Brassica Crops at a Glance

Cuong Truong

Writer’s Comment: “Brassica Crops at a Glance” was originally written as a term paper for Dr. Gept’s Evolution of Crop Plants course. Our assignment was to pick a species or group of species and to describe what is known about how it was domesticated and to explain what evidence supports those theories. I chose to do my paper on Brassica crops, a group of six related species and countless cultivars that provide common foods like cabbages, radishes, broccoli, and more. I worked on collecting as much of the available research as I could in the weeks before starting the paper. Although much had to be left out, it is my hope that what has been written here can be used as a good introduction to plants in the genus Brassica.

Instructor’s Comment: Domesticated plants (and animals) provide a handy experimental system to study the workings of evolution in biological systems in general. What is the role of mutation, selection, and other evolutionary processes in generating the incredible diversity of our living environment, from the deepest ocean to the highest mountain peaks, from the driest of deserts to the wettest of rainforests? In The Origin of Species (1859), Darwin had already used crop plants and farm animals as examples to illustrate how selection—whether human or natural—can lead species on the path to greater adaptation and survival. Cuong Truong, in his scientific review paper on Brassicas (or cabbages), very aptly describes how, where, and when human selection, combined with biological processes such as hybridization and tetraploidization (doubling of chromosome number), has developed a truly mind-boggling array of related crops, where many organs can be used for food: turnip, cabbage, cauliflower, broccoli, Brussels sprouts, and oilseed rape.
—Paul Gepts, Department of Plant Sciences


Brassica crops provide one of the most diverse arrays of cultivated plants across the world, and the study of Brassica evolution and domestication demonstrates the need for a multidisciplinary approach to understanding crop origins. Several botanical methods have been devised to identify Brassica and related plants within the Cruciferae family, but these may not always be reliable because traits common among all the species are rare. The main Brassica species that have been domesticated can be traced through historical records to the Mediterranean, though secondary and independent centers have been proposed. The exact geographical origin of some cultivars, such as that of Brassica oleracea var. gemmifera (Brussel sprouts) can be traced rather easily from historical records, but the origins and spread of other Brassica crops are often more difficult to establish. Dr. Nagaharu U’s groundbreaking research in 1935, however, revealed the genetic relationships between the cultivated Brassica species. A better understanding of these genetic relationships and the origin of the amphidiploid species has also helped researchers narrow down exactly where crosses may have occurred, as well as the direction of spread.


Brassica crops provide an amazingly wide variety of harvested products from a relatively small group of species. Because of this dazzling amount of diversity, there has been no shortage of disagreement over the origin of Brassica crops, the relationships between species, and even what could accurately be called a Brassica species. After much investigation into botanical, geographical, historical, and genetic factors, researchers have gained a better idea over the past few decades of how the common Brassica crops have come to be. 

The vast majority of Brassica products are derived from six main species within the genus: Brassica oleracea, Brassica rapa, Brassica nigra, Brassica juncea, Brassica napus, and Brassica carnita. From these six species, farmers have utilized or developed many subspecies and cultivars that provide a wide range of leaf, root, stem, flower bud, seed, and oil crops. Since they are so closely related, each Brassica species can easily breed with others and produce viable hybrids; in fact, B. juncea, B. napus, and B. carnita are naturally occurring hybrids of the other three species and have become stable populations (Dixon 2007).

Most Brassica species have a biennial life cycle. Brassica crops undergo vegetative growth until exposed to a period of cold temperature, and then proceed to flower when the temperature rises again. Flowering usually involves sudden stem elongation (known as “bolting”), production of flowers, and eventually seed pods. This pattern occurs in most types of Brassica crops and is the result of vernalization adaptations. Plants use the transition from cold to warm temperatures to determine when winter has transitioned into spring. The general mechanism for vernalization seems to be controlled by genes that are triggered by temperature and produce proteins that affect flowering (Amasino 2004). There is usually some degree of self-incompatibility in the diploids, promoting a high level of outcrossing. The hybrids species are mostly self-fertile, though.

B. oleracea exhibits a wide range of variation among its subspecies and cultivars. The extreme differences undoubtedly come from the fact that different harvested organs and traits have been amplified over time by human selection. The heading cabbages make up a cultivar group known as capitata (Latin for “having a head”). This “head” is made up of immature leaves tightly wrapped in a ball around the terminal bud. Capitata varieties are the most leaf-intensive B. oleracea crops. However, the kales (var. alboglabra and var. acephala) form much looser heads or no heads at all and have prominent stems that may elevate plants as high as two meters. B. oleracea var. acephala is also harvested as a leaf crop, but the fact that it is not as leaf-intensive as capitata crops (and because of its resemblance to wild B. oleracea plants) suggests that var. acephala may have been among the first B. oleracea crops domesticated (Christman 2007). On the other hand, the botrytis (cauliflower) and italica (broccoli) cultivar groups are grown for their flower heads. Botrytis can be distinguished from italica by the fact that 90% of the floral meristems abort at maturity, while most of the meristems on italica actually develop into functional flower buds (Gray 1982). Other specialized varieties include gongylodes (kohlrabi), which is grown for its shortened and thickened spherical stem, and gemmifera (Brussels sprout), which grow upright stems as high as one meter, capped with small cabbage heads (Dixon 2007).

Strong directional selection for certain traits and highly diverse cultivar groups can also be found in B. rapa crops. The rapa cultivar group is the oldest known of the B. rapa crops and is believed to have been the first within the genus to be domesticated (Gomez-Campo, et al. 1999). Plants in this group are usually grown for the large swollen portions of their taproots. Cultivar groups pekinensis and chinensis are two of the most widely grown B. rapa cultivars. They are distributed throughout Asia and are eaten as leaf vegetables much as capitata is in Europe. Pekinensis is a heading vegetable, producing a tightly wrapped bundle of leaves similar to capitata crops. Chinensis, on the hand, does not produce a cabbage head (Rakow 2004). 

B. juncea, B. napus, B. carnita, and B. nigra are mostly grown for seeds, and further processed into spices or oilseeds. There is also some cultivation as leaf crops in these species, but oilseed production is by far the most popular use, with B. napus crops being the third largest vegetable oil source after soybeans and cottonseed (FAO 2009). 

Overall, the six cultivated Brassica species have proven to be endlessly mutable in satisfying human taste and use. The exchange and movement of Brassica crops across the world has only accelerated with increased trade and migration. The Food and Agriculture Organization’s 2008 estimate of worldwide production of “cabbages and other Brassicas” (bok choy, kale, kohlrabi, and heading cabbages all fall under this category) is 64 million tons. That year, farmers around the world also grew 680 thousand tons of mustard seed (other Brassica and Cruciferae species are counted along with B. nigra), 61 million tons of rapeseed (B. napus var. oleifera), and 19 million tons of cauliflower and broccoli (FAOSTAT 2011). The immense scale of Brassica crop production is staggering, showing that understanding the complexity of Brassica domestication and evolution will not be an easy task.

Evidence on Brassica Origins
The Botanical Evidence

As a means of understanding species origins, scientists have conducted morphological studies of Brassica crops to compare them with other members of the Cruciferae family (of which Brassicas are members). Due to the genus’ high level of morphological variation, knowing a few common physical traits can be helpful in identifying what could be a species of Brassica and ruling out what is not. Gomez-Campo (1980) studied the Cruciferae family in detail and found several traits that could be used to distinguish a tribe called Brassiceae consisting of 51 genera—of which genus Brassica is part. Plants in the Cruciferae family can be identified easily by their flowers, which always have four petals arranged in a cross-like pattern. Among the Brassiceae, there is a tendency to have segmented fruit; however, because this does not occur in 20 out of the 51 genera, it is not always a reliable method of identification. It is generally present in genus Brassica, but not in all species (Gomez-Campo 1980). 

Somewhat more reliable morphological indicators can be found by looking at seed and seedling development. Gomez-Campo cites De Candolle’s work with cotyledon position in developing embryos. It has been found that cotyledons folding longitudinally around the radicale is an exclusive characteristic in the Brassiceae tribe. Again, there are exceptions, but Brassica is not among them. Another useful seed quality is color, which to some extent is helpful in determining geographical origin. Seeds are usually yellow in Middle Eastern and South Asian species, brown in Mediterranean species, and black in Eastern European and Russian plants (Gomez-Campo 1980). 

The use of the traits mentioned above does help to bring some order to the confusion, but newer morphological and molecular evidence has surfaced. Recent RFLP analysis has shown that many groups and genera within the tribe need to be revised or removed completely because actual evolutionary relationships have not been represented solely by morphological similarities (Hall, et al. 2009). Gomez-Campo, looking back at his original research from nineteen years earlier, has acknowledged that morphological evidence alone may not be reliable for mapping evolutionary relationships. For the domesticated Brassicas in particular, Gomez-Campo pointed out that over time the presence of segmented fruit and other identifying characteristics may take on regressive forms (Gomez-Campo 1999).  Botanical evidence may help narrow down the morphological traits unique to a certain grouping, but it is less reliable on its own in determining evolutionary history. 

The Difference Between Wild and Domesticated Brassicas

More reliable evolutionary relationships may be established by looking at wild relatives of Brassica crops. Figuring out whether two plants are related or not is fairly simple—one need only breed a domesticated variety with its wild ancestor. If the result is viable offspring, then an evolutionary relationship has been confirmed. Harlan (1975) pointed out that domestication of a crop from a wild ancestor usually does not lead to speciation, so viable offspring should not be hard to produce from a cross (Harlan 1975).

What exactly defines a wild Brassica? Generally, the more domesticated a crop is, the more adapted it is to human care and the less able it is to survive in the wild. Loss of seed through “shattering” is an important trait of domesticated plants. Usually, farmers of grain crops select for varieties that do not eject their seeds on their own, so they can be conveniently saved for human consumption. In this respect, Brassicas like B. napus that are grown for oilseed production are not fully domesticated since seed shattering has not yet been removed (ScienceDaily 2009). B. napus may have been taken up without extensive morphological or behavioral modification because it was and still is the highest yielding Brassica oilseed (Mendham 1995). Furthermore it has been grown as a crop in a relatively short time span of 400 years (Gomez-Campo, et al. 1999). Only recently has breeding taken place to select for lower levels of undesirable acids (like erucic acid) while increasing levels of more desirable seed content (Uppstrom 1995).

Another indicator of domestication is the amplification of traits that humans find useful. The heading habit in some leaf crops, for example, may have been favored over non-heading types because heads produce more leaf material faster and take up less space. In B. rapa var. pekinensis, Kato (1981) found that this heading cultivar produced up to three times more leaf material than non-heading chinensis plants in the same period of time (Kato 1981). Domesticated B. oleracea leaf crops differ significantly in the kinds of leaves found in wild B. oleracea. Wild varieties typically have smaller, thicker leaves with less chlorophyll and more cell wall layers to deal with dry coastal and even desert habitats (Dixon 2007). 

B. oleracea var. botrytis is grown for its floral organs, and in the course of its domestication it has paid a high cost in terms of species survival: the loss of the ability to reproduce effectively. Since the majority of its flowers abort before becoming functional, it is hard to imagine this trait surviving for long with all the selective pressure against it in the wild. The research on botrytis genetics from Boyles, et al. (2000) has shown that the high abortion rate of floral meristems is likely due to small mutations in genes determining flower development that were preserved and maintained by human cultivation.

Generally, a domesticated crop is adapted to rely on humans to some extent. It loses some of its ability to survive in the wild, and it changes over time as humans select for traits that benefit them. A wild plant is generally the opposite, a strict survivalist. If viable offspring can be produced from a cross, the wild plant can be tentatively considered an ancestor, and further investigation of other relatives in a geographical area might be made to develop a larger picture of the crop.

Geographical and Historical Evidence

Studying the geographical distribution of Brassica crops and their relatives along with the history of use and spread among human populations may be useful in gaining an idea of where domestication actually occurred. Sometimes, if records or common knowledge are detailed and reliable enough, they may already hold the answer. For example, it’s not hard to identify where Brussels sprouts (B. oleracea var. gemmifera) were developed. The first written records of gemmifera date back to the year 1587, when the city of Brussels was already known as a major grower of gemmifera, thus giving researchers a rough idea of its center of origin and the period of time since its emergence (Aggie Horticulture). The Russian botanist Vavilov theorized that centers where the greatest diversity of a crop occurred were also centers of origin and most likely the points of domestication. More recently, though, researchers have found that pinpointing where a crop was domesticated may not be so simple due to the fact that there may be secondary centers of diversity or none at all (Harlan). Centers of diversity are still worth looking at, though they should not be considered automatically the sites of original domestication. This makes a study of recorded human use necessary to complement geographical information. 

Most of the Brassica species seem to have spread from the Mediterranean coasts. Wild B. Oleracea occur throughout the coasts of the Mediterranean and the British Isles. Their natural habitats are mostly isolated cliffs where little plant competition exists. Being scattered along the wide coastlines of Europe led to the rise of many endemic varieties of B. Oleracea, perhaps providing the genetic diversity farmers later harnessed (Sauer 1993). What attracted humans to wild B. Oleracea was most likely its ability to survive rough habitats and store a large amount of nutrients (Tsunoda 1980). The philosopher Theophrastos (370-285 BC) gave one of the earliest written descriptions of the B. oleracea cultivated by the ancient Greeks. These early domesticates were acephala of some sort. The Roman statesman Cato (234-149 BC) also wrote about the cultivation of acephala and headed B. oleracea (Snogerup 1980). Botrytis and italica were developed at some time in the Roman period too. In 1860, wild capitata plants growing on a shore in England were used in breeding experiments, eventually resulting in very basic italica-like offspring (Aggie Horticulture). All of this indicates that the Mediterranean center of B. oleracea diversity is also the center of origin. B. nigra is also believed to have originated in the Mediterranean and has been known to infest slopes and wheat fields as a weed. It was used as a spice in Turkey, a medicinal plant in Ethiopia, and a vegetable crop in the Agean islands (Tsunoda). There is little difference between cultivated B. nigra and its wild counterparts, except that it is less branched and taller (Sauer).

B. rapa is now commonly accepted as having originated in the Mediterranean, although there has been some confusion about its ancestral origin. B. rapa is believed to have emerged on the cold highlands rather than the coastal areas of the Mediterranean (Rakow). As mentioned in the introduction, B. rapa var. rapa is believed to be the first Brassica ever domesticated—carbonized rapas have been found in Neolithic sites, and de Candolle suggested it might have been domesticated around 2500 to 2000 BC. What confused many researchers at first was the fact that some varieties of B. rapa moved east towards Eastern Europe, Central Asia, and eventually East Asia (Gomez-Campo, et al.). Another group of B. rapa also moved into Northern Europe, and the Romans encountered B. rapa crops for the first time mostly as rapas cultivated by the Northern European barbarians (Sauer). When B. rapa arrived in East Asia, it diversified into the unique leaf crops, var. pekinensis and var. chinensis. An ancient treatise on agriculture from around 3000 years ago mentions chinensis only briefly but proves that it was around at the time (Li 1969). Chinese horticulturalists have been confused as to why thousands of cultivars of these Brassicas were grown in China but wild forms could not be found (Chia 1981).

The amphidiploid Brassica species are generally found where their parent species’ territories overlap. The amphidiploid B. Juncea may have resulted from a mixing of B. rapa and B. nigra around Southwest Asia and India (Sauer). Today, India is a major producer of B. juncea as an oilseed and spice crop. Samples of carbonized B. juncea found at ancient sites date back to 2300 BC (Prakash 1980). Vaughn (1977) proposed that the B. juncea could have developed as two separate geographical races from centers of diversity in India and China. His research led him to suggest Indian B. juncea were more closely related to the rapa side, while Chinese B. juncea were more closely related to the nigra side (Vaughn 1977). B. carnita is a cross between B. nigra and B. oleracea that occurred in Ethiopia (Sauer). B. carnita has not expanded significantly from Ethiopia and surrounding countries, but it is beginning to get noticed for its lack of seed shattering (Mendham). B. napus comes from crosses between B. olercea and B. rapa (Sauer). 

Genetic Evidence

Analyzing genetic evidence is much more straightforward than the approaches previously considered. Genetic evidence is easier to work with because the material analyzed is readily available, and there is a great deal less guesswork and uncertainty about evolutionary relationships and trends.

One of the most significant breakthroughs in understanding Brassica genetics came from the Korean-Japanese botanist U. In the late 1920’s, T. Moringa completed genome analysis of the Brassica crop species and determined the haploid numbers to be n = 8 (B. nigra), n = 9 (B. oleracea), and n = 10 (B. rapa). It was apparent that the other three species were combinations of the first three. The numbers were: n = 17 (B. carnita), n = 18 (B. Juncea), and n = 19 (B. napus). In 1935, U confirmed this theory by successfully synthesizing B. napus, thus establishing the evolutionary relationship between the six species that is standard knowledge today (Chopra, et al. 1999). Since that first step, genetic analysis of crop evolution has continued to advance. One example of the power of genetic analysis in determining evolutionary relationships is the use of RFLP analysis by Gepts, et al. (1993) to determine that Phaseolus vulgaris was independently domesticated in the Andes and in Mesoamerica (Gepts, et al. 1993).

In recent years, genetic analysis of Brassica crops has raised new questions. RFLP analysis of nuclear, mitochondrial, and chloroplast DNA of the diploid Brassica species suggest that B. nigra may have evolved from a separate ancestor than B. oleracea and B. rapa, which are more closely related to each other (Chopra, et al. 1999). At present, the relationships between the diploid Brassicas and their evolutionary relationships with their progenitors remains a mystery in the same way that the relationships between the six Brassica crops once were.

Recommended Future Lines of Research

Through examining a combination of botanical, historical, and genetic evidence, a more complete picture of Brassica domestication and evolution has been presented. The progress that has been made has led researchers to new questions about the original diploid Brassica parents. At the same time, many unresolved problems and gaps remain in the current understanding of Brassica domestication. Although researchers have a general idea of Brassica radiation from the Mediterranean, there does not seem to be much research on this issue. The exact mechanisms and pressures that have encouraged the morphological diversity of domesticated Brassicas is also a neglected area of research.


“Kohlrabi and Brussel Sprouts are European.” Aggie Horticulture. Texas AgriLife Extension Service. Web. 26 May 2011. <

“Cabbage Flowers for Food.” Aggie Horticulture. Texas AgriLife Extension Service. Web. 26 May 2011. <

Amasino, Richard. “Vernalization, Compatence, and the Epigenetic Memory of Winter.” American Society of Plant Biologists Oct. 2004: 2553-2559. Jstor. Web. 26 May 2011.

Boyles, Abee L., Michael D. Purugganan, and Jane I. Suddith. “Variation and Selection at the CAULIFLOWER Floral Homeotic Gene Accompanying the Evolution of Demosticated Brassica Oleracea.” Genetics. June 2000: 855-862. Jstor. Web. 8 June 2011.

Chia, Wen Li. “The Origin, Evolution, Taxonomy and Hybridization of Chinese Cabbage.” Chinese Cabbage. Ed. Griggs, T. D., N. S. Talekar. Tainan: Hong Wen Printing Works, 1981. 3-10. Print.

Chopra, Virender L., Pulugurtha B. Kirti, Yoshihito Takahata, and Shyam Prakash. “Cytogenetics.” Biology of Brassica Coenospecies. Ed. Gomez-Campo, C. Amsterdam: Elsevier, 1999. 59-106. Print.

Christman, Steve. “Brassica oleracea var. acephala.” Floridata. Floridata. Web. 12 Dec. 2007, 26 May 2011. <> 

Dixon, Geoffrey R. Vegetable Brassicas and Related Crucifers. Wallingford, UK: CABI Pub., 2007. Print.

“Food Outlook, December 2009.” Food and Agriculture Organization. Web. 26 May 2011. <>
FAOSTAT. Web. 26 May 2011. <>.

Gepts, Paul and Viviana L. Becerra Velasquez “RFLP Diversity of Common Bean (Phaseolus Vulgaris) in Its Centres of Origin.” Genome vol. 37 no. 2 (1994): 256-63. NRC Research Press. Web. 26 May 2011.

Gomez-Campo, Cesar. “Morphology of the Morpho-Taxonomy of the Tribe Brassiceae.” Brassica Crops and Wild Allies. Eds. Gomez-Campo, C. and K. Hinata and S. Tsunoda. Tokyo: Japan Scientific Society Press, 1980. 3-31. Print.

Gomez-Campo, Cesar. “Taxonomy.” Biology of Brassica Coenospecies. Ed. Gomez-Campo, C. Amsterdam: Elsevier, 1999. 3-32. Print.

Gomez-Campo, Cesar and Shyam Prakash. “Origin and Domestication.” Biology of Brassica Coenospecies. Ed. Gomez-Campo, C. Amsterdam: Elsevier, 1999. 33-58. Print.

Gray, A. R. “Taxonomy and Evolution of Broccoli (Brassica oleracea var. italic).” Economic Botany. Oct.-Dec. 1982: 397-410. Jstor. Web. 26 May 2011.

Harlan, J. R. Crops and Man. Madison: ASA, 1975. Print.

Kato, Toru. “The Physiological Mechanism of Heading in Chinese Cabbage.” Chinese Cabbage. Ed. Griggs, T. D., N. S. Talekar. Tainan: Hong Wen Printing Works, 1981. 207-215. Print.

Li, Hui-Lin. “The Vegetables of Ancient China.” Economic Botany. Jul.-Sep. 1969: 253-260. Jstor. Web. 17 May 2011.

Mendham, N.J. and P.A. Salisbury. “Physiology: Crop Development, Growth and Yield.” Brassica Oilseeds: Production and Utilization. Eds. Kimber, D. S., and D. I. McGregor. Wallingford, Oxon, UK: CAB International, 1995. 11-65. Print.

Vaughn, J. G. “A Multidisciplinary Study of the Taxonomy and Origin of ‘Brassica’ Crops.” Bioscience. Jan. 1977: 35-40. Jstor. Web. 26 May 2011.
Prakash, Shyam. “Cruciferous Oilseeds in India.” Brassica Crops and Wild Allies. Eds. Gomez-Campo, C. and K. Hinata and S. Tsunoda. Tokyo: Japan Scientific Society Press, 1980. 151-167. Print.

Rakow, G. “Species Origin and Economic Importance of Brassica.” Biotechnology in Agriculture and Forestry 54: Brassica. Eds.Douglas, C.J. and E.C. Pua. Berlin: Heidelberg, 2004. 3-13. Print.

Sauer, Jonathan D. Historical Geography of Crop Plants: a Select Roster. Boca Raton: CRC, 1993. Print.

Snogerup, Sven. “The Wild Forms of the Brassica oleracea Group (2n = 18) and Their Possible Relations to the Cultivated Ones.” Brassica Crops and Wild Allies. Eds. Gomez-Campo, C. and K. Hinata and S. Tsunoda. Tokyo: Japan Scientific Society Press, 1980. 121-133. Print.