Brad Chapman Candelario Mondragon Jacobo Ronald A Bunch and Andrew H Paterson


Breeding and Biotechnology Research Objectives Expanding Markets Breeding

Breeding programs Germplasm Pools The Cultivated Gene Pool The Semidomesticated and Wild Gene Pool Germplasm Collections Breeding Systems and Techniques Controlled Pollination Clonal Propagation/Apomixis Future Directions Biotechnology

Overview of Plant Molecular Tools DNA and Protein Analysis Tissue Culture and Transformation Evolutionary Relationships Molecular Tools for Cacti Future Directions Conclusions Literature Cited


This chapter focuses on the highly important fruit-bearing group of the genus Opuntia, although the techniques, methods, and concepts presented are transferable to most of the Cactaceae. Because cactus germplasm can be improved via plant breeding and molecular techniques, genetic resources and current objectives for breeding and biotechnology are discussed. A summary of molecular genetic research involving cactus is followed by a plan for utilizing the work from other plant species to advance the biotechnology of cactus.

The previous chapters leave little doubt about the uniqueness and usefulness of the Cactaceae; unfortunately, the family has been underutilized as a target for improvement using plant breeding and biotechnological techniques. The multiple uses and the ability of cacti to thrive in arid and semiarid environments that many other species are unable to tolerate make it of immediate interest to breeders and molecular biologists seeking to develop crops for areas typically unsuitable for conventional agriculture. The most obvious trait of interest is the ability of cacti to be productive under water-limited conditions (Nobel 1988). Coupled to the drought-tolerant phenotype is the ability of cacti to tolerate high growth temperatures. Third, another well-studied feature of cacti is their ability to thrive under elevated atmospheric carbon dioxide levels (Nobel 1991a; Cui et al. 1993; Nobel and Israel 1994). In addition, cacti have an extremely high productivity under optimal conditions (Nobel 1991b). Cacti are economically important and physiologically interesting, making them a worthy and intriguing subject for breeding and molecular work. Because the cacti used today are primarily wild plants that have merely been domesticated, advances in plant breeding and biotechnology could extend the growing range into new environments and could generate new uses for cactus products. Defining objectives for these investigations will be an essential part of developing organized and productive breeding and molecular biological programs.

Breeding and Biotechnology Research Objectives

The broad goals of both biotechnology and breeding research for cacti are to understand the molecular mechanisms underlying interesting and useful traits and to use this understanding to develop crops that are better suited to human needs. Improvement must be intimately tied to the needs of growers, processors, and consumers. Because cacti lack the extensive breeding history of other crops, many avenues of research on the improvement of cactus germplasm are possible. The priority of research goals will vary depending on the production region and the local uses of cacti. Improvements to cactus varieties will fall into three main categories: (1) expanding the production area into new environments, (2) improving the quality and productivity of cacti to expand into new markets, and (3) adding new traits to allow the development of new uses for cacti.

Cacti are well adapted to the arid and semiarid regions where food and fodder crops are limited, although minimum temperatures substantially limit the growing range in some areas (Wang et al. 1997). Thus, a primary biotech-nological goal should be to expand the production area for cacti by producing cold-tolerant cultivars. Current high-yielding fruit, vegetable, and forage varieties are killed by temperatures of -5 to -8°C (Loik and Nobel 1991; Parish and Felker 1995; Wang et al. 1997). Nevertheless, wild opuntias have a broad geographic range, including southern Canada (Loik and Nobel 1993). Vast areas of the south-central United States and northern Mexico would be available to grow cacti if cold tolerance could be introduced to highly productive, high-quality varieties. Because researchers are beginning to elucidate the mechanisms for cold tolerance in Opuntia germplasm, the immediate objective should be to understand the genetic basis for this tolerance and begin to integrate this into commercially useful varieties with a minimal transfer of undesired characteristics. For fruit production, chilling requirements may be an important aspect for research. For instance, exposure to day/night temperatures of i5/5°C versus 35/25^ produces significantly more floral than vegetative buds for Opuntia ficus-indica (Nobel and Castañeda 1998). Genetic manipulation of the chilling requirements of temperate fruit crops is common. Employing similar techniques for Opuntia species will help expand fruit production areas to the tropics and subtropics. In addition, out-of-season production would be possible in established cactus production areas.

Increased disease and insect resistance is probably a prerequisite for introducing cactus varieties into new regions. In general, cacti are relatively tolerant of diseases and insects (Chapter 14), although several diseases attack them (Granata and Sidoti 1997; Saad et al. 1998), and more cactus diseases are likely to appear as cactus production spreads. Cultivars with both broad-based resistance and disease specific resistance will be essential for maintaining consistently high yields.

Various biological questions relating specifically to cacti potentially have long-term applied value. Understanding the molecular basis of adaptations conferring drought and high-temperature tolerance has implications for increasing drought and temperature tolerance in other crop species and is of general interest for understanding plant physiology. Similarly, cacti may prove to be important organisms to study the molecular genetics and evolutionary conservation of CAM (Cushman and Bohnert 1997). A greater understanding of cactus polyploidy will be useful for both basic and applied research. The basal number of chromosomes is n = 11 (Pinkava et al. 1985, 1992). Many wild species are diploids, although polyploids up to 2n = 19X = 209 or higher occur (Mondragon 1999). The amazing range of chromosome numbers for cacti presents an opportunity to study the basis of polyploidy that is paralleled among cultivated plants only by sugarcane. In ad dition, understanding the mechanisms behind this polyploid nature may help facilitate crosses between cacti of different ploidy levels, reducing some of the genetic barriers for hybridization. Immense opportunities exist for breeding and biotechnological studies of cacti. This chapter outlines specific research that can accomplish objectives and answer questions discussed above. Areas are emphasized where different disciplines can make important contributions.

Expanding Markets

Although cactus pears are a coveted fruit by many and a staple seasonal fruit in some areas of the world, their consumption is generally limited to ethnic groups with a historical association with the fruit. Seeds, which are typically swallowed along with the pulp, are the major deterrent to attracting new consumers to cactus pear. The seed count ranges from 80 to more than 300 per fruit (Pimienta 1990; Barbera et al. 1991), which means there are 3 to 8 grams of seeds per fruit, depending on fruit size and the cultivar (Mondragon and Perez 1996). Unbalanced gametes as a result of polyploidy can lead to the partial seed set commonly observed for O. ficus-indica (Nerd and Mizrahi 1994). The ratio between aborted and normal seeds is higher in Italian (0.44) than in Mexican cultivars (0.11; Pimienta and Mauricio 1987; Barbera et al. 1994). Seed content is positively correlated with fruit size, and large fruits command a premium price in the market. Thus the ideal fruit should have a large number of seeds to attain good size, but a high ratio of aborted to normal seeds.

The environmental and crop management factors influencing seed content have not yet been separated from genetic factors. Parthenocarpy (production of fruits without seeds because of lack of pollination or fertilization, or fertilization followed by embryo abortion) has also been mentioned as a solution to the problem of seediness in the fruit. Natural parthenocarpy was reported in BSi, a yellow fleshed accession studied in Israel that does not require pollination for fruit set and development (Weiss et al. 1993; Nerd and Mizrahi 1994). However, the degenerated seeds still contain hard arils, and the overall fruit quality is low. Several attempts have been made to reduce the size of the seeds by means of gibberellin application but with little success (Gil and Espinosa 1980; Aguilar 1987; Ortiz 1988).

Fruit size and shape as well as pulp color are important factors to consider when developing new cultivars of cactus pear. Large fruits are preferred. Also, oval or barrel-shaped fruits are easier to harvest than elongated fruits and therefore suffer less harvest damage to the stem end

(Cantwell 1991). In Mexico, green-white pulp is preferred, in Italy and northern Europe yellow-orange pulp is preferred, and in the United States and Canada red pulp is often favored. Pulp juiciness and the high content of soluble solids are considered important fruit quality factors (Wang et al. 1998). For some markets, fruits with higher acid content are favored (Saenz and Costell 1990). Post-harvest handling, packing, and storage need to be considered when developing new cultivars of cactus pear for commercial use (Corrales-Garcia et al. 1997).

Although spine density and size on the cladodes vary among accessions (collected wild varieties or cultivars), some spines are present even in so-called spineless cultivars. Spineless individuals do not occur in the wild, indicating that this trait was acquired through domestication. In regard to crop management, spines represent an inconvenience because they hamper routine operations, such as pruning, and represent an important obstacle at harvest time. Commercial varieties for fruit in Mexico are spiny (with the exception of 'Roja Lisa'). Other producing countries rely on spineless cultivars, and efforts to develop new varieties should be focused on producing spineless types.

The presence of glochids on the fruit peel is also a major constraint to increased consumption. Selection and breeding for glochid-free varieties should be a priority (Barbera 1995). Another solution is to develop cultivars with easily shed glochids. Genes for low numbers of areoles and short glochids are present in O. robusta, although the fruit of this species is not well accepted due to its low sugar content, bland flavor, and short shelf life.

For cactus used as a vegetable, the presence of spines and glochids is one of the main constraints for the development of a wider market outside Mexico. A solution to this problem is the development of spineless varieties. Another possibility is to implement early screening for low number of glochids on the cladodes as well as short, soft glochids. Moreover, cultivars of Nopalea species produce cladodes that are nearly free of spines and glochids, which can help introduce cacti as a vegetable into new markets. The ability to shed the spines and glochids before the cladodes age could be a selection criteria when breeding for new vegetable varieties. Another useful selection criteria could be the shape of the podarius (tubercle); large podaria can facilitate mechanical peeling. Suitability of new accessions to be consumed as a vegetable can be evaluated in the second year, because there is no need to wait until the reproductive stage is reached. Spines, if not properly removed, can also be a problem for feeding livestock. Breeding for increased nutritional content would be a wor thy goal to increase the availability of feed in arid and semiarid regions (Felker 1995a).


Cactus pear was domesticated in the highlands of Central Mexico. Reports of consumption of cacti date back to the ancient groups that inhabited Mesoamerica, and evidence exists that agaves and cacti have been part of the human diet for over 9,000 years (Nobel 1994). The transition to planned cultivation of Opuntia, a crucial step in the early domestication of wild plants, may have taken place before 8,000 years B.P. (Hoffman 1995). While some traditionally used species (e.g., Steneocereus spp., Hylocereus spp., and Selenicereus megalanthus) are only now being cultivated, cactus pear has been farmed for thousands of years. It is, together with corn, beans, and agave, among the oldest cultivated plants in Mexico.

Cactus pear developed into a formal crop in the 20th century as it evolved from collections of wild plants to exceptional hand-selected individuals grown in backyards. The transition of the rural population toward an urban semi-industrial society created a consumer base for the initial market of cactus pear in the 1960s. During the 1970s through the 1980s, programs initiated by the Mexican government provided an important boost to the cactus pear industry—increasing the area of commercial orchards and establishing research programs. Selection in the last few centuries has led to a number of outstanding cultivars with exceptional fruit quality and productivity, as well as drought and frost tolerance. The outcome of this long-term activity has resulted in six to eight commercial culti-vars as the basic stock for the Mexican and Italian markets, currently the most important producers and consumers worldwide.

Breeding Programs

Hybridization of cactus pears was claimed by Luther Burbank in the early 1900s, leading to the development of the so-called Burbank spineless cactus with an immense potential as cattle forage in desert areas. Several varieties produced from extensive crossing and selection among accessions from Mexico, Africa, Australia, and other countries were developed and aggressively marketed (Dreyer 1985). The lack of records and formal publications gave rise to questions about the sources of plant material and methodologies used for the released cultivars. Nevertheless, Burbank was responsible for the distribution of a tremendous amount of Opuntia germplasm throughout the world, such as the cactus pear cultivars found today in South Africa (Wessels 1988).

Modern-day cactus pear breeding in Mexico began in 1964 when the late Facundo Barrientos of the Colegio de Postgraduados, Chapingo, began his program for its genetic improvement. The ambitious program included germplasm collection and utilization, breeding of new cultivars, and development of new uses for cactus pear. He envisioned that the ideal cactus pear cultivar should be multipurpose and widely adapted, and he worked toward high fruit quality, high forage quality, drought and heat resistance, and early maturity. In the early years he introduced selections 'COPENA Vi' and 'Fi' for vegetable and forage, respectively. He developed a series of high-quality white (or green) pulped selections (the 'COPENA T' series) that had only modest commercial success as they did not present a clear advantage over the native white culti-vars. Later, he broadened his efforts to include other colored cactus pears and other species of fruiting cacti. The red pulp cultivar 'Frieda' or 'Torreoja' was his last released cultivar. Prior to his untimely death in 1993, Dr. Barrientos interacted with cactus growers and researchers throughout the world and shared with them germplasm and breeding methodologies. He created a spark that ignited the current interest in new cactus varieties and cactus breeding.

A fresh start at breeding is now underway in the United States, South Africa, Israel, and Mexico. Encouraged by the Food and Agriculture Organization of the United Nations, a renewed interest in the collection of wild and semi-domesticated accessions has developed. In addition, the number of publications with information on crop management practices and new uses of cactus pear are increasing.

Germplasm (gene pool) collection and characterization is the major field of interest in most of the Mexican institutions engaged in cactus pear research, reflecting the fact that Mexico is the main reservoir of variability of edible opuntias. A special concern is the rapid disappearance of valuable plant material from wild stocks in northcentral Mexico resulting from the intensive utilization by the cattle and dairy industry. Backyard orchards have been searched for outstanding individuals to widen the limited array of commercial cultivars for fruit and vegetable production. Three state universities and one technical school support these germplasm collection programs. Formal breeding programs, however, are conducted only by the Instituto Nacional de Investigaciones Forestales y Agropecuarias (INIFAP) and the Colegio de Postgraduados. Their breeding efforts are mostly focused on the development of spineless cultivars for fruit production. Potential expansion into new national and foreign markets emphasizes the development of colored, juicy fruits with low seed content, early fruiting habit, high soluble solids, and tolerance to indigenous pests and diseases. Vegetable and fodder cultivars are also a priority. The recurrent frosts and droughts observed during the last decade are prompting the search for new cultivars with sustained productivity under such conditions. Programs rely on hybridization and selection as breeding tools. The germplasm base for these projects is totally indigenous, and large populations of hybrids and segregants have been in field trials since 1994.

Two institutions in Mexico are pioneering the application of molecular techniques to cactus pear breeding, the Centro de Investigación y Estudios Avanzados (CINVES-TAV) in Guanajuato and the Instituto Tecnológico Agropecuario (ITA 20) in Aguascalientes. Their investigations are directed toward the study of enzymes involved in cell wall softening, with the goal of controlling ripening and thus avoiding temporal saturation of the national market. Other areas of research interest include seed storage proteins (Silos-Espino et al. 1999), cactus transformation, and gene transfer to increase protein content. Beginning in the 1980s, Texas A&M University-Kingsville became involved in agronomic research, extension work, and collection and introduction of cactus pear to the United States. The program aims to develop frost-tolerant cultivars to overcome the common limitation to growing cactus in this region. In 1996 the first round of crosses was initiated, marking the beginning of a long-term breeding program. This institution is also responsible for the popularization of the vegetable cultivar 'Spineless 1308' and various cactus products in Texas.

D'Arrigo Brothers, a produce company based in California, started a private breeding program in 1994 to improve their spineless commercial cultivar 'Andy Boy,' which is similar to the 'Rossa' cultivar grown and marketed in Italy and currently dominates the American out-of-season market. The main objective of this program is the improvement of fruit quality. The program utilizes germplasm obtained from Dr. Barrientos and has recently added a portion of the Texas A&M University-Kingsville collection as breeding material. Seedlings from the first crosses were planted into the field in 1997. Other active programs include one located in Sassari, Italy, that is involved in the improvement of fruit quality using naturalized accessions collected in the semiarid Mediterranean region of Italy. Field trials of crossed and open pollinated seedlings were initiated in 1994. In Argentina, evaluation trials of local and introduced accessions are being conducted. In addition, an ongoing collaborative project with the University of Georgia is developing a molecular marker map from progeny of a cross between two Texas A&M

accessions. In South Africa, extensive trials are being conducted to evaluate the best local cultivars. In addition, a study has begun utilizing random amplified polymorphic DNA (RAPD) markers to identify cultivars of plant material being sold to growers in that country. In Israel, evaluation of crosses and selections of fruit-producing Cereus, Hylocereus, and Selenicereus species is underway as part of a large project to develop new fruit crops for that country.

Germplasm Pools

Each of the major germplasm pools (cultivated, semido-mesticated, and wild) exhibits traits of interest for the genetic improvement of cactus pear. A wide range of ploidy levels exists among and within the Opuntia species. Pimienta and Muñoz (1995) compiled data from several authors and list reported ploidy levels of 2x, 3x, 4x, 5x, 6x, 8x, iox, iix, i2x, i3x, i9x, and 2ox. The cultivated types of cactus pear generally have higher chromosome numbers (2n = 6x = 66 and 2n = 8x = 88) than the wild populations (usually 2n = 2x = 22 and 2n = 4x = 44). Although large crossing studies have yet to be conducted, barriers to cross compatibility are minimal among the cultivated species. Natural hybrids are common in both cultivated and wild populations of cactus pear (Pimienta and Muñoz i995). Trujillo (i986) obtained viable seeds from crosses of wild O. streptacantha with O. robusta and O. leucotricha with O. cochinera. Wang et al. (i996) successfully crossed O. lind-heimeri with several cultivated cactus pears. In addition, O. robusta and cultivated cactus pear have been crossed.

The Cultivated Gene Pool

Cultivated species of cactus pear include Opuntia ficus-indica, O. albicarpa, O. streptacantha, and O. robusta plus hybrids between O. ficus-indica and the others (Pimienta and Muñoz i995). For breeding purposes, this germplasm pool is best sorted out by cultivar or specific traits of interest rather than by taxonomic species. Modern cultivars of cactus pear are the products of long-term informal, but effective, selection by growers. The most important selection criteria have been fruit size and quality, plant productivity, and tolerance to drought and frosts. The number of cultivars in each of the countries that grow cactus pear varies according to the intensity of usage and the size of the initial germplasm base, with Mexico accounting for the largest diversity.

Three main groups of cactus pear can be recognized according to the color of the peel and pulp: white (or light green), yellow (including deep orange), and red (from light red to deep purple). The most popular white cultivars in Mexico are 'Reyna,' 'Cristalina,' 'Esmeralda,' and 'Bur-

rona.' 'Reyna' sets the standard of quality and dominates the national market (Mondragon and Perez 1993). The national demand in Mexico is supplemented by the orange-pulped 'Naranjona' and 'Amarilla Montesa' and the red-purple 'Roja Lisa' (Mondragon and Perez 1993, 1996; Pimienta and Muñoz 1995). In the United States, the red-fruited cultivar 'Andy Boy' is available from September until April; it is produced only in California where irrigation and mild winters facilitate out-of-season production. In Chile, a cultivar of O. ficus-indica known as 'Verde' (Green; Sudzuki 1995) or 'Blanca' (White; Pimienta and Muñoz 1995) is the most common; since 1982, exports to the U.S. market take advantage of the summer harvest season in the Southern Hemisphere, which coincides with winter in the Northern Hemisphere. The cultivars available in the Sicilian area of Italy are 'Gialla' (Yellow), 'Bianca' (White), and 'Rossa' (Red), with 'Gialla' being the most common (Barbera et al. 1992). A 'seedless' cultivar is also known but its commercial cultivation has never been attempted because of the poor quality of the fruits. Production in Israel is based mostly on the cultivar 'Ofer,' which has yellow pulp. In South Africa the varieties available, 40% of which have light-green fruit, originated from 21 spineless types imported from the Burbank Nursery of California in 1914. Besides color, varieties also differ in other pulp characteristics, peel features, post-harvest physiology, and response to environmental factors.

Because the vegetable nopalitos are most widely consumed in Mexico, Mexico is the main source of germ-plasm. The Universidad Autónoma de Chapingo near Mexico City has assembled an extensive collection; 'Milpa Alta' (O. ficus-indica) is the most important cultivar and is cultivated mostly in the region of the same name near Mexico City. 'COPENA Vi' is perhaps the second most important vegetable cultivar. Another variety, 'COPENA Fi,' was selected for fodder production by Dr. Barrientos in the 1960s, but its tender pads are also suited to consumption as a vegetable. Both have intense green color, a thin epidermis, good flavor, and low acidity (Flores 1995). 'Moradilla,' 'Atlixco,' 'Polotitlan,' and 'Redonda' are further examples of locally selected varieties of O. ficus-indica used for vegetable production. In the southwestern states of Michoacan and Jalisco, the 'Nopal blanco' or white cactus pear is the cultivar of choice for nopalitos and tolerates humid conditions (up to 1,600 mm of rainfall annually). The cultivar 'Valtierrilla' is used as a vegetable in central Guanajuato if picked very young. The cultivar 'Spineless 1308' (O. cochellinifera or Nopalea cochellinifera), selected by Peter Felker from accessions collected in a humid tropical region of Tamaulipas, Mexico, is suitable for field cul tivation in the coastal rain-fed areas ofTexas and for greenhouse cultivation in frost-prone areas near San Antonio.

The Semi-Domesticated and Wild Gene Pool

A remarkable diversity of locally known semi-domesticated types exist in the native areas of cactus pear, some serving two or even three purposes: fruits, tender pads, and fodder. Traditionally cactus pear plantings are found on small family properties in dry regions. Opuntia hedges are concentrated near farmsteads, where they also protect fruit and vegetable gardens. Taxonomically these cactus pear plants are similar to various hybrids in the cultivated germplasm and represent the transition from wild plants to modern commercial cultivars. Their genetic variability is important as a source for new and valuable traits that may be necessary as breeding programs proceed (Hoffman 1995). Outstanding accessions in family orchards provide a source of individuals derived by chance from free natural outcrossings, whose germplasm has only been minimally collected and characterized.

The last reservoir of interesting individuals is wild populations. Several efforts have been made to collect representatives from the wild populations of fruit-producing Opuntia species in Mexico. In 1993, Mexico, Israel, and the United States joined efforts to collect in the highlands of northern Mexico and assembled 130 accessions of fruit-producing Opuntia species, selected primarily for cold hardiness (Felker 1995b). Similar endeavors have collected several accessions of the hardy Texas native Opuntia lindheimeri. Crosses among many of these species have been successful (Wang et al. 1996), although little information is available for crosses outside of this limited germplasm.

Germplasm Collections

Germplasm collections of cactus pear are maintained at several locations around the world (Table 15.1). The largest number of collections and of entries are located in Mexico, where the greatest diversity occurs for native cactus pear. In addition, each breeding project and cactus pear research program has a collection of at least some of this material (often combined with local accessions).

Due to the genetic makeup of cactus pear, its long-lived perennial habit, and large plant size, the maintenance of germplasm banks is difficult and costly. Collecting plant material based on morphological traits and common local names often leads to duplicated accessions within a collection. The morphology of Opuntia species is greatly influenced by the environment; their rapid growth means their reactions to environmental changes are also fairly rapid and more drastic than other cacti. For instance, traits

Major cactus pear germplasm collections


Collection locations

Approximate number of entries and description



Instituto Nacional de

Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP)

Palma de la Cruz, San Luis Potosi; San Luis de la Paz, Guanajuato; Sandovales, Aguascalientes; Tecamachalco, Puebla; Ensenada, Baja California Norte

800 entries, mostly from central Mexico, commercial cultivars, and backyards; 95% fruit types, 5% vegetable

Dr. Candelario Mondragon J. or M. C. Rafael Fernández M.

Universidad Autonoma de Chapingo

Chapingo, México; El Orito, Zacatecas

160 entries with a wide variety of cultivars from all over Mexico; mostly for fruit, ~24 for vegetable, and some for fodder

Ing. Claudio Flores Valdez and Dr. Clemente Gallegos F.

Secretaria de Educacion Publica (SEP) Universidad Autonoma de Nuevo Leon

Villa Hidalgo, Zacatecas Marin, Nuevo León

90 entries of various Mexican cultivars 110 entries; many for vegetable

Ing. Eloy Rodríguez Dr. Rigoberto Vázquez A.

United States

Texas A&M University, Kingsville

Kingsville, Texas

180 entries of cultivated and wild plants from northern Mexico, southwestern United States, and a few other countries; mostly fruit, some for vegetable and forage, including cold hardy types

Dr. Robert Morgan

South Africa

Department of Agriculture, Land and Environment

Pietersburg, Northern Province

80 entries of South African cultivars, selections, and chance hybrids from Burbank material

Mr. Johan Potgieter


University of Palermo


45 entries from Italy, South Africa, Chile, and Argentina; fruit

Dr. Paolo Inglese

Dipartimento di Colture

Arboree di Universita di Sassari



Dr. Innocenza Chessa

such as spininess, cladode shape and size, fruit characteristics, and plant productivity are affected by the environment (Weniger 1984). Duplication within collections is very common and complicates the evaluation and utilization of the genetic resources. A major cause of the duplication among collections is that each research group may maintain separate stocks to have access to as much genetic variability as possible. Increased characterization of these collections is vitally important to the success of cultivar improvement in cactus pear; however, data collection and analysis for these collections will be quite expensive. Only limited exchange of cactus germplasm has taken place since 1992, when the international cooperation regarding cactus pear formally began. More flexible exchange of germplasm among countries is restricted due partly to the lack of accessible data from the collections and partly to regulations. Despite these limitations, basic descriptions are available for the most important cultivars and outstanding accessions from Mexico (Pimienta and Muñoz 1995; Mondragon and Bordelon 1996), Italy, South Africa, Argentina, and Chile (Pimienta and Muñoz 1995).

An important component of developing successful cactus breeding programs will be utilizing the diversity available. Thus methods for classification of this diversity will need to be developed concurrently with cactus breeding programs (Mondragon 1999). Principal Component Analysis is a powerful multivariate statistical technique to group and visualize the relationship among cultivars based on a number of morphological measurements (Hillig and Iezzoni 1988). Such analysis of 32 cactus pear accessions from central Mexico based on 17 measured traits (Mondragon 1999) indicates a positive relationship between cladode width and fruit weight. This can be used in breeding programs to select for fruit size in juvenile plants, instead of waiting until a plant reaches maturity. A positive correlation also exists between the number of areoles on the cladodes and the number of areoles on the fruit (Mondragon 1999). Three groups incorporating eight of the measured traits are the most important in describing the cultivars: plant vigor, frequency of areoles, and spininess, which together encompass 71% of the variability among the plants. The 32 accessions fall into seven categories: wild accessions, small-fruited plants, "spineless," improved varieties, varieties cultivated for fruit production, very spiny accessions, and a single cultivar 'Cristalina,' that showed extremely vigorous growth. This approach for characterizing accessions will facilitate the incorporation of available germplasm into breeding programs by allowing researchers to narrow their focus to single accessions within each major category.

Breeding Systems and Techniques

Cactus flowers are typically relatively large and hermaphroditic with abundant stamens, a single style, and showy perianth parts (Nerd and Mizrahi 1997). The structure of these flowers suggests that they are cross pollinating and, in fact, insect pollination is quite common. For cactus pear under natural conditions, pollination by insects (mainly bees) ensures a high number of seeds once the flower opens. In the fruit-producing Opuntia species, self-compatibility and self-pollination occur, and bagged flowers can set fruit (Nerd and Mizrahi 1997). Protandry can explain self-pollination prior to flower opening. Little information is available regarding the ability of wild cactus species to self-pollinate. Bagged flowers of O. lindheimeri did not produce fruits in one study (Grant et al. 1979). However, five out of ten bagged flowers of O. lindheimeri set fruit and produced seed in another study (Wang et al. 1996). With regard to other species of fruiting cacti, clones of Cereus pe-ruvianus, Hylocereus costaricensis , and H. polyrhizus are self-incompatible and those of H. undatus and Selenicereus megalanthus are self-compatible. Sicilian cultivars of cactus pear are self-compatible, as problems of fruit set are seldom encountered in vegetatively propagated plantations composed of a single cultivar or in single plants grown in backyards (Nerd and Mizrahi 1997). Similarly in the Pyramids region of central Mexico, about 7,000 ha of the cultivar 'Reyna' are cultivated without apparent pollination problems.

Controlled Pollination

Short-term pollen storage for less than a week can be accomplished by collecting buds close to flowering and placing them in a cool and shaded location. Before use, the buds are exposed to full sun for a few hours to promote flowering. Unopened buds can also be used, but their pollen yield is lower. Pollen collected fresh or stored at room temperature for up to 6 days will remain viable and effect successful pollination (Bunch 1997). To facilitate germination, five milligrams of cactus pollen are suspended in 5 ml of germination medium prepared with 100 ppm H3BO4, 300 ppm Ca(NO3)2 • 4H2O, 200 ppm MgSO4, and 100 ppm KNO3 in a 40% sucrose solution.

The emasculation (removal of the stamens) of a cactus pear flower resembles a surgical operation and as such should be performed carefully (Fig. 15.1). The material needed includes rubber gloves, a brush, a sharp knife or razor blade, small scissors with a bent tip, a rinsing bottle, paper towels, glassine or paper bags, and rubber bands. When emasculating flowers of cactus pear, take the fol

Pollination Emasculated Flowers Image
Figure 15.1. Methodology for emasculating flowers of cactus pear: (A) emasculating an Opuntia ficus-indica flower; (B) washing the emasculated flower; (C) covering the emasculated flower with a bag; and (D) pollinating the emasculated flower with an open flower.

lowing steps: (1) clean the exterior of the buds with the brush to allow easy handling; (2) excise the corolla, using as few strokes as possible, avoiding wounds and mechanical damage to the style; (3) carefully remove the stamens and anthers, cutting close to the base; (4) rinse thoroughly with clean water to get rid of pollen residues and anthers; and (5) cover the emasculated flower bud with a bag to prevent unwanted pollination.

Although it is important to minimize damage to the flower during emasculation, the flowers are able to recover from some damage to the receptacle without significantly harming reproductive potential. The wound response of cactus involves the abundant flow of mucilage from the wounds. This mucilage covers the wound and dries, thereby reducing water loss and preventing infections. It is important that the stigma and style remain intact and undamaged. Even young flowers can be emasculated without appreciable loss of fertility. However, handling very young buds is difficult, because the stamens tend to be less exposed and the risk of mechanical damage to the stigma and style is greater. The application of Dithyocarbamate (Sevin, 2%) powder in the receptacular cavity after partial emasculation can prevent insect visitation and undesired pollination (Wang et al. 1996).

After emasculation, the stigma becomes receptive in 3 to 4 days; however, under warm (> 35°C) conditions, it is receptive earlier. When the stigma is receptive, it is shiny and sticky and the lobes are wide open. The most efficient way to pollinate is using a detached, fresh, fully open flower devoid of its style and corolla to allow close rubbing of the stamens with the stigma of the female flower. Stored pollen can be applied onto the stigma of an emasculated flower with a #3 camelhair paintbrush (Bunch 1997). Partially opened buds can also be used, taking advantage of the protandric nature of cactus pear pollination. Longer availability of flowers for crossings can be accomplished by eliminating the first flowering flush, which ensures that after 50 to 70 days a new round of crosses can be per formed. In general, crosses performed later in the flowering season reduce the number of normal seeds. Typically 100 to 250 normal seeds can be obtained from a single fruit, depending upon the crossing conditions and the cultivars involved in the cross. The difficulty involved in emasculation and the high number of seeds that can be expected from a single fruit underscores the importance of having a few carefully performed crosses rather than numerous potentially unsuccessful ones.

Self-pollination can easily be performed by bagging unopened cactus flowers to avoid cross pollination. Moving pollen from the anthers to the stigma of the same flower often results in increased seed set. Typically, inbreeding of a cross-pollinated and vegetatively propagated crop does not result in superior progeny, although this has not been confirmed for cacti. In addition, self-pollinated progeny will help elucidate genetic factors controlling important traits by unlocking genes hidden behind heterozygosity in these out-crossing species. Trials are currently underway to document differences among various breeding methods.

Seeds should be extracted from ripe and healthy fruits. Fruits are peeled and processed in a blender at low speed. Seed disinfection is accomplished by soaking the seeds in commercial bleach (5-6% sodium hypochlorite) for 10 minutes. Seeds can be dried in an oven for 2 to 3 hours at 55 to 60oC. Seeds obtained in the same season can germinate after slight scarification. Dry storage (6-14 months) increases the rates of seed germination of several types of cactus pear (Mondragon and Pimienta 1995). Seed viability is reduced after long-term storage. For instance, storage of seed lots for 9 years reduces the seed germinability by 50% (Muratalla et al. 1990). Temperature is the most important variable for cactus seed germination. For nineteen species of cacti the optimal temperature for seed germination ranges from 17 to 34°C with a mean of 25°C (Nobel 1988). Differences in germination have also been attributed to cultivar and seed condition. Seeds may have a physical dormancy due to the hard seed coat, which hinders germination. Several treatments have been reported to overcome this barrier. Scarifying seeds in hot water (80-90oC) twice and allowing them to cool to room temperature and then soaking them in distilled water for 24 hours promotes germination under greenhouse conditions in 10 to 17 days (Mondragon 1999). Seeds soaked in gibberellic acid (35 mg/liter) can germinate faster (Wang et al. 1996). Planting media should be kept moist for optimal germination of cactus pear seeds under greenhouse conditions.

Seedlings can be kept in germination trays until the first cladode grows to 10 to 15 cm. At this size, they can be transplanted to small pots (10 cm in diameter) or black plastic bags. Prior to transplanting, the entire tap root can be excised and the seedling allowed to dry for 5 to 7 days to promote healing and drying of the wounded tissue. Elimination of the tap root eases the task of transplanting and promotes formation of lateral roots, increasing the root volume and improving anchorage. Minimizing the time from crossed seed to a mature plant, ready for evaluation, is important for any breeding program. Cacti generally grow slowly, especially the columnar and spherical forms, under natural conditions. Most cacti respond to optimal conditions of water and nutrients (Nobel 1988). Other factors influencing this response include photope-riod (Sanderson et al. 1986) and atmospheric CO2 concentration. Six-month-old Opuntia seedlings grown in a greenhouse can be transplanted to the field. At this age they may bear two or three slender pads and can be managed as adult plants or used as a source of vegetative material for grafting.

Plants of cactus pear derived from seeds tend to grow in an upright slender manner, branching only in the upper cladodes. By comparison, plants grown from cuttings tend to have thicker and wider cladodes as well as more pads on the lower part of the plants. Branching can be promoted in seedlings by pinching at the one-cladode stage. This practice encourages thickening of the basal cladode as well as branching, thus increasing photosynthetic area and improving anchorage and vigor. Pinching also allows for an early expression of adult cladode shape. Grafting is a potential method of reducing juvenility. By grafting immature cladodes onto mature cladodes, the scion (the detached plant part used in a graft) may have at least one flowering cladode in the next growing season, in comparison with the 4 to 6 years needed for seedlings transplanted to the field directly from the greenhouse.

Grafting is a standard practice to maintain rare forms of cacti, such as those lacking chlorophyll, cristate forms (Haage 1963; Pilbeam 1987; Pimienta and Muñoz 1995), and endangered species. Most information available pertains to spherical forms being used as scions and sharp angled forms of Cereus, Hylocereus spp., Myrtillocactus geo-metrizans, Pereskia, Rhipsalis, and Trichocereus being used as stocks (Pizzeti 1985). Grafting platyopuntias is a little more complicated. The main concern for Opuntia species is the shape and thickness of the stocks and scions. The best species to use as rootstocks is O. ficus-indica, because the plants are fast growing and almost spineless. Cultivars such as 'Seleccion Pabellon' available in Mexico for vegetable and forage production and 'Gialla' of Italy are also well suited for this purpose. Grafting can be performed with young, full-size cladodes. The scion and stock should be selected and cut to match for size, shape, and thickness. To create matching shapes, both the donor cladode and the rootstock can be cut with the same device (e.g., an aluminum can with one edge sharpened to cut and act as a template at the same time). The scion should be placed carefully, trying to match the cambial tissues as closely as possible and maintained in place using rubber bands.

Clonal Propagation/Apomixis

New and superior cultivars need to be propagated as rapidly as possible for further trials and commercial production. Asexual propagation can be performed for cactus pear using stems (or portions thereof) of any age as well as flowers and unripe fruits. Commercial orchards typically rely on large pad fractions, whole pads, or short branches. A few specialized propagation nurseries have been attempted in Mexico with limited success, as cuttings are most often obtained from productive orchards using pruning residues. Efficient protocols for in vitro propagation are also available (Escobar et al. 1986; Villalobos 1995), but they are not used on a commercial scale due to the ease and lower cost of propagation from cladodes.

Apomixis, the formation of asexually derived embryos, is common in the genus Opuntia (Mondragon and Pimienta 1995). Polyembryonic seed percentages range from 0 to 50% for several Opuntia species and from 2 to 16% for several wild and cultivated cactus pear accessions. From a breeding perspective, polyembryony can be a problem during germination of crosses, because seedlings derived from somatic tissue will be identical to the female parent as opposed to a combination of the two parents. In general, identification of apomictic seedlings is relatively easy in the early stages of germination and seedling growth; however, some apomictic seedlings are indistinguishable from the zygotic seedling, and they tend to appear more similar as they become older. The apomictic seedling usually emerges later, is smaller, and is weaker than the zygot-ic seedling. Seedlings from polyembryonic seeds must be separated early in the germination phase of a program and some somatic seedlings will probably enter the evaluation phases of a breeding program (Mondragon and Bordelon 1996; Bunch 1997).

Future Directions

Cactus genetic research is in its infancy, and the vast majority of breeding work lies ahead. Several people who were fortunate enough to have worked with Dr. Barrientos have retold a story of his: you can cross two cactus pear varieties and obtain five species in the progeny. This story attests to the tremendous variability available among cactus pear and to the amount of work needed to sort out and harness this variability. With the increasing number of new and well-documented crosses performed in cactus programs around the world, the genetics behind many traits should be gradually revealed. Information from these crosses will help sort out taxonomic relationships and lend insight into the genetic control of some of the special characteristics and physiological systems that make cacti such remarkable plants (Nobel 1994).

As breeding efforts are continued, new and exciting research will unfold. Future projects may involve creating ploidy series of cacti of a specific genetic makeup. Also, in keeping with the sustainable nature of cacti and their potential to benefit regions of the world with limited resources, nutrient utilization of new cultivars must be investigated. Cultivars should be developed with enhanced ability to benefit from associations between rhizosphere organisms to increase productivity and reduce dependence on fertilizers. New biotechnological tools will help breeders meet their objectives as well as allow access to new sources of genetic variation. The ultimate goal will be to develop cacti into crops rivaling the best of other fruit, vegetable, and forage crops.


An important complement to cactus breeding programs is the development of molecular tools to work with cacti. As has been demonstrated for numerous crop species, molecular biological tools can speed up the breeding process (Paterson et al. 1991; Staub et al. 1996), elucidate genetic mechanisms that cannot be easily dissected through plant breeding techniques (Paterson 1995), and in many cases can accomplish goals that are not possible through breeding. This section reviews the molecular work done on cacti and proposes a plan for the development of molecular tools for cacti utilizing the work done on traditional crop species.

Overview of Plant Molecular Tools

Most plant molecular genetics has focused on economically important crop species, such as corn or rice, or on facile models, such as Arabidopsis. The work done on other species can serve as a springboard to speed development of such tools for cacti. In turn, genetic analysis of cacti would pave the way for developing molecular programs for alternative crop species. A well-developed genetic map is important in this regard (Paterson et al. 1991; Staub et al. 1996). Genetic maps are constructed by utilizing highly variable, polymorphic molecular markers to assess levels of recombination in a cross between two variable parents. The molecular markers serve as landmarks along the chromosomes of the organism and reflect underlying differences in the genomes of the two plants involved in the cross. Both hybridization-based markers, such as Restriction Fragment Length Polymorphisms (RFLPs), and PCR-based markers, such as RAPDs and Microsatellites, are routinely used. The ability of molecular markers to detect subtle changes in genome structure make them highly advantageous over traditional morphological markers for developing genetic maps of cacti. If a breeder attempts to move a cold-resistant phenotype from a wild Opuntia accession into a specific cultivar and lacks a molecular marker associated with cold resistance, all plants need to be grown to maturity and then tested for cold resistance. By contrast, if cold resistance is associated with a molecular marker, this marker can be screened in young plants, thus helping the breeder to make selections earlier in development.

Another important tool for molecular work is the ability to transform cactus plants with DNA fragments of interest. Although progress has already been made in this area (Zárate et al. 1998, 1999a), many hurdles remain. An important step is to generate stably transformed plants that can transmit the transgene of interest to subsequent generations through sexual crosses. For instance, if a gene conferring cold tolerance is isolated, it can be integrated into cultivated plant germplasm by transformation without having to worry about traditional breeding difficulties, such as making wide crosses or removing unwanted wild germplasm introduced into a cultivar when it is crossed with the wild plant. This is especially important for cacti, as the time from seedling to mature plant can be long, making the production of backcross populations time consuming. In addition, transformation allows genes to be transferred from plants that cannot be crossed with cacti, or even from other organisms. For instance, if one or a few genes controlling seediness can be identified, these genes could be inactivated through transformation of anti-sense constructs and screening for transformed plants that silence the seediness-related genes.

DNA and Protein Analysis

Molecular work done on cacti reveals a number of intriguing features. The polyploidy level varies, with a basal chromosome number of 11 (Pinkava et al. 1985, 1992). Analysis of nuclei from Opuntia TAM 1308 by flow cytometry indicates that they possess a DNA content similar to that of small genome crops such as sorghum and tomato. Isozyme analysis detecting variability among plants based on activity of known enzymes indicates a high level of variability among cacti, especially in pollen (Chessa et al. 1997). Other isozyme studies show a low variability among accessions but a curiously high variability among fruits and cladodes of the same cultivar (Uzun 1997). In contrast to isozyme studies, RAPD assays the variability of unknown DNA sequence fragments amplified using the Polymerase Chain Reaction (PCR). RAPD markers were used to verify the somatic origin of putative apomictic seedlings using seeds from two crosses of Mexican accessions (Mondragon 1999). The study indicated that seedling size and fresh weight are the only morphological features associated with the apomictic seedlings.

RAPDs have also been employed to compare DNA variability for relationships among Opuntia species with traditionally derived morphological classifications (Wang et al. 1998). Discrepancies occur between molecular and traditional classifications, although more work needs to be done to characterize these differences. A DNA extraction technique for cacti, which helps overcome the difficulties caused by mucilage, has been used for 32 accessions (Mondragon 1999).

Tissue Culture and Transformation

An efficient micropropagation protocol for cactus pear has been developed (Escobar et al. 1986). In particular, in vitro culture can produce virus-free plants and is a method for rapid propagation of new cultivars. The technique has not been used on a commercial scale due to the comparative ease of propagation through traditional methods. The production of somatic embryos (asexually formed embryos that arise from sporophytic cells unconnected with maternal tissues) occurs for a number of cactus species in tissue culture (Torres-Muñoz and Rodríguez-Garay 1996; Santacruz-Ruvalcaba et al. 1998). Ongoing work at the Federal University of Ceará in Brazil has led to a number of advances in both tissue culture and transformation techniques, such as callus and cell suspension cultures of Opuntia (Zárate et al. 1999c), cultivation of isolated shoot meristems, and regeneration of whole plants from the apical meristem (Zárate et al. 1999b). These tissue culture techniques have produced transient transformation of a P-glucuronidase reporter gene in callus and cell suspension cultures (Zárate et al. 1998) and in apical meristems (Zárate et al. 1999a). Analysis of seed reserve proteins in Opuntia ficus-indica (Uchoa et al. 1998) will be useful to research on improving cactus cultivars through transformation techniques.

Evolutionary Relationships

Much information from other plant species can be used for developing molecular tools for cacti. Understanding these relationships will be key to using comparative mapping techniques to speed the development of a molecular map for cacti. Molecular sequence data for the plastid gene encoding the large subunit of Rubisco (rbcL) were obtained for a number of plant species, including some in the Cac-taceae, from Genbank ( The sequences were aligned using the Clustal alignment program ( Top.html), and phylogenetic trees were calculated via the parsimony method using DNAPars (available as part of the Phylip molecular phylogeny package; A consensus tree created using the majority-rule method (CONSENE, also in the Phylip package) indicates that the Cactaceae are most closely associated with the Aizoaceae. A well-characterized member of this family is Mesembryanthemum crystallinum, which also utilizes the CAM photosynthetic pathway. Closely associated to the Cactaceae are the Poaceae (grasses) and the Solanaceae. This tree is biased by the use of a gene involved in the photosynthetic machinery, which likely varies between CAM, C3, and C4 species. No phy-logeny based on variation in a single gene can characterize the relationships between such a broad class of plant families. Hopefully, future molecular phylogenetic work will refine the position of the Cactaceae among other plant families.

Molecular Tools for Cacti

A first step in building molecular tools is to identify cactus sequences that relate to potentially useful genes in other taxa. This can be accomplished by hybridizing cactus clones with probes from other taxa and looking for expressed genes that are highly similar. To achieve this goal, a cDNA library will be constructed from Opuntia accessions from Texas A&M University-Kingsville. Genes recognized in other organisms involved in critical processes for cacti will be utilized. For example, cold-temperature-tolerance genes have been identified in Arabidopsis (Thomashow 1998) that might allow recognition of cactus genes involved in the same process. Other possibly useful genes to identify include those involved in drought/salt tolerance (Winicov 1998) and those uniquely involved in CAM (Cushman and Bohnert 1997).

The second step in developing cactus molecular tools will be to create a genetic map with the useful sequences providing the DNA markers. A mapping population of approximately 120 plants has been created by a cross between Opuntia accession TAM 1281 and Opuntia accession TAM 1250. Copies of this population are available in California, Georgia, and Argentina, and these plants will be used to create a genetic map of cacti with an average spacing of 5 cM between markers. The sequences identified will be used as RFLP probes to develop this map. Because of the morphological variability between the parents, and based on initial RAPD data, genetic polymorphism between the parents may be frequently encountered. This map will be publicly available, as will the clones from the mapping population, allowing other researchers to use and extend the map. Once an initial genetic map of cacti is created, morphological markers will be placed on the map. This will provide breeders a working framework and will also help characterize the interesting phenotypes associated with cacti.

Future Directions

Speeding the development of molecular tools for cacti using other well-studied plant genomes will provide workers interested in the molecular biology of cacti with a set of tools to accomplish their research goals. In addition, the localization of agronomic traits on the genetic map will allow breeders to use molecular techniques in their programs. However, the work described here is only an initial step toward the development and characterization of cacti at the molecular level. Another important goal should be the development of transformation systems for cacti. Because cacti may also be amenable to transformation using Agrobacterium, development of these systems could supplement and extend the work being done with particle bombardment. Research on tissue culture of cacti must be extended to provide a rapid system for the regeneration of transformed tissue into whole plants. Further work characterizing cactus germplasm using molecular markers can help create a better understanding of the relationships between available accessions. Cultivars have been recognized primarily by their morphological characteristics, which can be misleading (Wang et al. 1998). Molecular markers should be developed and utilized to evaluate germplasm resources, helping breeders determine the level of molecular similarity between two cultivars in a cross.

Another area of potential research is the study of chromosomal changes underlying the variable polyploid nature of the cactus genome. This has practical interest, because chromosomal imbalance associated with polyploidy may lead to partial seed set in O. ficus-indica (Nerd and Mizrahi 1994). Characterizing how polyploidy affects seed set may lead to the development of techniques to produce varieties with fewer seeds. In addition, physical maps of the cactus genome should be developed to allow researchers to identify and sequence genes of interest in cacti more easily. These tools should provide insight into the molecular basis for many of the interesting phenotypes of cacti. The molecular characterization of these phenotypes will not only be intriguing within the framework of understanding cactus and plant morphology, but may also be potentially useful for the improvement of other crop species.


The Cactaceae exhibit many unique phenotypes, making it an ideal family for study by plant breeders and molecular biologists. This chapter has focused on the improvement of cactus germplasm via plant breeding and molecular techniques. Many modern varieties of cactus pear are the products of long-term, informal, yet effective selection by growers from plants in family gardens. These cultivars were selected in Mexico or were obtained from Mexican germplasm and adapted to other countries. More formal cactus pear breeding was attempted in the early 20th century in California and again during the 1970s in Mexico. Only recently have many of the advances and more formal techniques been utilized by research groups around the world. The molecular work done on cacti can benefit from biotechnological work on other species. Increased work in both breeding and biotechnology of cacti should provide significant insights into the processes that contribute to their unique phenotypes. Using these tools and techniques to accomplish the major objectives outlined in this chapter will improve cactus cultivars available to both farmers and consumers. Ultimately, breeding and biotechnology will play an important role in increasing the understanding and usefulness of cacti.


B.C. thanks Peter Felker for useful discussions and the Howard Hughes Medical Institute for a fellowship. Research of A.H.P has been funded by the International Arid Lands Consortium and the USDA Foreign Agriculture Service.

Literature Cited

Aguilar, B. A. 1987. Efecto de la aplicacion de acido gibere-lico (GA3) y urea en el fruto del nopal (Opuntia amy-chlaea Tenore). Master's Thesis, Colegio de Post-graduados, Chapingo, Mexico.

Barbera, G. 1995. History, Economic and Agro-ecological Importance. In Agroecology, Cultivation and Uses of Cactus Pear (G. Barbera, P. Inglese, and E. Pimienta-Barrios, eds.). FAO Plant Production and Protection Paper 132. FAO, Rome. Pp. 1-11.

Barbera, G., F. Carimi, and P. Inglese. 1991. The re-

flowering of prickly pear (Opuntia ficus-indica (L.) Miller): Influence on removal time and cladode load on yield and fruit ripening. Journal of the American Society for Horticultural Science 5: 77-80.

Barbera, G., F. Carimi, and P. Inglese. 1992. Past and present pole of the Indian-fig prickly pear (Opuntia ficus-indica (L.) Miller, Cactaceae) in the agriculture of Sicily. Economic Botany 46: 10-20.

Barbera, G., P. Inglese, and T. La Mantia. 1994. Influence of seed content on some characteristics of the fruit of cactus pear (Opuntia ficus-indica Mill.). Scientia Horticulturae 58: 161-165.

Bunch, R 1997. Update on cactus pear breeding and new products at D'Arrigo Bros. Journal of the Professional Association for Cactus Development 2: 60-70.

Cantwell, M. 1991. Quality and postharvest physiology of "nopalitos" and "tunas." In Proceedings of the Second Annual Texas Prickly Pear Conference. McAllen, Texas. Pp. 50-66.

Chessa, I., G. Nieddu, P. Serra, P. Inglese, and T. L. Mantia. 1997. Isozyme characterizaton of Opuntia species and varieties from Italian germplasm. Acta Horticulturae 438: 45-52.

Corrales-Garcia, J., J. Andrade-Rodriguez, and E. Bernabe-Cruz. 1997. Response of six cultivars of tuna fruits to cold storage. Journal of the Professional Association for Cactus Development 2: 160-168.

Cui, M., P M. Miller, and P. S. Nobel. 1993. CO2 exchange and growth of the Crassulacean acid metabolism plant Opuntia ficus-indica under elevated CO2 in open-

Building Your Own Greenhouse

Building Your Own Greenhouse

You Might Just End Up Spending More Time In Planning Your Greenhouse Than Your Home Don’t Blame Us If Your Wife Gets Mad. Don't Be A Conventional Greenhouse Dreamer! Come Out Of The Mould, Build Your Own And Let Your Greenhouse Give A Better Yield Than Any Other In Town! Discover How You Can Start Your Own Greenhouse With Healthier Plants… Anytime Of The Year!

Get My Free Ebook


    Is opuntia spineless burbank's self pollinating?
    3 years ago

Post a comment