Volume 24, Horticultural Reviews

HORTICULTURAL REVIEWS Volume 24

Horticultural Reviews is sponsored by: American Society for Horticultural Science

Editorial Board, VoluDle 24 Matthew A. Jenks John \T. Possingham Margaret Sedgley

HORTICULTURAL REVIEWS Volume 24

edited by

Jules Janick Purdue University

John Wiley 8' Sons, Inc. NEW YORK / CHICHESTER / WEINHEIM / BRISBANE / SINGAPORE / TORONTO

This book is printed on acid-free paper.

0

Copyright © 2000 by John Wiley & Sons, Inc. All rights reserved. Published simultaneously in Canada. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4744. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 605 Third Avenue, New York, NY 10158-0012, (212) 850-6011, fax (212) 850-6008, E-Mail: PERMREQ @ WILEYCOM. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold with the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional person should be sought. Library of Congress Catalog Card Number: 79-642829 ISBN 0-471-33374-3 ISSN 0163-7851 Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1

Contents

Contributors

ix

Dedication: Haruyuki Kamemoto

xi

Richard A. Criley 1. Bioreactor Technology for Plant Micropropagation Meira Ziv

I. II. III. IV. V. VI.

Introduction Plant Developmental Pathways in Bioreactors Plant Cell and Tissue Growth in Bioreactors Physical and Chemical Factors in Liquid Cultures Cell and Aggregate Density, Foaming and Medium Rheology in Bioreactors Summary and Conclusions Literature Cited

2. Biogenesis of Floral Scents Natalia Dudareva, Birgit Piechulla, and Eran Pichersky I. II. III. IV. V.

Introduction Pathways and Site of Synthesis Molecular Genetic Control Variation in Biosynthesis and Emission Over Time Conclusions Literature Cited

1

1

6 12

14 22 23

24 31

32

34 38 42 50 51 v

CONTENTS

vi

3. Triazoles as Plant Growth Regulators and Stress

Protectants

55

R. Austin Fletcher, Angela Gilley, Narendra Sankhla, and Tim D. Davis I. II. III. IV. V. VI. VII. VIII.

Introduction Translocation and Efficacy of Application Methods General Plant Responses to Triazoles Mode of Action Stress Protection Potential and Current Applications A Novel Seed Treatment Technology Summary Literature Cited

4. Ecologically-based Practices for Vegetable Crops Production in the Tropics Hector R. Valenzuela I. II. III. IV. V.

Introduction Integrated Cultural Management Nutrient Management and Soil Conservation Ecologically-based Pest Management Conclusions and Future Prospects Literature Cited

5. Lettuce Seed Germination Daniel J. Cantliffe, Yu Sung, and Warley M. Nascimento

I. II. III. IV. V. VI. VII. VIII.

Introduction Seed Structure Germination Environmental Factors Affecting Germination Restriction of Lettuce Seed Germination at High Temperature Increasing Thermotolerance in Lettuce Seed Changes in the Embryo and Endosperm During Germination Summary and Conclusion Literature Cited

56 59 61 66 76 81 115 116 118

139 140 146 156 172 201 203

229 229 231 232 233 236 242 254 263 264

vii

CONTENTS

6. Viroid Dwarfing for High Density Citrus Plantings Ronald J. Hutton, Patricia Broadbent, and Kenneth B. Bevington

277

Introduction Causal Agent Use of Viroids for Tree Size Control Vegetative Growth Reproductive Growth Intensive Viroid-Dwarfed Plantings Economic Considerations Management Summary and Conclusions Literature Cited

278 280 284 291 295 297 303 306 311 312

7. Growth, Development, and Cultural Practices for Young Citrus Trees

319

I. II. III. IV. V. VI. VII. VIII. IX.

Frederick S. Davies and James J. Ferguson I. II. III. IV. V. VI. VII. VIII. IX.

Introduction Growth and Development Tree Selection and Planting Irrigation Fertilization Freeze Hardiness and Protection Pruning Biotic Factors Summary Literature Cited

8. Fruit Growth Measurement and Analysis Lin us U. Opara I. II. III. IV. V. VI.

Introduction Fruit Developmental Stages Indices of Fruit Growth Growth Measurement Techniques Approaches to Fruit Growth Analysis Applications of Fruit Growth Data

320 321 329 330 338 343 347 348 361 362

373 374 379 383 388 390 409

CONTENTS

viii

VII. VIII.

Some Problems in Fruit Growth Measurement and Analysis Summary and Prospects Literature Cited

413 414 419

Subject Index

433

Cumulative Index

434

Cumulative Contributor Index

456

Contributors Kenneth B. Bevington, NSW Agriculture, Agricultural Research and Advisory Station, Dareton, Australia Patricia Broadbent, NSW Agriculture, Elizabeth Macarther Agricultural Institute, Menangle, Australia Daniel J. Cantliffe, University of Florida, IFAS, Horticultural Sciences Department, Gainesville, FL 32611-0690 Richard A. CrUey, Department of Horticulture, University of Hawaii, Honolulu, HI 96822 Frederick S. Davies, Horticultural Sciences Department, University of Florida, Gainesville, FL 32611-0690 Tim D. Davis, Texas A & M University, Texas Agricultural Experiment Station, Dallas, TX 75252 Natalia Dudareva, Horticulture Department, Purdue University, West Lafayette, IN 47907 James J. Ferguson, Horticultural Sciences Department, University of Florida, Gainesville, FL 32611-0690 R. Austin Fletcher, Department of Environmental Biology, University of Guelph, Guelph, Ontario N1G 2W1, Canada Angela Gilley, Department of Environmental Biology, University of Guelph, Guelph, Ontario N1G 2W1, Canada Ronald J. Hutton, NSW Agriculture, Yanco Agricultural Institute, Yanco, Australia Warley M. Nascimento, University of Florida, IFAS, Horticultural Sciences Department, Gainesville, FL 32611-0690 Linus U. Opara, Bioproducts Quality Research, Center for Postharvest & Refrigeration Research, Institute of Technology and Engineering, Massey University, Palmerston North, New Zealand Eran Pichersky, Biology Department, University of Michigan, Ann Arbor, MI 48109 Birgit Piechulla, Department of Molecular Physiology of Plants and Microorganisms, Gertrudenstrasse 11a, University of Rostock, Rostock 18051, Germany Narendra Sankhla, Texas A & M University, Texas Agricultural Experiment Station, Dallas, TX 75252 Yu Sung, University of Florida, IFAS, Horticultural Sciences Department, Gainesville, FL 32611-0690 Hector R. Valenzuela, Department of Horticulture, University of Hawaii at Manoa, Honolulu, HI 96822-2279 Meira Ziv, Agricultural Botany & the Warburg Center for Biotechnology in Agriculture, The Hebrew University of Jerusalem, Rehovot, 76100, Israel ix

Haruyuki Kamemoto

Dedication: Haruyuki Kamemoto Haruyuki Kamemoto is a distinguished horticulturist who has made significant contributions to the floriculture industry of Hawaii and the tropical world. His life's work has concentrated on breeding and development of anthuriums and dendrobium orchids at the University of Hawaii, where he is affectionately known by all as Kammy. His "official" retirement in 1986 to Professor Emeritus status did not slow his pace, as he merely continued his efforts without the distractions of committee meetings and other academic duties on a contractual part-time basis (if every day is considered part-time!) until his "final" retirement on June 30,1999.

Born in Honolulu in 1922 and raised on a family farm not far from the present-day campus of the University of Hawaii, Kammy enrolled in the College of Tropical Agriculture and earned both his BS and MS degrees there. In 1947, he enrolled for Ph.D. studies at Cornell University, where he was a member of the three K's of floriculture: Harry Kohl, Tony Kofranek, and Harry Kamemoto. He took great pride in achieving his degree in floriculture-plant breeding under the tutelage of several faculty. Returning to Honolulu in 1950, he was hired by the University of Hawaii as an Assistant Professor in the Horticulture Department and launched far-sighted flower breeding programs. Four decades plus of research into the culture and breeding of anthuriurns and dendrobium orchids propelled these two flowers into the leading floricultural crops of Hawaii. On the anthurium side, an anthracnoseresistant pink cultivar, 'Marian Seefurth', was released from his breeding program in 1963, followed by more than 20 new cultivars, including many interspecific hybrids and the "tulip" type anthuriums. His interspecific anthurium breeding program stimulated others to follow, and a number of potted and cut varieties now grace florist shops and cut flower stands around the world. UH-developed anthuriums now constitute a major part of Hawaii's anthurium industry. More than two dozen cut flower and potted dendrobium cultivars have originated through his work, which produced the first seed-propagated dendrobium cultivar, 'Uniwai Blush' in 1972. Many of these cultivars have become standards of Hawaii's orchid industry because of their high yields, long vaselife, and array of colors from white through dark xi

xii

DEDICATION: HARUYUKI KAMEMOTO

purple-violet. His potted dendrobiums have become a mainstay of Hawaii's potted floricultural exports. Dr. Kamemoto's program has been productive in turning out fine graduate students, as well, and he credits them with doing much of the "nitty-gritty" that led to the many orchid and anthurium releases: counting chromosomes, making crosses, growing out and maintaining the plants, recording and crunching data, starting tissue cultures, and authoring publications. No one left his program without knowing basic research methods as well as how to repot orchids and anthuriums. He has advised twenty-four students in master's programs and eleven students in doctoral programs. They have come from many parts of the world and many of them have returned to academia. Their appreciation of his guidance and knowledge instigated many invitations to make presentations in other parts of the world, as well as referrals of potential graduate students to his program. For many years, he taught the floriculture production course at the University of Hawaii and an Orchidology course, one of the few available in any university setting. Kammy is deeply concerned about the floriculture industry and has made innumerable presentations to grower groups to inform, to educate, to cajole, to incite (on occasion), and to develop the dendrobium and anthurium industries of Hawaii and elsewhere. The Kamemoto Scholarship, established in 1986 to help outstanding students interested in careers in floriculture, has been well-supported by former students, colleagues, friends, and the Hawaii floriculture industry because of their admiration and respect for his contributions. Dr. Kamemoto's valued contributions of seed-produced dendrobium orchid hybrids that are free of disease and his novel interspecific anthurium hybrids have made Hawaii's orchid and anthurium industries successful and competitive worldwide. Demands for his expertise have led to consultantships in India, Thailand, Okinawa, and Taiwan, and he has contributed to conferences and other horticultural meetings throughout Asia, as well as in the Caribbean, South America, Australia, Israel, and Europe. His contributions have been recognized by many awards from industry, professional organizations of which he is a member, and the University of Hawaii. His honors include the Alex Laurie award (1984) and Society of American Florists Hall of Fame (1991), Fellow of the American Society of Horticultural Science (1979), ASHS Distinguished Graduate Teaching Award, Norman Jay Coleman Award of the American Association of Nurserymen (1984), University of Hawaii Board of Regents Medal for Excellence in Research (1978), and the College of Tropical Agriculture and Human Resources Outstanding Alumnus

DEDICATION: HARUYUKI KAMEMOTO

xiii

Award (1995). Societies have been profuse in recognizing his accomplishments, and he is a Fellow of the Orchid Society of Southeast Asia, Gold Medalist of Achievement from the Malayan Orchid Society and the American Orchid Society. Life Memberships in the Japan Orchid Society, American Orchid Society, and Orchid Society of Thailand, and the American Anthurium Society have been bestowed upon him. Breeding a new cultivar has never been the end product in itself. Kammy believed that you had to learn how to grow it, write about it, promote it, and use it to advance to the next level. He has authored over 220 papers in books, journals, newsletters, and, with Dr. Heidi Kuehnle, his successor, has co-authored Breeding Anthuriums in Hawaii (University of Hawaii Press, 1996). They have recently completed a second book, Breeding Dendrobiums in Hawaii, which highlights his unique breeding technique using amphidiploids to create bi- and tri-genomic combinations that can be readily reproduced from seed and features the many colorful and unique varieties that have been developed at the University of Hawaii, including the beautiful Dendrobium Ethel Kamemoto 'Splendor', named in honor of Kammy's wife. Kammy's life has been one of devotion to his profession, his students, his university, and his community. We proudly dedicate this volume of Horticultural Reviews to Haruyuki Kamemoto to honor a lifetime of accomplishments. Richard A. Criley Department of Horticulture University of Hawaii

1 Bioreactor Technology for Plant Micropropagation Meira Ziv Agricultural Botany and the Warburg Center for Biotechnology in Agriculture The Hebrew University of Jerusalem Rehovot 76100, Israel

I. Introduction II. Plant Developmental Pathways in Bioreactors A. Somatic Embryogenesis B. Organogenic Pathway C. Bud or Meristem Clusters D. Anomalous Plant Morphogenesis III. Plant Cell and Tissue Growth in Bioreactors IV. Physical and Chemical Factors in Liquid Cultures A. The Gaseous Atmosphere 1. Oxygen Level 2. CO 2 Effects 3. Ethylene B. Mineral Nutrients Consumption C. Carbohydrate Supply and Utilization D. pH Effects E. Growth Regulator Effects F. Temperature Effects V. Cell and Aggregate Density, Foaming and Medium Rheology in Bioreactors VI. Summary and Conclusions Literature Cited

Horticultural Reviews, Volume 24, Edited by Jules Janick ISBN 0-471 0

B

A

15

2

ea :.:s==

I

-48

•

I

-24

,

c

t:J

>-

~ -< .>to

"i:I

t.rJ n

:::c:

C

"'-

10

Z t:J

t""' t""'

?i

0

i

24

i

48

i

72

I0

96 Hours

Time post-anthesis

Levels of linalool emission, LIS activity (B), LIS mRNA and LIS protein (A) during flower development in petals of Clarkia breweri.

z>-

t:J

~ "i:I

n :::c: tTl

~ C/J ~

0-
50% initial reduction), goals which are also deemed realistic for implementation in the U.S. (Pimentel et al. 1991). Furthermore, low-input pest controls are necessary for the vast majority of farmers throughout the tropics who have little or no access to the capital needed to purchase pesticides or other high-cost capital inputs (such as fertilizers, hybrid seed, and machinery). However, for the development and implementation of alternative pest management practices, a greater knowledge is required of pest biology, environmental interactions, and consequently greater management skills for implementation (McSorley 1996). B. Nutrient by Pest Interactions 1. Diseases. Healthy and vigorous crop growth often promotes tolerance to pest attack (Marschner 1995). Plants exposed to an unbalanced

nutrient supply may become more susceptible to arthropod pest, disease, and/or weed infestations. For example, excessive N applications have resulted in a greater incidence of rust, powdery mildew, and bacterial diseases (van Bruggen 1995). Also, in Florida, high P application rates resulted in a greater incidence of Helminthosporium in sweet corn (Stoner 1951). Nutrient x disease interactions, being affected by microorganisms, environmental variables, and the host, are consequently highly dynamic and complex, and thus are poorly understood. However, a better understanding of host nutrition and its effects on disease development offer the potential for modification of fertilizer application programs for management of important diseases (Huber 1981). Nutrients such as N, K, P, Ca, and others have been reported to improve plant tolerance to disease attack (O'Rourke and Millar 1966; Canaday and Wyatt 1992; Yamazaki and Hoshina 1995). For example, modifications in the application rates of complete fertilizers resulted in less phytophthora root rot, soybean mosaic virus, and stem canker in soybean (Pacumbaba et al. 1997), and in a reduced incidence of early blight, Alternaria solani, in tomato (Jones and Jones 1986). Calcium fertilizer levels affect plant tolerance to disease, not only in the field, but also during the postharvest phase (Conway et al. 1994; Yamazaki and Hoshina 1995). Along these lines, the literature covering crop disease x nutrient interactions is extensive, but due to the many variables involved it is still difficult to make generalizations. Thus location-specific nutrient x disease interactions need to be evaluated on a case-by-case basis (Huber 1981).

174

H.VALENZUELA

Nutrient applications may enhance plant tolerance to pest attack (Marschner 1995). In some cases nutrients may directly inhibit pathogenic growth, or even promote activity of biocontrol organisms, as was shown with zinc for the biological control of tomato crown and root rot, Fusarium oxysporum (Duffy and Defago 1997). Nutrient foliar sprays may also induce systemic resistance to disease, as shown with phosphate salts, resulting in powdery mildew suppression and enhanced growth in cucumber (Reuveni et al. 1993). The mechanism for the induced disease resistance is unknown, but the potential exists for expanded use of these disease control approaches, as the control mechanism becomes better understood and as the application technology is adapted to commercial growing conditions (Huber 1981). 2. Arthropod Pests. Nutritional imbalances and excessive fertilization may result in increased arthropod pest numbers in plants. For example, artemisia plants fertilized with ammonium nitrate showed greater phloem- and seed-feeding insect numbers than unfertilized plants (Strauss 1987). Examples of other pests that showed increased numbers in response to higher N application rates include the corn leafhopper, Dalbulus maidis (Power 1987); the potato leafhopper (Roltsch and Gage 1990); caterpillar pests in head cabbage (H. R. Valenzuela, unpublished data; Jansson et al. 1991); and increased aphid population growth rates in potato (Jansson and Smilowitz 1986). In soybean, high P application rates resulted in greater larval velvetbean caterpillar, Anticarsia gemmatalis, numbers than at the lower Prates. Southern stink bug, Nezara viridula, levels were also greater with the higher P rates, but only in one of two years (Funderburk et al. 1991). The fertilizer source and farming technique may also affect arthropod pest levels. In New York, collards that received organic amendments had less arthropod pest numbers than plants receiving inorganic fertilizers (Culliney and Pimentel 1986), a result also observed in sorghum (Strauss 1987). In Ohio, soils from organic farms showed biological "buffering," resulting in fewer European corn borer oviposition levels, compared to comparable soils from conventional farms. Lower pest levels were found in plants grown in the soils from organic farms than from conventional farms even after application of several fertilizer sources and despite similar plant growth in both treatments (Phelan et al. 1995). However, additional work is necessary to evaluate the effect of organic amendments on arthropod pest pressure, to better understand existing ecological interactions that explain the observed results, and to corroborate the tentative data available from the few studies conducted to date in this area.

4. ECOLOGICALLY-BASED PRACTICES FOR VEGETABLE CROPS PRODUCTION

175

C. Habitat Manipulation Techniques 1. Polycultures Systems

Diseases. Some polyculture designs may reduce disease pressure. This was first reported in Europe in 1767 with the use of non-host aphid intercrop barriers, which reduced virus transmission on the cash crop (Batra 1982). This assumption was further backed up by early work with agronomic crops that showed less disease in fields planted with both multiline disease resistant and susceptible cultivars (Leonard 1969). Since then multi-line cropping was established in several areas for management of diseases such as Puccinia, Septoria, Rhynchosporium, and Pseudocercosporella in cereals (Wolfe 1990). This approach also showed potential in Colombia, as observed in Phaseolus, with respect to Uromyces rust (Panse et al. 1997). Commercial examples of this approach are seen in India as well as in Germany, where over 350,000 ha of multi-line spring barley is grown, resulting in less fungicide applications (Wolfe 1990). Polycultures may also be designed to minimize crop disease outbreaks. Work in Kenya showed less halo blight, common mosaic, anthracnose, common blight, and angular leaf spot in polyculture beans than in monocultures (Francis 1985). However, Boudreau and Mundt's (1992) studies showed reduced Uromyces bean rust pressure in bean/maize polyculture than in bean monocultures in some cases, but not in others, indicating the importance of season and site-specific effects. The mechanism for reduced disease levels found in some polycultures, rather than in monocultures, is yet to be unraveled. However, intimate interactions may exist between plant species grown in close association. For example, fungal isolates obtained from the zoysiagrass, Zoysia tenuifolia, rhizosphere enhanced growth and induced systemic resistance in cucumber to anthracnose attack (Meera et al. 1994). Arthropods. Research initiated in the 1940s showed reduced arthropod pest pressure (Willey 1979a; Batra 1982; Risch et al. 1983; Francis 1985; Risch 1987; Andow 1991a), a lower probability of exceeding economic injury levels (Andow 1991b) and thus a decreased incidence of arthropod (Risch 1987) pest outbreaks in some diversified systems than in monocultures. However, many exceptions do occur, as observed with the higher pod-sucking bug, Clavigralla spp., levels found in cowpea intercropped with maize, than in cowpea monoculture (Gethi and Khaemba 1991). In the U.S., strip intercropping was practiced before the advent of pesticides, to fight pest attack. Mixed plantings to prevent pest outbreaks

176

H. VALENZUELA

is also a common practice in forestry (Batra 1982). However, knowledge gaps are extensive in the area of habitat manipulation concerning pest population dynamics. From a regional perspective (Ricklefs 1987), little is understood about how larger landscape biotic and abiotic factors, such as those caused by extensive monocultures, affect smaller microhabitat biodiversity (Altieri and Letourneau 1984; Dennis and Fry 1992). Interplanting insect resistant and susceptible cultivars may also result in less damage to susceptible cultivars, as observed with ground arthropod pests in sweetpotato (Schalk et al. 1992), and with respect to Liriomyza leafminer damage in potato (Midmore and Alcazar 1991). By selecting compatible early- and late-maturing potato cultivars with differential tolerance to pest attack, overall pest pressure was reduced, and resource utilization apparently improved, resulting in a 20% overall increased productivity in the multi-line plots than with the singlecultivar treatments (Midmore and Alcazar 1991). The high level of potential biological interactions that exist in polycultures makes it difficult to explain pest dynamics resulting in lower pest levels or in possible pest outbreaks. The high degree of arthropod interactions observed in unsprayed crop monocultures alone (Ellington et al. 1997) give an indication of the challenge to develop a systematic understanding of the underlying mechanisms for possible pest management through enhanced biocontrol in polycultures. The level of complexity in monocultures, focused on the population dynamics of a single pest species on an area-wide basis, was shown by Brazzle et al. (1997), who evaluated ten cultural management practices that affected whitefly population dynamics in 56 commercial cotton fields in the Imperial Valley, California. Whitefly numbers were affected by planting dates, number of pesticide applications, plot size, and proximity of muskmelon fields. Similarly, with regard to black rot disease management, a survey of 27 strawberry farms in New York found several cultural factors that were correlated with disease severity, including soil compaction, soil texture, flat-bed culture, herbicide treatments, and years under monoculture (Wing et al. 1995). This intricate level of complexity rapidly increases, even in simple polycultures (Andow 1991b). Thus, a polyculture consisting of 2 crop species, 6 herbivore species, and 6 beneficial arthropod species results in a system with 91 potential two-way and 364 potential three-way ecological interactions (Andow 1991a). Similarly, the many facets that need to be evaluated, dealing with crop growth alone, in the highly complex agroforestry systems typically found in many tropical areas was described by Huxley (1985). Reduced pest loads in diverse systems may result due to the presence of alternative hosts, alterations of the abiotic environment such

4. ECOLOGICALLY-BASED PRACTICES FOR VEGETABLE CROPS PRODUCTION

177

as physical barriers or shade (Roltsch and Gage 1990), release of chemical signals that disrupt or delay host identification (Teulon et al. 1993; Dobson 1994), presence of alternative preys, as well as lower host numbers (resource concentration hypothesis), and increased natural enemy activity, as compared to monocultures (Risch 1987; Hunter and Aarssen 1988), as well as a combination of several factors. As reviews indicate (Risch et al. 1983), broad generalizations cannot be made with regard to arthropod population dynamics in response to habitat manipulation, but instead an understanding should exist of the pest's dispersal and demographic behavior, prior to making predictions regarding response patterns in polycultures (Kareiva 1987). Both plant species and the plant structural diversity existent in polycultures may explain the greater arthropod diversity and correspondent lower pest damage levels reported for some polycultures. Altieri (1984) evaluated this supposition in brussel sprout density and polyculture studies. Doubling the planting density alone did not alter the arthropod fauna, but increasing the plot complexity by intercropping brussel sprouts with either faba beans or wild mustard did increase the diversity of the arthropod population in the intercrops, compared to the monocultures. A more diverse arthropod community may have been supported by the greater plant structural diversity in the polycultures, and not by increased plot biomass production alone, since simply doubling the planting density did not increase the monoculture arthropod faunal composition as observed in the polycultures. The decreased pest pressure observed with many arthropod pests in polycultures is attributable, in part, to a greater activity of beneficials (Altieri and Letourneau 1984). For example, in Texas, relay intercropping of cotton with wheat, mustard, and sorghum resulted in predator conservation during fallow periods and in respective higher predator numbers and lower cotton aphid, Aphis gossypii, levels in the polycultures than in the isolated cotton monocultures (Parajulee et al. 1997). Also, greater generalist predatory spider numbers were found in squash/ cowpea/maize polycultures than in squash monocultures, supporting Root's "enemies" hypothesis, while specialist leafhopper Anagrus parasitoids showed no response to the habitat manipulation treatments (Letourneau 1990a). However, Letourneau's work (1990a,b) indicates that variables other than response to species diversity and density may be involved in arthropod response to polycultures. Additional variables such as recruitment, tenure time, oviposition patterns, reproduction, emigration rates, short-distance movements, in response to aspects of polycultures (such as structure, color contrasts, volatiles, and vegetation texture) other than species richness itself, need to be identified, so that

178

H. VALENZUELA

specific polyculture schemes may be designed that will result in greater natural enemy activity and in corresponding decreased pest levels (Letourneau 1990b). A greater activity of beneficiaIs is not always the mechanism responsible for the lower pest numbers found in polycultures, as determined with the lower Empoasca potato leafhopper numbers found in a tomato/ bean system compared to the numbers found in bean monocultures by Roltsch and Gage (1990) in Michigan. This latter study, which supports the resource concentration hypothesis, showed lower leafhopper numbers as the tomato density increased, thereby decreasing relative bean densities, plus the higher tomato densities may have also decreased the "quality" of the bean foliage, as indicated by parallel N-nutrient greenhouse studies (Roltsch and Gage 1990). Therefore, non-host intercrops may provide a physical barrier, or modify the microenvironment in other manners, disrupting the motility of target pests. For example, in Nicaragua, two-year replicated and large-scale on-farm (2 ha plots each for tomato/bean and tomato monoculture plots) validation trials, a tomato/bean system also resulted in lower pest levels in tomato with respect to several pests including Heliothis fruit worms, and Liriomiza leafminers, while a trend was observed toward lower Spodoptera armyworms numbers (Rosset 1989). The mechanism of action for the reduced pest numbers was not determined in this study, but presumably the bean plants provided a physical barrier, resulting in less pest oviposition on tomato. Natural enemy activity was minimal in these plots, perhaps due to the history of intensive pesticide applications in the area (Rosset 1989). In another example, shade and wind protection provided by maize plants in a bean/maize system were the likely causative factor resulting in greater aphid dispersal rates (lower residence-time) from the bean plants in the polycultures as compared to the bean monocultures (Bottenberg and Irwin 1991). The greater dispersal rates from bean plants may result in less virus transmission in the polycultures, since the next landing site for the departing aphids may be a non-host, either a maize plant in the polyculture or another plant outside of the production area (Bottenberg and Irwin 1991). The extensive chemical volatile richness that exists in monocultures (Charron et al. 1995) such as celery (Van Wassenhove et al. 1990) and squash (Peterson et al. 1994), is increased in polycultures (Dobson 1994). This is significant, considering the role that volatiles and alkaloids play in insect and plant interactions (Roltsch and Gage 1990; Peterson et al. 1994; Charron et al. 1995). Many of these volatiles attract beneficials and can also possibly delay, detract, or prevent pests from finding their

4. ECOLOGICALLY-BASED PRACTICES FOR VEGETABLE CROPS PRODUCTION

179

target hosts in polycultures (Dobson 1994), and thus could represent yet another tool for pest management through habitat diversification. Even though lower pest numbers have been reported in polycultures than in monocultures, little is understood about possible yield advantages resulting from the reduced pest loads in polycultures (Andow 1991b). Plant compensatory response to pest damage and the resulting effects on inter- and intra-species competition in the polycultures may confound yield effects of component crop species, which precludes making any broad generalizations.

Nematodes. Even though the evidence is scant, polycultures may also be effective for suppression of nematode pests, as observed in cowpea/ maize systems in Nigeria (Bridge 1996). Also, in the West Indies, intercropping carrots reduced M. incognita levels by 16% with onion, 26% with cabbage, and 27% with chives compared to the counts found under carrot monoculture (McDonald 1985). Weeds. Polycultures may be designed to effectively manage weed populations in small-scale production systems. In a two-year study conducted in Nkpolu, Nigeria, optimal weed control and greatest maize yield were obtained in monoculture maize with herbicide applications. However, overall economic returns were highest when weeds were managed through a combination of one hand-weeding in a maize/sweetpotato polyculture or one hand-weeding with a maize/peanut system than in the respective monocultures (Zuofa and Tariah 1992). In California, polyculture combinations of lettuce, faba bean, and pea had less weed biomass than the respective monocultures (Sharaiha and Gliessman 1992), but some crop combinations and planting patterns were more effective in smothering weeds than others. Most polyculture combinations also had an LER greater than one (Sharaiha and Gliessman 1992). The selection of the crop cultivar used may also have an effect on the level of weed control obtained in polycultures, as do other cultural factors such as N and irrigation application rates, as shown in a pea/barley/weedy mustard system (Liebman 1989). 2. Cover Crops and Living Mulches. Cover crops, which provide another mechanism for enhanced vegetational diversity in the farm, showed a positive effect on beneficial arthropod populations, resulting in less arthropod pest fruit damage in Mediterranean and temperate climate orchards (Fye 1983; Altieri and Schmidt 1985). The cover crops were effective in serving as a refuge for beneficial organisms, and also as a source of alternate prey for the target pests. In Georgia, several cover

180

H. VALENZUELA

crops enhanced predator populations in muskmelon (Bugg et al. 1991a) and in pecan orchards (Bugg et al. 1991b; Bugg and Dutcher 1993). It is plausible that similar mechanisms to build up beneficial populations through the use of cover crops may be established in annual tropical cropping systems. An understanding of pest population dynamics in response to cover cropping practices will be useful in the design of cover crop and residue management techniques that will minimize the incidence of pest outbreaks. For example, incorporated legume cover crop residues resulted in greater seedcorn maggot, Delia platura, numbers in soybean (Hammond 1990). However, gradually lower maggot numbers were found with incorporated grasses, and with unincorporated (no-till) legume and grass species, with no enhanced maggot numbers in the latter two treatments, as compared to the bare-ground soybean treatments (Hammond 1990). Legumes were effective as living mulches in Brassica plots to reduce pest incidence of Brevicoryne brassicae, Pieris rapae, and Erioischia brassicae. The reduction in P. rapae was attributed to an increase in the activity of the predacious ground beetle Harpalus rufipes (Altieri and Letourneau 1984). In other studies clover living mulches resulted in a 34% increase in cabbage root fly, Delia brassicae, predator levels. Similar results were obtained in one experiment by Ryan et al. (1980). In addition Ryan et al. (1980) observed lower root fly numbers in all experiments, and increased cabbage yields in experiments when water was not a limiting factor. In similar studies, fewer aphids were found in collards intercropped with lana vetch living mulches, compared to collard monoculture plots (Altieri and Letourneau 1984). In work with maize and tomato, plots with clover living mulches had a larger number of natural enemies than bare-ground plots, including more ground predators (Carabidae, Staphylinidae, and spiders) (Altieri et al. 1985). Broccoli grown in clover living mulches also had fewer cabbage aphids and flea beetles than monoculture broccoli, but differential broccoli growth in these treatments confounded the arthropod pest data (Altieri et al. 1985). Follow-up work, in which clover living mulch growth was suppressed through mowing, to minimize competitive effects, also showed fewer aphids in the living mulch treatments than in the broccoli monocultures (Costello 1994). Living mulches, by providing a barrier to insect movement, or through other undetermined mechanisms, may also be effective in delaying the onset of aphid-transmitted viral diseases, as observed in a zucchini experiment with buckwheat, mustard, and buckwheat/mustard living mulch treatments in Hawaii (Hooks et al. 1998).

4. ECOLOGICALLY-BASED PRACTICES FOR VEGETABLE CROPS PRODUCTION

181

An important feature of living mulches is their weed suppression characteristics (Ilnicki and Enache 1992), as observed in living mulch trials with eggplant in Hawaii (Valenzuela and DeFrank 1994). Living mulches may thus be used to reduce pest outbreaks, in addition to their contribution to soil fertility and to enhanced weed control (Enache and Ilnicki 1990). Cover crop rotational programs may also be effective for disease management. For example, a Sudan grass, Sorghum vulgare, cover crop effectively reduced verticillium wilt and increased potato yields, compared to fallow treatments and cover cropping with other species (Davis et al. 1996). The lower verticillium levels caused by the Sudan grass cover crop were attributable to an increased activity of biological control organisms in the rhizosphere. Similarly, cover crops can effectively be used in rotation programs for management of important nematode pests. A list of plant species with reported resistance or tolerance to particular nematode species buildup is presented in Table 4.2. 3. Trap Crops. Arthropods often show feeding preference among cultivars or crop species, a response which may be used to develop trapcropping systems as a tool for pest management (Hokkanen 1991). For example, Liberty Hyde Bailey observed that the squash vine borer, Melittia cucurbitae, oviposited more readily on some squash species than on others (Robinson 1992), an observation also recorded by J. DeFrank and H. R. Valenzuela (unpublished data) concerning melon fly (Dacus spp.) oviposition preference in cucurbits. Other examples include strawberry cultivar susceptibility to spider mites (Poe and Howard 1971), cucurbit cultivar response to Diaphania pickleworm damage (Peterson et al. 1994), and Brassica cultivar susceptibility to caterpillar damage (Valenzuela 1993; Mays and Kok 1997). Trap-crop systems have been used for pest management in several forestry and agronomic crops (Hokkanen 1991), such as the use of an alfalfa trap crop, which effectively reduced the green mirid, Creontiades dilutus in cotton (Mensah and Khan 1997). Commercial applications of trap crops, as part of an overall pest management program, were established in temperate areas for vegetables such as potato and cauliflower, and in tropical regions with field crops such as cotton and soybean (Hokkanen 1991). In Oklahoma, systemic insecticide-treated 'Lemondrop' squash plantings representing , C'l:l 30

c

o

O-+--,----,..---y-.-.--.-;

O-t-r-r-,.--.--r---r--l

100

• • wi seedcoats o wlo seedcoats

70

60

Q)

100 . . . . - - - - - - - - - , Sour Orange 90

20 10 10 15 20 25 30 35 40 45

5 10 15202530 354045

Temperature °C

Temperature °c

Fig. 7.1. Days to emergence of seeds of rough lemon, sour orange, 'Pineapple' orange, and trifoliate orange with and without seed coats. Source: Wiltbank et al. (1995).

7. GROWTH, DEVELOPMENT, AND CULTURAL PRACTICES FOR CITRUS TREES

323

In addition, removal of the seed coats greatly reduced the time to first emergence. Again, days to emergence were increased considerably below 15° or above 40°C. Minimum temperatures for germination varied slightly from previously reported values, but were generally in the same range. Recent work by Rouse and Sherrod (1997) expanded upon Wiltbank's study using the same carefully controlled temperatures. The authors determined optimum germination temperatures for 17 citrus cultivars and related species. As expected, a range of optima occurred from 25°C for Poncirus trifoliata to 33°C for 'Rangpur' x 'Troyer'. Days to germination also varied from 5 to 28 depending on species and temperature. Stoffella et al. (1995) tested the effects of CO 2 -enriched irrigation water on seedling emergence and growth of four citrus rootstocks (rough lemon, 'Carrizo' citrange, 'Cleopatra' mandarin, and sour orange). Germination and seedling growth differences were observed among rootstocks, with rough lemon seedlings having the lowest mean days to germination. 'Carrizo' citrange seedlings had the greatest stem diameters and were tallest 72 days after planting. Daily stem diameter growth increased at 0.03 mm and shoot elongation daily rate was 0.15 mm at 27°/17°C (day/night) temperatures. Citrus seeds have hypogeous germination and usually the radicle is the first organ to emerge (Fig. 7.2A). As seeds germinate, they rapidly take up water for the first 24 to 36 h, with uptake rates reaching a plateau after that time (Mobayer 1980b). Light is not required for germination. The primary shoot is next to emerge (Fig. 7.2B). Initial growth emanates from the shoot and root apical meristem. Development of first foliage leaves and lateral branches (Fig. 7.2C) occurs a few weeks after emergence, depending largely on temperature. Initial shoot growth occurs as a single stem emanating from a single apical meristem until lateral branches form later in development (Fig. 7.2D). Temperature also has a pronounced effect on seedling growth. Peltier (1920), who was actually interested in temperature effects on growth of citrus canker, also measured growth of citrus and citrus hybrids. Optimum growth occurred at 20° to 30°C depending on species, with limited growth below about 15°C or above 35°C. He also astutely identified differences between temperature optima and minima for citrus compared with Poncirus and Poncirus hybrids. Of interest, growth of grapefruit was inhibited above 35°C, whereas Poncirus, a more temperate genus, continued to grow. However, a limited number of trees was used in each of five experiments. Similarly, Girton (1927) found that seedling stem and leaf growth rates of sweet oranges increased as temperature increased from 13° to 31°C. Temperature ranges for seedling growth in this study generally were in agreement with those of previous studies

F. DAVIES AND

324

J. FERGUSON

Apical meristem

Seed,hypogeous germination

A.~ ~

-

Cotyledons

Primary root (radicle) First foliage leaves

I

B.

c.

D.

Fibrous roots •

.0

Fig. 7.2. Germination and early growth and development of citrus seedlings. (A) Seed with primary root, (B) seed with primary shoot, (C) seedling with first foliage leaves and secondary (lateral) roots, (D) seedling during early development.

for seed germination. The minimum temperature for growth was 12°C and the maximum about 37°C. Young and Peynado (1962) examined effects of constant daytime and variable nighttime temperatures on growth and freeze hardiness of 30 citrus and related species. Minimum nighttime growth temperatures ranged from 13.8° to 15°C for Poncirus to below 8.8°C for C. macrophylla, limes, and lemons. Growth of lemon types at low soil temperatures may contribute to their greater freeze sensitivity compared to Poncirus species and hybrids (Wilcox and Davies 1981). Several studies have demonstrated the important effect of soil (root) temperature on growth of citrus trees. In a classic study, Girton (1927) exposed sour orange seedlings to several root temperatures. Growth rates were low at 13°C; optimum total root length and root elongation

7. GROWTH, DEVELOPMENT, AND CULTURAL PRACTICES FOR CITRUS TREES

325

rate occurred between 23° and 27°C. Root hair index also increased significantly with increasing temperatures until 37°C was reached. In a similar study, Halma (1935) grew rooted cuttings of sweet orange, lemon, and grapefruit at root temperatures of 16° to 27°, 12° to 22°, and 3° to 20°C for nearly a year. Lemon cuttings grew considerably more than the other cultivars and growth increased with increasing root temperature. Orange and grapefruit cuttings showed decreased growth in the 3° to 20°C range. Optimum soil temperatures were estimated to be 31° to 34°C for sweet orange and grapefruit, 34°C for rough lemon, and 32° to 34°C for sour orange. B. Budded Trees

In many commercial citrus areas, seedlings are grown in the field or the greenhouse until the stem attains 5 to 10 mm in diameter. This occurs 3 to 6 months after planting, depending on temperature, light, water, and adequate pest and disease control. The seedling (rootstock) is then budded or grafted using the desired scion cultivar. Inverted or regular T budding is the most commonly used method in most citrus regions (Davies and Albrigo 1994). A vertical slit about 2 to 4 cm long is made in the rootstock stem followed by a horizontal cut at right angles, thus completing the T shape. The bud is then removed from the scion budstick, inserted into the slit, and wrapped with plastic tape. After 1 to 2 weeks the bud has "taken," i.e. the cambium of the rootstock and scion grow together (Jackson 1991). The scion is "forced" to grow in several ways because the new bud will not grow or will grow poorly if the rootstock is allowed to grow normally. Consequently, the stem above the bud is removed entirely, cut partially through and bent over (bending) or just bent over and tied (lopping). Rouse (1988) found that the new bud grew most rapidly if the rootstock was lopped or bent. Williamson et al. (1992) found growth was best when shoots were bent rather than lopped or cut (Fig. 7.3). Photosynthate translocation from rootstock leaves was also greatest for bent vs. cut or lopped treatments, which likely contributed to increased growth. The pattern of shoot growth after forcing is interesting. From day 8 to 32 there was a rapid increase in shoot length, with an average daily growth rate of 0.66 cm for the bent treatment (first growth flush). Shoot growth ceased from day 32 to 75 and then resumed from day 75 to 120 at about 0.62 cm (second growth flush). Guazzelli et al. (1995) also measured shoot growth rate in a greenhouse nursery as part of a fertilizer study. In this instance, daily growth rates ranged from 0.21 to 0.33 cm probably due to differences in temperature between the two studies.

F. DAVIES AND

326

J. FERGUSON

48 44

Bending

40 36

E 32

Q

:5 28 O'l

c

(!)

- 24 c o

;X

20 16 12 8

4

8

16

24

32

40

48

56

64

72

80

88

96 104 112 120

Days after bud forcing

Fig. 7.3. Scion length of citrus nursery trees following bud forcing by cutting off, lopping, or bending. Data points and bars represent means ± SE (n ranged from 5 to 22 depending on measuring date). Source: Williamson et al. (1992).

Media (root) temperatures less than about 15°C inhibit shoot growth in the nursery, particularly for trees on 'Swingle' citrumelo rootstock. Increasing media temperatures from 15° to 25°C significantly increased percentage budbreak (Al-Jaleel and Williamson 1993). Thus, the authors suggested that bottom or container heating may be useful for nurserymen to improve growth of trees on 'Swingle' citrumelo rootstock. The influence of root temperature on growth was further supported in field studies using budded trees of 'Washington' navel and 'Eureka' lemon. Generally, growth ceased during the season when soil temperatures were below 18°C (Halma and Compton 1936). Air (canopy) temperature was not controlled in either study; obviously it also has an effect on tree growth. Liebig and Chapman (1963) grew 'Washington' navel oranges in the greenhouse on three rootstocks at root temperatures of 14 0, 22° or 30°C. Vegetative growth increased with increasing root temperature, but flowering was greatest at 14°C, with no flowering occurring at 30°C. When plants were moved from 30° to 14°C, they flowered. Rootstock and soil type did not affect growth, possibly due to the short duration of the study (9 months).

7. GROWTH, DEVELOPMENT, AND CULTURAL PRACTICES FOR CITRUS TREES

327

It has long been observed that citrus trees grow in flushes (Reed 1928; Reed and MacDougal 1937). The pattern and periodicity of lemon shoot growth was first defined mathematically by Reed (1928), who identified three growth flushes for mature lemon trees growing in California. Each flush showed a characteristic sigmoidal growth curve. About the same time, Crider (1927) showed that there is a distinct alternation of root and shoot growth. Young citrus trees typically have 3 to 5 growth flushes per year in subtropical areas, with almost continuous growth flushing in low tropical regions due to high heat unit accumulation (Davies and Albrigo 1994). Bevington and Castle (1985) clearly demonstrated the alternation of shoot and root growth using 13-month-old 'Valencia' orange trees growing in large chambers placed in the field; their data further suggest that root and shoot growth occurs in flushes (Fig. 7.4). In winter and spring, shoot and root growth alternate, but by late summer both occur simultaneously, probably due to high average soil temperatures. Mean root elongation rate increased linearly from 17° to 30°C.

....... 35

i

>0-

c

"C

"-

~20

-

.c ~ 0

...J

o

>

>-100 a..

o z