---

COFFEE
Coffea spp., family Rubiaceae

Worldwide about 90 percent of the coffee produced comes from C. arabica L., 9 percent from C. canephora Pierre ex. Froehner, and less than 1 percent from C. Iiberica Bull. ex. Hiern. Brazil produces about half of the world supply of coffee, about 1.5 million tons. The most important cultivar in Brazil is 'Mundo Novo' with about three-fourths of a million acres planted (Monaco and Carvalho 1969). The combined African states are second with about 0.8 million tons. Colombia produces about 0.5 million, and El Salvador, Guatemala, and Mexico about 0.1 million tons each. In the United States, coffee is produced to an extremely limited extent only in Hawaii and Puerto Rico. In the mid-1920's, coffee production in Puerto Rico exceeded 9,000 tons, but production diminished during World War II. It is now being rehabilitated. Production of 'Koona' coffee in Hawaii is declining.

In Africa (primarily Angola, Congo, Ivory Coast, Uganda, and Madagascar), the principal coffee is C. canephora, which is now in great demand for the manufacture of instant coffee.

Yield per acre varies enormously from over 2,000 pounds of clean coffee per acre in Hawaii to only 360 lb/acre in Brazil (Chandler 1958*, Haarer 1962, Purseglove 1968*, Wellman 1961). In the United States, consumption is approximately 16 pounds per head per annum. We import annually about 1.5 million tons.

Crane and Greene (1948-53) made an extensive review with pertinent abstracts of the literature on coffee.

Plant:

Coffee trees may grow to a height of 20 to 30 feet if unpruned, but in cultivation they are pruned to about 6 feet to facilitate harvesting of the berries. Much of the coffee in Brazil is grown in the full sun although it prospers in the shade of taller trees. Considerable care is required to keep the trees in proper productive condition. The plant is susceptible to frost, intense heat, drought, and diseases. It grows best at average temperatures of about 60 deg to 75 deg F, with 75 inches of rainfall but needs a 2- to 3-month dry period to initiate flowering. The trees are usually planted about 8 feet apart; they begin bearing at about 4 years, reach full bearing by 15 years, and may bear 6 to 100 years longer (Hearer 1962).

The fruit is a drupe (or stone fruit), but is often called a cherry or berry. It is individually picked at the proper stage of ripeness. About 5 pounds of fruit yield 1 pound of clean coffee. One cherry, usually bearing two seeds, develops from a blossom. The less-desired, one-seeded cherry is referred to as a "peaberry." The fruit is dried or processed in water to remove the skin, cleaned, and bagged for shipment and use. The beans are roasted, ground, and brewed with hot water to produce a stimulating beverage (Wellman 1961).

The taxonomy of Coffea is confused, but doubtless the three previously mentioned species are the most important. There are also many botanical varieties, mutants. and cultivars.

Inflorescence:

The fragrant white flowers occur in clusters of two to 20 in the leaf axils (fig. 81). Each flower is about an inch long by an inch deep, the tubular corolla is about 1/3-inch long and the calyx is usually made up of five flared petals, forming a starlike inflorescence. There are usually five long stamens on short, curving filaments and a long style with a two-part stigma. The stamens are attached to the corolla between the lobes so that the anthers are about the height although not necessarily close to the stigma. Pollen is shed immediately after the flower opens, and the stigma is immediately receptive. Nectar is secreted at the base of the tubular corolla, but accessible to honey bees and many other insects. Both nectar and pollen are attractive to many kinds of insects.

On sunny days, the flower generally opens early in the morning and pollen shedding starts soon afterwards. The pollen is produced in comparatively small quantities and is not sticky. It may be transported by wind and insects (Carvalho and Krug 1950). If the day is cloudy, the flower may remain closed but self-pollination can occur within the flower. Two days after opening, or fertilization within the closed flower, the parts begin to wither and fall, leaving the ovary (Hearer 1962). Krug (1935) and Montealegre (1946) suggested that the lingering of the withered blossoms on the tree is an indication of an absence of insect pollinators, whereas if the petals fall freely and soon, they have been pollinated and a good crop should be expected.

Purseglove (1968*) stated that approximately 40 percent of the flowers set fruit and are harvested as mature fruit, a certain number of buds never swell, but may persist until harvest; others fall in the early stages of growth, mainly in the first 10 weeks. Ferwerda (1951) stated that a tree may produce 10,000 to 50,000 flowers but 70 to 90 percent of them fall. This fall may be due to self-incompatibility in the flower, incompatible pollen in general, absence of pollination, or defective embryo sac. Mayne (1934) kept close observation on about 20,000 original coffee buds over a 3-year period and reported that 37.6, 41.4, and 38.6 percent of them were harvested as mature coffee.

[gfx]
FIGURE 81. - Coffee in full bloom.

Pollination Requirements:

Wellman (1961) stated that C. arabica is self-fertile, yet at times some insect pollination occurs but it is not necessary. However, he stated that the other two species, C. canephora and C. Iiberica, are self-sterile and require action of wind or insects. Ferwerda (1936) stated that C. excelsa Cheval., C. Iiberica, and C. robusta Linden [= C. canephora] were self-sterile. Haarer (1962) is in agreement with Wellman on this. Amaral (1952) showed that caged C. arabica plants produced 39 percent less coffee than open plants. Later, Amaral (1960) conducted another experiment using C. arabica, cv. 'Caturra KMC', in two flowering seasons, in which he recorded the set of fruit on branches caged to exclude bees and compared production with branches freely visited by bees. Fruit setting on the protected branches was 61.7 percent, whereas on branches visited by bees it was 75.3 percent, indicating a slightly beneficial effect. According to Free (1970*), Sein (1959) had a set of 60 percent and 70 percent on bushes caged to exclude bees and bushes not caged, respectively.

Later experiments by Amaral (1972) leave no doubt that C. arabica is definitely benefited by bee pollination. He showed that bees increased set of C. arabica cv. 'Mundo Novo' (in cages with bees) about 82 percent over trees in cages without bees.

Carvalho and Krug (1950) studied the effect of cross-pollinating agents on C. arabica and concluded that 7.3 to 9.0 percent of seed resulted from cross-pollination. They did not indicate the intensity of activity of the pollinating agents. Lower (1911) claimed that honey bees are of great value in pollinating coffee in Puerto Rico during rainy seasons. Mendes (1961) showed that tubes from foreign pollen grew faster than self pollen, thus insuring crossing.

Montealegre (1946) believed that insect pollination played a much more important role in the production of coffee in Puerto Rico than was commonly thought, and he believed that honey bees increased coffee yields; however, no data was presented to support his beliefs. Nogueira- Neto et al. (1959) concluded that the part played by insects in pollinating C. arabica cv. 'Bourbon' was of only secondary importance. Sein (1923) concluded that bees are beneficial to coffee in Puerto Rico, and Rudin (1942) reported that coffee plantations were installing colonies of honey bees for pollination of coffee in Puerto Rico. Zimmerman (1928) believed that honey bees played only a minor role in the pollination of coffee, at Least in the case of the larger plantations, and no reference is made to supplementing the number of pollinating insects in the area. The quite limited evidence indicates that C. arabica is not dependent upon pollinating insects, but under some conditions at least insects can be beneficial, possibly to a substantial degree to this species of coffee. C. canephora is self-sterile (Devreux et al. 1959, Purseglove 1968*) as is the selection reported by Krug et al. (1950) and Mendes (1949), made up of C. arabica X C. dewevrei Wildem. & Dur.

Pollinators:

Ferwerda (1948) stated that pollen transfer of the pronouncedly cross-pollinated C. robusta Linden is accomplished by wind. Carvalho and Krug (1950) concluded that insects and wind were of about equal importance in the cross-pollination of coffee in Brazil. Carvalho et al. (1969) reported 7.3 to 9.05 crossing of which 4.8 to 5.3 percent was accredited to gravity, 2 to 5 percent to wind, and 0 to 2 percent to insects. McDonald (1930) suggested that growers in East Africa keep honey bee colonies on their plantations. Lower (1911) indicated that bees benefited coffee in Puerto Rico and that colonies should be placed in the coffee plantations. Montealegre (1946) also indicated that honey bees benefited coffee in Costa Rica. Nogueira-Neto et al. (1959) stated that larger bees such as honey bees and Melipona quadrifasciata Lepeletier were more efficient pollinators of coffee in Brazil, but the overall benefit was considered rather insignificant. Sein (1923, 1959) showed that honey bees were beneficial to coffee in Puerto Rico, and Rudin (1942) stated that Costa Ricans were installing colonies of honey bees in their coffee plantations for pollination purposes. Amaral (1972) stated that honey bees were the dominant pollinating agent in the area of his studies. He further showed that colonies in the coffee groves collected predominantly coffee pollen (80 percent of the pellets identified) during the peak of flowering.

In most instances where pollination of coffee was studied, the honey bee was the most important pollinating insect visiting the flowers.

Pollination Recommendations and Practices:

The use of bees as pollinators of coffee has not been recommended on C. arabica, although the evidence indicates that a beneficial effect is obtained when pollinators were concentrated on this crop during the brief period of its flowering. Amaral recommended that honey bee colonies be placed every 100 m in the coffee grove just before flowering starts. The coffee specialist might be agreeably surprised to discover the increase derived from a large-scale community-type honey bee pollination program. The other two important species, C. canephora and C. Iiberica, are self- sterile, and they would appear to be greatly benefited by bee pollination. Considering the recent increased importance of the self-sterile African C. canephora in the production of instant coffee, the use of bees in its pollination would appear to be highly profitable.

LITERATURE CITED:

AMARAL, E.
1952. [ESSAY ON THE INFLUENCE OF APIS MELLIFERA L. ON THE POLLINATION OF THE COFFEE PLANT (PRELIMINARY NOTE).] Esc. Super. de Agr. "Lutz de Queiroz" (Sao Paulo, Brazil) Bul. 9, 6 pp. [In Portuguese.]

______ 1960. [INFLUENCE OF INSECTS ON POLLINATION OF CATURRA COFFEE.] Rev. de Agr. (Sao Paulo) 35(2): 139-147. [In Portuguese, English summary.]

______ 1972. lNSECT POLLINATION OF COFFEA ARABICA L., AND RADIUS OF ACTION OF APIS MELLIFERA LINNAEUS 1758, IN THE COLLECTION OF POLLEN IN COFFEE PLANTATIONS IN BLOOM.] 82 pp., Dept. Ent., "Lutz de Quiroz" Superior School of Agr., Sao Paulo Univ., Piracicaba S.P. Brazil. [In Portuguese, summary and general conclusions in English. ]

CARVALHO, A., and KRUG, C. A.
1950. [POLLINATING AGENTS FOR COFFEE (COFFEA ARABICA). ] Bragantia 9: 11 - 24. [In Portuguese.]

______ FERWERDA, F. P., FRAHM-LELIVELD, J. A., and others.
1969. COFFEE. In Ferwerda, F. P., and Wit, F., eds. Outlines of Perennial Crop Breeding in the Tropics, pp. 189-241. H. Veenman and Zonen, N. V., Wageningen, The Netherlands.

CRANE, J. C., and GREENE, L.
1948-53. ABSTRACTS OF SOME OF THE LITERATURE PERTAINING TO COFFEE. 2 v., 150 pp. U.S. Dept. Agr. Off. Foreign Agr. Relat. Tech. Collab. Branch.

DEVREUX, M., VALLAEYS, G., POCHET, P., and GILLES, A.
1959. RESEARCH ON THE SELF-STERILITY OF ROBUSTA COFFEE (COFFEA CANEPHORA PIERRE). Pub. lnst. Nat. Agron. Congo Belge: Ser. Sci. 78: 44. 4262 (Abstract) Plant Breed. 29(4): 837.

FERWERDA
1936. [POLLINATION IN COFFEE SPECIES GROWN IN THE DUTCH EAST INDIES.] Zuchter 8: 92 - 102. [In German.] See Crane and Green 1948 for abstract.

______ 1948. COFFEE BREEDING IN JAVA. Econ. Bot. 2(3): 258-272.

______ 1951. [FRUIT DROP IN ROBUSTA COFFEE AND ITS RELATION TO POLLENIZATION AND FERTILIZATION.] Vakblad voor Biologen 31: 123 - 130. [In Dutch.] Abstract in Euphytica 1(3): 232. 1952.

HAARER, A. E.
1962. MODERN COFFEE PRODUCTION. 495 pp. Leonard Hill, London.

KRUG, C. A.
1935. HYBRIDIZATION OF COFFEE, A PRELIMINARY STUDY OF FLOWERING HABITS AND METHODS OF CROSSING. Jour. Hered. 26: 325 - 330.

______MENDES, J. E. T., CARVALHO, A., and MENDES, A. J. T.
1950. [A NEW TYPE OF COFFEE.] Bragantia 10(1): 11-25. [In Portuguese, English summary.]

LOWER, W. V.
1911. BEEKEEPING IN PUERTO RICO. Puerto Rico Agr. Expt. Sta. Cir. 13, 31 pp.

MAYNE, W. W.
1934. ANNUAL REPORT OF THE COFFEE SCIENTIFIC OFFICER 1933-34. Mysore Coffee Expt. Sta. Bul. 12. See Crane and Green 1948 for abstract.

MCDONALD, J. H.
1930. COFFEE GROWING: WITH SPECIAL REFERENCE TO EAST AFRICA. 205 pp. East Africa Ltd., London.

MENDES A. J. T.
l919. [CYTOLOGICAL OBSERVATIONS IN COFFEE.] Bragantia 9(1-4): 25-34. [In Portuguese, English summary.]

______ 1961. [SPEED OF POLLEN TUBE PENETRATION IN COFFEE ARABICA.] Bragantia 20(1): 495 - 502. [ln Portuguese, English summary.]

MONACO, L. C., and CARVALHO A.
1969. COFFEE GENETICS AND BREEDING IN BRAZIL. Shell Pub. Health and Agr. News 12(2): 74-77.

MONTEALEGRE, M. R.
1946. [THE FERTILIZATION OF COFFEE FLOWERS.] Revista del lust. de Defensa del Cafe de Costa Rica 15: 337 - 340. [ln Spanish.]

NOGUEIRA-NETO, P., CARVALHO, A., and FILHO, H. A.
1959. [THE EFFECT OF THE EXCLUSION OF POLLINATING INSECTS ON THE YIELD OF BOURBON COFFEE.] Bragantia 18: 441 - 468. [In Portuguese, English summary.]

RUDIN, J.
1942. [BEES AS POLLINATION AGENTS IN FRUIT TREES.] Rev. Inst. Defensa Cafe de Costa Rica 12(96): 490 - 491. [ In Spanish. ]

SEIN, F., JR.
1923. [BEES IN COFFEE PLANTATIONS.] Puerto Rico Insular Sta. Cir. 79, 6 pp. [ln Spanish]

______ 1959. [DO BEES HELP COFFEE?] Hacienda 55: 36-50. [ln Spanish.]

WELLMAN, F. L.
1961. COFFEE; BOTANY CULTIVATION AND UTILIZATION. 488 pp. Leonard Hill, London.

ZIMMERMANN, A.
1928. [POLLINATION OF COFFEE TREES.] His Kaffee, ed. 2, pp. 31-34. Deutscher Auslandverlag, Hamburg. [In German.]

---