Embrapa Gado de Leite has the main Elephant Grass Germplasm Active Bank (BAGCE) in the country. However, the main species of grasshopper (Hemiptera: Cercopidae) and the genetic variability for resistance to these pest insects were not known. Therefore, the objective of this work was to study the population dynamics of spittlebugs and to evaluate the accessions of the elephant grass germplasm bank, regarding resistance to Mahanarva liturata. The population dynamics of cercopids in Coronel Pacheco, MG, was investigated in eight accessions of the BAGCE. Adults and nymphs were quantified and correlated with plant phenology and climatic factors. In a second stage, resistance mechanisms in elephant grass germplasm were studied, selected based on studies of genetic and morphological diversity. For antibiosis, survival and viability of the mature phase of the insect pest were verified in these accessions, as well as the feeding time by the EPG technique (Electrical Penetration Graphic). For evaluation of no preference, the eight accessions were placed in screened enclosures and 200 adult spittlebugs were released, with the chance of choice in four replicates. After 12, 24, 48 and 72 hours, the number of adults/elephant grass accession was quantified. For tolerance, spittlebug adults were released and kept for ten days on the plants. In these plants, growth, stem diameter and number of tillers were evaluated, and for symptoms of damage, scores from 1 to 5 were assigned using a visual scale. The predominance of Mahanarva spectabilis, Mahanarva fimbriolata and Deois schach was observed in elephant grass during the sampling period. The elephant grass genotypes Roxo de Botucatu and Pioneiro were selected because they were candidates for susceptible and resistant controls due to the antibiosis mechanism, respectively. The genotypes Cameroon and Cameroon de Piracicaba were the most promising as resistant due to the non-preference mechanism. By the EPG technique, records were obtained that characterize the moment when the insects were feeding; however, it was not possible to characterize the feeding sites (xylem or phloem) due to a lack of standard procedure in the literature. In the tolerance study, the number of insects used (0, 4 or 8) per plant did not cause significant injuries to the tested elephant grass genotypes. Low population density of spittlebug nymphs and adults was recorded in the field experiment, both in the accession plots selected as resistant and in those selected as susceptible; thus, the resistance was not confirmed in the field. Through this research we have a general picture of the spittlebug species occurring in elephant grass; a definition of elephant grass genotypes to be indicated as susceptible and resistant controls in trials; in addition to the selection of promising elephant grass accessions to be used by the farmer, as they were less favorable to the attack of spittlebugs, which is the main biotic problem of forage crops. It is important to note that all this information is unprecedented in the scientific literature.