Robarbica (Robusta x Arabica) Hybrids: Indian CoffeeSelection-6

Arabica coffee (Coffea arabica) possesses the quality characteristics of fine aroma, flavour and taste. Hence, it is commercially very important. However, this species is also the most susceptible to pests and diseases. Of the hundred species of Coffea, only two other species, C. canephora (Robusta) and C. liberica (Liberica) were of some commercial interest. These species produce coffee deficient of quality attributes. Of these two, Robusta is more popular and is produced by many coffee-growing countries. This species possesses resistance to some very important adversaries of Arabica such as leaf rust (caused by Hemileia vastatrix), white stem borer (Xylotrechus quadripes) and root lesion nematode (Pratylenchus coffeae)(Anonymous, 2000). On account of these characters, Robusta is considered an important source of resistance genes for improving Arabica. Arabica is a tetraploid carrying four sets of chromosomes (character bearers) while Robusta is a diploid with only two sets of chromosomes (C. arabica, 2n=4x=44; C. canephora, 2n=2x=22). This is an important and significant difference between these two species (Ram, 2001). Another important difference is that Arabica is self-fertile; producing seed upon self-pollination while Robusta is self-sterile producing hardly any seed when self-pollinated (DeVreaux et al., 1959). These differences render hybridization of these two species very difficult. However, small numbers of hybrid seed were obtained in various efforts of interspecific hybridization across the world (Sybenga, 1961). These seeds produced triploid plants with an imbalanced chromosome complement (2n=33). The crossing was said to be more successful in producing seed when Arabica is involved as the seed bearing mother plant (♀ parent) and Robusta as the pollen donor (♂ parent).

Breeding Efforts to Develop Robusta-Arabica Hybrids

As mentioned above, Robusta carries active and expressing genes conditioning resistance to important adversaries, but does not possess quality attributes. Another important aspect of Robusta is that it has greater caffeine content in its beans. These biological differences besides the others mentioned above render the interspecific hybridization efforts very difficult. However, these differences are useful in setting the breeding priorities and goals to produce synthetic types. Thus, the evolution of an Arabica type plant that produces coffee with the outstanding quality attributes besides possessing the resistance characters of Robusta and even possibly its productivity became the goal of all these exercises. Breeding efforts to evolve Robusta-Arabica hybrids were made with these objectives since the first half of 19th Century in Indonesia, India and Brazil. Later, this line of work was also taken up in Ivory Coast. These efforts can be classified into two major types: the first is prospecting for natural hybrids of Robusta-Arabica parentage and the second is artificial hybridization of Robusta and Arabica in reciprocal combination.

Prospecting for Natural Robusta-Arabica Hybrids

The early work in Indonesia was primarily prospecting for natural hybrids of Robusta-Arabica parentage. The earliest report on such hybrids mentions “Arla” hybrids discovered in Indonesia. However, not much attention was given to these hybrids and their further improvement as Indonesian coffee industry moved towards total Robusta cultivation (Cramer, 1957). Other interspecific hybrids like “Kawisari” (Arabica-Liberica parentage) also faded into history.

Devamachy hybrids discovered in India in 1949 are also considered to be of Robusta-Arabica parentage. These are utilized in the breeding programme to evolve Selection-5 that is commercially exploited in India (Srinivasan and Vishveshwara, 1980; Srinivasan and Ramachandran, 1997).

The most important natural hybrid of Robusta-Arabica parentage is Hibrido de Timor (HDT) discovered in Portuguese Timor. Hibrido de Timor (HDT) introduced into the germplasm bank of Central Coffee Research Institute (CCRI), India in 1961 originated in a C. arabica field on the Timor Island in 1917 (Bettencourt, 1973). HDT is known to manifest resistance to all known races of the rust fungus and is classified as “Type-A” (Rodrigues et al., 1975). It is a putative interspecific hybrid of Arabica (C. arabica L.) and Robusta (C. canephora Pierre ex Froehner) and is referred to as an Arabicoid (Eskes, 1989; Ram and Sreenath, 2000). HDT produces segregates manifesting a spectrum of resistance reactions (Bettencourt et al., 1992). These segregates are susceptible to a few races of the rust but manifest resistance to many races, especially the prevalent ones.

Artificial Hybridization of Robusta and Arabica

In India, the first crossing of these species was effected in 1937 with Robusta as the mother parent (♀) and Kents Arabica as the pollinator (♂). The few F1 hybrids obtained were all found to be triploid and the progeny highly sterile (Sreenivasan, 1987). This F1 was repeatedly backcrossed to the Arabica parent as shown in the genealogy chart. Of the three backcross progenies BC-II was found to possess an optimal combination of characters of both parents such as high resistance of Robusta and quality attributes of Arabica. However, the plants of BC-II were still unstable with a variety of cytological and reproductive abnormalities (Sreenivasan, 1987). Self- and open pollinated seed from the selected individual plants of BC-II were utilized to raise succeeding generations. An F2 descendant line S.2357 (F2 of BC-II) possessing high resistance to leaf rust and good cluster characters as well as the cup quality characters similar to Arabica was distributed to growers. Two progenies of the third generation derived from S.1156 line of BC-II (S.2827 and S.2828) were also found to be morphologically similar to Arabica carrying the high rust resistance and tight fruit cluster characters similar to Robusta with cup quality character of Arabica. Seed from these lines was also distributed for field trials as Selection-6 and its productivity and quality were assessed over the past many years. In the course of this time, this selection was adopted by several growers who, in turn, aided its evolution into a perfect material for commercial exploitation. This document enumerates the experiences of those growers with this material and its evolution under varied cultural practices.

Artificial interspecific hybrids involving C. arabica and induced autotetraploid C.canephora were produced in Brazil as early as 1950 (Monaco, 1977).

Selection of Parents and Breeding Protocol

Breeders at CCRI selected S.274 1/11 (Robusta) as the mother plant (♀parent) on account of the high yield of this plant (double the yield of family mean), its bold fruits and beans (a character that is said to be inherited from the ♀-parent in coffee) and wide adaptation of S.274. The Kents Arabica was selected as the pollen donor (♂parent) on the basis of the quality character and high yield potential.

An important difference between the interspecific hybrid programme of India and other coffee coffee growing countries of the world is that in India, a diploid plant of Robusta was involved in crossing as the mother parent in its native chromosome condition (2n=22) while in Brazil, Ivory Coast and other countries the Robusta parent was rendered tetraploid (2n=44) by colchicine treatment before it is used as the male parent in crossing. On account of this, the Robarbica hybrids of India are different from the Icatu hybrids of Brazil and Arabusta hybrids of Ivory Coast as they carry the cytoplasm of Arabica. Indian Robarbicas carry the cytoplasmic endowments of Robusta and are likely to be important in specific breeding contexts like the resistance against coffee berry disease (CBD). However, the apprehension that beverage quality of interspecific hybrids is compromised stands void as can be seen from the literature (Narasimhaswamy, 1960; Ganesh et al., 2002; Petracco, 2000; Bertrand et al., 2003; Fazuoli et al., 1977).

Backcross breeding method was adopted as it leads to the restoration of the genotype of the recurrent parent by the third backcross generation to an extent of 93% with the introgression of a variable proportion of genes from the donor parent. This variable proportion usually does not exceed 5-7% in the case of normal crossings (diploid-diploid crosses).

Pedigree of Robarbica (Selection-6)

Robusta (S.274 1/11) x Arabica (Kents)

Triploid F1 (S.594)

BC-I Tetraploid (S.905)
BC-II Tetraploid (S.1156)

1st Generation        S.2088                                        S.2089                                     S.2090

2nd Generation      S.2399                                         S.2383                                     S.2386

S.2354                                       S.2384


S.2357 (Distributed to growers)

3rd Generation     S.2828 (Distributed to growers)

4th Generation     S.4374   S.4376


Hibrido de Timor (HDT)  S.2769 x S.2828  >>> S.4369, S.4375

S.2828 x S.1156   >>>>  S.4370, S.4371, S.4372, S.4373

Agro-Climatic Requirements of Sln.6

Sln.6 being an interspecific hybrid of Robusta and Arabica possesses wide ranging adaptability in different agro-climatic conditions. Arabica, in general, prefers high altitude, cooler conditions with high relative humidity and hence is grown under the shade canopy of tall trees. Whereas, Robusta thrives well and is cultivated successfully at lower elevations with warm and humid conditions. The hybrid Sln.6 has the advantage of the genetic contributions of both these species and this is responsible for the adaptability of this selection to different conditions. An elevation of 2800 to 3200 ft. above M.S.L is found to be highly suitable for an adequate vegetative growth and moderate to high crop of this selection. It is also observed that this selection performs well under low rainfall of 40-60” as well as high rainfall of 100-125”. This selection is best grown under medium shade to maintain its vigour, productivity and resistance/tolerance to leaf rust.

Soils and Soil Management

Selection-6 has shown better growth and yield potential in deep, well-drained and fertile soils. Soil pH ranging from 5.8-6.8 is considered ideal for the successful cultivation of this selection. In case of higher acidity (pH 5.0 –5.5), application of lime (1.0 to 5.0 tons per ha.) has improved the yield. Cover digging (to a depth of 15-18”) at the end of monsoon (October – November) is advised for young clearings. On steep slopes, opening cradle pits is found to be more useful to conserve moisture as well as weed control. Scuffling is practiced in plantations to control weeds and to provide aeration to the sub-surface soil.


Spacing provides the criterion to judge the performance of a variety in a given location and climatic conditions by assessing the yield and resistance to pests and diseases under a specific regime of nutrition.

Sln.6 is a tall variety of coffee that has a higher plant vigour and wider bush spread. This necessitates wider spacing like 7×7 ft. or 7×8 ft. for good growth, adequate aeration and light to the bush. A spacing of 6×6 ft. is also adopted when the soil is not rich enough or if the plot is located in a marginal area.


Healthy plants with vigorous growth, dark green foliage and thick stem are selected for planting in the field. Pits of 45x45x45 cm were opened at the required spacing and are filled with topsoil. Insecticide Thimet-10G or Chlorpyriphos is also added to the pit to protect the plants from cockchafers and cut worms that prevail in new clearings. Six-month old seedlings are preferable for new planting to establish a vigorous and healthy plantation. Planting older plants (18-month old) may lead to the development of bent root and other similar phenomena resulting in poor and slow growth, moisture stress during dry period and exhaustion of the plants even when their crop is not heavy.


Conservation of moisture in a coffee plantation is of paramount importance in the establishment of young plants and maintaining the established plantations in a good condition. Mulching is the practice of covering the drip circle area of the plant with leaf litter at the end of monsoon. This is done in bearing as well as non-bearing areas of the plantation to conserve moisture and add organic matter to the soil to make it porous and fertile. Mulching is practiced for all coffee varieties and suits well for Sln.6 also.


There are two modes of weed control: manual weeding and chemical weeding. For Sln.6, three to four rounds of manual weeding are useful for better growth and yield. Use of weedicides is generally practiced during the initial years of establishment (1-5 yrs).

Case Studies of Planters Growing Sln.6 in Considerable Area

Sln. 6 is the Indian equivalent of Hibrido de Timor that was extensively used in the Arabica Coffee breeding programmes all over the world as a source of rust resistance genes. Of the various F2 and F3 lines derived from BC-II, two lines became commercially important. These are S.2357 and S.2828. S.2357 is a large spreading bush while S.2828 is relatively compact.  These lines became popular on account of their high rust resistance and high yield with acceptable quality.  A case study of the performance of this selection was carried out in six estates growing it on a reasonably large scale.

Table 1. Sln. 6 materials grown in the sample Estates  

Average Rainfall Age of plants Estate Name Identity of the material Elevation from MSL
60” 1972-75 Swarnagiri Estate, Sidapur S. 2828 2900-3000 ft.
50” 20 Years Chowdirange Estate, Sidapur S. 2828 2850 ft.
100-120” 22 Years Raxidi Estate, Hanbal S. 2828 3200 ft.
90” 19 Years Divangudda Estate, Sakleshpur S. 2357 2900-3000 ft.
90-100” 19 Years Basaveshwara Estate, Jambandi S. 2357 3200 ft.
60 “ 25 Years Coodanhally Estate, Balagodu S. 2357 3000 ft.

Data presented in Table-1 shows that Sln. 6 is in Coffee growing areas located in a range of elevations (2850-3200 ft. from MSL) with different rainfall patterns (50-120”/annum).  Best performance of Sln. 6 in these locations with overlapping climatic and edaphic characters required for Robusta and Arabica cultivation is an indication of the adaptation of this selection in Robusta as well as Arabica growing areas.

Table 2. Disease control in the Case Study Estates 

Estate Name Important diseases Main Control Measures Other methods of Disease control
Swarnagiri Leaf Rust on about 5% plants No chemical control was applied No other methods also were used in disease control
Chowdirange Leaf Rust on about 1% plants After 1999 chemical controls stopped completely No other methods also were used in disease control
Raxidi Leaf Rust on about 2% plants 0.5% Bordeaux spray in 2 rounds (pre and post-monsoon) No other methods were used in disease control
Dewangudda Leaf Rust on

10-20 % plants

1 % Bordeaux spray on pre-monsoon application Systemic fungicide Contaff spray in 2 rounds
Basaveshwara Leaf Rust on

10-15 % plants

0.5 % Bordeaux spray in 2 rounds (pre and post-monsoor) Systemic fungicide Bayleton spray in 2 rounds
Coodanhally Leaf Rust on

5-10 % plants

0.5 % Bordeux spray in 2 rounds (pre and post-monsoon) Systemic fungicide Contaff spray in 1 round (post-monsoon)

 Note : Black rot, the other disease of considerable spread is found to be infecting Sln. 6 also.  However, it could be effectively controlled by cultural operations like centring and handling.

From the above table, it can be seen that the incidence of leaf rust in Sln. 6 is only on a small percentage of plants (1-5 % in S.2828/2827 and 10-20 % in S.2357). However, many of the susceptible plants are highly susceptible like Kents that is one of the parents of this selection. These plants can be converted into resistant and productive ones through top working.  Such an exercise leads to a minimization of the requirement for fungicide sprays and finally its total elimination as practiced in Swarnagiri and Chowdirange Estates.

Table 3. Yield and Quality details

Estate Name Yield of Clean Coffee/acre % of A-grade % of Pea berry Cup Quality Out turn Processing method
Swarnagiri 680-700 kg 64-65 % 11-13 % FAQ Fruit  38% Parchment-Bean 84 % Wet
Chondirange 750 kg 64-65 % 10-15 %  – Fruit 38 % Parchment Bean 84 % Wet
Raxidi 740-800 kg Wet + Dry
Dewangudda 650-700 kg 50-55 % 45-48 % & 85 % Wet + Dry
Basaveshwara 750 kg 60-65 % 45 & 85 % Wet
Coodanhally 750 kg 40 % Wet
CCRI 750-800 kg 60-65% 10-15% FAQ 45% & 85% Wet

Agro-techniques adopted for Sln. 6

Estate Name Fertilizer dosage Weed Control Shade pattern Pest Control  Organic manures Irrigation Desuckering/Cemting/Pruning
Swarnagiri                            NPK/acre in 2 rounds Manual Medium Tracing for WSB incidence 3-4 plants/acre removed in a year Only to support blossom showers, if inadequate 2 times/year Light pruning after harvest in December to remove whippy wood
Chondiragi 60:40:60 kg NPK/acre/year. N and K applied in 3 splits while P is applied once as broad cast Manual Medium Tracing for WSB incidence 3-4 plants/acre removed in a year 8 kg/plant Only to support blossom showers, if inadequate 2 times/year Light pruning after harvest in December to remove whippy wood
Raxidi 110:112:96 kg NPK/acre/year in 2 rounds Manual Medium Lindane spray stopped since 3 years.  Borer incidence on 0.2-0.3 % plants      
Diwangudda 103:97:150 kg NPK/acre/year in 2 rounds Manual Medium 3-4 % plants infested (Lindane spray not done due to low prices) Poultry manure Irrigation during march every year Handling in 4 rounds pruning once
Basaveshwara 126:126:126 kg NPK/acre/year in 3 rounds Manual Medium to low 2 Plants/acre removed due to borer 200 gms Akshya/plant Irrigated Higher harvest reported  
Coodanhally 80:80:80 kg NPK/acre/year in 4 rounds (reduced to 40:40:40 in 2 rounds for the past 2 years Manual Medium 0.2-0.3 Not irrigated Light handling once in year (Pruning/handling became irregular due to price situation)

Breeding Efforts to Further Improve Robarbica

A generally reported shortcoming of Selection-6 is its relatively poor out-turn. This is a result of the presence of empty locules in the fruits of this selection of interspecific hybrid descent. Empty locules in the produce can be reduced through selection and presently, efforts are devoted to identify mother plants that produce least empty locules (i.e. under 5%), low peaberries (also under 5%) and triage (2-3% or less). This selection is expected to improve out-turn of the Selection-6 in just one generation of advancement as the presently exploited materials are in F2 and F3 generations.

Further, crosses were effected to render this selection more resistant to leaf rust as follows:

New Crosses of Selection-6

Sl. No. Parentage Accession No.
01 S.1156 10/4 (BC II ) x  S.2769 25/16  (HDT) S.4369
02 S.1156 10/4  x S.2828-5 S.4370
03 S.2828-5 x S.1156 10/4 S.4371
04 S.1156 10/4  x S.2828-33 S.4372
05 S.2828-33 x S.1156 10/4 S.4373
06 S.2828-5 x S.2828-33 S.4374
07 S.2828-33 x S.2769 25/16 S.4375
08 S.2828-33 x S.2828-5 S.4376

Plants obtained from the above crosses were vigorous and spreading types. They manifest high resistance to leaf rust in the field. Productivity of S.4369 and S.4375 is reasonably high. Individual plants in these progenies were marked for further breeding.


Anonymous. 2000. Coffee Guide. Central Coffee Research Institute, India.

Bertrand B, Guyot B, Anthony F, Lashermes P. 2003. Impact of the Coffea canephora gene introgression on beverage quality of C. arabica. Theor. Appl. Genet. 107: 387-394.

Bettencourt AJ, Lopes J, Palma S. 1992. Factores geneticos que condicionum a resistencia as racas de Hemileia vastatrix Berk. et Br. dos clones tipo dos grupos 1, 2 e 3 derivados de Hibrido de Timor. Broteria Genetica XIII (LXXX): 185-194.

Bettencourt AJ. 1973. sCpnsideracoesn gerais sobre o Hibrido de Timor. Instituto Agronomico, Campinas, Sao Paulo, Brazil.

Cramer PJS. 1957. A Review of Literature of Coffee Research in Indonesia. Inter American Institute of Agricultural Sciences, Turrialba.

DeVreaux M, Vallayes G, Pochet P, Gilles A. 1959. Recherches sur l’autosterilite du cafeier robusta (Coffea canephora Pierre). INEAC Serie Scientifique 78. 48p.

Eskes AB. 1989. Resistance. In: Coffee Rust: Epidemiology, Resistance and Management. (Eds. Kushalappa AC, Eskes AB), pp. 171-292. CRC Press, Boca Raton.

Fazuoli LC, Carvalho A, Monaco LC, Teixeira AA. 1977. Qualidade de bebida do café Icatu. Bragantia 36: 165-172.

Ganesh D, Ram AS, Prakash NS, Mishra MK, Ahmed J, Jagadeesan M, Reddy AGS, Srinivasan CS. 2002. Evaluation of Coffea liberica x Coffea eugenioides and its progenies for yield, leaf rust resistance and quality. In: PLACROSYM XV (Eds. Sreedharan K, Vinod Kumar PK, Jayarama, Basavaraj MC), pp. 72-77. Central Coffee Research Institute, India.

Monaco LC. 1977. Consequences of the introduction of coffee rust in Brazil. Annals New York Acad. Sci. 287: 57-71.

Narasimhaswamy RL. 1960. Arabica selection S.795: Its origin and performace – A study. Indian Coffee 24: 197-204.

Petracco M. 2000. Organoleptic properties of espresso coffee as influenced by botanical variety. In: Coffee Biotechnology and Quality (Eds. Sera T, Soccol CR, Pandey A, Roussos S), pp. 347-353. Kluwer Academic Publishers, Dordrecht, London, Boston.

Ram AS, Sreenath HL. 2000. Genetic fingerprinting of coffee leaf rust differentials with RAPD markers. In: Coffee Biotechnology and Quality (Eds. Sera T, Soccol CR, Pandey A, Roussos S), pp. 197-208. Kluwer Academic Publishers, Dordrecht, London, Boston.

Ram AS. 2001. Breeding for rust resistance in coffee: The gene pyramid model. J. Plantn. Crops. 29(1): 10-15.

Rodrigues Jr. CJ, Bettencourt AJ, Rijo L. 1975. Races of the pathogen and resistance to coffee rust. Annu. Rev. Phytopathology 13: 49-70.

Sreenivasan MS. 1987. Cyto-embryogical studies of Robusta-Arabica coffee hybrids. Ph.D. Thesis. University of Mysore, Mysore.

Srinivasan CS, Ramachandran M. 1997. Selection 5B – S.2931 (S.333 x Devamachy Hybrid) – An old Arabica hybrid rediscovered with promising features. Indian Coffee 61(11): 4-6.

Srinivasan CS, Vishveshwara S. 1980. Selection in coffee: Some criteria applied and results obtained in India. J. Coffee Res. 10: 53-62.

Sybenga J. 1960. Genetics and cytology of coffee: A literature review. Bibl. Genet. 19: 217-316.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: