before R7 growth stage and the times of irrigation withdrawal were different within genotypes according to growth stages of peanut genotypes and reliable data from preliminary trials and the simulated data of Girdthai et al. (2010). At the terminal drought plots, irrigation was withheld and soil moisture was allowed to decrease gradually to meet the drought condition, 1/3 available water (AW) at R7 growth stage of each genotype. And from then on, drought condition (1/3 AW) was maintained until harvest. Rainout shelters were used to protect the water-stressed plots from the unseasonal rain at 90-93 days after planting (DAP) in 2011-2012.

In order to maintain the specific water regimes, the amount of water that was applied was based on crop water requirements using the Doorenbos and Pruitt (1992) methodology along with water loss from surface evaporation which


299

was calculated as described by Singh and Russell (1981).

Data collection Meteorological conditions, soil properties and soil moisture Relative humidity (%), water evaporation (mm), rainfall (mm), maximum and minimum air temperature (ºC), and solar radiation (Cal cm-2) were recorded daily from sowing until harvest by a weather station located 100 m away from the experimental field. Physical and chemical properties of soil covering all treatments and replications were determined before planting. Soil samples were taken from 5 points per main plot at depths of 0-5, 5-15, 15-30, 30-45, 45-60, 60-75 and 75-90 cm and they were air-dried. After drying and bulking, representative soil samples were analyzed to determine the physical and chemical properties. In addition, soil moisture content; field capacity (FC) and permanent wilting point (PWP) also were determined with the pressure plate method.

Soil moisture was measured by gravimetric method at depths of 0-5, 10-15, 25-30, 40-45, 55-60, 70-75 and 85-90 cm at planting, the day irrigation was withdrawn, at R7 growth stage, and at harvest.

Plant water status Relative water content (RWC) was recorded to evaluate plant water status and it was measured on the day irrigation was withdrawn, R7 growth stage and at harvest following Kramer (1980). It was measured at one leaflet of the second fully expanded leaf from the top of the main stem of 5 plants for each plot at 10:00-12:00 am (Clavel et al., 2006 and Girdthai et al., 2010). These five leaflets were put into a vial with a rubber stopper and the vial was sealed with parafilm. The vials were suddenly kept inside the ice box to prevent moisture loss. After measuring the field weights, the leaflets were soaked in distilled water for 8 hours and turgid weights were determined again. Then, these leaflets were oven-dried at 80 ºC for 48 hours or until the dry weight became constant.

Finally, RWC was determined as follows: RWC (%) = [(FW-DW) / (TW-DW)] x 100, where FW = sample field weight, TW = sample turgid weight (saturated weight) and DW = sample dry weight. Nodule dry weight, biomass and pod yield At full maturity (R8), for each plot excluding border plots, plants in an area of 8 m2 were harvested. They were uprooted, and soil was gently removed from the root by washing them on a 0.5-mm screen. Nodules were then removed from each root by hand. These nodules were oven-dried at 80 ºC for 48 h or until dry weight became constant and nodule dry weight was recorded for each genotype. At the same time, leaves, stems and pods were separated and they were oven-dried at 80 ºC for 48 h or until dry weight became constant and their dry weights were recorded. Biomass production was calculated as the sum of the dry weight of shoots and pods. The pods were removed from the plants and they were air-dried to obtain approximately 8% moisture content. Then, pod yield was calculated and pods were shelled. Nitrogen fixation (NF) and its reduction (%) Fixed N2 was determined at harvest by the same reliable N-difference method following McDonagh et al. (1993), Bell et al. (1994) and Pimtrach et al. (2008). Separated leaves, stems and pods, which were already oven-dried at 80 ºC for 48 h and pods were used. First, dry pods were hand-shelled. These dried leaves, stems, shell and seed were ground using a hammer mill. Then, total N was individually analyzed for each shoot, shell and seed. It was measured using the automated indophenol method (Schuman et al., 1973) and read on a flow injection analyzer model 5012 (Tecator Inc., Hoganas, Sweden). Finally, total N (g plant-1) was calculated as the sum of N in shoot (leaf + stem), shell, and seed. Fixed N2 (NF) content was calculated as follows:

Htoon et al. (2013)

300

NF of genotype = total N of genotype – total N of non-nodulating line at the respective main plot. Percentages of reduction in NF and nodule dry weight (NDW) from drought stress were used to evaluate the sensitivities of the genotypes to drought stress. Percentages of the reduction in these traits were calculated for each genotype as follows:

Percentage of reduction in NF = [1-(NF under stress/ NF under non stress)] x 100, and, Percentage of reduction in NDW = [1-(NDW under stress/NDW under non stress)] x 100.

Statistical analysis The data were subjected to analysis of variance according to a split plot design (Gomez and Gomez, 1984) and all calculation was performed using MSTAT-C package (Bricker, 1989). Combined analyses of variance were done for those characters where error variances for the two years were homogeneous. Due to the significance of water x genotype interaction for NF and BM (Table 3), data of each water regime were separately analyzed according to RCBD, and Duncan’s multiple range test (DMRT) was used to compare the means (Gomez and Gomez, 1984). Correlation coefficients between NF, NDW, BM and PY were calculated separately for each genotype at each water regime of each replication to assess their relationship. RESULTS AND DISCUSSION Meteorological conditions, soil properties and plant water stress The two experiments were conducted from October 2010 to February 2011 and from October 2011 to February 2012. Although there was total rainfall, (30.2 mm) during the drought-stress period in 2011/2012, rainout shelters were used to protect the drought-stressed main plots from the rain. So there was no interference from the rain and the desired soil moisture contents could be controlled. There was no rainfall during

the drought-stress period in 2010/2011. Maximum temperature (T-max) and minimum temperature (T-min) did not differ much between two seasons. Mean air temperatures ranged from 18.92 to 30.25 °C and 19.38 to 31.15 °C during 2010/2011 and 2011/2012, respectively. The seasonal means of solar radiation were 13.23 Cal cm-3 and 19.18 Cal cm-

3 in the first and second season, respectively (Figures 1b and d). The relative humidity values ranged from 69 to 95% and 71 to 91%. Mean daily pan evaporation was 4.47 mm and 5.14 mm in the first and second season, respectively (Figures 1a and c).

Soil properties analyzed in 2010/2011 and 2011/2012 were described in Table 1. Physical properties were not much different between two years. Soil pH (1:1 H2O) in 2010/2011 and 2011/2012 were 6.08 and 6.18, respectively. Soil analysis data in the 2 years showed a range of 0.018-0.021% for total N, 23.95-40.74 ppm for available P, 33.09-38.34 ppm for exchangeable K, and 418.33-446.67 ppm for C, clearly indicating that N was inadequate (Marschner, 1995). Total N in the reference genotype (non-nodulating line) was lower than those in other genotypes (data not shown). There were differences in growth of non-nodulating and nodulating genotypes. Pimratch et al. (2008) explained that such differences in growth of the non-nodulating genotype and nodulating genotypes indicated that the uptake of soil N was not sufficient, and growth of the test peanut lines depended largely on fixed N2. Soil moisture percentages at soil depths of 30 cm and 60 cm at sowing, the last day of irrigation, R7 growth stage, and harvest time under well-watered and terminal drought conditions are presented in Table 2. Soil moisture contents at 30 cm were consistent under well-watered conditions and ranged from 9.74% to 10.51% and 10.84% to 10.99% in 2010/2011 and 2011/2012, respectively. But soil moisture contents measured at 60 cm in the 2011/2012 experiment were closer to the desired soil moisture contents at each data collection compared with the 2010/2011 experiment.

Table 1. Physical and chemical properties of the soil at the 0-30 cm depth in the experimental fields, 2010/2011 and 2011/2012. Property 2010/2011 2011/2012 Sand (%) 85.08 83.98 Silt (%) 7.30 8.32 Clay (%) 7.62 7.70 pH (1:1 H20) 6.08 6.18 EC (1:5 H20) (dS m-1) 0.03 0.05 CEC (c mol kg-1) 5.22 5.93 OM (%) 0.44 0.41 Total N (mg kg-1) 185.00 203.00 Available P (mg kg-1) 23.95 40.74 Exchangeable K (mg kg-1) 33.09 38.34 Exchangeable Ca (mg kg-1) 418.33 446.67 Table 2. Soil moisture content (%) at sowing, the last day of irrigation, R7 growth stage and harvest at 0-30 cm and 0-60 cm under well-watered and terminal drought conditions in 2010/2011 and 2011/2012.

Year Treatment Soil Soil moisture content (%)

depth (cm) Sowing Last day of irrigation R7 stage Harvest

2010/2011 Well-watered 0-30 10.51 9.91 9.74 10.25 Terminal drought 0-30 10.24 9.87 6.16 6.21 Well-watered 0-60 10.13 8.77 8.64 9.61 Terminal drought 0-60 10.26 8.47 5.91 6.43 2011/2012 Well-watered 0-30 10.88 10.84 10.99 10.92 Terminal drought 0-30 10.52 10.54 6.11 6.09 Well-watered 0-60 10.81 10.63 10.83 10.76 Terminal drought 0-60 10.42 10.42 6.12 6.44 2010/11; FC = 10.14%, PWP = 4.47, 1/3 AW= 6.33 using pressure plate method. 2011/12: FC = 10.18%, PWP = 4.50, 1/3 AW= 6.37 using pressure plate method. *FC=field capacity, PWP=permanent wilting point, AW=available water. well-watered = full irrigation since sowing until final harvest. Terminal drought = 1/3 AW at R7 until final harvest. Table 3. Mean square from the combined analyses of variance for fixed N2 (NF) and nodule dry weight (NDW) of five peanut genotypes at final harvest grown under well-watered and terminal drought in the dry seasons of 2010-2011 and 2011-2012. Source of variation df NF NDW Year (Y) 1 121680** 0.0180** Reps within year (Y*R) 6 7046 0.0011 Water regimes (W) 1 2386353** 0.9176** Y*W 1 6611 0.0018 Error Y*R*W 6 11351 0.0034 Genotypes (G) 4 371520** 0.2981** Y*G 4 8121 0.0013 W*G 4 439822** 0.0146** Y*W*G 4 4229 0.0022 Error Y*R*W*G 48 4209 0.0011 *, ** = significant at P ≤ 0.05 and P ≤ 0.01, respectively. df = degrees of freedom. Well-watered = full irrigation since sowing

until final harvest. Terminal drought = 1/3 AW at R7 until final harvest. .

Figure 1. Rainfall, relative humidity (RH), evaporation (E0), maximum (Tmax) and minimum (Tmin) temperature and solar radiation during October-January 2010/2011 (a, b) and 2011/2012 (c, d) at the meteorological station, Khon Kaen University, Thailand.

80

85

90

95

100

Last day irrigation R7 stage Harvest

Well-watered

Terminal droughtRW

C (%

)

80

85

90

95

100


Well-watered

Terminal drought

RW

C (%

)

(b)

Figure 2. Relative water content (RWC) on the last day of irrigation, R7 stage, and harvest stage of five peanut genotypes grown under well-watered and terminal drought in (a) 2010/2011 and (b) 2011/2012. *Well-watered = full irrigation since sowing until final harvest. Terminal drought = 1/3 AW at R7 until final harvest.

(a) (b)

(c) (d)

(a) (b)


303

Table 4. Effects of terminal drought stress on fixed N2 (NF) and nodule dry weight (NDW) and their reduction (%) of five peanut genotypes at final harvest grown under well-watered and terminal drought conditions, 2010-2011 and 2011-2012.

Genotype NF (g N plant-1) NDW (g plant-1)

Well-watered Drought Reduction (%) Well-watered Drought Reduction

(%) ICGV 98308 1. 22 a 0.51 b 58 b 0.84 a 0.58 a 31 b ICGV 98324 0.82 b 0.71 a 13 cd 0.66 b 0.48 b 27 b ICGV 98348 0.94 b 0.24 c 74 a 0.59 c 0.36 c 39 ab Tainan 9 0.55 c 0.50 b 9 d 0.55 c 0.38 c 31 b Tifton 8 0.60 c 0.45 b 25c 0.45 d 0.23 d 49 a Mean 0.83 0.48 42 0.62 0.41 34 Different letters adjacent to data in the same column show significance at P ≤ 0.05 by Duncan’s multiple range test. Well-watered = full irrigation since sowing until final harvest. Terminal drought = 1/3 AW at R7 until final harvest. Table 5. Correlation coefficients between fixed N2 (NF), nodule dry weight (NDW), biomass (BM) and pod yield (PY) at well-watered and terminal drought treatments, 2010-2011 and 2011-2012. Trait BM PY

Well-watered condition NF 0.80** 0.57* NDW 0.79** 0.51* Terminal drought NF 0.47* 0.10 NDW 0.61** 0.22 *, ** = significant at P ≤ 0.05 and P ≤ 0.01, respectively. Well-watered = full irrigation since sowing until final harvest. Terminal

drought = 1/3 AW at R7 until final harvest.

Htoon et al. (2013)

304

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

R5 R7 Harvest

NDW

g p

lant

-1ICGV 98308 Well water

Drought

Growth stage

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

R5 R7 Harvest

NDW

g p

lant

-1

ICGV 98324 Well waterDrought

Growth stage

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

R5 R7 Harvest

NDW

g p

lant

-1

ICGV 98348 Well waterDrought

Growth stage

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

R5 R7 Harvest

NDW

g p

lant

-1

Tainan 9 Well waterDrought

Growth stage

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

R5 R7 Harvest

NDW

g p

lant

-1

Tifton 8 Well waterDrought

Growth stage

Figure 3. Nodule dry weight (NDW) g plant-1. (a) ICGV 98308, (b) ICGV 98324, (c) ICGV 98348, (d) Tainan 9 and (e) Tifton 8 at R5 stage, R7 stage and harvest under well-watered and terminal drought conditions. (Bars represent ± SE, well-watered = full irrigation since sowing until final harvest, terminal drought = 1/3 AW at R7 until final harvest.)

(a) (b)

(c) (d)

(e)

The effect of drought on peanut could also be seen from the RWC values of peanut leaves under well-watered and terminal drought conditions on the last day of irrigation, at R7, and at harvest (Figure 2). The same RWC values were observed on the last day of irrigation in both years but they were significantly lower in the stressed treatment than the well-watered treatment at R7 and harvest in both years. The results indicated that RWC was decreased by terminal drought treatment. The highest RWC was observed at R7 under well-watered treatment in both years, followed by RWC at well-watered treatment on the last day of irrigation and at harvest. Plants under terminal drought stress wilted showing water deficit symptoms a few days after withholding water. They than showed severe wilting during the treatment period but well-watered plant were just normal. Reduction of NF and NDW due to water stress effects of water stress on N2-fixation (NF), nodule dry weight (NDW) and their reduction The combined analysis of variance showed that the differences between water regimes (W) for fixed nitrogen (NF) and nodule dry weight of NF responses to terminal drought (NDW) were significant (P ≤ 0.01). Comparisons of NF responses to terminal drought across genotypes have shown variations, indicating a genetic control for this trait (P ≤ 0.01) (Table 3). The year effects (Y) were also significant for both NF and NDW (P ≤ 0.01), but they were relatively low for these traits compared to W and G main effects. The W x G interaction effects were significant for both traits (P ≤ 0.01).

Fixed nitrogen (NF) of 5 peanut genotypes at harvest, which were grown under well-watered and terminal drought conditions, are presented in Table 4. The results clearly showed that terminal drought reduced NF in every genotype. ICGV 98324 was the best genotype it could maintain high NF even when it was subjected to terminal drought and NF, reduction was relatively low (13%). The NF of ICGV 98324 at the well-watered and terminal drought were 0.82 g N plant-1 and 0.71 g N plant-1, respectively. Pimratch et al. (2008)

studied the response of NF to prolonged drought and reported that ICGV 98324 could not maintain high NF under prolonged drought. Tainan 9 also had the lowest reduction for NF (9%) but it was not the best genotype because it had low NF (0.55 g N plant-1) even under normal condition. In a recent research, Tainan 9 also performed poorly in terms of total biomass and pod yield under terminal drought and had the highest reduction in total biomass (Girdthai et al., 2010). Tainan 9 was designated to be a drought-sensitive genotype. NF of Tifton 8 at well-watered condition was 0.60 g N plant-1 and it decreased to 0.45 g N plant-1 under terminal drought. The NF of Tifton 8 was lower than these other varieties. The result supported Pimratch et al. (2008), who observed that Tifton 8 had low NF with high reduction under water stress. Interestingly, drought-tolerant genotype ICGV 98308 had 58% reduction of NF under terminal drought, although it was the highest (1.22 g N plant-1) for this trait under well-watered condition. ICGV 98348 also showed high NF reduction 74%, indicating that terminal drought seriously reduced NF from 0.94 g N plant-1 into 0.24 g N plant-1. It is evident from the literature that NF was affected by early season drought (Htoon et al., 2009 and Puangbut et al., 2011) and prolonged drought (Pimratch et al., 2008) and the responses of peanut genotype to drought were different. Our result was in agreement with the previous finding because in our study, peanut genotypes had varying, NF in responses to terminal drought. In fact, estimates of NF in peanut suggest that symbiosis could provide between 100 and 200 kg N ha-1 under a range of different field conditions (Giller et al., 1987 and Boddey et al., 1990), although both the quantity of fixed and the proportion of total crop N derived from NF can be influenced by factors such as cultivar and water stress (Giller et al., 1987; Peoples et al., 1992; Pimratch et al., 2008 and Htoon et al., 2009). Pimratch et al. (2008) reported that in peanut, NF was decreased by 67.2% 1/3 available water. Thus, differences in NF between genotypes closely reflected differences in total N accumulation. Nambiar and Dart (1983) revealed that in peanut genotype Kadiri 71-1, NF attained a peak during the pod-filling stage and declined at maturity. They reported that the pattern of NF

Htoon et al. (2013)

306

in the post-rainy season was quite different, increasing until about 75 days after planting, and then decreasing rapidly, with differences developing between cultivars. Kadiri 71-1 was a cultivar with late maturity and so other cultivars with early maturity might attain an earlier peak in NF. In our experiment, irrigation was withheld since R5 stage of five peanut genotypes, ranging from 50 to 64 days after planting (DAP). It was clear from our result that, peanut plants subjected to terminal drought, even though the plants were almost at the peak of NF, were affected by low soil moisture content and eventually, biological NF was reduced. As soon as soil moisture decreased, NF was reduced immediately because NF rates are more sensitive to reductions in soil water content than are other processes such as photosynthesis, transpiration, leaf growth rates or nitrate assimilation (Serraj et al., 1999). Furthermore, the negative effect of drought on NF is the sum of three different responses: effects on the interaction of legumes by rhizobia, effects on nodule growth and development, and finally, direct effects on nodule function (Arrese-Igor et al., 2011). High N2-fixing genotypes usually have more nodules and greater weight (Phillips et al., 1989 and Toomsan et al., 1991). In our experiment, NDW of five peanut genotypes and their reductions were recorded at well-watered and terminal drought (Table 4). NDW of all genotypes were reduced by terminal drought. Drought-resistant line ICGV 98308 had the highest NDW (0.84 g plant-1) under well-watered condition but NF decreased reduced to 0.58 g N plant-1 with 31% reduction. ICGV 98308 also showed the best performance for NDW (0.58 g plant-1) when this genotype was subjected to terminal drought. ICGV 98324 also had high NDW under both well-watered (0.66 g plant-1) and terminal drought conditions (0.48 g plant-1). Although such traits of ICGV 98324 decreased under terminal drought, this genotype still ranked high compared with ICGV 98348, Tainan 9 and Tifton 8. ICGV 98348 had medium NDW under both treatments. Tainan 9 and Tifton 8 showed the lowest NDW not only under well-watered but also under terminal drought conditions. Our results were not in accordance with Venkateswarlu et al. (1989) who reported

that NDW did not decrease until the stress was severe (-2.3 MPa) and nodule number also remained constant in peanut with increasing water stress. In contrast, Pimratch et al. (2008) observed that NDW decreased under prolonged drought with reductions at 90 DAE (34.2% for 2/3 AW and 56.1% for 1/3 AW) and at 60 DAE (21.0% for 2/3 AW and 42.6% for 1/3 AW). But the reduction of NDW at 90 DAE was higher than that at 60 DAE. Similar results were reported by Htoon et al. (2009) who studied the response of peanut genotypes to NDW under pre-flowering drought. They reported that NDW decreased in some peanut genotypes when plants were subjected to pre-flowering drought. The data for NDW were collected three times, at R5 (the day irrigation was withdrawn), R7 (terminal drought), and harvest. Figure 3 clearly showed that NDW of each genotype at 3 stages were different. On the last day of irrigation, the NDW values of each genotype under well-watered and terminal drought conditions were almost the same because water was applied to plants equally. Nodule growth still continued until R7 under both well-watered and drought but growth rates were significantly different for every genotype between the 2 treatments, except for ICGV 98324. It had the same NDW under both conditions Figure 3b. As soon the NDW as of the plants were subjected to terminal drought, every genotype decreased NDW. But under well-watered condition, nodule growth still went on, and the highest NDW was observed at final harvest, except ICGV 98348 which showed lower NDW than NDW at R7. Our results confirmed those of Nambiar and Dart (1983) who said that nodule weight could increase until harvest under normal condition. Similarly, Wahhab and Bhuiya (1984) stated that the establishment and formation of nodules in peanut are continuous processes and, under well-watered conditions, the most nodules and the most effective nodules per plant occurred 100 days after sowing. NDW could be increased because of continuous growth of nodule from R7 to harvest, but it was found that terminal drought apparently significantly reduced NDW, which is a trait closely related to NF (Pimratch et al., 2008 and Htoon et al., 2009).


307

When irrigation was withheld at R5, five peanut genotypes in this study continued their nodule growth until R7 but NF might be decreased soon after soil water depletion due to higher sensitivity of NF to water deficit rather than other mechanisms in the plant (Serraj et al., 1999). From R7 to harvest, NDW decreased because severe terminal drought inhibited nodule growth and, in addition, prolonged drought promoted nodule decay (Mulongoy, 1992). Furthermore, it should be noted that the size of nodule did not have any effect on NF and NDW. For instance, the size of nodule of Tifton 8 was the biggest under both water regimes, but when NF and NDW were recorded, they became the lowest, whereas ICGV 98308, which had the smallest nodules showed the highest NF and NDW (Table 4). Our results also confirmed the finding of Hungria and Bohrer (2000) who also explained that some strains of Rhizobium can fix higher N even with smaller nodule number and weight. Correlation between fixed nitrogen (NF), nodule dry weight (NDW), biomass production (BM) and pod yield (PY) Significant correlations between NF and BM were found at both well-watered condition (r = 0.80, P = 0.01) and terminal drought (r = 0.47, P = 0.05) (Table 5). NDW, which is characterized as a trait related to N2-fixation, was significantly correlated to BM (r = 0.79, P = 0.01) under well-watered condition and terminal drought (r = 0.61, P = 0.01). The results clearly indicated that NF is very important for BM under every condition. Previous research has reported that positive relationships between NF, NDW and BM production were found at both levels of drought stress, and the relationship was stronger the more severe the drought stress is (Pimratch et al., 2008).

The correlation coefficient between NF and PY was highly significant only at well-watered condition (r = 0.57, P = 0.01) but it was not significant under terminal drought (r =0.10). The significant correlation (r = 0.51, P = 0.05) was observed between NDW and PY only under well-watered condition. Htoon et al. (2009) reported that NDW was highly correlated to BM

rather than to PY under pre-flowering drought. Our results supported an assumption that the greatest source of N for peanut is symbiotic fixed N2 and a large quantity of N was partitioned to the peanut seeds: 62% to 76% of total accumulated N was located in the seeds (Van Rossum et al., 1993). Even kernel N might directly be derived from N (Williams et al., 1990). Nitrogen fixation in peanuts should thus have a major influence on pod yield. But our results imply that when peanut genotypes were subjected to terminal drought condition, PY was not supported by NF. This may be due to the decreasing NF and their importance in plant growth (BM), especially under severe water stress. Additional evidence was found in the research work of Venkateswarlu et al. (1990) who reported that water stress at the pod-filling stage reduced kernel yield and N partitioning to reproductive parts. In general, the amount of NF has exceeded the N requirements for pod growth; thus, residual fixed N2 was present in vegetative parts to be contributed to soil after pod harvest (Dakora et al., 1987 and McDonagh et al., 1993). But under severe water stress, NF was decreased and this NF was very important for plant growth because of the inability of the plant to use soil N effectively, arising from the difficulty of roots to mine soil N in drying soil (Pimratch et al., 2008). The current study clearly indicated that NF highly contributed to BM production and PY under well-watered condition but contribution to BM was higher. However, under the terminal drought condition, the decreasing NF with increasing drought intensity might be able to support BM only. In this way, drought stress occurring during the seed-filling phase or otherwise, terminal drought has been observed to cause the greatest reduction in pod yield (Ravindra et al., 1990 and Girdthai et al., 2010). Therefore, reduction of NF and NDW and their lack of contribution to PY might be one of the reasons for the low PY under terminal drought.

Htoon et al. (2013)

308

CONCLUSIONS The intensity and length of drought and their effects on peanut productivity, including biological NF have been studied to formulate an efficient breeding program under drought condition. Terminal drought during pod-filling stage until harvest greatly reduced the NF and NDW in peanut but peanut genotypes did differ in NF and NDW under terminal drought condition. Drought-tolerant genotypes ICGV 98308 and ICGV 98324 should be selected for high NF and NDW across water regimes. During terminal drought period, fixed N2 contributed to biological yields only and there was no contribution to economic yields. Since the correlation coefficients of NF to economic yields were low under terminal drought, the ability to fix high N2 under terminal drought stress could aid peanut genotypes in maintaining high yield under water-limited conditions. ACKNOWLEDGEMENTS The authors are grateful for the financial support of the Higher Education Research Promotion and National Research University Project of Thailand, Office of Higher Education Commission, through the Food and Functional Food Research Cluster of Khon Kaen University, Thailand, the Peanut and Jerusalem Artichoke Improvement for Functional Food Research Group and the Plant Breeding Research Center for Sustainable Agriculture. Grateful acknowledgement is made to the Thailand Research Fund, the Commission for High Education and Khon Kaen University for providing financial support to this research through the Distinguished Research Professor Grant to Professor Dr. Aran Patanothai. This work was also partially supported by Peanut CRSP NMS 72 project by USDA through the University of Georgia ECG-A-00-0700001-00. Peanut genotypes for this study were kindly provided by ICRISAT. We also appreciate the contribution of many people who participated in field data collection and sample preparation and processing. REFERENCES Araus JL, Slafer GA, Reynolds MP, Royo C (2002).

Plant breeding and drought in C3 cereals: what should we breed for? Ann. Bot. 89: 925-940.

Arrese-Igor C, Conzalez EM, Marino D, Landrera R, Larrainzar E, Gil-Quintana E (2011).

Pyhsiological responses of legume nodules to drought. Plant Stress 5: 24-31.

Bell MJ, Wright GC, Suryantini, Peoples MB (1994). The N2-fixing capacity of peanut cultivars with differing assimilate partitioning characteristics. Aust. J. Agric. Res. 45: 1455-1468.

Boddey RM, Urquiaga S, Neves MCP (1990). Quantification of the contribution of N2 fixation to field grown grain legumes-a strategy for the practical application of the 15N isotope dilution technique. Soil Biol. Biochem. 22: 649-655.

Boote KJ (1982). Growth stage of peanut (Arachis hypogaea L.). Peanut Sci. 9: 35-40.

Bricker, A.A. 1989. MSTAT-C user’s guide. Michigan State University, East Lansing, Michigan.

Clavel D, Diouf O, Khalfaoui JL, Braconnier S (2006). Genotype variations in fluorescence parameters among closely related groundnut (Arachis hypogaea L.) lines and their potential for drought screening programs. Crop Sci. 34: 92-97.

Dakora FD, Aboyinga RA, Mahama Y, Apaseku J (1987). Assessment of N2 fixation in groundnut (Arachis hypogaea L.) and cowpea (Vigna unguiculata L. Walp.) and their relative N contribution to a succeeding maize crop in northern Ghana. MIRCEN J. 3: 389-399.

DeSilva M, Purcell LC, King CA (1996). Soybean petiole ureide response to water deficits and decreased transpiration. Crop Sci. 36: 611-616.

DeVries JD, Bennett JM, Albrecht SL, Boote KJ (1989). Water relations, nitrogenase activity and root development of three grain legumes in response to soil water deficits. Field Crops Res. 22: 33-44.

Doorenbos, J., W.O. Pruitt. 1992. Calculation of crop water requirements. Irrigation and Drainage Paper. No. 24. FAO, of the United Nations, Rome, Italy. pp. l-65.

FAOSTAT. 2008. Statistical databases. (cited on Nov. 25, 2008) Available at URL: http://faostat.fao.org.

Giller KE (2001). Nitrogen fixation in tropical cropping systems. 2nd ed. CAB International, Wallingford. 423 p.

Giller KE, Nambiar PTC, Srinivasa Rao B, Dart PJ, Boote KJ (1987). A comparison of nitrogen fixation in genotypes of groundnut (Arachis hypogaea L.) using the 15N isotope dilution. Biol. Fert. Soils 5: 23-25.


309

Girdthai T, Jogloy S, Vorasoot N, Akkasaeng C, Wongkaew S, Holbrook CC, Patanothai A (2010). Associations between physiological traits for drought tolerance. Plant Breed. 129: 693-699.

Gomez KA, Gomez AA (1984). Statistical procedures for agricultural research. 2nd ed. John Wiley and Sons. New York, USA. 335 p.

Guerin V, Trinchant JC, Rigaud J (1990). Nitrogen fixation by broad bean (Vicia faba L.): nodules and bacteroids under water-restricted conditions. Plant Physiol. 92: 595-601.

Hungria M, Bohrer TRJ (2000). Variability of nodulation and dinitrogen fixation capacity among soybean cultivars. Biol. Fert. Soils 31: 45-52.

Htoon W, Jogloy S, Toomsan B, Sanitchon J (2009). Response to early drought for traits related to nitrogen fixation and their correlation to yield and drought tolerance traits in peanut (Arachis hypogaea L.). Asian J. Plant Sci. 8: 138-145.

Johensen C, Baldev B, Brouwer JB, Erskine W, Jermyn WA, Lang LJ, Malik BA, Miah AA, Silim SN (1994). Biotic and abiotic stress containing productivity of cool season legume in Asia, Africa and Oceania, In: F.J. Muehlbauer and W.J. Kaiser, eds., Expanding the production and use of cool season food legume. Kluwer, Academic Publishers, Dordrecht, the Netherlands. pp. 175-194.

Kramer PJ (1980). Drought stress and the origin of adaptation. In: N.C. Turner and P.J. Kramer, ed. Adaptation of plant to water and high temperature stress, John Wiley & Sons, New York, pp. 7-20

Marschner H (1995). Mineral nutrition of higher plants. 2nd ed. Academic Press, London, 889 p.

McDonagh JF, Toomsan B, Limpinantana V, Giller KE (1993). Estimates of the residual nitrogen benefit of groundnut to maize in northeast Thailand. Plant Soil 154: 267-277.

Mulongoy K (1992). Technical paper 2. Biological nitrogen fixation, In: B.R. Tripathi and P.J. Psychas, eds., The AFNETA Alley Farming Training Manual Volume. Alley Farming Network for Tropical Africa. International Institute of Tropical Agriculture, Ibadan.

Nageswara Rao RC, Williams JH, Sivakumar MVK, Wadia KDR (1988). Effect of water deficit at different growth phase of peanut. II Response to drought during preflowering phase. Agron. J. 80: 431-438.

Nageswara Rao RC, Wright GC (1994). Stability of the relationship between specific leaf area and carbon isotope discrimination across environments in peanut. Crop Sci. 34: 98-103.

Nambiar PTC, Dart PJ (1983). Factors influencing nitrogenase activity (acetylene reduction) by root nodules of groundnut, Arachis hypogaea L. Peanut Sci. 10: 26-29.

Nautiyal PC, Ravinda V, Zala PV, Joshi YC (1999). Enhancement of yield in peanut following the imposition of transient soil-moisture-deficit stress during the vegetative phase. Exp. Agric. 35: 371-385.

Pallas JE Jr, Stansell JR, Bruce RR (1977). Peanut seed germination as related to soil water regime during pod development. Agron. J. 69: 381-383.

Peoples MB, Bell VJ, Bushby HVA (1992). Effect of rotation and inoculation with Bradyrhizobium on nitrogen fixation and yield of peanut (Arachis hypogaea L., cv. Virginia Bunch). Aust. J. Agric. Res. 43: 595-607.

Phillips TD, Wynne JC, Elkan GH, Schneeweis TJ (1989). Field evaluations of peanut cultivar Bradyrhizobium specificities. Peanut Sci. 16: 54-57.

Pimratch S, Jogloy S, Vorasoot N, Toomsan B, Kesmala T, Patanothai A, Holbrook CC (2008). Effect of drought stress on traits related to N2-fixation in eleven peanut (Arachis hypogaea L.) genotypes differing in degrees of resistance to drought. Asian J. Plant Sci. 7: 334-342.

Puangbut D, Jogloy S, Vorasoot N, Akkasaeng C, Kesmala T, Patanothai A (2009). Variability in yield response of peanut (Arachis hypogaea L.) genotypes under early season drought. Asian J. Plant Sci. 8: 254-264.

Puangbut D, Jogloy S, Vorasoot N, Akkasaeng C, A Patanothai (2011). Association of transpiration efficiency with N2 fixation of peanut under early season drought. Int. J. Plant Prod. 5: 381-394.

Ravindra V, Nautiyal PC, Joshi YG (1990). Physiological analysis of drought resistance and yield in groundnut (Arachis hypogaea L). Trop. Agric. Trinidad 67: 290-296.

Reddy TY, Reddy VR, Anbumozhi V (2003). Physiological responses of peanut (Arachis hypogaea L.) to drought stress and its amelioration: a critical review. Plant Growth Regul. 41: 75-88.

Schuman GE, Stanley MA, Knudson D (1973). Automated total nitrogen analysis of soil and plant samples. Soil Sci. Am. Proc. 37: 480-481.

Serraj R, Sinclair TR, Purcell LC (1999). Symbiotic N2 fixation response to drought. J. Exp. Bot. 50: 143-155.

Htoon et al. (2013)

310

Sinclair TR, Serraj R (1995). Dinitrogen fixation sensitivity to drought among grain legume species. Nature 378: 344.

Singh S., Russell MB (1981). Water use by maize/pigeonpea intercrop on a deep Vertisol. In: Proceedings of the International workshop on pigeonpeas. December 1980. ICRISAT Center Patancheru, AP, India. pp. 271-282.

Toomsan, B., V. Limpinuntana, J. Homchan, P. Sripa, P. Paopuree, M. Srichantawong, S. Manikan, S. Saranun, J. Sanitchon, S. Theerapongthanathorn, J.F. McDonagh, K.E. Giller (1991). Measurement of nitrogen fixation in some groundnut lines using N-15 isotope dilution technique. In: A. Patanothai, ed. Proceedings of the 10th Thailand National Groundnut Research Meeting. The Hinsuay Namsai Resort Hotel, Rayong, Thailand. Oct. 16-19. pp. 311-316. (in Thai with English Summary).

Van Rossum D, Muyotcha A, Verseveld HWV, Stouthamer AH, Boogerd FC (1993). Effects of Bradyrhizobium strain and host genotype, nodule dry weight and leaf area on groundnut (Arachis hypogaea L. ssp. fastigiata) yield. Plant Soil 154: 279-288.

Venkateswarlu B, Maheswari M, Saharan N (1989). Effects of water deficit on N2 (C2H2) fixation in cowpea and groundnut. Plant Soil 114: 69-74.

Venkateswarlu B, Saharan N, Maheswari M (1990). Nodulation and N2 (C2H2) fixation in cowpea

and groundnut during water stress and recovery. Field Crops Res. 25: 223-232.

Vorasoot N, Songsri P, Akkasaeng C, Jogloy S, Patanothai A (2003). Effect of water stress on yield and agronomic characters of peanut (Arachis hypogaea L.). Songklanakarin J. Sci. Technol. 25: 283-288.

Wahhab MA, Bhuiya ZH (1984). Abundance and distribution of nodules in root system of groundnut. Bangladesh J. Agric. Res. 9: 97-102.

Williams JH, Dutta M, Nambiar PTC (1990). Light interception as a source of variation for nitrogen fixation in groundnut genotypes. Plant Soil 121: 83-88.

Wright GC, Hubick KT, Farquhar GD (1991). Physiological analysis of peanut cultivar response to timing and duration of drought stress. Aust. J. Agric. Res. 42: 453-470.

Wright GC, Nageswara Rao RC (1994). Groundnut water relations. In: J. Smartt, ed. The groundnut crop. a scientific basis for improvement. Chapman and Hall: London. pp. 281-325.

Wright GC, Nageswara Rao RC, Basu MS (1996). A physiological approach to the understanding of genotype by environment interactions a case study on improvement of drought adaptation in groundnut. In: M. Cooper, and G.L. Hammer, eds. Plant adaptation and crop improvement. CAB International: Wallingford. pp. 365-380.


RELATIONSHIPS BETWEEN ROOT TRAITS AND NUTRIENT UPTAKE AND NITROGEN FIXATION IN PEANUT UNDER TERMINAL DROUGHT

W. HTOON1, W. KAEWPRADIT1*, S. JOGLOY1, N. VORASOOT1, B. TOOMSAN1,

C. AKKASAENG1, N. PUPPALA2 and A. PATANOTHAI1

1Department of Plant Science and Agricultural Resources, Faculty of Agriculture, Khon Kaen University, Muang, Khon Kaen 40002, Thailand

2Agricultural Science Center, New Mexico State University, Clovis, New Mexico 88101, USA. *Corresponding author’s email: [email protected]

SUMMARY Information on physiological traits that can be potentially use as surrogate traits for drought tolerance is very important to minimized yield reduction under drought. This study investigated the correlations between root traits and nutrient uptake and N2 fixation (NF) in peanut under terminal drought condition. Five peanut genotypes–ICGV 98308, ICGV 98324, ICGV 98348, Tainan 9 and Tifton 8–were tested under well-watered and terminal drought conditions at beginning maturity (R7) growth stage until final harvest. A split-plot experiment in randomized complete block design with four replicates was used. The data were recorded for nutrient uptake, NF, nodule dry weight (NDW) and root traits. Root surface area was significantly correlated with uptake of N, P, K and Mg but it did not correlate with uptake of Ca under terminal drought. Strong correlations between root volume, root length density (RLD), % RLD and uptake of N, P, K and Mg were observed under terminal drought. However, root dry weight (RDW) did not correlate with nutrient uptake under both environments but correlated with uptake of Ca significantly under terminal drought. Despite the deep RLD of peanut genotypes under terminal drought, the plants could not take up and contribute sufficient Ca. Only large root systems at 0-50 cm (about fruiting zone) might be able to accumulate Ca. A deeper root system could help maintain nodule growth and NF even under severe terminal drought condition. Deep root systems favor nutrient uptake under terminal drought because of strong correlations between these traits. Moreover, drought avoidance by enhancing root surface area, root volume, RLD and % RLD are the main mechanisms that help peanut to maintain nodule growth and NF under terminal drought condition. . Keywords: Plant nutrition, Rhizobium, soil water, correlation coefficient, end-season drought

Manuscript received: November 18, 2012; Decision on manuscript: February 6, 2013; February 20, 2013. © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Drought during the pod- and seed-forming stages has been shown to reduce pod yield of peanut (Arachis hypogaea L.) by 56-85% (Nageswara Rao et al., 1988). Up to 35% of pod yield and 21% of biomass were reduced under terminal drought (Girdthai et al., 2010).

Breeding for drought resistance has been an important strategy for alleviating yield reduction under drought condition. While direct selection for pod yield under drought conditions could be effective, the limitations of this approach were the large genotype x environment (G x E) interaction and low heritability of such trait resulting in slow breeding progress (Wright et

RESEARCH ARTICLE


312

al., 1996). Information on physiological traits for achieving higher yields under drought stress could not be known completely and gaps still exist.

Rooting depth, root distribution, root mass and root length density (RLD) have been identified as drought avoidance traits (Matsui and Singh, 2003; Taiz and Zeiger, 2006). According to Boote (1983), root growth rate of peanut was reduced by compact soil. When peanut plants encounter drought conditions, plants improve root distribution at the deeper soil layers (Wright and Nageswara Rao, 1994). Songsri et al. (2008) reported that peanut genotypes with higher RLD at lower soil depths enhanced drought tolerance and that higher RLD in deeper soil layers contributed to high pod yield and harvest index under long-term drought. Besides RLD, Rucker et al. (1995) revealed that some peanut genotypes with large root system (root dry weight) under nonstress conditions gave higher yield under drought conditions.

Since a plant obtains its water and mineral requirements from its roots, such an ability of the roots is important to gain high plant productivity. Root development is fundamentally involved in the response to many plant stresses; in particular, drought and mineral deficiency (Rucker et al., 1995; Maiti et al., 2002) and drought resistance might be enhanced by improvements in soil water extraction capability (Wright and Nageswara Rao, 1994). Like all other crops, the growth of peanut plants mostly relies on the supply of plant nutrients originating from the soil (Marschner, 1995). Fageria et al. (2002) mentioned that water stress affected root growth, nutrient mobility in the soil and nutrient uptake. Moreover, Kulkarni et al. (1988) observed some evidence that N, P and K uptake of peanut was reduced by drought stress. However, the mechanisms underlying the co-responses of root growth and nutrient uptake to drought stress have not been well understood. Especially, there is lack of information on the contribution of root traits to nutrient uptake during terminal drought conditions, which occurred at sensitive growth stages.

The peanut crop is capable of obtaining significant quantities of atmospheric nitrogen (N2) from its association with diverse strains of Rhizobium spp. or Bradyrhizobium spp. (Van-

Rossum et al., 1993). The greatest source of NF for peanut is symbiotic NF and it accounts for 50-80% of total N consumed by the plant (Boddey et al., 1990). Within root nodules, NF was done by the bacteria, and the NH3 produced was absorbed by the plant (Marschner, 1995; Giller, 2001). Water stress affected nodulation, nodule growth and weight, as well as N2-fixing activity in peanut (De-Vries et al., 1989; Venkateswarlu et al., 1989; Pimratch et al., 2008). Mulongoy (1992) mentioned that deep-rooted legumes exploiting moisture in the lower soil layers could continue N2 fixing when the soil is drying. In peanut, nodules form at the junctions of the lateral roots; besides being firmly attached to the root, they were close to the root vasculature. Nodules with deep-seated origin generally resisted stress more effectively (Sprent, 1981). Effective nodules are generally located with a higher density in tap roots than in lateral roots (Hungria and Bohrer, 2000). Therefore, both lateral and vertical root growth play an important role for nodule growth and NF in peanut. This is true especially when peanut plants are subjected to drought conditions; the responses of root traits and the trait variations among peanut genotypes might be involved in nodule development and maintaining NF. In addition, so far, previous reports have independently reported the effect of drought on root traits and NF, and there is still a need to investigate the relationships between root traits and NF in peanut.

Root traits can be used as selection criteria for drought tolerance (Ludlow and Muchow 1990; Matsui and Singh, 2003; Songsri et al., 2008). Similarly, Samarah et al. (2004) stated that nutrient uptake in soybean under drought stress might have an important role in drought tolerance mechanisms. In addition, NF and related traits could be used as drought tolerance traits (Pimratch et al., 2008; Htoon et al., 2009). In peanut, so far, information on several physiological traits that contribute to crop productivity under drought and their correlations are available in the literature. Unfortunately, the relationships between root traits, nutrient uptake and NF under both well-watered and drought stress are still in question. Whether or not root responses contribute to higher nutrient uptake and NF under terminal

Htoon et al. (2013)

313

drought stress needs to be established in peanut because this period is crucial in achieving high productivity. A better understanding of the mechanisms of plant adaptation to drought is important to improve drought resistance in peanut. Therefore, the objective of this study was to investigate the relationship between root traits and nutrient uptake and NF under terminal drought. MATERIALS AND METHODS Experimental design and plant materials The 2-year field experiments were conducted on loamy-sand, Oxic Paleustults (Yasothon soil series) at the Field Crop Research Station of Khon Kaen University, Thailand, from October 2010 to January 2011 and from October 2011 to January 2012. Five peanut genotypes–ICGV 98308, ICGV 98324, ICGV 98348, Tainan 9 and Tifton 8–were used based on previously measured differences in drought tolerance index (DTI) for pod yield and biomass (Girdthai et al., 2010). The ICGV lines are from the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) and they were identified as drought-resistant with different drought tolerance levels. ICGV 98324 had high DTI for total biomass in both years, whereas ICGV 98348 also had high DTI for pod yield across years. ICGV 98308 had low DTI for pod yield (Girdthai et al., 2010). Tainan 9 performed poorly in terms of total biomass and pod yield under terminal drought and had the highest reduction in total biomass. Tifton 8 exhibited the highest biomass production under terminal drought but reduction in pod yield of Tifton 8 was also high (low DTI) (Vorasoot et al., 2003; Girdthai et al., 2010). Moreover, a non-nodulating line which is used as a check variety for NF was also grown in each replication. This genotype was only used to estimate the fixed N2 in each genotype at each treatment but it was not included when the data were subjected to analysis of variance.

Each genotype was grown under two soil moisture conditions: well-watered condition and terminal drought [1/3 available water (1/3 AW)] at R7 growth stage (Boote, 1982) to final

harvest. The water-stressed period in this experiment was defined as terminal drought (Girdthai et al., 2010). A split-plot in a randomized complete block design with four

The soil was disk-plowed thrice, leveled and harrowed. During land preparation, lime (CaCO3) was incorporated at a rate of 625 kg ha-

1. A basal dose of P and K fertilizer was incorporated into the topsoil at the rate of 24.7 kg P ha-1 and 31.1 kg K ha-1, respectively, shortly before planting. In order to increase the Rhizobium population in the soil, inoculation of Rhizobium was accomplished by applying a water-diluted commercial peat-based inoculum of Bradyrhizobium (mixture of strains THA 201 and THA 205; Department of Agriculture, Ministry of Agriculture and Cooperatives, Bangkok, Thailand) at the rate of 13 g kg-1 seed on rows of peanut seeds soon after planting. Gypsum (CaSO4) at the rate of 312 kg ha-1 was applied 40 days after emergence (DAE). Irrigation A subsurface drip irrigation system was installed with a spacing of 50 cm midway between peanut plant rows to supply water. Drip lines with a distance of 20 cm between emitters were installed at 10 cm below the soil surface and fitted with a pressure valve and water meter to make sure that water supply is efficient and uniform across each plot. For the well-watered treatment, water was applied daily from planting to harvest. However, the growth stage of plants was checked regularly in water-stressed plots in order to determine the right time to stop irrigation. In general, the drought treatment was initiated at 15-20 days before R7 growth stage and the times of irrigation withdrawn were different within genotypes according to growth stages of peanut genotypes and reliable data from preliminary trials and simulated data of Girdthai et al. (2010). At the terminal drought plots, irrigation was withheld and soil moisture was allowed to decrease gradually to meet the predetermined drought condition (1/3 AW) at R7 growth stage of each genotype. And from then on, the drought condition was maintained until harvest.

To maintain specific water regimes, the amount of water applied was based on crop


314

water requirements using the Doorenbos and Pruitt (1992) methodology, along with water loss data from surface evaporation described by Singh and Russell (1981). Meteorological conditions, soil properties and soil moisture Relative humidity (%), water evaporation (mm), rainfall (mm), maximum and minimum air temperature (ºC), and solar radiation (Cal cm-2) were recorded daily from sowing until harvest by a weather station located 100 m away from the experimental field. Physical and chemical properties of soil covering all treatments and replications were determined before planting. In addition, soil moisture content, field capacity (FC) and permanent wilting point (PWP) also were determined with the pressure plate method. Soil moisture was also measured by gravimetric method at depths of 0-5, 10-15, 25-30, 40-45, 55-60, 70-75 and 85-90 cm at planting, on the day irrigation withdrawn, at R7 growth stage and at harvest. Plant water status Relative water content (RWC) was recorded to evaluate plant water status and it was measured on the day irrigation was withdrawn, at R7 growth stage and at harvest following Kramer (1980). It was measured in one leaflet of the second fully expanded leaf from the top of the main stem of five plants for each plot at 10:00-12:00 am (Clavel et al., 2006; Girdthai et al., 2010). RWC was determined as follows: RWC (%) = [(FW-DW) / (TW-DW)] x 100, where FW = sample field weight, TW = sample turgid weight (saturated weight) and DW = sample dry weight. Plant nutrient analysis At their full maturity (R8), for each plot excluding boarder plots, plants in an area of 8 m2 were harvested and partitioned into leaf, stem, pods and roots prior to drying (80 ºC for 48 h) or until the dry weight became constant. Dried leaf, stem, shell and seed were then ground using a

hammer mill. In this case, dry leaf and stem samples were proportionally taken according to leaf and stem dry weight ratio. Then, individual nutrients were analyzed for each shoot, shell and seed. Nitrogen contents were measured using the automated indophenol method (Schuman et al., 1973; Pimratch et al., 2008) and read on a flow injection analyzer model 5012 (Tecator Inc., Hoganas, Sweden).

Concentration of P was determined by a spectrophotometor and K concentration was measured by a flame photometer (Kaewpradit et al., 2009). Concentrations of Ca and Mg were determined by atomic absorption spectroscopy (Al-Karaki and Al-Raddad, 1997). Finally, nutrient uptake (g plant-1) was calculated individually by multiplying dry weight and nutrient concentration. Nodule dry weight and nitrogen fixation At their full maturity (R8), for each plot excluding boarder plots, five plants were uprooted and soil was gently removed from the roots by washing them on a 0.5 mm screen. Nodules were then removed from each root by hand. These nodules were oven-dried at 80 ºC for 48 h or until the dry weight became constant. Nodule dry weight was recorded for each genotype.

Fixed N2 was determined at harvest by the same reliable N-difference method following McDonagh et al. (1993), Bell et al. (1994) and Pimtrach et al. (2008). Fixed nitrogen (NF) content was calculated as follows:

NF of genotype = total N of genotype – total N of non-nodulating line at the main plot Root traits Root length density (RLD) was measured at final harvest (R8) using an auger. In this method, soil samples were taken by a sampling core sampler tube 76 mm in diameter and 1.15 m long (Welbank et al., 1974). For each genotype, root samples were collected from two positions–i.e. center of the plant and between two rows at a distance of 25 cm from each plant. Plant parts above the ground were removed prior to root sample collection. Root samples were taken

Htoon et al. (2013)

315

from a depth of 90 cm and separated into six layers consisting of 0-15, 15-30, 30-45, 45-60, 60-75 and 75-90 cm. The samples were individually kept in plastic bags and stored inside a cold room. Root samples of each layer were washed manually with tap water to remove soil from the roots. The root samples were then scanned and analyzed with the Winrhizo program (Winrhizo Pro (s) V. 2004a, Regent Instruments, Inc.) to determine total root length per sample, root surface area (cm2), root diameter (mm) and root volume (cm3). RLD was calculated as the ratio of root length (cm) to soil volume (cm3). RLD from the first (0-15 cm) and second (15-30 cm) layers were combined and

defined as a single 0-30 cm layer or upper soil layer, while the RLD for the deeper layers (third to sixth) were combined to form a single 30-90 cm layer or lower soil layer (Jongrungklang et al., 2011). The percentage of RLD in the 30-90 cm layer (% RLD30-90cm) was calculated as: %RLD30-90cm = [RLD30-90cm / (RLD30-90cm + RLD0-30cm )] x 100

Finally, the average between two positions (plant center and between rows) were calculated for all root traits abovementioned.

Table 1. Soil moisture content (%) at sowing, on the last day of irrigation, at R7 growth stage and at harvest at 0-30 cm and 0-60 cm under well-watered and terminal drought conditions during 2010/2011 and 2011/2012. Year Treatments Soil Soil moisture content (%) depth

(cm) Sowing Last day of

irrigation R7 stage Harvest

2010/2011 Well-watered 0-30 10.51 9.91 9.74 10.25 Terminal drought 0-30 10.24 9.87 6.16 6.21 Well-watered 0-60 10.13 8.77 8.64 9.61 Terminal drought 0-60 10.26 8.47 5.91 6.43 2011/2012 Well-watered 0-30 10.88 10.84 10.99 10.92 Terminal drought 0-30 10.52 10.54 6.11 6.09 Well-watered 0-60 10.81 10.63 10.83 10.76 Terminal drought 0-60 10.42 10.42 6.12 6.44 2010/11; FC = 10.14%, PWP = 4.47, 1/3 AW= 6.33 using pressure plate method. 2011/12; FC = 10.18%, PWP = 4.50, 1/3 AW= 6.37 using pressure plate method. *FC: field capacity, PWP: permanent wilting point, AW: available water. Well-watered = full irrigation since sowing until final harvest. Terminal drought = 1/3 available water (AW) at R7 until final harvest.


316

Figure 1. Rainfall, relative humidity (RH), evaporation (E0), maximum (Tmax) and minimum (Tmin) temperature and solar radiation during Oct-Jan 2010/2011 (a, b) and 2011/2012 (c, d) at the meteorological station, Khon Kaen University, Thailand.

Figure 2. Relative water content (RWC) on the last day of irrigation, at R7 stage and harvest stage of five peanut genotypes grown under well-watered and terminal drought during 2010/2011 (a) and 2011/2012 (b). * Well-watered = Full irrigation since sowing until final harvest. Terminal drought = 1/3 available water (AW) at R7 until final harvest.

Root dry weight was also determined at R8 using the monolith method for one plant per plot. The size of the monolith was 50 × 20 cm with a depth of 50 cm. The roots were removed from the monolith soil sample using the same

method described previously for the core sample. The root samples were oven-dried at 80 °C for 48 h or until constant weight and root dry weight (g plant-1) was determined.

80

85

90

95

100


Well-watered

Terminal droughtRW

C (%

)

80

85

90

95

100


Well-watered

Terminal drought

RW

C (%

)

(b)

(a) (b)

(a) (b)

(c) (d)

Htoon et al. (2013)

317

Statistical analysis Simple correlation was used to determine the relationships between root traits and nutrient uptake, NF and NDW under well-watered and terminal drought conditions and all calculations were performed using MSTAT-C package (Bricker, 1989). RESULTS Meteorological conditions, soil properties and plant water stress The first experiment was conducted from October 2010 to January 2011. Average air daily temperature ranged from 18.92 to 30.25 °C during the growing season (Figure 1). There was no rainfall during the terminal drought period, while total rainfall before the drought-stressed period was 86.1 mm. The second experiment was carried out from October 2011 to January 2012. The average air daily temperature ranged from 19.38 to 31.15 °C during the growing season. Total rainfall during the growing season was 41.4 mm and total rainfall during the terminal drought period was 30.2 mm. Each rain-out shelter, covering each subplot, prevented rain from reaching the experimental plots in the case of unseasonal rain. Some characteristics of the soil analyzed in 2010/2011 and 2011/2012 were as follows: soil pH (1:2.5 H2O) in 2010/2011 and 2011/2012 were 6.08 and 6.18, respectively. Total N in 2011/2012 was 203.00 mg kg-1 while it was 185.00 mg kg-1 in 2010/2011. Higher available P (40.74 mg P kg-1), exchangeable K (38.34 mg K kg-1) and exchangeable Ca (446.67 mg Ca kg-1) were observed in 2011/2012. Predetermined drought stress levels (1/3 AW) measured by pressure plate method in the first and second year were 6.33% and 6.37%, respectively. Soil moisture percentages at soil depths of 30 cm and 60 cm at sowing, on the last day of irrigation, at R7 growth stage and harvest under well-watered and terminal drought conditions are presented in Table 1. In the well-watered treatment, soil moisture contents measured at 30 cm on the day of sowing

(10.51%), last day of irrigation (9.91%), R7 (9.74%) and at final harvest (10.25%) were more consistent than those at 60 cm in 2010-2011. In general, soil moisture contents at 30 and 60 cm in 2011-2012 were closer to the desired soil moisture contents both in well-watered and terminal drought conditions (1/3 AW) compared with those measured in 2010-2011.

The RWC decreased under terminal drought in both years (Figure 2). Values of RWC on the last day of irrigation were almost the same as those under well-watered and terminal drought conditions. However, RWC at the water-stressed treatment was significantly lower than that at the well-watered treatment in both years, indicating that degrees of drought stress were reasonably controlled at the predetermined levels. The highest RWCs were observed at R7 under well-watered treatment in both years, followed by RWC at well-watered treatment on the last day of irrigation and harvest. Water-stressed plants exhibited severe wilting symptoms (e.g., leaf folding), but leaf senescence and abscission were not observed in every genotype during these terminal drought periods. Relationship between root traits and nutrient uptake According to Songsri et al. (2008), peanut genotypes that have a higher RLD in the deeper soil layers should have enhanced drought tolerance and this can aid peanut genotypes to obtain higher pod yield and harvest index under long-term drought conditions. They also reported that deeper roots might enhance the partitioning of assimilates to developing pods. Moreover, in our experiment, peanut genotypes did not show any significant root trait responses at 0-30 cm and there were no correlations with nutrient uptake and NF under terminal drought (data not shown). Therefore, we only focused on the root traits at soil depths between 30 and 90 cm.

The correlations between root surface area (cm2) and uptake of N, P, K, Ca and Mg were not significant under well-watered condition (Table 2). When peanut genotypes were subjected to terminal drought condition, root surface area significantly correlated with


318

uptake of N (r = 0.51, P ≤ 0.01), P (r = 0.43, P ≤ 0.01), K (r = 0.33, P ≤ 0.05) and Mg (r = 0.32, P ≤ 0.05), but there was no correlation with uptake of Ca.

The correlations between root diameter and nutrient uptake were not significant under well-watered condition. Similarly, there were no correlations between root diameter and uptake of N, P, K, Ca and Mg under terminal drought condition.

Strong correlations between root volume and uptake of N (r = 0.58, P ≤ 0.01), P (r = 0.51, P ≤ 0.05), K (r = 0.58, P ≤ 0.01) and Mg (r = 0.46, P ≤ 0.01) were observed under terminal drought, while the correlation coefficients between root volume and nutrient uptake were not significant under well-watered condition. The correlation between root volume and uptake of Ca was not significant under both conditions.

Similarly, RLD significantly correlated with uptake of N (r = 0.81, P ≤ 0.01), P (r = 0.76, P ≤ 0.01), K (r = 0.71, P ≤ 0.01) and Mg (r = 0.53, P ≤ 0.01) under terminal drought and there was no significant correlation between these traits under well-watered condition.

In addition, % RLD showed similar correlations with nutrient uptake under both conditions. The % RLD did not correlate with nutrient uptake under well-watered condition but it was obvious that % RLD significantly correlated with uptake of N (r = 0.35, P ≤ 0.05), P (r = 0.40, P ≤ 0.01), K (r = 0.39, P ≤ 0.05) and Mg (r = 0.33, P ≤ 0.05) under terminal drought. except for Ca uptake.

However, RDW did not correlate with nutrient uptake under well-watered conditions. But RDW only correlated with uptake of Ca significantly (r = 0.36, P ≤ 0.05) under terminal drought.

Table 2. Correlation coefficients between root surface area (cm2), root diameter (mm), root volume (cm3), root length density (RLD), percentage of root length density (% RLD) at 30-90 cm, root dry weight (RDW) (g plant-1) at 0-50 cm, and uptake of N, P, K, Ca and Mg (mg plant-1) across 2010/2011 and 2011/2012. Nutrient uptake (mg plant-1)

Root surface area (cm2)

Root diameter (mm)

Root volume (cm3)

RLD

% RLD RDW (g plant-1)

Well-watered N 0.28 -0.12 0.09 -0.03 0.05 0.08 P 0.17 -0.10 0.00 -0.06 0.06 0.14 K 0.12 -0.08 0.02 -0.08 0.05 0.16 Ca 0.12 -0.02 -0.01 -0.10 -0.09 0.13 Mg 0.30 0.06 0.25 0.14 0.20 0.20 Terminal drought N 0.51** 0.28 0.58** 0.81** 0.35* -0.19 P 0.43** 0.06 0.51** 0.76** 0.40** -0.20 K 0.33* 0.25 0.58** 0.71** 0.39* -0.20 Ca 0.16 0.29 0.25 0.10 -0.27 0.36* Mg 0.32* 0.16 0.46** 0.53** 0.33* -0.26 *, ** = significant at P ≤ 0.05 and P≤0.01, respectively. Well-watered = Full irrigation since sowing until final harvest. Terminal drought = 1/3 available water (AW) at R7 until final harvest.

Htoon et al. (2013)

319

Relationship between root traits, N2 fixation and nodule dry weight Under well-watered condition, significant correlations were observed between NF and root surface area (r = 0.63, P ≤ 0.01), root volume (r = 0.48, P ≤ 0.01), RLD (r = 0.46, P ≤ 0.01) and %RLD (r = 0.40, P ≤ 0.01) (Table 3). Their correlations were consistent, even though peanut genotypes were subjected to terminal drought as significant correlations were also found between NF and root surface area (r = 0.53, P ≤ 0.01), root volume (r = 0.31, P ≤ 0.05), RLD (r = 0.65, P ≤ 0.01) and % RLD (r = 0.64, P ≤ 0.01). There was no correlation between root dry weight and

NF under well-watered condition. The correlation became significant but negative (r = -0.52, P ≤ 0.01) under terminal drought. Root surface area significantly correlated with NDW under well-watered condition (r = 0.47, P ≤ 0.01) and under terminal drought (r = 0.32, P ≤ 0.05). The correlation coefficients between NDW with root volume (r = 0.31, P ≤ 0.05), RLD (r = 0.55, P ≤ 0.01), and % RLD (r = 0.35, P ≤ 0.05) were significant under terminal drought. RDW correlated with NDW significantly (r = 0.34, P ≤ 0.05) only under well-watered condition.

Table 3. Correlation coefficients between root surface area (cm2), root diameter (mm), root volume (cm3), root length density (RLD), percentage of root length density (%RLD) at (30-90 cm), root dry weight (RDW) (g plant-1) at (0-50 cm), and N2-fixation (NF) (mg plant-1) and nodule dry weight (NDW) (mg plant-1) across 2010/2011 and 2011/2012. NF and NDW Root

surface area (cm2)

Root diameter (mm)

Root Volume (cm3)

RLD %RLD RDW (g plant-1)

Well-watered NF 0.63** 0.20 0.48** 0.46** 0.40** 0.30 NDW 0.47** 0.04 0.27 0.23 0.19 0.34* Terminal drought NF 0.53** -0.23 0.31* 0.65** 0.64** -0.52** NDW 0.32* -0.50 0.31* 0.55** 0.35* -0.24 *, ** = significant at P ≤ 0.05 and P ≤ 0.01, respectively. Well-watered = Full irrigation since sowing until final harvest. Terminal drought = 1/3 available water (AW) at R7 until final harvest.

DISCUSSION According to Boote (1983), peanut had the highest root distribution at the 0 to 30 cm soil depth. Nevertheless, root distribution was moved towards the deeper soil layer with decreasing moisture availability in the soil (Wright and Nageswara Rao, 1994). The responses of root mass to drought conditions were observed in some peanut genotypes, but not in all (Del-Rosario and Fajardo, 1988). Despite the controversial observations on root growth responses such as RLD and root mass to water deficit conditions (Robertson et al., 1980; Boote, 1983; Pandey et al., 1984), root traits have been

identified as drought avoidance traits (Turner, 1986; Matsui and Singh, 2003; Taiz and Zeiger, 2006). Under drought condition, peanut might be able to adapt by increasing root length to mine more available water (Alycmeny, 1997; Mayaki et al., 1976). However, a deeper root system that contributes to maintaining nutrient uptake and NF has not been clearly demonstrated. In our study, root traits did not correlate with uptake of N, P, K, Ca and Mg under well-watered conditions. In fact, root surface area, root diameter, root volume, RLD and % RLD mentioned in these experiments were measured at 30 to 90 cm soil depth.


320

However, interestingly, there was no association between such root traits measured at a soil depth of 0 to 30 cm and nutrient uptake under both well-watered and terminal drought conditions (data not shown). The negative relationship between root diameter and uptake of N, P, K, Ca and Mg under well-watered conditions and their non-significant correlations under terminal drought indicated that root diameter was not an important root trait to determine nutrient uptake. Increasing effective root diameter would have little effect on nutrient uptake (Fitter and Hay, 2002).

Root surface area might be important in N, P, K and Mg uptake under terminal drought conditions. However, Nye (1973) and Russel and Clarkson (1976) observed that nutrient uptake would be related more to root volume than to root surface area. Nevertheless, our experiment revealed that root surface area was equally related to nutrient uptake as much as root volume was because there were strong correlations between root volume, RLD and % RLD and uptake of N, P, K and Mg under terminal drought. The large root surface area might have greater chances to make contact with soil nutrients and consequently take them up.

Boote (1983) reported that root elongation rate decreased as soil compaction increased. Soil water deficit might affect not only root depth but also root distribution in peanut. However, under terminal drought conditions, the five peanut genotypes tested in our experiments had increases in root volume, RLD and RLD (%) (data not shown). Songsri et al. (2008) also reported that RLD of peanut genotypes increased under prolonged severe drought conditions. Nevertheless, under pre-flowering drought, peanut genotypes showed a differential response in terms of root quantity and distribution. In our experiments, the large root system (root dry weight) did not correlate with nutrient uptake under both well-watered and terminal drought conditions. In fact, root dry weights were measured at 0 to 50 cm by monolith method and such a large root system alone could not be able to take up nutrients from water-deficient soil around this root system. The result was evident as Jongrungklang et al. (2011) concluded that a larger root system alone might not contribute much to pod yield if the large root

portion is not distributed into moist soil. Based on the findings of Songsri et al. (2008), the responses of RLD into deeper soil layers might allow plants to be able to mine more available water and to take up nutrients in the subsoil. Therefore, the possession of a deep and large root system in the lower soil profile, which allows access to water, was crucial in determining uptake of nutrients under terminal drought. Soon after the drought treatment started, such ability to modify root distribution to exploit deeper soil water could not only help mine water but also to take up soil nutrients while roots at the upper soil layers were unable to take up nutrients with decreasing soil moisture.

However, interestingly, the contra-dictory results revealed that RDW was correlated with uptake of Ca only under terminal drought. The result was reasonable because sufficient soil moisture in the fruiting zone was very important for Ca uptake. Boote et al. (1982) reported that effective uptake of calcium from the fruiting zone was enhanced by adequate soil water. Unlike other crops, due to the subterranean nature of peanut fruits, the peanut plant did not receive sufficient Ca from the xylem flow into the fruit (Skelton and Shear, 1971). Peanut plant requires more soil Ca or more adequate soil moisture in the fruiting zone. Therefore, despite root length density of peanut genotypes under terminal drought, the plants could not take up and contribute sufficient Ca and only a large root system at the 0 to 50 cm (about fruiting zone) might be able to take up Ca.

NF and its related traits such as biomass, nodule number and nodule dry weight in peanut were also affected by drought (Sinclair and Serraj, 1995; Pimtrach et al., 2008). Nevertheless, there is still insufficient knowledge on the correlations between NF and root depth and distribution under water deficit conditions. Strong and consistent correlations between NF and NDW and root surface area, root volume, RLD, and % RLD were observed under both well-watered and terminal drought conditions, indicating that a deeper root system could help maintain nodule growth and NF, even under severe terminal drought conditions. Despite the inability of the roots to mine soil

Htoon et al. (2013)

321

water from the upper dry soil, NF and consequently nodule growth were decreased, although some peanut genotypes might produce large root systems under terminal drought conditions. Rucker et al. (1995) observed that some peanut genotypes with large root systems under non-stressed conditions provided high yield under drought conditions, and these genotypes were supposed to possess drought avoidance traits. However, in our experiments, peanut genotypes with deep root systems consistently gave high NF and NDW, even under terminal drought conditions. On the other hand, the previous report mentioned that drought stress reduced aerial growth and limited assimilate supply to nodules (Osman et al., 1983; Minchin, 1997). However, peanut genotypes with deep root systems might be able to access moisture to maintain plant mechanisms such as photosynthesis, which, in turn, supports photosynthates and gives energy to nodule growth and NF under terminal drought conditions.

In conclusion, the responses of deep root systems favor nutrient uptake under terminal drought conditions because of their strong correlations. A large root mass was important for Ca uptake only and it alone was not effective to take up other nutrients when large root systems in the upper soil were unable to mine soil water. Root diameter might not be an important trait for nutrient uptake and NF. Moreover, drought avoidance by enhancing root surface area, root volume, RLD and % RLD are some mechanisms that help peanut maintain nodule growth and NF under terminal drought conditions. A better understanding of the correlations between root traits and nutrient uptake and biological NF will be useful to explain the underlying peanut responses under terminal drought. Consequently, the present findings suggest that selection of peanut genotypes with deep root systems would improve nutrient uptake and NF under terminal drought conditions. ACKNOWLEDGEMENTS The authors are grateful for the financial support of the Higher Education Research Promotion and National Research University Project of Thailand, Office of Higher

Education Commission, through the Food and Functional Food Research Cluster of Khon Kaen University, Thailand, the Peanut and Jerusalem Artichoke Improvement for Functional Food Research Group and the Plant Breeding Research Center for Sustainable Agriculture. Grateful acknowledgement is due the Thailand Research Fund, the Commission for Higher Education and Khon Kaen University for providing financial support to this research through the Distinguished Research Professor Grant of Professor Dr. Aran Patanothai. This work was also partially supported by the Peanut CRSP NMS 72 project by USDA through the University of Georgia ECG-A-00-0700001-00. Peanut genotypes for this study were kindly provided by ICRISAT. We also appreciate the contribution of many people who participated in the field data collection and sample preparation and processing. REFERENCES

Al-Karaki GN, Al-Raddad A (1997). Effects of arbuscular mycorrhizal fungi and drought stress on growth and nutrient uptake of two wheat genotypes differing in drought resistance. Mycorrhiza 7: 83-88.

Alycmeny MN (1997). The effect of drought on growth and dry matter allocation in seedlings of Vigna ambacensis L. J. King Saud Univ. 1: 41-51.

Bell MJ, Wright GC, Suryantini, Peoples MB (1994). The N2-fixing capacity of peanut cultivars with differing assimilate partitioning characteristics. Aust. J. Agric. Res. 45: 1455-1468.

Boddey RM, Urquiaga S, Neves MCP (1990). Quantification of the contribution of N2 fixation to field grown grain legumes–a strategy for the practical application of the 15N isotope dilution technique. Soil Biol. Biochem. 22: 649-655.

Boote KJ (1982). Growth stage of peanut (Arachis hypogaea L.). Peanut Sci. 9: 35-40.

Boote KJ (1983). Peanut. In: I.D. Teare and M.M. Peet, eds. Crop-water relations. John Wiley and Sons, USA. pp. 255-286.

Bricker AA (1989). MSTAT-C user’s guide. Michigan State University, East Lansing, Michigan.

Clavel D, Diouf O, Khalfaoui JL, Braconnier S (2006). Genotype variations in fluorescence parameters among closely related groundnut (Arachis hypogaea L.) lines and their potential for drought screening programs. Crop Sci. 34: 92-97.

Del Rosario DA, Fajado FF (1988). Morphophysiological responses of ten peanut


322

(Arachis hypogaea L.) varieties to drought stress. Phillip. Agric. 71: 447-459.

Devries JD, Bennett JM, Albrecht SL, Boote KJ (1989). Water relations, nitrogenase activity and root development of three grain legumes in response to soil water deficits. Field Crops Res. 22: 33-44.

Doorenbos J, Pruitt WO (1992). Calculation of crop water requirements. FAO: Irrigation and Drainage Paper. No. 24. FAO, the United Nations, Rome, Italy. pp. l-65.

Fageria NK, Baligar VC, Clark RB (2002). Micronu-trients in crop production. Adv. Agron. 77: 185-268.

Fitter, A., Hay R.. 2002. Environmental physiology of plants. 3rd ed. Academic Press, London, UK. 367 p.

Giller, K.E. 2001. Nitrogen fixation in tropical cropping systems. 2nd ed. CAB International, Wallingford, 423 p.

Girdthai T, Jogloy S, Vorasoot N, Akkasaeng C, Wongkaew S, Holbrook CC, Patanothai A (2010). Associations between physiological traits for drought tolerance. Plant Breed. 129: 693-699.

Htoon W, Jogloy S, Toomsan B, Sanitchon J (2009). Response to early drought for traits related to nitrogen fixation and their correlation to yield and drought tolerance traits in peanut (Arachis hypogaea L.). Asian J. Plant Sci. 8: 138-145.

Hungria M, Bohrer TRJ (2000). Variability of nodulation and dinitrogen fixation capacity among soybean cultivars. Biol. Fert. Soils 31: 45-52.

Jongrungklang N, Toomsan B, Vorasoot N, Jogloy S, Boote KJ, Hoogenboom G, Patanothai A (2011). Rooting traits of peanut genotypes with different yield responses to pre-flowering drought stress. Field Crops Res. 120: 262-270.

Kaewpradit W, Toomsan B, Cadisch G, Vityakon P, Limpinuntana V, Saenjan P, Jogloy S, Patanothai A (2009). Mixing groundnut residues and rice straw to improve rice yield and N use efficiency. Field Crops Res. 110: 130-138.

Kramer PJ (1980). Drought stress and the origin of adaptation. In: N.C. Turner, and P.J. Kramer, eds. Adaptation of plant to water and high temperature stress. John Wiley & Sons, New York. pp. 7-20

Kulkarni JH, Ravindra V, Sojitra VK, Bhatt DM (1988). Growth, nodulation and N uptake of groundnut (Arachis hypogaea L.) as influenced by water deficits stress at different phenophase. Oleagineux 43: 415-419.

Ludlow MM, Muchow RC (1990). A critical evaluation of traits for improving crop yields in water-limited environments. Adv. Agron. 43: 107-153.

Maiti RK, Wesche-Ebeling P, Nunez-Gonzalez A, Sanchez-Arreola E (2002). Root system and mineral nutrition, In: R.K. Maiti, and P. Wesche-Ebeling, eds. The peanut (Arachis hypogaea) Crop. Science Publishers, Inc., New Hampshire. pp. 125-146.

Marschner, H. 1995. Mineral nutrition of higher plants. 2nd ed. Academic Press, London. 889 p.

Matsui T, Singh BB (2003). Root characteristics in cowpea related to drought tolerance at the seeding stage. Exp. Agric. 39: 29-38.

Mayaki WC, Teare ID, Stone LR (1976). Top and root growth of irrigated and non-irrigated soybeans. Crop Sci. 16: 92-94.

McDonagh JF, Toomsan B, Limpinantana V, Giller KE (1993). Estimates of the residual nitrogen benefit of groundnut to maize in Northeast Thailand. Plant Soil 154: 267-277.

Minchin FR (1997). Regulation of oxygen diffusion in legumes nodule. Soil Biol. Biochem. 29: 881-888.

Mulongoy K (1992). Biological nitrogen fixation, In: B.R. Tripathl and P.J. Psychas, eds. Technical paper 2. The AFNETA alley farming training manual. Alley Farming Network for Tropical Africa International Institute of Tropical Agriculture, Ibadan.

Nageswara Rao RC, Williams JH, Sivakumar MVK, Wadia KDR (1988). Effect of water deficit at different growth phase of peanut. II. Response to drought during preflowering phase. Agron. J. 80: 431-438.

Nye PH (1973). The relation between the radius of a root and its nutrient absorbing power (α). J. Exp. Bot. 24: 783-786.

Osman AK, Wynne JC, Elkan GH, Schneeweis TJ (1983). Effect of leaf removal on symbiotic nitrogen fixation in peanut. Peanut Sci. 10: 107-110.

Pandey PK, Herrera WAT, Villegas AN, Pendleton JW (1984). Drought responses of grain legumes under irrigation gradient. III. Plant growth. Agron. J. 76: 557-560.

Pimratch S, Jogloy S, Vorasoot N, Toomsan B, Kesmala T, Patanothai A, Holbrook CC (2008). Effect of drought stress on traits related to N2-fixation in eleven peanut (Arachis hypogaea L.) genotypes differing in degrees of resistance to drought. Asian J. Plant Sci. 7: 334-342.

Robertson WK, Hammond LC, Johnson JT, Boote KJ (1980). Effects of plant-water status on root

Htoon et al. (2013)

323

distribution of corn, soybeans, and peanuts in sandy soil. Agron. J. 72: 548-550.

Rucker KS, Kvien CK, Holbrook CC, Hook JE (1995). Identification of peanut genotypes with improved drought avoidance traits. Peanut Sci. 22: 14-28.

Russel RS, Clarkson DT (1976). Ion transport in root systems. In: N. Sunderland, ed., Perspectives in experimental biology, Vol. 2. Pergamon, Oxford. pp. 401-411.

Samarah N, Mullen R, Cianzio S (2004). Size distribution and mineral nutrients of soybean seeds in response to drought stress. J. Plant Nutr. 27: 815-835.

Schuman GE, Stanley MA, Knudson D (1973). Automated total nitrogen analysis of soil and plant samples. Soil Sci. Am. Proc. 37: 480-481.

Sinclair TR, Serraj R (1995). Dinitrogen fixation sensitivity to drought among grain legume species. Nature 378: 344.

Singh, S., and M.B. Russell. 1981. Water use by maize/pigeonpea intercrop on a deep Vertisol. In: Proceedings of the International Workshop on Pigeonpeas, December 1980. ICRISAT Center, Patancheru, AP, India. pp. 271-282.

Skelton BJ, Shear GM (1971). Calcium translocation in the peanut (Arachis hypogaea L.). Agron. J. 63: 409-412.

Songsri P, Jogloy S, Vorasoot N, Akkasaeng C, Patanothai A, Holbrook CC (2008). Root distribution of drought-resistant peanut genotypes in response to drought. J. Agron. Crop Sci. 194: 92-103.

Sprent, JI (1981). Nitrogen fixation. In: L.G. Paleg, and D. Aspinal, eds. Physiology and biochemistry of drought resistance in plants. Academic Press, London, pp. 131-143.

Taiz L, Zeiger E (2006). Stress physiology. In: L. Taiz and E. Zeiger, eds. Plant physiology. 4th ed. Sinauer Associates, Inc.: Massachusetts. pp. 671-681.

Turner NC (1986). Adaptation to water deficit: A changing perspective. Aust. J. Plant Physiol. 13: 175-190.

Van Rossum D, Muyotcha A, Verseveld HWV, Stouthamer AH, Boogerd FC (1993). Effects of Bradyrhizobium strain and host genotype, nodule dry weight and leaf area on groundnut (Arachis hypogaea L. ssp. fastigiata) yield. Plant Soil 154: 279-288.

Venkateswarlu B, Maheswari M, Saharan N (1989). Effects of water deficit on N2 (C2H2) fixation in cowpea and groundnut. Plant Soil 114: 69-74.

Vorasoot N, Songsri P, Akkasaeng C, Jogloy S, Patanothai A (2003). Effect of water stress on yield and agronomic characters of peanut (Arachis hypogaea L.). Songklanakarin J. Sci. Technol. 25: 283-288.

Welbank PJ, Gibb MJ, Taylor PJ, Williams ED (1974). Root growth of cereal crops, In: Rothamsted Experimental Station Annual Report 1973 part 2, Rothamsted Experimental Station, Harpenden, U.K., pp. 26-66.

Wright GC, Nageswara Rao RC (1994). Groundnut water relations. In: J. Smartt, ed. The groundnut crop. A scientific basis for improvement. Chapman and Hall: London. pp. 281-325.

Wright GC, Nageswara Rao RC, Basu MS (1996). A physiological approach to the understanding of genotype by environment interactions–A case study on improvement of drought adaptation in groundnut. In: M. Cooper and G.L. Hammer, eds. Plant adaptation and crop improvement. CAB International: Wallingford. pp. 365-380.


EVALUATION OF SEEDLING AND ADULT PLANT STAGES RESISTANCE TO Sclerotium rolfsii Sacc. IN JERUSALEM ARTICHOKE

(Helianthus tuberosus L.)

RATTIKARN SENNOI1,2, SANUN JOGLOY1,2, WEERASAK SAKSIRIRAT1, PORAMATE BANTERNG1,2*, THAWAN KESMALA1

and ARAN PATANOTHAI1


Khon Kaen, 40002, Thailand 2Plant Breeding Research Center for Sustainable Agriculture, Faculty of Agriculture, Khon Kaen University,

Khon Kaen 40002, Thailand *Corresponding author’s email: [email protected]

SUMMARY

Southern stem rot caused by Sclerotium rolfsii is an important problem in Jerusalem artichoke production in Thailand. Chemical control is not suitable to suppress the disease because chemicals can contaminate the crop. The use of host plant resistance is the most appropriate means to control the disease. Screening of Jerusalem artichoke for resistance to S. rolfsii will be much easier and effective if resistance at seedling stage is correlated with resistance at adult stage. Two plant ages at seedling stage (20 days after transplanting (DAT)) and at mature stages (85 DAT) were assigned as main plots, and 10 genotypes of Jerusalem artichoke were assigned as sub-plots in split-plot design with four replications for two years. Data were recorded for lesion length, disease incidence, days to permanent wilting and area under disease progress curve (AUDPC). The results showed that some genotypes exhibited resistance at seedling stage but they were not resistant at adult stage. The correlations between seedling and adult stages were low for all traits. JA 98 took a longer time to permanent wilting and had lower AUDPC at seedling and adult stages. JA 6 had lower disease incidence at seedling and adult stages. Key words: chemical control, correlation, disease incidence, host plant resistance, southern stem rot

Manuscript received: November 26, 2012; Decision on manuscript: February 28, 2013; Manuscript accepted: March 4, 2013 © Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION Southern stem rot caused by Sclerotium rolfsii is a serious disease worldwide. The fungus pathogen has a wide range of hosts including vegetables, flowers, legumes, cereals, forage plants and weeds (Agrios, 2005). The disease on Jerusalem artichoke (Helianthus tuberosus L.) has been reported in both temperate and tropic regions (Koike, 2004; Sennoi et al., 2010). Records show that this disease caused 60% plant

damage and resulted in serious yield losses (Mccarter and Kays, 1984).

Initial symptoms consisted of wilting of new shoots and leaves, followed by browning and collapse of all foliage. Crown and lower stem tissues were colonized internally and externally by white, cottony mycelia. Tan, spherical sclerotia that measured approximately 1 mm in diameter formed on surfaces of the affected crowns and stems (Koike, 2004).

Screening for S. rolfsii resistance has been reported in many plant species, including

RESEARCH ARTICLE


325

crops and ornamental plants (Shew et al., 1987; Fery and Dukes, 2002; Akram et al., 2008; Xu et al., 2009). To the best of our knowledge, screening for S. rolfsii resistance in Jerusalem artichoke has not been reported. Attempts have been made to identify Jerusalem artichoke genotypes with resistance to S. rolfsii at the seedling stages, and the putatively resistant and susceptible genotypes were identified (Sennoi et al., 2012). The relationship between resistance at seedling stage and resistance at maturity has not been verified in Jerusalem artichoke. Relationships between seedling and adult plant resistance to other pathogens in Brassica napus (Li et al., 2006), wheat (Wang et al. 2005; Hovmøller, 2007) and barley (Prashant et al., 2009) have been reported.

The question is whether the Jerusalem artichoke genotypes with resistance to S. rolfsii at the seedling stage are also resistant at the mature stage. If resistant genotypes can be readily identified at the seedling stage and resistance persists at maturity, the screening of Jerusalem artichoke for resistance to S. rolfsii will be much easier and effective. The objective of this work was to evaluate the relationship between seedling and adult plant resistance to S. rolfsii in Jerusalem artichoke. MATERIALS AND METHODS Experimental design Two experiments were conducted at the Khon Kaen University Agronomy Farm, Khon Kaen, Thailand, from May to August 2011 and repeated during March to June 2012. The first experiment was undertaken in an open-side greenhouse, and the repeated experiment was carried out in an open environment. Pot experiments in split-plot design with four replications were used in both experiments. Two plant ages at seedling stage (20 days after transplanting (DAT)) and at mature stages (85 DAT) were assigned as main-plots, and 10 genotypes of Jerusalem artichoke obtained from the Plant Gene Resource of Canada, Saskatoon, Saskatchewan, Canada and the Leibniz Institute of Plant Genetics and Crop Plant Research, Gatersleben, Germany were assigned as sub-

plots. Genotypes HEL 280, HEL 278, HEL 246, JA 98 and HEL 65 were in the resistant group, and JA 122, JA 6, HEL 62, JA 2 and JA 102 were identified as susceptible genotypes (Sennoi et al., 2012). All genotypes were identified based on days to permanent wilting in the seedling stage. The experimental unit was five plants or five pots for each Jerusalem artichoke genotype in a replication. The pots were arranged with spacing of 0.7 m between genotypes and 0.5 m between plants within the same genotype. Weeds were manually controlled throughout the experiment, and N, P2O5, and K2O fertilizer at the rate of 0.3 g per pot was given to adult plants at 30 DAT. Preparation of plant materials Tubers were cut into small pieces with 3-4 buds and incubated at ambient temperature in charred rice husks for 7 days for pre-germination. The tuber pieces were then transferred to plastic trays for a week until the first two leaves were fully expanded. The seedlings were further transferred to plastic pots with 28 cm in height and 31 cm in diameter containing steamed soil and charred rice husks at the ratio of 1:1 by volume. The soil was Roi-et series (fine-loamy, mixed, subactive, isohyperthermic Aeric Kandiaquults) and obtained from a local farmer's farm in Khon Kaen province. Seedlings for adult plant treatment were transplanted earlier by 55 days, followed by the seedlings for young plants, so the treatments were applied 85 days DAT for adult plant treatment and 20 DAT for young plant treatment on the same day. On this day, the treatments were ready for simultaneous inoculation by S. rolfsii. Preparation of S. rolfsii inoculum Isolate 1 of S. rolfsii, was obtained from the Khon Kaen University Agronomy Farm. The isolate was cultured on potato dextrose agar (PDA) in an incubator at 25 ± 2°C. Single sclerotia from stored test tube slants were surface-sterilized using 70% methyl alcohol for 1 min, and then rinsed with sterilized water for 1 min. The sclerotia were then transferred to PDA and incubated at room temperature (25 ± 2°C). After 3 days, mycelia of S. rolfsii from PDA in

Sennoi et al. (2013)

petri dishes were cut using a 0.5 cm-diameter cork borer and transferred to sterilized sorghum seeds (Sennoi et al. 2010) in bottles containing 30 g steam-sterilized sorghum grains. The bottles were regularly shaken to facilitate thorough colonization by the fungus. After 2 weeks of incubation, the inoculum was ready for use. Evaluation of seedling and adult plant resistance to S. rolfsii Seedling and adult plants were inoculated at the same time. Plants were inoculated by cutting a 0.5-cm-long and 1-mm-deep vertical wound into the crown of each plant and placing a single infested sorghum seed close to the wound. The wounds were covered with moist cotton with sprayed water after inoculation to facilitate infection. Water was supplied to the experiment regularly to avoid stress. Data collection and statistical analysis Days to permanent wilting of the plants (defined as all leaves wilted) were observed every day. Plants were evaluated every other day after

inoculation for lesion length (cm) and number of plants with permanent wilting symptom. The number of plants with permanent wilting symptoms was later converted into disease incidence. The area under the disease progress curve (AUDPC) was calculated from disease incidence according to a formula suggested by Marcel et al. (2008).

Data for the first and repeated experiments were analyzed statistically for each parameter and error variances between two trials were tested for homogeneity. Data with homogeneity of variance were subjected to combined analysis of variance for the two trials according to a split-plot design for factor main plots and subplots effects (Hoshmand, 2006). Least significant difference (LSD) was used to compare mean differences. Data on plot basis were used to calculate correlation coefficients between seedling growth stage and mature growth stage for disease incidence, lesion length, days to permanent wilting and AUDPC according to Spearman’s rank correlation (Hoshmand, 2006). All calculations were done using STATISTIX eight software program (Analytical software, Tallahassee, Florida).

Table 1. Mean squares from combined analysis of variance for lesion length (cm), disease incidence (%), days to permanent wilting and area under disease progress curve (AUDPC) of Jerusalem artichoke. Source of variation df Lesion length

(cm) Disease incidence (%)

Days to permanent wilting

AUDPC

Experiment (E) 1 2.2* 180.6ns 75.4** 14497.0ns Rep within experiment 6 0.2 98.9 2.0 6441.6 Stage of plant (S) 1 44.8** 600.6ns 5535.4** 25560000.0** E × S 1 0.7* 1050.6* 5.7ns 20407.8ns Rep within stage of plant and experiment

6 0.1 152.3 3.7 36564.7

Genotypes (G) 9 1.7** 2361.2** 48.4** 581801.0** E × G 9 0.3* 89.0ns 2.4** 8009.8ns S × G 9 0.2ns 947.9** 29.9** 319940.0** E × S × G 9 0.1 120.1ns 1.8* 23191.3ns Pooled error 108 0.1 134.0 0.9 14186.2 C.V. (%) 30.1 15.7 8.6 13.8 *Significant at 5%; ** significant at 1%.


327

Table 2. Correlation of Sclerotium rolfsii resistance traits for 10 Jerusalem artichoke genotypes between seedling and adult plant stages. Seedling Adult plant

Lesion length

(cm)

Disease incidence

(%)

Days to permanent wilting AUDPC

Lesion length (cm) 0.16ns

Disease incidence (%) 0.24*

Days to permanent wilting 0.28*

AUDPC 0.16ns

* Significant at 5%; ns = not significant. RESULTS Seedling and adult plant stages were statistically different for lesion length, days to permanent wilting and area under disease progress curve (AUDPC) but were not different for disease incidence (Table 1). The interaction between experiment and plant age were found for lesion length and disease incidence. Genotypic differences were observed for lesion length, disease incidence, days to permanent wilting and AUDPC. Genotype by experiment interactions were found for lesion length and days to permanent wilting. In addition, the interactions between plant age and genotype were statistically different for disease incidence, days to permanent wilting and AUDPC.

The correlations between two plant ages for lesion length, disease incidence, days to permanent wilting and AUDPC were low, although they were positive and significant, especially for disease incidence (r = 0.24) and days to permanent wilting (r = 0.28) (Table 2). Seedling stage had longer lesion length (1.8 cm), spent less time until permanent wilting (5 days) and had higher AUDPC (1262.7) than adult plant stage (0.7 cm for lesion length, 17 days for days to permanent wilting and 463.3 for AUDPC). For disease incidence, there was no statistical difference between seedling and adult plant stages (data not shown).

Lesion lengths of 10 Jerusalem artichoke genotypes at the seedling stage ranged from 1.1 to 2.6 cm, and genotypes JA 6, JA 122 and JA 102 had longer lesions than the others. Shorter lesions were detected for HEL 246 and HEL 65. With adult plants, lesion ranged from 0.2 to 1.3 cm, and longer lesions were observed for HEL 62, JA 102 and JA 2, whereas shorter lesions were found for JA 98 and HEL 246 (Table 3).

Disease incidence ranged from 56.2 to 87.5% at the seedling stage. More severe disease was observed for HEL 62 and JA 102 (Table 3). Lower disease incidences were observed for HEL 280, HEL 65 and JA 6. Disease incidence among adult plants ranged from 50.0 to 97.5%, and JA 6, HEL 246 and JA 98 had lower disease incidence than the others. Genotypes with higher disease incidence included JA 102, JA 2 and JA 122.

The differences between two plant ages and among Jerusalem artichoke genotypes were observed in days to permanent wilting (Table 3). Days to permanent wilting ranged from 3 to 7 days at the seedling stage, whereas days to permanent wilting among adult plants ranger from 13.6 to 19.9 days. The data indicated that adult plants were more resistant than young plants. HEL 62, JA 6, HEL 278 and JA 122 took fewer days to reach permanent wilting, whereas HEL 65, HEL 280 and JA 98 took more days to reach permanent wilting, indicating that


they were more resistant to the disease at the seedling stage. For adult plant resistance, HEL 62, HEL 280 and JA 102 spent fewer days to reach permanent wilting date, whereas JA 98, JA 6 and HEL 246 took more days to reach permanent wilting, indicating that they were more resistant to the disease at the mature plant stage.

The AUDPC ranged from 1,132.9 to 1,562.5 at the seeding stage and ranged from 31.3 to 850.0 at the adult plant stage. The genotypes with higher AUDPC in seedling stage were HEL 62, HEL 278 and JA 102; HEL 280, JA 122 and JA 6 had lower AUDPC. However, there was not much difference among genotypes in the lower AUDPC group. At the adult plant stage, higher AUDPC values were observed in JA 102, HEL 62, JA 122 and HEL 280, and lower AUDPC was seen in JA 6, HEL 246 and JA 98 (Table 3). DISCUSSION To the best of our knowledge, the relationship of resistance to S. rolfsii in Jerusalem artichoke between seedling and adult growth stages has not been reported in the literature. Other studies have been done for other kinds of pathogen and plant hosts such as powdery mildew caused by Blumeria graminis f. sp. tritici in wheat (Wang et al., 2005), Puccinia hordei in barley (Golegaonkar et al., 2009), net blotch caused by Pyrenophora teres f. maculate in barley (Williams et al., 2003).

S. rolfsii can cause the disease in Jerusalem artichoke at both seedling and adult stages. The disease, however, was different between the two growth stages as indicated by low correlations between seedling and adult plant stages for lesion length, disease incidence, days to permanent wilting and AUDPC. The results, thus, pointed out that the mechanisms controlling resistance to S. rolfsii at seedling and adult growth stages might be different. Therefore, selection at the seedling stage will not be effective.

In broccoli, resistance to downy mildew caused by Peronospora parasitica at seedling stage and mature stage was not correlated (Dick and Petzoldt, 1993). Similar results were also

reported in peanut resistance to limb rot caused by Rhizoctonia solani (Franke et al. 1999). Rank correlations between seedling and adult plant data for Mycosphaerella graminicola isolates IPO323 and IPO290 were significant, adult plants appeared to be more susceptible than seedlings, so evaluation of resistance may require seedling as well as adult plant tests (Kerma and van Silfhout, 1997).

In this study, resistance at seedling stage was not related to resistance at mature plant stage. This was possibly due to the low number of Jerusalem artichoke genotypes. However, it might be possible to find Jerusalem artichoke genotypes with resistance to S. rolfsii at seedling stage and mature plant stage, if large numbers of accessions are screened. The hypothesis of our study was that resistance to S. rolfsii stem rot in adult plant stage would be high positively correlated with resistance to seedling disease. If this hypothesis were true, screening of seedling could be used as a primary procedure to identify the potential of Jerusalem artichoke germplasm. This study may be valuable for further breeding of Jerusalem artichoke resistance to S. rolfsii.

In the present study, HEL 65 and HEL 280 had shorter lesion length, lower disease incidence revealed to be lower severity at seedling stage based on shorter lesion length, longer days to permanent wilting, and lower AUDPC at seedling stage than at mature stage. However, differences among Jerusalem artichoke for these disease parameters were rather low at the seedling stage. Jerusalem artichoke genotypes that showed higher resistance at mature plant stage were HEL 246, JA 98 and JA 6 as indicated by lower disease incidence, longer days to permanent wilting and lower AUDPC.


329

Table 3. Seedling and adult plant resistance in 10 genotypes of Jerusalem artichoke inoculated with Sclerotium rolfsii * Genotype Lesion length (cm) Disease incidence (%) Days to permanent wilting AUDPC

Seedling Adult Seedling Adult Seedling Adult Seedling Adult

HEL 280 1.4 ± 0.2 0.7±0.3 56.3 ± 7.4 67.5 ± 10.4 6.4 ± 0.5 13.7 ± 1.6 1132.9 ±84.7 680.0 ± 104.4

HEL 278 1.9 ± 0.2 0.5±0.1 67.5 ± 14.9 77.5 ± 12.8 3.9 ± 0.7 16.3 ± 1.8 1478.1 ± 113.5 437.5 ± 151.0

HEL 246 1.1 ± 0.2 0.4 ± 0.5 77.5 ± 12.8 57.5 ± 12.8 5.0 ± 1.0 18.5 ± 1.1 1218.8 ± 173.4 103.1 ± 62.7

JA 98 1.7 ± 0.3 0.2 ± 0.1 70.0 ± 10.7 52.5 ± 14.9 5.7 ± 1.0 19.9 ± 1.5 1265.6 ± 161.9 143.8 ± 119.4

HEL 65 1.2 ± 0.4 1.1 ± 0.6 62.5 ± 12.8 80.0 ± 10.7 6.9 ± 2.0 17.9 ± 2.0 1171.9 ± 141.2 431.3 ± 187.5

JA 122 2.4 ± 0.6 0.6 ± 0.4 70.0 ± 10.7 92.5 ± 10.4 4.1 ± 0.7 17.8 ± 1.2 1143.8 ± 125.8 706.3 ± 94.3

JA 6 2.6 ± 0.5 0.7 ± 0.6 62.5 ± 12.8 50.0 ± 10.7 3.5 ± 0.9 18.9 ± 1.3 1148.1 ± 77.8 31.3 ± 9.4

HEL 62 1.6 ± 0.3 1.3 ± 0.5 87.5 ± 10.4 87.5 ± 14.9 3.0 ± 1.0 10.4 ± 1.1 1562.5 ± 45.3 825.0 ± 138.2

JA 2 1.8 ± 0.6 0.9 ± 0.5 77.5 ± 12.8 92.5 ± 10.4 5.5 ± 0.5 17.7 ± 1.7 1164.4 ± 99.8 425.0 ± 139.7

JA 102 2.3 ± 0.5 1.2 ± 0.5 85.0 ± 9.3 97.5 ± 7.1 4.3 ± 0.4 15.0 ± 1.7 1341.3 ± 152.8 850.0 ± 69.0

Max. 3.5 2.3 100.0 100.0 9.0 22.5 1600.0 950.0

Min. 0.5 0.1 40.0 40.0 2.0 7.3 1000.0 25.0

Mean 1.8 0.8 71.6 75.5 4.8 16.6 1262.7 463.3

*Data are presented as mean ± standard deviation (n = 8).


The difference in genotype identification based on days to permanent wilting at the seedling stage between our previous study and the present study was found. This might have been caused by the difference in inoculation procedure. The inoculation method used in this experiment seems to be better than the one followed in the previous experiment because the level of resistance could be identified within several days(for the seedling stage). However, an effective inoculation method that will give consistent results in the identification of Jerusalem artichoke resistance to the disease should be further developed.

Wheat cultivars with low infection of yellow rust caused by Puccinia striiformis f. sp. tritici at seedling stage were moderately resistant under field conditions, and two cultivars of wheat showed good resistance in both seedling and adult stages (Torabi and Nazari, 1998). Ninety-two genotypes of barley were evaluated for their reaction to Puccinia hordei pathotypes at seedling and adult plant stages. Seven genotypes with resistance to the pathogen at the seedling stage had very high levels of resistance at adult plant stage (Golegaonkar et al., 2009).

The genotype difference in terms of days to permanent wilting was greatest when compared with other characters as indicated by the highest F-test ratios for genotypes. Therefore, this character could be useful for screening of the crop for resistance to S. rolfsii. In addition, this trait was also found to be useful for screening 91 genotypes of Jerusalem artichoke in the previous work (Sennoi et al., 2012).

Genotypes of Jerusalem artichoke that exhibited resistance both seedling and adult plant stages were also identified in the present study, based on days to permanent wilting, JA 98 and HEL 65 were more resistant to S. rolfsii at both seedling and adult plant stages and based on disease incidence, JA 6 was also more resistant at adult and seedling stages. These genotypes are useful for further breeding for S. rolfsii resistance in Jerusalem artichoke.

ACKNOWLEDGEMENTS This research was funded by a grant from the program Strategic Scholarships for Frontier Research Network for the Joint Ph.D. Program Thai Doctoral Degree from the Office of the Higher Education Commission, Thailand. Grateful acknowledgement is due to the Thailand Research Fund (TRF), the Commission for Higher Education (CHE) and Khon Kaen University (KKU) for providing financial support to this research through the Distinguished Research Professor Grant of Professor Dr Aran Patanothai and the Peanut and Jerusalem Artichoke Improvement for Functional Food Research Group, KKU and Plant Breeding Research Center for Sustainable Agricultural, Khon Kaen, Thailand. The Plant Gene Resource of Canada and Leibniz Institute of Plant Genetics and Crop Plant Research, Germany, are acknowledged for their donation of Jerusalem artichoke germplasm. REFERENCES Agrios GN (2005). Plant Pathology 5th ed. Academic

Press, London, 599 p. Akram A, Iqbal SM, Rauf CA, Aleem R (2008).

Detection of resistant sources for collar rot disease in chickpea germplasm. Pak. J. Bot. 40(5): 2211–2215.

Dickson MH, Petzoldt R (1993). Plant age and isolate source affect expression of downy mildew resistance in broccoli. Hort. Sci. 28(7): 730–731.

Fery RL, Dukes Sr. PD (2002). Southern blight (Sclerotium rolfsii Sacc.) of cowpea: yield-loss estimates and sources of resistance. Crop Prot. 21: 403–408.

Franke MD, Brenneman TB, Holbrook CC (1999). Identification of resistance to Rhizoctonia

limb rot in a core collection of peanut germplasm. Plant Dis. 83: 944–948.

Prashant GG, Davinder S, Robert FP (2009). Evaluation of seedling and adult plant resistance to Puccinia hordei in barley. Euphytica. 166: 183–197.

Hoshmand, A.R. 2006. Design of experiments for agriculture and the natural sciences. 2nd ed. Chapman & Hall/CRC, Boca Raton, Florida, 19 p.

Hovmøller MS (2007). Sources of seedling and adult plant resistance to Puccinia striiformis f.sp. tritici in European wheats. Plant Breed. 126: 225–233.

Kema GHJ, van Silfhout CH (1997). Genetic variation for virulence and resistance in the wheat-Mycosphaerella graminicola pathosystem. III. Comparative seedling and


331

adult plant experiments. Phytopathology 87: 266–272.

Koike ST (2004). Southern blight of Jerusalem artichoke caused by Sclerotium rolfsii in California. Plant Dis. 88: 769.

Li H, Smyth F, Barbetti MJ, Sivasithamparam K (2006). Relationship between Brassica napus seedling and adult plant responses to Leptosphaeria maculans is determined by plant growth stage at inoculation and temperature regime. Field Crops Res. 96: 428–437.

Marcel TC, Gorguet B, Ta MT, Kohutova Z, Vels A, Niks RE (2008). Isolate specificity of quantitative trait loci for partial resistance of barley to Puccinia hordei confirmed in mapping populations and near-isogenic lines. New Phytol. 177(3): 743–755.

Mccarter SM, Kays SJ (1984). Disease limiting production of Jerusalem artichoke in Georgia. Plant Dis. 68: 299–302.

Golegaonkar PG, Davinder S, Robert FP (2009). Evaluation of seedling and adult plant resistance to Puccinia hordei in barley. Euphytica 166: 183–197.

Sennoi R, Jogloy S, Saksirirat W, Patanothai A (2010). Pathogenicity test of Sclerotium rolfsii, a causal agent of Jerusalem artichoke (Helianthus tuberosus L.) stem rot. Asian J. Plant Sci. 95: 281–284.

Sennoi R, Jogloy S, Saksirirat W, Kesmala T, Patanothai A (2012). Genotypic variation of resistance to southern stem rot of Jerusalem artichoke caused by Sclerotium rolfsii. Euphytica. Online, DOI: 10.1007/s10681-012-0813-y

Shew BB, Wynne JC, Beute MK (1987). Field, microplot, and greenhouse evaluation of resistance to Sclerotium rolfsii in peanut. Plant Dis. 71: 188–191.

Torabi M, Nazari K (1998). Seedling and adult plant resistance to yellow rust in Iranian bread wheats. Euphytica. 100: 51–54.

Wang ZL, Li LH, He ZH, Duan XY, Zhou YL, Chen XM, Lillemo M, Singh RP, Wang H, Xia XC (2005). Seedling and adult plant resistance to powdery mildew in Chinese bread wheat cultivars and lines. Plant Dis. 89: 457–463.

Williams KJ, Platz GJ, Barr AR, Cheong J, Willsmore K, Cakir M, Wallwork H. (2003). A comparison of the genetics of seedling and adult plant resistance to spot form of net blotch (Pyrenophora teres f. maculata). Aust. J. Agric. Res. 54: 1387–1394.

Xu Z, Gleason ML, Mueller DS (2009). Development of rapid method using oxalic acid to assess resistance among host cultivars to petiole rot caused by Sclerotium rolfsii var. delphinii. Plant Health Progress. Online, DOI: 10.1094/PHP-2009-0128-01-RS.


RESPONSE TO FIVE CYCLES OF MODIFIED MASS SELECTION FOR EAR LENGTH IN WAXY CORN

K. SENAMONTRY1, K. LERTRAT1, 2 and B. SURIHARN1,2*

1Department of Plant Science and Agricultural Resources, Faculty of Agriculture,

Khon Kaen University, Khon Kaen, 40002 Thailand 2Plant Breeding Research Center for Sustainable Agriculture,

Khon Kaen University, Khon Kaen, 40002 Thailand *Corresponding author’s e-mail: [email protected].

SUMMARY

The objectives of this study were to evaluate the response to modified mass selection for long ears of a waxy corn population, Khon Kaen Composite #1, and to investigate correlations between ear length and yield and other important traits. Five cycles of modified mass selection for ear length were evaluated in the rainy season of 2011 and the dry season of 2011/12. A randomized complete block design with four replications was used in this study. Five cycles of modified mass selection could increase ear length from 17.7 cm to 19.3 cm, and the selection gain per cycle was 0.39 cm. The correlation coefficient between ear length and unhusked ear yield was positive and significant (r = 0.85**); that between ear length and husked ear yield was also positive and significant (r = 0.50**). The results indicate that selection for ear length could increase ear yield in Khon Kaen composite #1 population. The improved population can be released as an open-pollinated variety and used as a germplasm source for yield improvement in waxy corn. Keywords: Zea mays L. var. ceratina, crop breeding, population improvement, response to selection.

Manuscript received: November 7, 2012; Decision on manuscript: April 3, 2013; Manuscript accepted: April 26, 2013© Society for the Advancement of Breeding Research in Asia and Oceania (SABRAO) 2013


INTRODUCTION

Waxy or glutinous corn (Zea mays L. var. ceratina) is commercially grown as green corn in several Asian countries such as Thailand, China, Japan, Vietnam and Korea. Glutinous corn is similar to glutinous rice for its glutinous starchy endosperm and their taste is very similar. In Thailand and other countries in Asia, small- scale farmers have been growing waxy vegetable corn as a cash crop after rice harvest for more than a century. Currently, most glutinous corns and varieties grown in Asia are

open-pollinated local cultivars. These cultivars differ in agronomic characteristics, ear size, ear shape, kernel color and eating quality (Lertrat and Thongnarin, 2006). More recently, Thailand has exported hybrid seeds and frozen waxy corn products. The market for corn products is expanding locally and internationally. For the best table quality, however, waxy corn needs a very short harvest period and shorter un-processed storability compared with sweet corn (Simla et al., 2009).

RESEARCH ARTICLE


333

Corn breeding has been one of the main growing techniques to meet the increasing demand for this cereal for the past 50 years. Through the years, farmers have grown different types of cultivars based on new technology developments for corn improvement and cultural practices to maximize corn production. For example, farmers in different countries have used different types of corn varieties that suit their socioeconomic situations. Unlike field corn and super sweet corn, less attention has been given to waxy vegetable corn improvement. However, agronomic practices for this crop have been improved to meet the demand for this crop. Furthermore, the incorporation of genes that control sweetness, tenderness, kernel colors, ear size and other useful characteristics can diversify waxy corn products and increase market potential.

Selection for yield per se is difficult because of the complex inheritance of multiple genes that control yield and the interaction of numerous physiological processes. Selection for traits with simple inheritance that are closely correlated with yield should improve yield. Mass selection is the simplest and most inexpensive method in population improvement in cross- pollinated crops and this method could be effective for those traits that can be identified before or at the time of flowering (Vasal et al., 2004). Modified mass selection for prolificacy has been used effectively for yield improvement in waxy corn (Kesornkeaw et al., 2009). Ear length is an important yield component in some maize populations (Robinson et al., 1951; William et al., 1965). Studies have shown that yield could be improved by selection for ear length in field maize and super sweet corn (Jadhav et al., 1995; Fountain and Hallauer, 1996; Ali and Saleh, 2003). Positive association between yield and yield components indicates the correlated responses of yield to selection for ear length, and, therefore, improvement of yield by selecting longer ears is possible. However, Salazar and Hallauer (1986) reported that selection for long ears was not effective for improving yield in maize.

The contrasting results among previous investigations lead us to hypothesize that the response to selection may be due to differences in genetic variation in different populations. As

the waxy corn population Khon Kaen Composite #1 has good variation for ear length, selection for ear length in this population might improve yield. Therefore, the objectives of this study were to evaluate selection response to five cycles of modified mass selection for long ears of Khon Kaen Composite #1 waxy corn population and correlated responses in yields and yield components and to study the correlations between long ears and yield and yield-related traits. MATERIALS AND METHODS Plant material The waxy corn population Khon Kaen Composite #1 was developed by combining germplasm from five sources, including three commercial varieties from Vietnam (DALAD, NU58 and TN), a hybrid from China, an open-pollinated population 919-Knw previously developed from our breeding program, two local varieties in Thailand and a commercial hybrid in Thailand. The germplasm from Vietnam represents good eating quality. The hybrid from China has good yield, good stand ability, good eating quality and long ears. The open-pollinated population 919-Knw has good yield, good adaptation, good stand ability and long ears. The local varieties have good adaptation and good eating quality. The commercial hybrid in Thailand has good yield and good stand ability. Population development and selection procedures Khon Kaen Composite #1 (C0) (hereafter referred to as KKU-Vwx) was developed in the 2007 rainy season and 2007 dry season at the Faculty of Agriculture, Khon Kaen University, Thailand. The F2 generation of the hybrids was used in the crossing program. Each source of waxy corn population was grown at interrow spacing of 0.75 m and intrarow spacing of 0.25 m. The plots consisted of four rows 5 m long and 3.2 m wide for each population. Three days after silking, bulked pollens from all plants were used to pollinate all ears of all plants. At harvest, every pollinated ear from hand pollination was

K. Senamontry et al (2013)

334

harvested. This population was used as an original population for further mass selection.

Modified mass selection of KKU-Vwx population was initiated in 2008 (Figure 1). Selection for five cycles was conducted by growing plants in the breeding nursery at the Research Farm, Faculty of Agriculture, Khon Kaen University. The selected plants were initiated by bulked pollen collected from long- ear plants with good stand ability and are disease- free. The plants were harvested and the seeds from long ears with good husk cover were bulked to produce the next population cycle. The fifth cycle (C5) was completed in 2010 and a total of six populations, including the initial population and five improved populations, were made available for evaluation. Remnant seeds of these early populations were stored in a cold room to minimize differences in seed vigor. This experiment had only five improved populations because yield and ear length of the fifth cycle did not differ from those observed in the fourth cycle.

Field experiment Six populations (C0 – C5) of waxy corn were evaluated for two seasons in the rainy season of 2011 and the dry season of 2011/2012 at the Research Farm of Khon Kaen University (16o

47′ N, 102o81′ E, 199 m asl). A randomized complete block design with four replications was used. Each plot consisted of 4 - 5 rows m long; spacing was 0.75 m between rows and 0.25 m between plants, giving a plant population density of 5.33 plants m-2 (80 plants/plot).

Conventional tillage was practiced for soil preparation, and 15-15-15 fertilizer as a basal dose at the rate of 171 kg ha-1 was incorporated into the soil during soil preparation. The seeds were over-planted and the seedlings were later thinned to obtain the desired stand at seedling stage. Two splits of 15-15-15 fertilizer at the rate of 93.75 kg ha-1 plus urea (46-0-0) at the rate of 93.75 kg ha-1 for first split and 15-15-15 fertilizer at the rate of 125 kg ha-1 plus urea at the rate of 62.5 kg ha-1 for second split were applied to the crop 14 days after planting (DAP) and 30 DAP, respectively. At flowering stage, 13-13-21 fertilizer was applied at the rate of 156.25 kg ha-1. Therefore, the total dose of

fertilizers was 150.65 kg ha-1 N, 78.78 kg ha-1 P and 91.27 kg ha-1 K, respectively. Sprinkler irrigation was supplied regularly to avoid drought stress, and insect pests, diseases and weeds were properly managed to optimize growth and yield. Data collection Data from each plot were recorded for un-husked ear weight (kg/ha), husked ear weight (kg/ha), ear diameter (cm), ear length (cm), days to tasseling, days to silking, ear height (cm), plant height (cm), days to tasseling, days to silking and days to harvest. All plants in each plot were used to evaluate days to tasseling, days to silking and days to harvest. After tasseling, plant height and ear height were also recorded from 10 randomly chosen plants in each plot. Plant height was measured from ground level to leaf collar, whereas ear height was measured from ground level to the node of the top ear. Harvest was done 18 days after silking in the rainy season. Number of days to harvest was calculated as days to 50% silking plus 18-20 days by harvesting two center rows of each plot. Data analysis Analysis of variance according to a randomized complete block design (Gomez and Gomez, 1984) was performed for husked ear weight, un-husked ear weight, ear diameter, ear length, days to tasseling, ear height and plant height to evaluate the effects of five cycles of modified mass selection for ear length on these traits. Mean separation was obtained by applying a LSD (P ≤ 0.05) test using MSTAT-C software (Russel 1994). Regression analysis was used to evaluate correlated responses to selection of studied traits.

Variances of two seasons were tested for variance homogeneity. The error variances of two seasons were tested for homogeneity, and the results indicate that variances for all characters under investigation were homogeneous as determined by F ratios, which were not larger than 3 (data not shown). Combined analysis of variance was then performed for each trait, and, when the main effect was significant, mean comparison was


335

carried out using an LSD (P ≤ 0.05). Response to selection was determined by simple linear regression. Difference for zero of b values was determined by T test. RESULTS Combined analysis of variance Significant differences between seasons were found for unhusked ear weight (kg/ha), husked ear weight (kg/ha), days to tasseling and days to silking, whereas differences between seasons

for ear length, ear diameter, ear height and plant height were not statistically significant (Table 1). Differences among cycles were significant for ear length, unhusked ear weight (kg/ha), husked ear weight (kg/ha), ear diameter and days to tasseling, whereas the differences were not statistically significant for days to silking, ear height and plant height. The interactions between season and cycle were significant for unhusked ear weight (kg/ha) and husked ear weight (kg/ha), whereas the interactions between season and cycle were not significant for ear diameter, days to tasseling, days to silking, ear height and plant height.

Table 1. Mean squares for ear length, unhusked ear weight, husked ear weight, ear diameter, days to tasseling, days to silking, ear height and plant height of five cycles of waxy corn population KKU-Vwx evaluated in the 2011 rainy season and 2011/12 dry season. 1 5 5 30 DF Characters Season (S) Cycles (C) S x C Pooled error C.V.% Ear length 1.2ns 5.2** 0.5ns 0.4 3.6 Unhusked ear weight 17,190.0** 9,996.2** 1,805.6* 18,056.80 8 Husked ear weight 16,001.0** 13,600.0* 1,654.8* 19,352.60 12.4 Ear diameter 0.3ns 0.1* 0.4ns 0.8 3.9 Days to tasseling 208.3** 3.1* 2.6ns 1.3 2.5 Days to silking 645.3** 2.6ns 3.1ns 2.3 3.2 Ear height 0.3ns 74.1ns 22.5ns 85.6 11.7 Plant height 1.4 ns 174.2ns 35.5 ns 153.6 7.4 ns, *, ** not significant and significant at 0.05 and 0.01 probability levels, respectively Comparison among cycles The data of two seasons were combined because error variances were homogeneous. Mean comparison among the cycles was conducted for yield traits, yield components and agronomic traits. However, where a large cycle and season interaction was presented, the analysis of separate seasons was reported.

Ear lengths ranging from 17.7 cm of the original population to 19.8 cm of cycle 4 were observed among cycles (Table 2). However, cycle 4 was not significantly different from cycle 5 (19.3 cm). Most improved cycles (cycles 2-5) were significantly higher than the original population, except for cycle 1 (18.0 cm). The

results indicated that five cycles of selection for long ears could increase ear length. Unhusked ear weights ranging from 15,793 kg ha-1in the original population to 19,025 kg ha-1 in cycle 4 were observed among populations. Cycle 2 (17178 kg ha-1), cycle 3 (17,414 kg ha-1), cycle 4 (19,025 kg ha-1) and cycle 5 (17,512 kg ha-1) were significantly higher than original population, whereas cycle 1 (16,309 kg ha-1) was not. For this result, cycle 5 was significantly lower than cycle 4, this could be explained by variation of yield trait in this experiment.

Similarly, husked ear yields ranging from 9,462 kg ha-1 in the original population to12,905 kg ha-1 in cycle 4 were recorded among the populations. Cycle 2 (11,042 kg ha-1), cycle


336

3 (11,369 kg ha-1), cycle 4 (12,905 kg ha-1) and cycle 5 (11,927 kg ha-1) were significantly higher than original population, whereas cycle 1 (10,429 kg ha-1) was not. The results indicated that selection for ear length could also increase unhusked ear weight and husked ear weight.

Ear diameters ranging from 4.4 cm in the original population to 4.7 cm in cycles 4 and 5, and all improved populations (4.6-4.7 cm) were significantly higher than original population (Table 2). The results indicated that selection for ear length could increase ear diameter.

Differences among populations were not significant for days to tasseling (47.0 days), days to silking (47.3 days), ear height (79.1 cm) and plant height (166.9 cm) (Table 2). The results indicated that selection for ear length in this study did not significantly affect these traits.

Response to mass selection Responses to five cycles of modified mass selection for ear length were evaluated for two seasons. In this study, direct response for ear length and correlated responses for yield and other agronomic traits were also evaluated. Significant b values indicated significant selection gains per cycle for the traits under investigation (Table 2).

It is clear that five cycles of selection for ear length could increase ear length per se of 0.40 cm per cycle as indicated by the significant b value (0.40*) (Table 2 , Fig. 2a). Selection for ear length could also increase unhusked ear weight and husked ear weight by 485.1 kg ha-

1cycle-1and 573.8 kg ha-1 cycle -1, respectively (Table 2; Figure 2c, 2d). However, selection for ear length did not significantly affect the best unhusked ear weight, best husked ear weight, kernel fresh weight and shelling percentage.

Table 2. Ear length, un-husked ear weight, husked ear weight, ear diameter, days totasseling, days to silking, ear height and plant height of five cycles of waxy corn population KKU-Vwx evaluated in the rainy season 2011 and dry season 2011/12. Seasons Ear

length (cm)

Unhusked ear weight (kg/ha)

Husked ear weight (kg/ha)

Ear diameter (cm)

Days to tasseling

Days to silking

Ear height (cm)

Plant height (cm)

Rainy 18.5 14,813b 9,362b 4.6 44.9b 43.7b 79.1 166.7 Dry 18.9 19,598a 13,015a 4.7 49.0a 51.0a 79 167.1 F-test Ns ** ** ns ** ** ns ns Cycles C0 17.7e 15,793c 9,462d 4.4b 46.7 47 79.1 163.7 C1 18.0de 16,309bc 10,429cd 4.6a 45.8 46.8 76.4 165.7 C2 18.6cd 17,178b 11,042bc 4.7a 47.6 47.7 77.3 162.4 C3 18.9bc 17,414b 11,369bc 4.6a 47.2 48.3 77.8 167.5 C4 19.8a 19,025a 12,905a 4.7a 47.1 47.1 78.9 166.7 C5 19.3ab 17,512b 11,927ab 4.7a 47.37 47.1 84.9 175.6 Mean 18.7 17,205 11,189 4.6 47 47.3 79.1 166.9 F-test2 ** ** ** ** ns ns ns ns LSD0.05 0.7 1,372.20 1,420.50 0.18 1.19 1.57 9.44 12.65 C.V (%) 3.6 7.81 12.43 3.91 2.5 3.26 11.7 7.42 b-value 1 0.40** 485.11* 573.76** 0.04** 0.18ns 0.06ns 1.08ns 1.93ns

ns, *, ** non-significant and significant at 0.05 and 0.01 probability levels, respectively.1/ b-values indicate difference from zero.

Selection for ear length could increase ear diameter by 0.04 cm per cycle (Table 2 and Fig. 2b), and it could also increase number of kernels per row of kernels per cycle (Table 2). Selection for ear length was not able to increase kernel depth and crop diameter. Selection for ear length did not significantly affect days to tasseling, days to silking, ear height and plant height.

Correlations Correlation coefficients among the 8 traits under investigation are presented in Table 3. The correlation coefficients indicated the correlated responses as affected by five cycles of modified mass selection for ear length.

Ear length was positively and significantly correlated with unhusked ear weight (0.60**), husked ear weight (0.50*), ear diameter (0.49*), kernel fresh weight (0.41*) and days to silking (0.48*). Unhusked ear weight was closely related to husked ear weight (0.60**) and ear diameter (0.48*); husked ear weight was closely related with ear diameter (0.47*) and days to silking (0.42*). The results indicated that most traits related to yield were interrelated, and selection for ear length had positive effects on these traits to some extent, depending on the trait.

Although the differences among populations were not significant for agronomic traits, there were significant correlation coefficients between ear length and days to silking (0.48*) (Table 3). Days to tasseling was significantly and positively correlated with days to silking (0.53*). The results indicated that selection for ear length or yield might increase maturity. Ear height was significantly and positively correlated with plant height (0.89**). Selection for low ear placement might result in shorter plants.

DISCUSSION The assumption tested in this investigation is that longer ears should provide higher number of kernels in the ears and ultimately provide higher yield if other yield components are not significantly affected. Therefore, the assumption

was tested in a waxy corn population, Khon Kaen Composite #1 in evaluating the direct response to five cycles of modified mass selection and the correlated responses of other yield components.

The results supported the assumption that selection for long ears could increase ear length by 0.39 cm per cycle. Many previous studies confirm our results, except for a few investigations. These findings were in agreement with previous studies in field maize (Cortez-Mendoza and Hallauer, 1979; Jadhav et al., 1995; Fountain and Hallauer, 1996; Rahman et al., 2007) and super sweet corn (Ali and Saleh, 2003). Furthermore, mass selection has been used successfully for improvement of corn yield (Maita and Coors, 1996; Bletsos and Goulas, 1998; Kesornkeaw et al., 2009), maturity (Troyer and Brown, 1976), prolificacy (Leon de and Coors, 2002; Kesornkeaw et al., 2009), leaf angle (Ariyanayagam et al., 1974), plant height and ear length (Cortez- Mendoza and Hallauer, 1979; Rahman et al., 2007).

However, the results were inconsistent with what Salazar and Hallauer (1986) found: that selection for ear length to improve yield was not effective in maize. The contrasting results of different studies could be mainly due to differences in genetic background of the corn populations. Low genetic variation and environmental feature can also be the reason for selection failure.. The success in selection would be due mainly to genetic variation in the populations because the method has been used successfully in many crops such as maize (Cortez-Mendoza and Hallauer, 1979; Bletsos and Goulas, 1998 Salazar and Hallauer 1986; De Leon and Coors, 2002), sweet corn (Ali and Saleh, 2003) and waxy corn (Kesornkeaw et al. 2009).

Ear length was also positively correlated with husked ear yield and unhusked ear yield of waxy corn. The results are rather convincing for the use of modified mass selection to improve yield in waxy corn because selection for long ears is much easier than selection for yield per se.

In conclusion, five cycles of modified mass selection for ear length were completed in KKU-Vwx waxy corn population. Ear lengths ranged from 17.7 cm in the initial population


338

(C0) to 19.8 cm in the fourth population cycle (C4). The selection gain was estimated at 0.39 cm per cycle. Positively correlated responses to selection were also observed for best unhusked ear weight (0.58*), best husked ear weight (0.45*), unhusked ear weight (0.60**), husked ear weight (0.50*), ear diameter (0.49*), kernel fresh ear weight (0.41*), days to silking (0.48*)

and days to harvest. The changes are favorable as selection for ear length can ultimately increase yield. This information is useful for corn breeders, who want to improve ear yield by using ear length as a surrogate trait. Moreover, this population can be released as an open-pollinated variety and used as a source of the long-ear trait in waxy corn breeding.

Table 3. Correlations among ear length, unhusked ear weight, husked ear weight, ear diameter, days to tasseling, days to silking, ear height and plant height of five cycles of waxy corn population KKU-Vwx evaluated in the 2011 rainy season and the 2011/12dry season.

Ear length Unhusked ear weight

Husked ear weight

Ear diameter

Days to tasseling

Days to silking

Ear height

Un-husked ear weight 0.60** Husked ear weight 0.50* 0.61**

Ear diameter 0.49* 0.48* 0.47*

Days to tasseling 0.17 -0.04 0.35 0.14

Days to silking 0.48* 0.21 0.42* 0.01 0.53*

Ear height 0.28 -0.12 0.08 0.17 0.02 0.06

Plant height 0.22 -0.12 0.03 0.17 0.11 0.01 0.89**

*, ** significant at 0.05 and 0.01 probability levels, respectively Figure 1. Diagram for population improvement of KKU-Vwx waxy corn population through modified five-cycle mass selection for long ears from 2008 to 2010.

KKU-VWX waxy corn population (C0)

Cycle 1 (C1)

Cycle 2 (C2)

Cycle 5 (C5)

Cycle 4 (C4)

Cycle 3 (C3)

2 stages of selection 1) Silking stage: long eared plants, good

standability and disease-free 2) Dry ear stage: long ears with good filled tip


339

Figure 2. Responses to five cycles of modified mass selection for ear length (a), ear diameter (b), unhusked ear weight (c) and husked ear weight (d) in KKU-VWX waxy corn population in the rainy of season 2011. ACKNOWLEDGEMENTS We are grateful for the financial support provided by the National Agricultural and Forestry Research Institute, the Swedish International Development Agency, and the National Science and Technology Development Agency, Thailand and the Plant Breeding Research Center for Sustainable Agriculture, Faculty of Agriculture, Khon Kaen University, Thailand. REFERENCES Ali ES, Saleh GB (2003). Response of two cycles of

phenotypic mass selection and heritability on two tropical sweet corn (Zea mays

L. saccharata) populations. Asian J. Plant Sci. 2: 65–70.

Ariyanayagam RP, Moore CL, Carangal VR (1974). Selection for leaf angle in maize and its effect on grain yield and other characters. Crop Sci. 14: 551–556.

Bletsos E, Goulas KK (1998). Mass selection for improvement of grain yield and protein in maize population. Crop Sci. 39: 1302–1305.

Cortez-Mendoza H, Hallauer AR (1979). Divergent mass selection for ear length in maize. Crop Sci. 19: 175–178.

De Leon N, Coors JG (2002). Twenty-four cycles of mass selection for prolificacy in the golden glow maize population. Crop Sci. 42: 325–333.

a) Ear length

cm

17.0

17.5

18.0

18.5

19.0

19.5

20.0b) Ear diameter

cm

4.3

4.4

4.5

4.6

4.7

4.8

c) Un-husked ear weight

Cycles

C0 C1 C2 C3 C4 C5

Yie

ld (k

g ha

-1)

15000

16000

17000

18000

19000

20000d) Husked ear weight

Cycles

C0 C1 C2 C3 C4 C5

Yie

ld (k

g ha

-1)

9000

10000

11000

12000

13000

14000

y = 0.40x + 17.34

r² = 0.862

y = 0.04x + 4.446

r² = 0.604

y = 485.11x + 15507

r² = 0.659

y = 573.76x + 9181

r² = 0.810


340

Fountain MO, Hallauer AR (1996). Genetic variation within maize breeding populations. Crop Sci. 36: 26–32.

Gomez KA, Gomez AA (1984). Statistical procedures for agricultural research. 2nd ed., John Wiley and Sons, Singapore. pp. 137-186.

Jadhav BS, Bhosaleand AS, Patil BR (1995). Correlation studies in irrigated rabi maize. J. Maharashtra Agric. Univ. 20: 94–96.

Kesornkeaw P, Lertrat K, Suriharn B (2009). Response to four cycles of mass selection for prolificacy at low and high population densities in small ear waxy corn. Asian J. Plant Sci. 8: 425–432.

Lertrat K, Thongnarin N (2006). Novel approach to eating quality improvement in local waxy corn: improvement of sweet corn taste in local waxy corn variety with mixed kernels from super sweet corn. Acta Hortic. 769: 145–150.

Maita R, Coors JG (1996). Twenty cycles of biparental mass selection for pollinated in the open-pollinated maize population golden glow. Crop Sci. 36: 1527–1532.

Rahman H, Khalil IH, Durrishahwar NI, Rati A (2007). Comparison for original and selected

maize populations for grain yield traits. Sarhad J. Agri. 23: 641–644.

Robinson HF, Comstock RE, Harvey PH (1951). Genotypic and phenotypic correlation in corn and their implications in selection. Agron. J. 43: 282–287.

Russel OF (1994). MSTAT—C v.2.1 (a computer based data analysis software). Crop and Soil Science Department, Michigan State University, USA.

Salazar AM, Hallauer AR (1986). Divergent mass selection for ear length in maize. Brazil. J. Genet. 9: 1-14.

Simla S, Lertrat K, Suriharn B (2009). Gene effects of sugar composition in waxy corn. Asian J. Plant Sci. 8: 417–424.

Vasal SK, Singh NN, Dhillon BS, Patil SJ (2004). Population Improvement Strategies for Crop Improvement. In: Jain HK and Kharkwan MC, eds, Narosa Publishing House, New Delhi, India pp 391- 406.

Williams JC, Penny LH, Sprague GF (1965). Full-sib and half-sib estimates of genetic variance in an open-pollinated variety of corn. Crop Sci. 5: 125–129.

x

SABRAO BOARD 2010-2013 (Executive positions in bold)

Prof. Sang-Nag Ahn, Vice President 2. Department of Crop Science, College of Agricultural and Life Sciences, Chungnam National University, Daejeon 305-764, Republic of Korea.

Dr. P. Banks, Leslie Research Centre, P.O. Box 2282, Toowoomba QLD 4350, Australia.

Dr. D.S. Brar, Associate Secretary General, IRRI, DAPO Box 7777, Metro Manila, Philippines.

Dr. Bui Chi Buu, Institute of Agricultural Sciences for Southern Vietnam, 121 Nguyen Binh Khiem, District I, Ho Chi Minh City, Vietnam.

Dr. Bertrand C. Y. Collard, Editor-in-Chief, International Rice Research Institute, DAPO Box 7777, Metro Manila, Philippines.

Dr. Georgia Eizenga, USDA-ARS Dale Bumpers National Rice Research Center, 2890 Hwy. 130 East, Stuttgart, AR 72160, USA.

Dr. Y. Fukuta, Associate Secretary-General, Japan International Research Center for Agricultural Sciences, 1-1, Ohwashi, Tsukuba, 305-8686, Japan

Dr. Tianfu Han, Institute of Crop Science, Chinese Academy of Agricultural Sciences, 12 Zhongguanchun South Street, Beijing 100081, China.

Prof. H. Ikehashi. Laboratory of Plant Genetics and Breeding, Nihon University, Kameino 1866, Fujisawa, Kanagawa 252-8510, Japan.

Dr. T. Imbe, National Institute of Crop Science, 2-1-18, Kannondai, Tsukuba, Ibaraki, 305-8518, Japan.

Dr. E. Javier, AVRDC – The World Vegetable Center, P.O. Box 42, Shanhua, Tainan 74151, Taiwan.

Dr. S. Jogloy, Deputy Editor-in-chief, Department of Plant Science and Agricultural Resources, Faculty of Agriculture, Khon Kaen University, Khon Kaen 40002, Thailand.

Dr. Kyu-Seong Lee, National Institute of Crop Science (NICS), 151, Seodun-dong Gwonseon-gu, Suwon, Gyeonggi-do, Republic of Korea.

Dr. G.S. Khush, University of California, Davis, California 95616, USA.

Prof. F. Kikuchi, Past President, 1077-31 Yatabe, Tsukuba, Ibaraki 305-8572, Japan.

Prof. S. Lamseejan. Department of Applied Radiation and Isotopes Kasetsart University, Bangkok 10900, Thailand.

Dr. K. Lertrat, Treasurer, Department of Plant Science and Agricultural Resources, Faculty of Agriculture, Khon Kaen University, Khon Kaen 40002, Thailand.

Dr. David J. Mackill, Vice President 1, IRRI, DAPO Box 7777, Metro Manila, Philippines.

Dr. H.P. Moon, Past President, Hanjin-Hyundai A. 106-604, Hwaseo 2-dong, Paldalgu, Suweon 440-152, Republic of Korea.

xi

Dr. R. Ohsawa, Institute of Agriculture and Forestry, University of Tsukuba, 1-1-1, Tennodai, Tsukuba, Ibaraki, 305-8572, Japan.

Dr. K. Okuno, Laboratory of Plant Genetics and Breeding Science, Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba 305-8572, Japan.

Dr. Mohamad bin Osman, School of Environmental and Natural Resource Sciences, Faculty of Science and Technology, Universiti Kebangsaan Malaysia (UKM), 43600 Bangi, Selangor, Malaysia.

Dr. E.D. Redoña, Secretary General, IRRI, DAPO Box 7777, Metro Manila, Philippines.

Dr. L.S. Sebastian, Philippine Rice Research Institute, Maligaya, Munoz, Nueva Ecija, Philippines.

Prof. Dr. Peerasak Srinives, President, Department of Agronomy, Kasetsart University, Kamphaeng Saen, Nakhon Pathom 73140, Thailand.

Mr. T. Takatoshi, Faculty of Agriculture, Kyoto University, Oiwake, Kitashirakawa, Sakyo-ku, Kyoto, 606-8502, Japan.

Dr. B.C. Viraktamath, Vice-President 3, Directorate of Rice Research, Rajendranagar, Hyderabad-500 030, AP, India.

xii

INSTRUCTIONS FOR AUTHORS Articles submitted to the SABRAO Journal of Breeding and Genetics must be original reports of merit dealing with any phase of plant breeding or genetics not previously or simultaneously submitted to, or published in, any other scientific or technical journal. It is an essential requirement that all of the authors have agreed to submit the manuscript to the journal.

Articles should be submitted directly to the Editor-in-Chief by email with the words ‘manuscript submission’ in the subject heading. The reviewing process will only be initiated after terms and conditions (described below) have been accepted by the corresponding author. Every manuscript will be assigned a 4 digit number usually within 7 days (e.g. MS13-26).

Publication fees The journal mainly publishes articles for SABRAO members and it is strongly preferred that at least one author should be a current member of the society. From January 2012, there is a US$50 publication fee FOR ALL ARTICLES (including SABRAO members), which must be paid before publication after acceptance of the article. This publication fee is used to pay for journal printing costs and to maintain the website. Non-members may also publish in the journal for a publication fee of US$ 100 per article. Types of articles The following types of articles are acceptable to the SABRAO Journal of Breeding and Genetics:

• research articles (describing research that expands the existing knowledge in a specific area) • short communications (concise articles describing preliminary results) • review papers (thorough review of literature with interpretations) • opinions (personal reflections) • tutorials (clear descriptions of topics to communicate specific research topics to a broad audience)

Review articles should be discussed with a member of the Editorial Board prior to submissions. Language The official language of the Journal is English. It is expected that manuscripts are clearly written with a high standard of English. Manuscripts may be written using British or US English, provided there is consistency for the entire article. A compulsory fee of US$25 per article will be charged for articles requiring extensive editing by the Publication Team prior to publication. Format Authors should follow the Journal format as closely as possible with respect to headings, formatting and references. The corresponding author's email addresses should be included on the manuscript. After acceptance of the article, it is expected that the corresponding author will re-submit the manuscript following the SABRAO J. Breed. Genet. template format, which will be provided by the Communicating Editor, and is available from the website.

xiii

A research article or short communication manuscript will usually contain the following parts: 1. TITLE - as concise and descriptive as possible, usually less than 20 words. Include the scientific names of the species studied if one or two are involved. 2. AUTHOR'S NAMES - each followed by a superscript number referring to the respective addresses of the author(s). 3. KEYWORDS - Six to eight keywords allowing the subject to be classified in retrieval systems. The words may occur in the title, and may occur in pairs, e.g., acid soils. 4. SUMMARY - should be concise and be completely self-explanatory and should cover (under 300 words) the aim, methods, major findings and at least one conclusion of the study. The significance of the findings for plant breeders or scientists should be clearly indicated. 5. INTRODUCTION - should briefly describe the subject area and context of research, with a summary of previous reports, including citations of the most significant ones. Sufficient background information should be included about the crop species for a broad readership. Point out the deficiencies in knowledge left by previous studies, then state which experiments have been designed and conducted to add new knowledge. 6. MATERIALS AND METHODS - describe the origin and nature of the materials used. Generally this section should be concise. Procedures used, experimental design, and methods of data analysis should be presented. This section can be concise citing appropriate references instead of lengthy descriptions of methods used. Statistical software packages used for data analysis (indicating the version and software distributor) must be indicated in this section. 7. RESULTS - present the key parts of the experimental data, referring to figures and tables as necessary. Do not repeat information in the text if it is shown in a table or figure. Use only the metric system of measurements. Place figures and tables on separate pages at the end of the paper, giving captions for figures and headings for tables which make them self-explanatory. 8. DISCUSSION - should be separate from the Results section, and should not repeat information already presented elsewhere. It should start with a sentence or two stating the main new findings of the research. This section should also include comparisons made with the results and inferences of previous, related studies. Criticisms of earlier studies are appropriate if they clarify the field. The remaining gaps in knowledge may be briefly pointed out, with or without an outline of future experiments which may provide some of the answers. 9. ACKNOWLEDGEMENTS - should be included if they are due to any person or organization (especially for funding support). 10. REFERENCES - Examples of text citations are: (Yoshida, 1996); Smith and Jones (1993); (Lucas et al., 1997). In the References section, the citations should be arranged alphabetically by first author, then by second and later authors, and then by year. The references should be given in the format shown below. Note that the author’s surname (or family name) should always be indicated first, followed by initials with no full-stops or periods. Abbreviations should follow abbreviations described by ISI Thomson Reuters Journal articles The format for references is as follows:

xiv

Authors (Year). Title of article. Journal title using standard abbreviations. Vol: 101-112. Nair RM, Schafleitner R, Kenyon L, Srinivasan R, Easdown W, Ebert AW, Hanson P (2012). Genetic improvement of mungbean. SABRAO J. Breed. Genet. 44 (2): 177-190. Finlay KW, Wilkinson GN (1963). The analysis of adaptation in a plant breeding program. Aust. J. Agric. Res. 14: 742–754. Book chapters: Yoshida M, Smith KJ, Jones D B (1989). Title of chapter. In: A. Lucas. B. Mason, and C. Baker, eds., Title of Book. Publisher, City, p. 45-70. Conference Proceedings (should only be cited if widely available): Yoshida M, Smith KJ, Jones DB (1989). Title of paper. In: S. Iyama, and G. Takeda, eds., Proc. Sixth Inter. Cong. SABRAO, August 21-25, 1989, Tsukuba, Japan. National Organizing Committee, Tsukuba. pp. 209-212. Book: Yoshida M, Smith KJ, Jones DB (1989). Title of Book. Publisher, City. Theses: Jones AB (1989). Title of thesis. Ph. D. Thesis. University, City. Write out one-word journal titles in full. Use standard abbreviations for multiple-word journal titles. Articles that have been accepted for publication can be included and designated “in press”. Unpublished data, submitted articles, and personal communications may be included in the text in parentheses. REPRINTS No printed reprints are provided but a pdf file can be freely-obtained from the website. FIGURES Colour figures and photographs can be included free of charge in the pdf file but will incur additional costs in the printed issue. Authors should ensure that colour figures and photos are compatible with black and white printing. REVIEWERS Papers will be refereed to international standards usually by two independent experts, so content must be novel, well proven by careful examination, clearly expressed, and concise as possible. In order to speed up the reviewing process, the corresponding author may nominate up to 5 potential reviewers by provide their email and institutional address. The senior and corresponding author should not have published with these reviewers within the last 3 years. The nomination of at least one overseas reviewer is encouraged but not essential. SUBMISSION Manuscripts should be submitted by email to the Editor-in-Chief: [email protected]

xv

MANUSCRIPT REVIEWING/PROCESSING TIME Generally it takes about 6 months for a decision on a manuscript to be made from the date of submission which is comparable to many other journals in the same area. The time required depends on many factors including the topic of the article and promptness of reviewers. Generally articles are accepted in revised form within 1 month after a decision is made and published in the next issue of the journal. Authors should be aware of typical processing times before submitting manuscripts in the journal. For further information, please contact: Dr. Bertrand (Bert) Collard Editor-in-Chief SABRAO Journal of Breeding and Genetics Society for the Advancement of Breeding Research in Asia and Oceania Email: [email protected] Website: http://www.sabrao.org/ Current address: Plant Breeding, Genetics and Biotechnology Division (PBGB) International Rice Research Institute (IRRI), DAPO Box 7777, Metro Manilla, Philippines T: +63 2 580 5600 2478; E: [email protected]

Volume 45 No. 2 June 2013

Documents