Coffee is one of the most popular beverages worldwide and a highly traded commodity. In order to maintain a high yield of the perennial crop, weed competition for resources needs to be reduced. For this purpose herbicides are commonly applied, with glyphosate being one of the most prominent herbicides in coffee fields. Glyphosate [N-(phosphonomethyl)glycine], the active ingredient in Roundup®, is characterized as systemic, non-selective, with a broad weed spectrum, inhibiting the enzyme 5-enolpyruvyl shikimate 3-phosphate synthase (EPSPS). The inhibition leads to an accumulation of shikimic acid. In coffee plantations glyphosate symptoms are commonly observed, probably caused by an unintentional spray drift. Root uptake represents an additional uptake route. Glyphosate can be applied several times per year thus a potential accumulation within the plant would lead to higher tissue concentrations than from single applications. The aim of this PhD thesis was to investigate the interaction of glyphosate with the coffee plant. An optimized coffee production, by minimizing the risk for a potential yield loss, is important due to the high commercial value of the crop. However, only limited data is available on the interaction of glyphosate with coffee trees and perennial crops in general; especially with respect to extended time periods after the exposure and multiple applications within the commercial life span of the individual plant. In order to determine glyphosate and aminomethylphosphonic acid (AMPA), the main glyphosate metabolite, in leaf material a method using liquid chromatography coupled to mass spectrometry (LC-MS) was developed. Variations of the method were tested using a full or simplified sample clean-up procedure with single quadrupole MS detection. Additionally the simplified sample treatment was further developed by the use of ultra performance liquid chromatography (UPLC) coupled to a triple quadrupole MS/MS system. All tested variations showed sufficient accuracy (recovery: 80-120%) and precision (˂ 20 relative standard deviation in percentage, RSD%). The full method and the MS/MS system provided a limit of quantification (LOQ) below 0.1 mg/kg; the commonly used maximum residue limit (MRL) for glyphosate in plant derived food products. Glyphosate was found in all samples analyzed from different coffee fields, regardless of management practices. AMPA was not detected above the LOQ in any greenhouse or field sample analyzed. Glyphosate levels seemed to increase in leaf material with multiple applications, indicating the risk for an accumulation in the coffee tree. In an experimental field trial glyphosate was, however, not detected after several applications. The biological response of young coffee plants to single glyphosate applications was evaluated using doseresponse curves (DRC) and shikimic acid levels. Furthermore the effect of multiple exposures on the viii severity of plant damage was investigated. The DRC revealed the presence of glyphosate symptoms to be the most sensitive response variable evaluated. Consequently, when symptoms are absent the risk for plant damage and yield loss is low. The field rate used in the greenhouse study, causing symptoms in 50% of the plants was realistic for field exposure scenarios. When symptoms are present, presumably caused by glyphosate, shikimic acid accumulation can be used for confirmation purposes. The best sampling time was 1 to 2 weeks after treatment, using only the youngest leaves of the plant. Shikimic acid levels were correlated with glyphosate levels 14 to 28 days after treatment (DAT). Earlier and later sampling did not show this correlation within the greenhouse study. A negative effect of glyphosate on the yield, correlated to plant leaf coverage was observed. The yield reduction was only observed for the first application carried out and not distinctively related to the number of applications or field rate. Most likely because of the plant age, possible other reasons are discussed. Biological response variables were not significantly increased by multiple applications compared to a single exposure. In conclusion, glyphosate can damage young coffee plants and under certain conditions reduce the yield. The herbicide can be analyzed with the developed method and is found in commercial coffee fields. Shikimic acid can be used as biomarker for a previous glyphosate exposure. Glyphosate potentially accumulates in coffee plants, yet the risk seems relatively low in a field situation. However, the use of glyphosate in young plantations should be avoided.