Soil pH, Caffeine, and Disease in Coffea arabica Lochner 1 A c omparison of s oil pH, l eaf c affeine c oncentration and d isease p revalence in Coffea a rabica Sofia Lochner Department of Molecular, Cellular, Biology University of California Berkeley EAP Tropical Biology and Conservation Program, Fall 2016 16 December 2016 ABSTRACT This study investigates the relationship between soil pH, caffeine content in leaves and disease prevalence in Coffea arabica Caffeine is a secondary metabolite that acts as an insecticide and antimicrobial to protect the coffee lea ves shoots, and fruits from fungus and herbivory I used caffeine w ater extraction and UV/V spectrophotometry to analyze the caffeine content in leaf samples from a total of forty trees in four different sites at the Life Monteverde farm in Ca itas, Guanacaste At each tree a soil sample was collected for pH analysis and a disease survey was conducted for Hemileia vastatrix, Mycena citricolor and herbivory Additionally I measured temperature, soil moisture, and light levels at each tree and compared between plots to see if there were significant differences in these factors depending on location. I compared d isease prevalence, caffeine content and soil pH amongst the forty sampled trees and found n o significant difference between any of these three factors The results from this study suggest that leaf caffeine concentration is dependent on other factors Una comparacin del pH del suelo, la concentracin de cafena en la h oja, y prevalencia de la e nfermedad en Coffea arabica RESUMEN Este estudio investiga la relacin entre el pH del suelo, el contenido de cafena en las hojas y la prevalencia de la enfermedad en Coffea arabica La cafena es un metabolito secundario que acta como un insecticida y antimicrobiano para proteger las hoja s, brotes y frutos del caf de hongos y herbivora. Utilic la extraccin con agua de la cafena y el espectrofotmetro UV / V para analizar el contenido de cafena en muestras de hojas de un total de 40 rboles en cuatro sitios diferentes en la finca Life Monteverde en Caitas, Guanacaste En cada rbol, tom una muestra de suelo para el anlisis del pH y realic un muestreo de presencia de Hemileia vastatrix, Mycena citricolor y herbivoria. Adems, med la temperatura, la humedad del suelo y los niveles d e luz en cada rbol para ver si haba diferencias significativas en estos factores dependiendo de la ubicacin. No hubo diferencias significativas de prevalencia de la enfermedad, contenido de cafena y pH del suelo entre los cuarenta rboles muestreados. Estos resultados sugieren que la concentracin de cafena foliar depende de otros factores no estudiados aqu ____________________________________________________________________________ __ The coffee industry is valued at 20 billion dolla rs and coffee is the main source of income for 25 million small producers worldwide (FAO, 2007). With millions of dollars in losses due to diseases and herbivory it is important to understand the inherent mechanisms that promote plant
Soil pH, Caffeine, and Disease in Coffea arabica Lochner 2 health, in an effort to reduce loss of profits and jobs. This study attempts to understand the relationship between caffeine, disease, and soil pH in Coffea arabica plants. Coffee plants produce caffeine, a sec ondary metabolite that has well documented effects on t he human central nervous system. W ith our global population consuming 2.25 billion cups of coffee every day coffee ranks as the most widely us ed psychoactive drug in the world (Dicum and Luttinger 1999 ) Besides its ability to stimulate the human body caffeine is shown to be a n important insecticide and antimicrobial H igh levels of caffeine found in the leaves are toxic to insects that attempt to eat the plant (Freeman, 2008) Caffeine at a dietary concentration of 0.3% kills nearly all larvae of the tobacco hornworm within 24 hours and similar results were obtained with other insects including butterfly larvae, mealworm larvae, milkweed bug nymph, and mosquito larvae. It has been s hown that the caffeine toxicity is primarily caused by inhibition of phosphodiesterase activity ( Nathanson, 1984). However caffeine can also be used to attract insects; for example bees get a beneficial energy boost from the low caffeine dose found in the nectar of coffee flowers enticing them to return and potentially boosting pollination rates (Fr eeman et al, 2008) Caffeine has also been shown to have antimicrobial and antifungal properties; one study found that caffeine significantly inhibited the growth of E.Coli (Ibrahim et al, 2006) and the fungus Xyleborus fornicates (Kumar et al, 1995) Caffeine is an energetically costly metabolite for the plant to produce but helps improve the overall fitness of the coffee plant in many ways ( Frischknecht, 1986). The leaves and seeds of Coffea arabica are about 1% caffeine by dry weight and this secondary metabolite is produced through a multistep biosynthetic process (Ashihara et al, 2008) The main pathway fo r caffeine production is a four step sequence of three methylation reactions and one nucleosides reaction Purine nucleotides contain four nitrogen atoms and are the starting material for this pathway ; t hus for this process to proceed the plants need a n ample source of nitrogen for adequate caffeine production (Ashihara et al, 2008). Since coffee plants attain nitrogen and many other essential compounds from the soil, the quality and nutrient content of soil is very important for overall health of the coffee tree. Coffea arabica thrives in volcanic, slightly acidic, and fertile soil (Zuchowski, 2007). Among the many factors that affect the health of cof fee plants, soil pH is critical; coffee plants can grow in neutral soil, but the optimum pH for best overall health and growth of a tree is between 5.0 and 6.0. However when pH is too low it can cause aluminum toxicity and deficiencies in critical nutrients such as phosphorous, calcium, and magnesium (Department of Agriculture, Forestry, and Fisheries 2012 ). Soil pH is hypothesized to be a major component in ino rganic nitrogen p roduction in agricultural soils (Kemmitt et a l, 2006). Total gaseous emissions of N 2 O, NO, and N 2 have repeatedly been shown to be less in acidic soil than in slightly alkaline soils This may be attributable to smaller amounts of organic carbon and mineral nitrogen available to the denitrifying p opulation under acid condition (Cooper, 2002). A possible factor influencing caffeine concentration could be suboptimal soil pH causing a deficiency in soil nitrogen reserves and thus limiting the caffeine biosynthetic pathway. Although it is known amongst farmers that soil pH is important for the health of coffee plants there is very little research about the correlation between soil pH and caffeine content. W hen soil pH is too low coffee plants are more prone to disease and predation ; since caffeine is a defense mechanism ( Ceja Navarro, 2015) it might be that soil pH affects the caffeine synthesis pathway. My central question is: does suboptimal soil pH affect the production of caffeine in Coffea arabica leaves? I predict ed that if a coffee plant grows in soil with a low pH it will
Soil pH, Caffeine, and Disease in Coffea arabica Lochner 3 produce less caffeine and thus be more prone to disease and predation. Presented here is an analysis of leaf caffeine content, soil pH, and disease prevalence in 40 coffee plants from four different farm sites at the Life Monteverde farm METHODS ONSITE SAMPLE COLLECTION: The study site was a sustainable coffee farm in Ca itas, Guanacaste called Life Monteverde Farm The farm plots were dispersed through secondary fo rest with somewhat variable conditions with respect to elevation, slope of field, integration of shade plants, and wind exp osure. I randomly selected ten trees to sample f rom each of the four selected farm sites. These four sites were selected because previous nutrient and pH data from the Instituto del Caf de Costa Rica ( ICAFE ) existed for these areas. Each sample consisted of information about the health of one coffee tree and the surrounding soil. To control f or other factors that might affect disease and caffeine concentration I recorded soil moisture, light, and temperature. At each tree the soil moisture content was me asured with a moisture meter probe, sun light reaching the tree was measured with the Rapite st 4 way meter, and air temperature was measured with a thermometer (following the manufactures instructions for each instrument) I also measured time of day and weather conditions when the sample was taken These metrics were not recorded for intensive s tudy of their e ffect on caffeine concentration and disease, but more to get a general sense of differences between farm sites. To test for disease incidence I randomly sampled three branches from the bottom, middle, and top of each tree and recorded how many leaves on each branch were affected by the fungi Hemileia vastatrix (Roya) and Mycena citricolor (ojo de gallo) and general herbivory (Figures 9 11) I recorded all incidences of disease on these nine branches if more than one affliction occurred on a single leaf each was recorded separately Finally soil and leaf samples were collected. I collected a soil core about 10 cm deep at the base of the tree to later be analyzed for pH in the lab. I also picked five healthy medium sized leaves from the bottom, middle, and top of the tree. If the re were enough healthy, similar sized leaves I did not pick leaves from the same branch. From the forty trees I collected 600 leaves and checked 5,464 leaves for disease. In total, f or each tree I collected a sample of 15 healthy leaves and one soil core and recorded the soil moisture, temperature, light levels, and disease incid ence. TESTING SOIL PH: First I measured and recorded the pH of the distilled water. I then put 20 g of the soil sample into a beaker and filled the beaker to 40 ml with distilled water. I agitated the mixture for one minute and then filtered the mud with a fine mesh strainer twice. Once the sediments had settled I put the Extech stick in the solution and waited for the pH reading to stabilize for five seconds before recording the value. Since I did not have access to an accurate soil pH meter I used this method with a water pH meter. Thus the pH reading stated here are relative to each other and the water.
Soil pH, Caffeine, and Disease in Coffea arabica Lochner 4 DETERMINATION OF CAFFEINE STANDARD CURVE: The Albino Rodriquez from IC To find the equation relating concentration of caffeine in a solution to absorbance I measured the absorbance of serial dilution s of a pure caffeine solution I then applied the y=mx+b equation for this serial dilution regression line to solve for concentration in the coffee leaf solutions. To make the serial dilution I mixed 10 mg of pure dry caffeine into one lit er of distilled water. I put 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 ml of th is solution into different graduated cylinders and then filled each one to 10 ml with distilled water (no additional water was added to the 10 ml cylinder ) and shook vigorously to combine. The quartz cuvette was cleaned with tap water twice and distilled w ater once before being filled with the caffeine solution. Before each absorbance recording I blanked the spectrophotometer with distilled water at 274 nm against a cuvette of distilled water. I then measured the absorbance of the d ilution twice at 274 nm. I plotted the average absorbance against the known concentration to find the regression lin e for caffeine absorbance (Figure 1 ) Using the equation of the line A=0.0828C 0.0016 (where A is absorbance and C is concentration) I solved for concentration (Equ ation 1). Given the R 2 value was greater than 0.995, I believe that this standard curve provides a reliable equation to determine the concentration of caffeine from the coffee leaf samples. Figure 1. The caffeine standard curve generated by a serial dilution of a pure caffeine solution. The equation A=0.0828C 0.0016 is the y=mx+b equation for the regression line between the ten dilutions. This equation is used to solve for concentration (C) in Equation 1. EQUATION 1: Absorbance to concentration caffe ine = Concentration mg/L (A = average absorbance at 274 nm) SPECTROPHOTOMETER SAMPLE PREPARATION: I r efrigerated t he leaves collected from the farm until they were placed in a drier no more than 24 hours after collection Leaves were heated until crisp and dry. I then blended e ach of the forty samples of 15 leaves until finely shredded (about 15 seconds) I then placed 5 grams of the shredded leaves into a 100 ml Erlenmeyer flask and added 100 ml of water. Caffeine is very A= 0.0828C 0.0016 R = 0.99603 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0 5 10 15 Absorbance Concentration of Caffeine mg/L
Soil pH, Caffeine, and Disease in Coffea arabica Lochner 5 soluble in boiling water (66 g/100 ml) (Sigma Aldrich). I heated flasks in a water bath for one hour to extract the caffeine. Once the flasks were cooled in a refrigerator I strained the solution with a coffee filter and refilled the aliquot to 100 ml of distilled water Since the spectrophotometer could not detect large concentration of caffeine particles I had to greatly dilute the coffee leaf samples. To prepare the sample for spectrophotometer analysis I used a syringe to add 0.2 ml of the leaf solution to 40 ml of deionized water. I rinsed the syringe with deionized water and flushed the syringe with the solution three times before removing the 0.2 ml. Once I diluted the leaf solution in a graduated cylinder I used a clean glass rod to mix the n ew solution to ensure a uniform concentration. SPECTROPHOTOMETER ANALYSIS: I used the same UV 200 RS spectrophotometer for all samples and the standardized curve. For the caffeine samples I rinsed the quartz cuvette with tap water twice, with deionized water once and then twice with the prepared sample. Since the peak absorb ance for caffeine is around 274 nm, I measured the absorbance of each sample between 280 nm and 270 nm. I tested each sample twice, zeroing the blank at 280 nm with the same sample o f distilled water each time and refilling the cuvette with a fresh sample. Using Equation 1 and Equation 2 I calculated the percent caffeine of the original dry sample from the average absorbance of each sample. From Equation 1 I found the concentration (mg/L) of caffeine in the sample and then entered this into Equation 2 to incorporate the dilution factor, leaf mass, and unit conversions to get a final percentage of caffeine per sample. EQUATION 2: Concentration caffeine to percent caffeine = Concentration x I averaged the two samples to generate the spectrophotometry curve for each of the forty samples. If any peak absorbance were at wavelengths greater tha n +/ 2 nm from 274 nm I excluded these points from the data set because there were presumably contaminated. I used StatPlus to run ANOVA tests to compare farm site to each variable (all three diseases and soil moisture, temperature, and light) RESULTS The main focus of this study is the relationship between soil pH and caffeine content in coffee leaves I found n o statistical ly significant correlation between the pH of the soil in which a tree grows and the caffeine content in its lea ves (Figure 2). Caf feine averaged 2.72% of the
Soil pH, Caffeine, and Disease in Coffea arabica Lochner 6 sample weight and ranged from 3.97% to 1.97%. There was a significant difference ( F 3,36 =3.87, p=0.017) in soil pH between the farm sites with an average soil pH range of 4.107 to 4.705 between farm sites and with a whole farm average of 4.506 Additionally the R 2 values for the comparison of caffeine concentration to the prevalence of Hemileia vastatrix (R = 0.0177) Mycena citricolor (R = 2.8E 06) and herbivory (R = 0.02007) suggested caffeine was not a good indicator of disease prevalence (Figures 3 5) One unusual finding is that Hemileia vastatrix prevalence was significantly different between sites ( F 3,36 =16.80, p<0.01 ). At site 1, 2, and 3 the percent of average affected leaves per tree was 1.09%, 3.04%, and 3.63% respectively. Site 4 was an outlier with 20.22% of its lea ves affected by Hemileia vastatrix. There was no significant correlation between soil pH and disease prevalence (Figures 6 8). Furthermore there was no si gnificant difference between the plots for moisture and temperatur e but there was a significant difference between plots for light levels ( F 3,36 =9.75, p <0.01 ). Figure 2: Percent caffeine in leaf sample compared to soil hydrogen ion concentration ([H + ]) A comparison of [H + ] to percent caffeine found in each five gram sample of dried coffee leaves No correlation was found between [H + ] and percent caffeine concentration, as supported by the low R 2 value of 0.011 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 0.00E+00 5.00E+04 1.00E+05 1.50E+05 2.00E+05 2.50E+05 3.00E+05 Percent Caffeine [H + ] Farm Site 1 Farm Site 2 Farm Site 3 Farm Site 4 Fig. R = 0.0177 0 5 10 15 20 25 30 35 40 45 0 2 4 6 % of leaves affected by H. vasatrix Percent Caffeine Fig. 3 R = 3E 06 0 10 20 30 40 50 60 0 2 4 6 % of leaves affected by M. citricolor Percent Caffeine Fig. 4
Soil pH, Caffeine, and Disease in Coffea arabica Lochner 7 Figure 3 5. Hemileia vastatrix Mycena citricolor and Herbivory vs Percent Caffeine of Leaf Sample T he percent of disease affected leaves per tree compared to the percent caffeine in the five gram sample generated from the 15 leaves from each tree. Figures 6 8. Hemileia vastatrix Mycena citricolor, and H erbivory vs soil Hyrdrogen Ion C oncentration. Site four was a significantly different outlier for H vas t atrix and was not included Note these are logarithmic curves because hydrogen ion concentration is measured on a logarithmic scale. R = 0.0201 0 10 20 30 40 50 0 2 4 6 % of leaves affected by herbivory Percent Caffeine Fig. 5 R = 0.1604 0 2 4 6 8 10 12 14 16 18 0.0E+00 1.0E+05 2.0E+05 3.0E+05 % of leaves affected vastatrix [ H + ] Fig. 6 R = 0.00112 0 10 20 30 40 50 60 0.0E+00 1.0E+05 2.0E+05 3.0E+05 % of leaves affected citricolor [ H + ] Fig. 7 R = 0.00206 0 5 10 15 20 25 30 35 40 45 50 0.0E+00 1.0E+05 2.0E+05 3.0E+05 % of leaves affected herbivory [ H + ] Fig. 8
Soil pH, Caffeine, and Disease in Coffea arabica Lochner 8 DISCUSSION The purpose of this study was to explore the relationship between soil pH, disease, and caffeine content of in Co f fea arabica leaves I predicted that plants growing in a lower soil pH would have relatively less caffeine in their leaves thus mak e the plants more prone to disease. Over the soil pH range sampled, the caffeine concentration remains relatively constant. T here was no significant difference between caffeine content and soil pH, suggesting that differe nces in caffeine content are due to other factors. Interesting ly there was no correlation between disease prevalence and caffeine concentration found in this study Other research findings support the theory that caffeine is a potent protective alkaloid (Kim et al, 2006). However there are a few studies that show that caffeine is not effective at preventing certain insects from eating the coffee plant tissues such as with the leaf miner Perileucopter coffeella (Filho and Mazzafer, 2000). While conducting the herbivory sample, I noticed the presence of leaf miner tracks in some of the leaves Another of the most destructive source s of herbivory not addressed in this s tudy (but present on the farm) is the coffee berry borer, Hypothenemus hampei These beetles have gut micro bes that can break down the caffeine, rendering the insects unharmed from t he normally lethal doses (Ceja Navarro et al, 2015). Insects specially adapted to tolerate caffeine could have skewed by herbivory count. Although there are expectation s such as P. coffeella and H. hampei caffeine is generally considered a highly effective insecticide (Frischknecht, 1986) Thus I find it is curious that there is no correlation between disease and caffeine content was found in this study The lack of correlation might stem from my sampling tech nique: I selected only healthy leaves to test for caffeine There is research showing that plants can redirect their second ary metabolites to injured tissues (Coley and Barone, 1996) so I avoided damaged leaves to control for this phenomenon However this method may have caused me to miss trends in caffeine and disease because coffee tree s may not systemically increase total caffeine content when under attack, but is rather a localized redistribution of caffeine (Frischknecht et al, 1986) This could be a way to allocate secondary metabolites in an energy efficient manne r (Madera, 20014 ). The interaction between detrimen tal insects and fungi systems is a complex phenomenon that could provide valuable insigh ts for better pest management. If we understand how to plant distribute its secondary metabolites while we could avoid removing the plant tissues with the most caffeine or cultivate strains that produce the most caffeine in the most is an important direction for future research. Hemileia vastatrix is a massive problem for coffee farmers, destroying up to 90% of crops in some regions of the world (McCook, 2006). One unusual finding in this study was that farm site four had a relatively large amount of H vastatrix, s ince there is no significant correlation between soil pH and in this location the high H. vastatrix prevalence must be due to other factors. The spores of this fungus are spread through wind and rain, but also through human movement (McCook, 2006) This plot was also located next to the roaster, parking lot, and road ; i t is possible that human traffic helps spread H vastatrix faster there than in other parts of the farm To decrease infection rates it is essential we critically examine how our behaviors promote the spread of disease not only in regards to H. vastatrix but to all afflictions of the coffee tree
Soil pH, Caffeine, and Disease in Coffea arabica Lochner 9 I conducted t he brief survey of light, soil moisture, and temperature conditions to develop a general sense of what other factors were actively affecting the health of the coffee trees. There was no significant difference between the plots for moisture and temperature but there was a significant difference between plots in light levels ( F 3,36 = 9.85, p <0.01 ). Differences in light levels could be related to varying amount of shad e from windbreaks, time of day and weather conditions. However there was no apparent pattern between disease prevalence or caffeine concentration and li ght level s It would be necessary to collect light level data on many more days and times to get a reliable estimate of lights effects on caffeine and disease For these three variables I assumed they were similar enough amongst the forty trees to have negligible affects on leaf caffeine content and disease prevalence. Various sources of error could have skewed my results. I t is possible that I could have biasedly selected the trees, branches, and leaves I randomly sampled Additionally for light, soil moisture, soil pH, and temperature I only sampled once at each tree in the morning ; to increase the quality of this data I could sample many times and take an average over time. With more time and more accurate instruments I could collect more reliable and detailed data. There are many avenues for future research in the realm of caffeine, soil, and disease. By studying other factors (i.e. age, pruning times, water contaminants etc.) that might affect caffeine cont ent in coffee plants coffee growers can get a better sense of how to provide an id eal environment in which coffee plants can produce the most secondary metabolites R esearch on soil components such as soil nutrient levels and texture could provide valuable insights into the best growth su bstrates to obtain higher coffee yields. If people wish to minimiz e profit losses due to disease it is important to understand how the many environmental factors affect the growth of the coffee plant. With coffee being a billion dollar industry it is esse ntial that scientists and farmers learn how to control the spread of fungi such as Hemileia vastatrix and Mycena citricolor and reduce the impacts of herbivory to minimize job and profit losses. Although this study found no correlation between soil pH, leaf caffeine content, and disease, further research is needed to understand the factors that influence the health of Coffea arabica plants. By promoting the production of the coffee plant s protect ive secondary metabolites and other defense mechanism s the use of traditional insecticides, herbicides and fungicides may be reduced : increasing production and minimizing loses through environmentally friendly methods. ACKNOLEDGEMENTS Many thanks to Sofa Arce Flores for the immense amount of time and guidance you put into this project. Also thank you to Frank Joyce and all the UCEAP staff that helped make this project p ossible.
Soil pH, Caffeine, and Disease in Coffea arabica Lochner 10 WORK CITED Ashihara, H., Sano, H., Crozier, A. 2008. Caffeine and related purine alkaloids: Biosynthesis, Catabolism, function and genetic engineering. Phytochemistry, Vol. 69. 841 856. Ceja Navarro, J. A., Vega, F. E., Karaoz, U. Hao, Z., Jenkins, S., Lim, H. C., Kosina, P., Indante, F., Northern, T. R., Brodie, E. L. 2015. Gut microbiota mediate caffeine detoxification in primary insect pest of coffee. Nature Communications, Vol. 6. Coley, P. D. Barone, J. A. 1996. Herbivory and plant defenses in tropical forests. Department of Biology, University of Utah. Vol 27. 305 335. Cooper, J. E. 2002. The influence of soil pH on denitrification: progress towards the understanding of this interaction over the last 50 years. European Journal of Soil Science, Vol. 53. 3:345 354. Deparment of Agriculture, Foresty, and Fisheries. 2012. Production Guideline Coffee. 1 28. Dicum, G., Luttinger, N. 1999. The coffee book: anatomy of an industry from the crop to the last drop. The New York Press, New York. Filho, O. G., Mazzafer P. 2000. Caffeine does not protect coffee against the leaf miner Perileucopter coffeella Journal of Chemical Ecology, Vol. 26. 6:1447 1464. Food and Agriculture Organization of the United Nations (FAO). 2007.The state of food and agriculture. Freeman, B. C., Gwyn, G. A. 2008. An overview of plant defenses against pathogens and herbivores. Plant Pathology and Microbiology, 1 12. Frischknecht, P. M. Ulmer Dufek, J. Baumann, T. W. 1986. Purine alkaloid formation in buds and developing leaflets o f Coffea arabica : expression of an optimal defense strategy? Phytochemistry. Vol. 25. 3:613 616. Ibrahim, S. A., Slameh, M. M., Phetsomphou, S., Yang, H., Seo, C. W. 2006. Application of caffeine, 1,3,7 trimethylxanthine, to control Escherichia coli O157: H7. Food Chemistry, Vol. 99. 4:645 650. Kemmitt, S.J., Wright. D., Goulding, K. W.T., Jones, D. L. 2006. pH regulation of carbon and nitrogen dynamics in two agricultural soils. Soil Biology and Biochemistry, Vol. 38. 5:898 911. Kim, Y.K., Uefuji, H., Og ita, S., Sano, H. 2006. Tran sgenic tobacco plants producing caffeine: a potential new strategy for insect pest control. Transgenic Research, Vol. 15. 6: 667 672.
Soil pH, Caffeine, and Disease in Coffea arabica Lochner 11 Kumar, N. S., Hewavitharanage, P., Adikaram, N. K. B. 1995. Attack on tea by Xyleborus fornicates : Inhibition of the symbiote, Monacrosporium ambrosium, by caffeine. Phytochemistry, Vol. 40. 4:1113 1116 Madera, M. 2014. Comparison of caffeine concentrations in pairs of Coffea arabica leaves exhibiting insect damage and Hemileia vastatrix UCEAP Spring 2014. McCook, S. 2006 Global rust belt: Hemileia vastatrix and the the ecological integration of world coffee production since 1850. Journal of Global History, Vol. 1. 2:177 195. Nathanson, J.A. 1984. Caffeine and related methylxanthines: po ssible naturally occurring pesticides. Science, Vol. 226. 184 187 Simga Aldrich. Product information for caffeine. Sigma Aldrich publications. Zuchowski, W. 2007. Tropical Plants of Costa Rica. A Zona Tropical Publication. 194 195. APPENDIX Table 1. The raw data for percent caffeine, percent of leaves affected by disease and herbivory, and soil hydrogen ion concentration for each tree sampled. Note two samples from site one and site two were omitted because their spectrophotometer curve was presumably contaminated. Figure 9. A leaf with signs of herbivory. Figure 10. A leaf a ffected by Mycena citricolor Figure 11. A leaf affected by Hemileia vastatrix