Microhistology analyses are well-supported techniques useful for studying herbivore diet composition, but they involve the correct identification and validated relative quantification of the plant species consumed, as well as the knowledge of possible biases inherent to methods. We analyzed two methods for quantifying the relative frequency of plant cuticles using an experimental procedure from laboratory- prepared samples. The factors affecting the precision of quantification were analyzed, the minimum sampling effort to detect all plants in the experimental mixture was estimated and the practical application on a complex real sample discussed, in order to describe a simple, standardized, and reliable procedure for studies of herbivore diet composition in Mediterranean ecosystems. Ten random mixtures were prepared with combinations of 40 known plant species, in different number and dry weight percentages. A total of 100 microscopic fields, distributed across 5 slides per sample, were analyzed using both quantification methods. Statistical analysis concluded that the two methods differed significantly. Higher precision was achieved with the method based on annotating the presence/absence of each species in each microscopic field, instead of the total number of times each plant was detected in each field. The important predictors of accuracy were the complexity of the sample and the proportion of plant material in the mixtures. The number of slides and microscopic fields did not significantly impact the quantification results, and no additional species were identified beyond 25 fields in the most complex samples. Due to its greater reliability, shorter analysis time, and reduced visual effort, the first method proved to be more efficient. This method is validated and useful in diet studies of overabundant populations of deer and extensive and transhumant livestock to evaluate the impacts of herbivores on ecosystems and generate sustainable management criteria in the use of pastures toward environmental and economic sustainability. (c) 2024 The Society for Range Management. Published by Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
