Flight of the eye gnat, Hippelates pallipes (Diptera: Chloropidae): correlation with temperature, light, moisture and wind velocity.

The flight activity, based on collections from humans, of Hippelates palliPes (Loew) eye gnats was measured under varying field conditions. Thresholds for flight were 18°C, 5.38 milliphots (5 ft-c) of light, and wind speed below 60-70 mfmin. No gnats were captured at a vapor pressure deficit of 0.0 mm Hg. Graphs are presented for flight activity at various values of temperature, light, VPD, and wind speed. Comparisons are made between the responsesof H. pusio and H. palliPes to temperature, light, VPD, and wind speed. HippelatespalliPes (Loew) is an annoying pest of man and animals in North Carolina. Like H. pusio Loew, it is attracted to body secretions and wounds. H. palliPes has not been directly implicated in the transmission of any disease, but the similarity of its biology to that of other HiPpelates makes it suspect. H. pusio has been implicated as a possible vector in the spread of conjunctivitis, anaplasmosis and bovine mastitis (Dow & Hines 1957, Roberts 1968, Sanders 1940). H. flavipes Loew has been implicated in the transmission of yaws in Jamaica by Kumm & Turner (1936), who incorrectly used the name H. pallipes according to Sabrosky (1951). H. flaviPes has been incriminated in the spread of staphylococcal skin infections in Panama (Taplin et al. 1967). H. flavipes, H. currani Aldrich and H. peruanus Becker have been associated with staphylococcal skin infections in Trinidadian children (Bassett 1970). Generally H. palliPes is less abundant than H. pusio in North Carolina, but on some occasions it is the dominant Hippelates species annoying man (Axtell & Edwards 1970). Other investigators in the southeastern United States either have not reported any data on H. palliPes (Jay 1962) or have reported only occasional captures (Bigham 1941, Dow & Hutson 1958). In Wisconsin, Hummadi & DeFoliart (1967) reported H. pallipes associated with various domestic and wild animals but did not collect from humans. Womeldorf & Mortenson (1963) collected 6 specimens in Cali'This research was supported in part by training grant ESOO069 from the National Institute of Environmental Health Sciences. Paper No. 3899 of the Journal Series of the North Carolina State University of the Agricultural Experiment Station. 'Research Associate and Professor, respectively, Department of Entomology, North Carolina State University, Raleigh, N. C. 27607, U.S.A. r fornia. The low numbers of H. pallipes reported may be a reflection of the collection methods in some cases. Axtell & Edwards (1970) found that the baited traps designed to capture H. pusio were poor collectors of H. pallipes. The factors determining the flight activity of H. pusio have been published (Gerhardt & Axtell 1972) but not those for H. pallipes. The thresholds of flight for H. pusio were determined to be 17 ac, 0.25 Hg vapor pressure deficit (VPD), 5.38 milliphots (5 ft-c) of light and wind speed below 67 m/min. Data on other species of Hippelates are fragmentary. In the laboratory, H. collusor (Townsend) was shown to have varying flight activity when the wave length of light was changed while the light intensity was kept constant (Dorner & Mulla 1961). The flight activity of H. collusor ceased at wind speeds in excess of 53.6 m/min. in a laboratory chamber. H. collusor preferred low relative humidity at low temperatures and high relative humidity at high temperatures under laboratory conditions (Dorner & Mulla 1962). The average number of H. flavipes observed in the field was reduced 82 % when the wind increased from 0-91 m/min. to 92-213 m/min. No H. flavipes were caught at wind speeds over 213 m/min. (Kumm 1935). The objectives of this investigation were to determine how the physical environmental factors of temperature, light, atmospheric moisture, and wind influence the flight activity of H. pallipes when responding to human bait; specifically, (1) to determine what environmental conditions must exist in the field before H. pallipes will actively seek man, and (2) to determine how the changes in flight activity that have been observed in the course of a day are correlated with the naturally occurring changes in the physical environment. In addition, comparisons were made between the responses of H. pusio and H. pallipes to human bait. MATERIALS AND METHODS The methods and equipment were the same as used in the previous study on H. pusio (Gerhardt & Axtell 1972). Field temperature measurements were made with a YSI@ Model 47 Scanning Tele1973 Gerhardt & Axtell: Flight studies of HiPpelates palliPes Thermometer3 and recorded on a YSI Model 80 Laboratory Recorder. The heat sensitive elements were YSI #401 thermistor probes. The probes were placed at 0.1 and 2 m above the ground in an open sunlit area and in an adjacent shaded area. All temperatures were recorded in degrees Celsius. Light was measured and recorded in the field with an automatic recording photometer (Berry & Raney 1968). Two photo-sensitive cells were placed 2 m above the ground, one in the open sunlight and the other in the shaded area. Recording was continuous and in foot candles. Atmospheric moisture was determined by taking wet and dry bulb temperatures with a Model 566-2 Bendix Psychron@4. Wet and dry bulb temperatures were taken immediately after each gnat collection at 0.1 and 2 m above the ground. Wind speed was measured with a No. 3AM680 Slow Speed AnemometerS (range 6 to 300 m/min.) at the same time and place as the wet and dry bulb readings. Two adult male humans were used as bait. The subjects sat on the ground across from each other and collected the gnats that came to them with a portable vacuum aspirator modified from a hand vacuum cleaner. The collection periods were 5 min. and made every 15 min., with the subjects moving alternately between the sun and the shade. Collections usually started at 0530 EST and continued for 12-14 hr. The gnats were collected in vials and taken to the laboratory for counting and identification. The analysis of the data for H. pallipes was made from collections on 8, 17, and 21 September 1970 at Whispering Pines, Moore County, N. C. and 15 July, 4 and 6 August 1971 at Raleigh, Wake County, N. C. The data were analyzed by multiple regression using a standard statistical program (Barr & Goodnight 1971) on the IBM 360/165 computer in Research Triangle Park, N. C. The ,Jx+ 1 transformation was used on the dependent variable (H. pallipes) to stabilize the variance. TABLE 1 contains a list of the parameters used in the analysis and their symbols which will be used in the text. All of the parameters (independent variables) shown in TABLE 1 were used in the regression analysis of the dependent variable. The final regression model was arrived at by reducing the sYellow Springs Instrwnent Co., Yellow Springs, Ohio 45387, U.S.A. 'Bendix Environmental Science Division, Baltimore, Md. 21204, U.S.A. "G. M. Mfg. & Instrument Corp., 2417 Third Avenue, New York, N. Y. 10451, U.S.A. 291 PARAMETER TABLE 1. List of parameters and corresponding symbols used in the multiple regression analysis. Temperature Temperature X Temperature Temperature X Vapor Pressure Deficit Temperature X Wind X Vapor Pressure Deficit Wind Temperature X Wind Wind X Wind Wind X Vapor Pressure Deficit Vapor Pressure Deficit Vapor Pressure Deficit X Vapor Pressure Deficit Reciprocal of the Vapor Pressure Deficit Relative Hwnidity Relative Humidity X Temperature Relative Humidity X Wind Temperature X Wind X Relative Humidity Temperature X Reciprocal of Vapor Pressure Deficit Wind X Reciprocal of Vapor Pressure Deficit Wind X Wind X R VPD Temperature X Wind X Wind Light Square Root of Light SYMBOL" TEMP STEMP TVPD TWV WIND TW SWIND WVPD VPD SVPD RVPD RH TRH RHW TWRH TRVPD WRVPD SWRVPD TSWIND LIGHT ROOTLI number of independent variables. This was accomplished by the step down method; i.e., variables with the most non-significant partial regression coefficients were rejected. The analysis was repeated several times, each time eliminating more non-significant variables, until only significant variables were left in the model. This investigation was conducted entirely under natural field conditions using wild populations of gnats. This was done to eliminate the hazards of extrapolating laboratory behavior data, using inbred laboratory strains, to field conditions and wild populations.