F1 hybrids

Yield in the Central American trials#

A comparison published in 2011 ran 15 trials in Costa Rica, El Salvador and Honduras, at 750–1,580 m. Of these, 10 were under agroforestry and 5 in full sun.¹ About 20 hybrids from crosses of American cultivars with Ethiopian and Sudanese accessions (ET6, ET15, ET25, E41, E416, E531, Anfilo and Rume Sudan) were grown beside pure-line cultivars such as Caturra, Catuai, Pacas and CR95.¹ Yield was recorded over three to seven harvests per trial and expressed in grams of green coffee per tree per year.¹

The hybrid mean was above the line mean in every trial, by 8% to 127%; the difference was significant in 14 of the 15 trials.¹ Over all trials, the hybrid mean was 460 g against 290 g under agroforestry (+58%) and 754 g against 562 g in full sun (+34%).¹ In nine trials, the best hybrid, named Centroamerica in the paper, produced on average about 37% more than the best line of each trial.¹ The authors conclude that hybrids could encourage producers who had moved to full sun to return to agroforestry.¹

Stability and earliness also favoured the hybrids. The coefficient of variation of yield ranged from 29% to 70% for hybrids and from 46% to 94% for lines; the difference was significant at eight sites.¹ The hybrids’ earlier yields were statistically significant in nine of the 15 trials.¹

A later trial in seven contrasting environments compared new F1 clones with conventional varieties. The conventional varieties were the least productive, producing 50% less than the best hybrid.⁵ For Starmaya, a multi-site trial in Nicaragua found 30% more coffee per plant than Caturra and Marsellesa. In a commercial field at Matagalpa, Starmaya averaged 41.1 bags (46 kg) of green coffee per hectare over three harvests (2015–2018), against 30.4 for Marsellesa.³

Cup quality and other traits#

The 2011 paper reports earlier sensory evaluations in which the hybrids performed as well as or better than the traditional lines.¹ In the 2020 trial, H1-Centroamericano and Starmaya contained more d-limonene than Caturra, and both scored better than the control in sensory analysis.⁵ These are results of named trials, not general properties of F1 hybrids.

Heterosis has a cost in fertility. The 2005 study found the hybrids significantly more sterile than their parental lines.² Selection on an index of yield, bean weight and fertility improved yield and bean weight, but the gain in fertility was not significant.² Rust resistance depends on the parents: in Starmaya, field observations found 8% of plants susceptible to leaf rust, against 1% in Marsellesa.³

Hybrids and leaf rust in a controlled study#

A 2017 study tested whether hybrid vigour affects leaf rust in arabica that carries no specific resistance. It compared the inbred line Caturra with two F1 hybrids, GPFA 109 and GPFA 124. All three are susceptible, and the hybrids were not expected to carry specific resistance to prevalent races.¹³ Six-month-old potted plants were grown in a phytotron, a climate-controlled growth room, under eight treatments.¹³ These combined two day/night temperature regimes (27–22 °C and 23–18 °C), two light levels and two nitrogen doses.¹³ The light levels, 300 and 1,000 µmol m⁻² s⁻¹ of photosynthetically active radiation, were meant to mimic agroforestry and full sun.¹³ The nitrogen doses totalled 134 and 266 mg per plant over 11 weeks.¹³ Four plants per treatment were inoculated with a mixture of rust spores collected in Colombia.¹³

Sporulation was measured as the weight of spores per infected leaf area, 43 days after inoculation.¹³ Under the warmer regime, it was significantly lower in the hybrids than in Caturra in all four combinations of light and nitrogen. Under the cooler regime, the difference was significant only with high light and low nitrogen.¹³ The paper reports slower infection in the hybrids in every treatment, although with the warmer regime and high light the course in GPFA 109 was very similar to that in Caturra.¹³ With the warmer regime, low light and low nitrogen, the latent period lasted about 21 days in Caturra and 37 days in the hybrids.¹³ The authors suggest that hybrid vigour may have reduced disease incidence.¹³

Temperature, light and nitrogen also changed sporulation. In Caturra, the warmer regime gave about 2,000 times more sporulation than the cooler one according to the abstract; the results section gives 10,000 times more spores per infected leaf area.¹³ In Caturra, high light with low nitrogen gave significantly more spores than the other treatments under both regimes.¹³ Under shade and sufficient nitrogen, rust incidence on the hybrids was very low.¹³ These results come from young plants, two hybrids and one line in controlled conditions. For the field, the authors cite only their own unpublished preliminary observations that susceptible F1 hybrids are less severely attacked than pure lines.¹³

Somatic embryogenesis and rooted cuttings#

In the coffee process described in a 2018 review, the whole mass-propagation cycle takes about 24 months.⁴ Embryos are multiplied in liquid culture, at 15,000–50,000 somatic embryos per gram of embryogenic suspension depending on the genotype, and pre-germinated in temporary immersion culture.⁴ Acclimatisation in the nursery took 20–30 weeks in the 2010 process.⁴

Cloning in culture can introduce mutations (somaclonal variation). The review reports rates of about 1.3%, and 0.9% in a sample of 800,000 plants; adjusted protocols brought it below 0.5%, with over 99% of trees conforming to the mother plant.⁴ Horticultural rooted mini-cuttings taken from the laboratory plantlets multiplied each plant 14 times in six months and halved the production cost.⁴

According to a coffee science textbook, cuttings of upright shoots grown on mother plants have been used on a large scale in Central America to propagate arabica hybrids.¹² Since 1997, producers across Central America have had access to some F1 hybrid clones.¹ As of 2018, more than 20 million arabica hybrid plants had been planted in the region.⁴

Male sterility and seed purity#

The pollen-free tree, CIR-SM01, is a natural mutant with a tall growth habit.³ Its sterility comes from a malfunction of the tapetum, the tissue that feeds developing pollen, at an early stage.³ The trait is recessive: in four F2 populations with 1,250 plants, 18.3% were sterile, which the authors suggest may reflect a two-gene (epistatic) 9:4:3 ratio.³ The sterility was complete and stable across environments and plant ages.³

Seed gardens rely on natural pollination by the pollen parent.³ Purity was the weak point. In 2016 fields, a mean of 16% of Starmaya plants were off-types, against 3% for Marsellesa.³ The authors trace this to stray pollen and to residual heterozygosity in both parents, and aimed for at most 5% off-types with fixed parents, windbreaks and insect-proof netting.³

The method limits the number of hybrids that can be made by seed to the number of available male-sterile plants.³ Ruiru 11 shows the older seed route: hand pollination. WCR notes that this has made Ruiru 11 seed hard to supply in sufficient volume.¹⁰

Reference values#

Cost and access#

Cloned plants cost more than seedlings. In 2011, the extra cost per tree was USD 0.5–0.6, an investment of USD 2,500–3,000 per hectare.¹ According to the 2018 review, adding rooted mini-cuttings cut the production cost from USD 0.50–0.60 to 0.25–0.30 per plant.⁴ The same review gives a sales price of about USD 0.8–1.0 for a cloned F1 plant against USD 0.3–0.5 for a seed-derived plantlet, and expects the extra cost to be repaid after three to four harvests.⁴

Even with mini-cuttings, a cloned hybrid cost about 1.5 times as much as a traditional seedling in 2019.³ The Starmaya authors estimated a hybrid seed at about USD 0.07, half the price of a cloned plantlet.³ According to a coffee science textbook, F1 varieties need vegetative propagation, which raises plantlet costs, and plantlet cost remains a barrier to wider use of in vitro methods.^11,12 According to WCR, only a handful of F1 hybrids had become commercially available to producers, and only in select countries.⁶

Open questions#

• Seed purity at scale. Whether seed gardens can keep off-types at or below the 5% the Starmaya authors set as commercially acceptable is not yet shown in a published field study opened for this entry.³
• More male-sterile mothers. Seed-propagated hybrids are limited to crosses with the male-sterile plants available.³
• Performance outside Central America. Starmaya was introduced to Vietnam and Cameroon for comparison with local varieties; no results were published in the sources opened for this entry.³
• Rust on F1 hybrids in the field. The lower sporulation of two hybrids was shown in young plants in a phytotron; the authors’ field evidence is unpublished preliminary observation.¹³

Key studies#