Statistical Correlations Between GLC Assay and Smaller European Elm Bark Beetle1 Bioassay2

American elm trees, Ulmus americana L., were sprayed with methoxychlor by helicopter or mist blower, and twig crotches were collected from sprayed trees for bio­ assay of Scolytus multistriatus (Marsham) and GLC as­ say. Correlations established between the 2 assays were Dutch elm disease, caused by the fungus Cerato­ cystis ulmi (Buisman) C. Moreau, is a serious shade tree problem in the United States. The smaller European elm bark beetle, Scolytus multistriatus (Marsham), the primary vector of C. ulmi, inocu­ lates healthy elms with fungal spores by feeding on twig crotches. Generally, arborists and city foresters control Dutch elm disease with sanitation programs and with spray applications of methoxychlor. In this study, 2 laboratory assays were used to quantify deposits on elm trees sprayed with methoxy­ chlor: (1) smaller European elm bark beetle bio­ assays, and (2) gas-liquid chromatography (GLC) assay. The objective was to establish correlations between beetle feeding failure5 and amount of methoxychlor residue.6 A correlation between the 2 assay methods is needed for predicting the degree of beetle feeding failure by estimating methoxychlor residues on living elms. METHODS AJ1;D MATERIALs.-Description of Treat­ ments.-In spring 1969, 762 American elm trees, Ulmus americana L., were sprayed with an average of 21⁄2 gal/tree of 12f2% methoxychlor emulsion, using a John Bean Model 300G Rotomist mist blower equipped with 3 no. 5 nozzles.7 The 2 outer and 1 center nozzles had no. 45 and no. 46 cores, respec­ tively. Also, 854 elm trees were sprayed with an average of 1 gal/tree 121⁄2% methoxychlor emulsion, using a Bell 47G-2 helicopter equipped with a 32-ft boom and 34 no. D6 nozzles with no. 45 cores. For 1 Coleoptera: Scolytidae. 2 Received for publication Mar. 9, 1972. 3 Research Entomologist and Research Plant Physiologist, re­ spectively, Northeastern Forest Experiment Station, Forest Service, USDA, Delaware, Ohio 43015. � Biological Statistician, Northeastern Forest Experiment Station, Forest Service, USDA, Upper Darby, Pa. 19082. 5 Mean percentage failure of the smaller European elm bark beetle to feed to the xylem of methoxychlor-treated elm-twig crotches. 6 Mean μg of methoxychlor/mm2 of elm-twig crotch bark sur­ face. 7 Mention of a proprietary product does not constitute endorse­ ment by the USDA. dependent on method of application and amount of methoxychlor residue. Correlations were low and variable for helicopter-sprayed trees and high and uniform for mist-blower-sprayed trees. both groups of trees, the mean DBH was 1.3 ft and the mean height was 45 ft. Sampling Plan.-Fifteen twigs were collected from each of 16 sectors on each sample tree (Fig. 1). Ten of these twigs were used in the bioassay and 5 were used in the chemical assay of methoxychlor. Twigs were collected in May, June, July, and September 1969. Six trees in each treatment were selected randomly each time. Assay.-The bioassay and GLC assay were per­ formed as described by Barger et al. (1971) and Cuthbert et al. (1973) .8 The criterion for the bio­ assay is the beetle-feeding-failure rate; for the GLC assay it is the methoxychlor residue. Analysis.-The feeding-failure rate (q) and the average methoxychlor residue (m) were calculated for each of the 16 tree sectors for every sample tree and then averaged over the 24 trees in each treat­ ment. Then the correlation between the feeding failure rate and methoxychlor residue was computed within each treatment and tree sector. The correla­ tions were calculated from the 24 values of (m, g) in each sector. RESULTS AND DISCUSSION.-Mist blower.-Methoxy­ chlor residues on trees sprayed by mist blower were relatively heavy and variable (Table 1) . On some sectors, residue values were ca. 50 times more than on other sectors. Both assays showed that some tree sectors caught no spray whereas others received much. By tree sectors, feeding failure ranged from 0.0 to 100%,9 and the methoxychlor residue ranged from 0.0 to 3.18 μg/mm2 of bark surface. Generally, resi­ dues were heavier and beetle feeding failures were greater at the bottom of the tree crowns than at the top. Helicopter.-On helicopter-sprayed trees (Table 2), feeding failure and methoxychlor residue were ca. 1⁄2 and 1⁄4, respectively, those of trees sprayed by mist 8 R. A. Cuthbert, A. C. Lincoln, J. H. Barger, and P. A. Reed. Formulation and application of methoxychlor for spraying elms. Manuscript in preparation. 9 0.0% = all beetles fed; 100% = no beetles fed. 80 ] OURNAL OF ECONOMIC ENTOMOLOGY T'ol. 66, no. I Fie. I .-Schematic of an elm tree, showing the 4 vertical levels and 4 quadrants from which twig-crotch samples were taken for bioassay and GLC assay. blower. Trees sprayed by helicopter had a small variation of methoxychlor residues. Methoxychlor residues and beetle feeding failures were heaviest at tree tops and progressively lower at each lower crown level. Correlation Between Assay and Bioassay.-The data suggest a correlation between the assay and bioassay. Table 3 gives correlations by tree sector for each Table 1.-Mean feeding-failure rates of bark beetles in bioassays and mean methoxychlor deposits by GLC assay. Trees sprayed with 121⁄2% methoxychlor emulsion by mist blower. Milwaukee, Wis. Spring 1969.