Gubar ( Lymantria dispar ) je jedan od najvažnijih šumskih štetnika na području Republike Hrvatske. Iz sustavnog praćenja gradacija u kontinentalnom području uočeno je da se gubar masovno pojavljuje svakih 10 do 11 godina, dok su u mediteranskom području gradacije češće i nepravilnije u periodicitetu. Kako bi odredili je li potrebno suzbijati gubara, važno je utvrditi intenzitet napada. Intenzitet napada ovisi o gustoći populacije ličinki, a gustoća populacije je povezana s raspoloživom masom lišća u sastojini. S obzirom da za hrast lužnjak postoji metodologija izračuna intenziteta napada, cilj ovoga rada je preispitati i redefinirati kritične brojeve jajnih legala gubara, primjenjujući rezultate recentnih istraživanja. S obzirom da za običnu bukvu do danas nisu poznati kritični brojevi, cilj je bio definirati ih. Za izračun kritičnih brojeva potrebno je izračunati prosječnu masu lišća koju pojede jedna ličinka te prosječnu masu lišća jednog stabla. Laboratorijski pokus prehrane lišćem hrasta lužnjaka i obične bukve postavljen je kako bi se utvrdila prosječna masa pojedenog lišća, a prosječna masa lišća jednog stabla je utvrđena korištenjem dostupnih modela biomase lišća. Rezultati izračuna kritičnih brojeva jajnih legala za hrast lužnjak pokazuju kako je taj broj u odnosu na do sada korišteni viši, što znači da će površina koju treba tretirati biti manja. Rezultati izračuna kritičnih brojeva jajnih legala za običnu bukvu predstavljaju potpuno novi i originalan podatak. Kritične brojeve jajnih legala prikazane u ovome radu potrebno je provjeriti prilikom sljedeće gradacije gubara kroz terensku evaluaciju, kako bi ih potvrdili. Daljnja razrada kritičnih brojeva moguća je korištenjem dostupnih prirasno-prihodnih tablica (bonitet i EGT) te podacima iz nacionalne inventure šume. Korištenje trenutno dostupnih modela biomase ukazuje na potrebu istraživanja mase lišća u krošnji stabala različite starosti za područje Republike Hrvatske, kako bi se usavršila preciznost izračuna kritičnih brojeva jajnih legala gubara. The gypsy moth ( Lymantria dispar ) is one of the most important forest pests in Croatia and a primary biotic factor responsible for oak decline, especially when tree defoliation during mass outbreaks is followed by infections with the oak powdery mildew ( Erysiphe alphitoides ). Population dynamics of gypsy moth differs between the two main regions of Croatia. In the Continental part, outbreaks occur in cycles every 10 to 11 years, the last one was recorded from 2012 – 2014. In the Mediterranean part, outbreaks are more frequent and less synchronized between sites. Key elements for sustainable forest management are forest pest monitoring programs and the assessment of pest population densities in time and space. In the case of gypsy moth, this is especially true since protection measures should be applied properly and only when prognoses of outbreaks predict severe negative impacts on the forest stands. Gypsy moth population density assessment in Croatia is carried out by counting the egg masses along transects in the forest stand. Predictions of infestation levels, through the number of gypsy moth egg masses per tree or hectare, uses five classes. Calculations include the number of trees with at least one egg mass, expressed in percentage of total. In the case of defoliators like gypsy moth, the basis for the prediction of defoliation is the average leaf mass consumed per one larvae. According to Androić (1965) a single larva consumes 12 g leaf material, however, detailed descriptions of the experimental setup and the tree species used for the feeding trials are missing. The aims of this paper were to (1) redefine the critical numbers of gypsy moth egg masses used so far for pedunculate oak ( Quercus robur ) and (2) define critical numbers for common beech ( Fagus sylvatica ). Egg masses for the laboratory experiments were collected in the continental region (forest area near city of Koprivnica) in December 2015. First, larvae were fed in groups from hatching to the end of the third instar with pedunculate oak or common beech leaves. Newly emerged fourth instars were then separated and kept individually. Larvae were weighed daily from the start of the fourth instar until the prepupal stage. Mass of pupae were determined three days after pupation. Fresh leaves used for feeding were weighed daily, the unconsumed leaf remains were collected, dried and also weighed. The leaf material was changed weekly. The fresh/dry mass ratio were calculated for each week to assess the dry mass of freshly given leaves. The average dry leaf mass consumed per larva was used to calculate the critical numbers of gypsy moth egg masses per tree and stand. The average leaf mass per one tree was calculated from biomass models on a dry mass basis. The recalculated critical numbers of gypsy moth egg masses for pedunculate oak are higher in comparison with the ones used until recently. Data relating to critical numbers of gypsy moth egg masses for common beech represent new findings for Croatia. Until recently, all defoliation predictions in common beech stands were based on data used for pedunculate oak. The higher critical numbers of gypsy moth egg masses is confirmed for the management units Ključevi (pedunculate oak) and Petrinjčica (common beech). In this study we also redefined the the average leaf mass consumed per one larvae. New critical numbers of egg masses still require field validation during the next gypsy moth outbreak. Future research about critical numbers of gypsy moth egg masses could implement data from yield tables (site quality and ecological and economic types (EGT)) and data from the national forest inventory. Usage of currently available biomass models points out the need for further studies about average leaf mass per one tree in Croatia (Table 5 and 6), in order to calculate critical numbers of gypsy moth egg masses more precisely.