MIcronutrIent AccuMulAtIon In conIlon coffee BerrIes wIth DIfferent MAturAtIon cycles

the number of days between anthesis and maturation of conilon coffee berries varies according to the genotype. thus, it is believed that periods of greater nutrient demand for fruit formation also vary according to the genotype, directly influencing fertilizer management. the goal of this study was to establish accumulation curves for the micronutrients boron, copper, iron, manganese, and zinc in conilon coffee trees with different maturation cycles. the experiment was conducted in nova Venécia, state of espírito santo, Brazil, during the reproductive cycle of the 2010/2011 crop year. four coffee genotypes with different maturation cycles (early, intermediate, late, and super-late) were studied. A completely randomized experimental design was used with five replications. The treatments correspond to the accumulation of B, cu, fe, Mn, and Zn in the berries every 28 days in the period from flowering to harvest. The early, intermediate, and late genotypes accumulated Fe, Cu, and Mn in a similar manner, with sigmoid curves, whereas the super-late genotype accumulated these nutrients exponentially. Zn was accumulated by all four genotypes following a sigmoid curve. the early, intermediate, and late genotypes accumulated B linearly, whereas the super-late genotype accumulated B following a sigmoid curve. the maturation cycle of the genotype must be taken into account to apply the correct rate of micronutrient fertilization in coffee plantations.


IntroDuctIon
More than 120 species have been described for the Coffea genus; however, only C. arabica, C. canephora, and C. liberica are commercially cultivated (Davis et al., 2011).C. arabica (arabica coffee tree) stands out as the main commercial crop, but the cultivation of C. canephora (robusta or conilon coffee tree) has significantly contributed to an increase in global coffee production (Conab, 2013).
The conilon coffee tree is characterized by reproduction through allogamy.Using asexual propagation is necessary to obtain productive varieties with a well-defined maturation cycle (Bragança et al., 2001).These varieties are polyclonal, consisting of sets of genotypes that are usually grouped according to their maturation cycles, which vary from early to super-late (Fonseca et al., 2004).
Knowing the maturation periods of coffee berries is essential for agricultural planning and to predict harvest yield, quality, and commercialization (Bardin-Camparotto et al., 2012).There have been important studies of gene expression at different stages of fruit development (Budzinski et al., 2011).The variation attributed to the fruit maturation cycle is related to the climate conditions for cultivation and/or the coffee genotype being used (Pezzopane et al., 2003;Petek et al., 2009).
Although varieties are formed by grouping clones with similar characteristics, the genotypes composing a single variety may display different behaviors regarding nutrient uptake and accumulation in their tissues.For the thirteen genotypes that compose the clonal variety 'Vitória Incaper 8142 ', Covre et al. (2013) reported different nutrient contents in the shoot and roots of seedlings 210 days after staking, a fact that further suggests differences among genotypes in the efficiency with which nutrients are taken up and accumulated.
Monitoring the accumulation of nutrients in fruits during the reproductive period has been one of the main tools used to estimate the crop's nutritional needs and to identify the optimal moments to apply fertilizers.Partelli et al. (2014) studied the nutrient accumulation behavior in fruits of different genotypes of conilon coffee and suggested that those with the shortest maturation cycles are the quickest to accumulate dry matter, N, P, K, Ca, Mg, and S.These authors also reported that fertilizers must be specifically managed for each genotype.
Given the scarcity of information on the micronutrient contents and rates of uptake in coffee berries, since plants have different aging cycles have a different nutrient absorption curves, the goal of this study was to establish accumulation curves for the micronutrients B, Cu, Fe, Mn, and Zn in conilon coffee plant fruits with different maturation cycles.

MAterIAl AnD MethoDs
The experiment was carried out in Nova Venécia reproductive cycle of the 2010/2011 crop year.The area is located at the geographical coordinates 18° 43' 46" S, 40° 23' 10" W, at a mean altitude of 100 m.At the study site, the minimum temperature ranges from 11.8 to 18 °C, and the maximum temperature ranges from 30.7 to 34 °C, with rainfall of approximately 1,200 mm per year (Incaper, 2012).During the experiment, the crop was duly irrigated, so there was no water deficit.
To fertilize production, 110 g superphosphate and 80 g of 20-00-20 fertilizer were applied, per plant, in November and December 2010, respectively, whereas in March and May 2011, respectively 100 and 120 g of 20-00-20 fertilizer were applied.
The selected genotypes were, respectively, the 12 V, 10 V, and 13 V clones of conilon Vitória 8142 variety and Ipiranga 501 variety.Four coffee tree genotypes (clones) with different maturation cycles (early, intermediate, late, and super-late) were used.Treatments corresponded to periods in which fruits were gathered to ascertain nutrient accumulation during each genotype's period of fruit formation.
The experimental design was completely randomized with five replications.Initially, plagiotropic branches were marked with the same pattern in 70 random plants of each genotype.The experimental sample consisted of one plant from which a plagiotropic branch was removed every 28 days of the period from flowering to fruit maturation to assess fruit dry matter, as well as the Fe, Zn, Cu, Mn, and B concentration and accumulation in the fruit.The branches had, on average, 30 productive berry clusters.
Sampling began on August 14, 2010, 20 days after anthesis for the early and intermediate genotypes (12 V and 10 V) and on September 11, 2010 for the late and super-late genotypes (13 V and Ipiranga 501).Prior to that, 65 uniform plagiotropic branches were randomly marked for each genotype, and five branches were removed/sampled at random every 28 days.Sampling ended on March 6, April 8, May 7, and July 3, 2011 for the early, average, intermediate, and late genotypes, respectively.Sample fruits were extracted and dried in a forced-air oven at 70 °C until the fruits reached constant weight; then, their dry matter was determined with a 0.001 g precision scale.The fruits were ground using a Willey stainless steel mill and sieved through a 0.841 mm fine mesh to conduct chemical analysis of Fe, Zn, Cu, Mn, and B. The analyses were carried out using the methods described by Silva (2009), in triplicate.
Nutrient accumulation in the berries found on the branches was calculated, based on the dry matter and the concentrations of the respective nutrients.Subsequently, the percentage of accumulation at different times was computed, considering the value from the last batch as 100 %, when over 80 % of the fruits on the branches were completely mature.
Data were subjected to regression analysis, and mathematical models were chosen according to the significance of the regression coefficients of the equations and the F-test for regression (p≤0.05).The best fits were indicated by higher values of the coefficient of determination (R 2 ).Statistical analysis was performed using Assistat 7.6 (Silva, 2013).Data plots were created using mean values and their standard deviations.

results AnD DIscussIon
Early, intermediate, and late genotypes accumulated the nutrients Fe, Cu, and Mn similarly, with 3-parameter sigmoid curves, whereas the super-late genotype accumulated these nutrients exponentially (Figures 1, 2, and 3).The sigmoid curves displayed an initial phase of slower accumulation, followed by a phase of strong intensification (with higher rates) and a phase with lower rates at the end of the cycle of berry formation.In contrast, the exponential increase displayed by super-late genotype suggests lower accumulation rates at the beginning of berry formation and an exponential increase up to the final phase of berry formation, when this rate reached its highest values.
High Fe, Cu, and Mn accumulation rates were found in the intermediate phase, starting on the 76 th day after anthesis (Figures 1, 2, and 3).All studied genotypes accumulated Zn following a sigmoid curve, displaying a phase of slow accumulation at the beginning of berry formation, followed by a phase of fast accumulation in the intermediate period of the berry formation/maturation cycle.Despite the nutrient accumulation curves being similar, the beginning of the fast accumulation phase depended on genotype; moreover, the late and super-late genotypes displayed reduced accumulation rates near the end of the cycle, whereas the early and intermediate genotypes displayed high accumulation rates until the fruits were mature (Figure 4).The early, intermediate, and late genotypes accumulated B linearly, Micronutrient AccuMulAtion in conilon coffee Berries with Different... whereas the super-late genotype accumulated B according to a sigmoid curve (Figure 5).This fact suggests that the genotypes have different needs, requiring different management.
The sigmoidal behaviors displayed by early, intermediate, and late genotypes (Figures 1, 2,  and 3) are similar to results reported for arabica coffee plants (Coffea arabica L.) grown at different altitudes (Laviola et al., 2007), i.e., the main commercial species of the Coffea genus show similar behavior.However, the difference between the super-late clone and the other genotypes confirms the great genetic diversity that exists among genotypes of the C. canephora species, as suggested by Souza et al. (2013).
High Fe, Cu, and Mn accumulation rates in the intermediate phase, starting on the 76 th day after anthesis (Figures 1, 2, and 3), are similar to results reported for C. arabica var.'Caturra', for which the highest percentages of the accumulation of these nutrients were found between 90 and 120 days after anthesis (Ramírez et al., 2002).This period coincides with fruit's rapid expansion phase (Laviola et al., 2008), a period that requires more water flow to the fruit.The fruit, consequently, receives greater amount of Fe, Cu, and Mn, which participate as enzyme activators in several metabolic processes.Super-late genotype displayed greater accumulation of Fe, Cu, and Mn starting on the 160 th day.
28.12 (DAA -105.77)-Ŷ = 98.00/ ;r 2 = 0.98** Ŷ = 99.34/The behavior of late and super-late genotypes in regard to Zn is similar to reported for arabica coffee grown at different altitudes (Laviola et al., 2007), in which a significant increase was observed until the 134 th day after anthesis, after which Zn concentration in the berries became stable until the end of the 266-day cycle.Corroborating these results, 'Caturra' arabica coffee grown in Costa Rica accumulated all of its Zn by the 210 th day after anthesis, i.e., there was no increase during the last 30 days of fruit formation, namely from the 210 th to the 240 th day (Ramírez et al., 2002).Thus, it is believed that greater Zn accumulation in the initial/intermediate phase of fruit formation is related to the rapid expansion phase, in which there is intense cellular elongation, an event that depends on auxin (Taiz and Zeiger, 2010).The accumulation patterns of the early and intermediate genotypes suggest that fruit formation phases may overlap or be very brief, presumably due to the short period of time between anthesis and fruit maturation.
The period during which fruits require the most Zn coincides with the period of vegetative growth, which begins in September and extends to May (Partelli et al., 2010;2013), and this overlap can lead to competition for the nutrient between the reproductive and vegetative parts.Because of this competition, special attention must be given to this Zinc fertilization management should be studied in more detail given that the beginning of the growth season is the recommended period for phosphate fertilization, in a single application, and this fertilization may lead to Zn deficiency, due to antagonism between P and Zn (Dechen and Nachtigall, 2006).However, there was no evidence of this antagonism in this crop, under the soil and climatic conditions of the present study.
As for B (Figure 5), the results for the early, intermediate, and late genotypes differed from those reported for arabica coffee at different altitudes (Laviola et al., 2007).The super-late genotype displayed a long phase of slow accumulation, initiating a rapid expansion phase only after the 132 nd day.

conclusIons
The early, intermediate, and late genotypes accumulated Fe, Cu, and Mn similarly, with sigmoid curves, whereas the super-late genotype accumulated these nutrients exponentially.
Zinc was accumulated by all the genotypes studied according to the sigmoid curve, though with varying accumulation rates among the genotypes during maturation.early in the cycle and accelerated from advancing maturity according to a sigmoid curve.
The maturation cycle of the genotype must be taken into account to apply the correct rate of micronutrient fertilization in coffee plantations.
figure 2. copper accumulation in berries (as a percentage of total accumulation) of four genotypes (clones) of 'conilon' coffee, from anthesis to fruit maturation.Bars represent standard deviation of the mean.**: significant at 1 %.
figure 1. Iron accumulation in berries (as a percentage of total accumulation) of four genotypes (clones) of 'conilon' coffee, from anthesis to fruit maturation.Bars represent standard deviation of the mean.**: significant at 1 %.
figure 4. Zinc accumulation in berries (as a percentage of total accumulation) of four genotypes (clones) of 'conilon' coffee, from anthesis to fruit maturation.Bars represent standard deviation of the mean.**: significant at 1 %.

(
figure 5. Boron accumulation in berries (as a percentage of total accumulation) of four genotypes (clones) of 'conilon' coffee, from anthesis to fruit maturation.Bars represent standard deviation of the mean.**: significant at 1 %.