*2.2. Molecular Identification*

For the molecular and phylogenetic characterization, fungal mycelia were disrupted using a minibead beater homogenizer (BioSpec, Bartlesville, OK, USA) with glass beads (0.5 mm diameter) as previously described [25]. Genomic DNA was extracted with Tri Reagent ® (Molecular Reagent Centre, Cincinnati, OH, USA) and used as template in PCR amplifications to detect sequence-characterized amplified regions (SCAR), ribosomal internal transcribed spacer (ITS), and elongation factor 1- α (EF1- α). Primers used are shown in Table 1. A SCAR marker specific for *B. bassiana* strain GHA was amplified with the thermal profile described by Castrillo et al. [26]. The ITS fragment (~600 bp) and EF1- α amplicon (~1200 bp) were amplified employing the touchdown PCR procedure described by Don et al. [27] and modified by Rehner and Buckley [28]. PCR products were visualized in 1% agarose gels stained with ethidium bromide. A 100-bp ladder standard (Productos Bio-Logicos, Quilmes, Argentina) was also used. Amplicons were ligated and cloned into a pGEM-T Easy vector (Promega, Madison, WI, USA) and transformed into *E. coli* JM109. Ampicillin-resistant colonies were isolated and their plasmid purified (Qiagen, Hilden, Germany). Inserts were sequenced (Macrogen, Seoul, Korea) to confirm their identity and use in phylogenetic analysis. Both ITS and EF1- α sequences corresponding to Bb-C001 were aligned along with sequences from several ARS Collection of Entomopathogenic Fungal Cultures (ARSEF) isolates (including GHA) belonging to the genus *Beauveria* [28]. The sequences from *Cordyceps* cf. *scarabaeicola* (ITS, GenBank AY532058; EF1- α, GenBank AY531967) were selected as out-group. Alignments were generated using ClustalW [29]. Phylogenetic analysis was carried out, and maximum parsimony tree was constructed with the program Mega 6.0 [30]. Gaps were excluded from the analysis. A bootstrap analysis with 1000 replicates was used to infer branch support.



## *2.3. Phenotypic Characterization*

Total conidia were collected from PDA cultures at di fferent time periods (see below) and suspended in distilled water containing 0.01% Tween 80. Conidia concentration of each initial suspension was estimated with a Neubauer haemocytometer using serial dilutions up to a factor of 10−7. For viability assays, three aliquots (5 μL each) were taken from this mother solution, plated on PDA and incubated at 27 ◦C. Twenty-four hours later, they were observed under a microscope using a 100× objective. Germination percentage was calculated as the number of germinated conidia/total number of conidia × 100. Both procedures (conidia concentration and viability) were repeated in the di fferent postsowing times (10, 20, and 30 days) to evaluate whether both parameters were time-a ffected. For radial growth, each strain was inoculated in the center of the 80 mm diameter Petri dish with PDA medium. Photographs were taken every 48 h, and the radial growth was measured using the software Image Tool 3.0, considering four perpendicular diameters. Observations were recorded for 30 days after inoculation. The slope was calculated by lineal regression and expressed as radial growth (mm/h) as described in [8]. For each strain, 12 replicates were done.
