750 gpm where it empties into Conasauga Lake. 

 The treated section of the Brasstown Bald stream 

 is three-fourths of a mile long and falls approx- 

 imately 900 feet enroute . The small spring from 

 which this stream originates flows at about 25 

 gpm from which point the volume increases to 

 approximately 1,350 gpm where the stream 

 leaves the sprayed area. Both streams fall and 

 cascade over mica schist boulders and are bor- 

 dered with heavy growths of laurel, rhododen- 

 dron, small hemlock, basswood and sassafras. 



which lay immediately above and below the lake . 



Two stations (1 and 2) were established 

 in the Brasstown Bald stream within the treated 

 area. An additional station was located approx- 

 imately one -half mile below the perimeter of the 

 spray area . Three stations (4, 5, and 6) were 

 established in a control stream. Bottom samples 

 from these control stations were comparable to 

 prespray samples from stations 1, 2, and 3 in 

 the experimental stream. 



The streams differ from each other most- 

 ly in that the Conasauga Lake stream empties 

 into Conasauga Lake within the spray area, while 

 the Brasstown Bald stream empties into a larger 

 stream several miles from the experimental 

 area. 



TREATMENT OF THE STUDY STREAMS 



The DDT was appUed to the Conasauga 

 Lake stream on May 15, 1959, and May 24, 1960, 

 and to the Brasstown Bald stream on May 22, 1959, 

 and May 26, 1960. First, the helicopter laid 

 narrow swaths parallel to the creek until a strip 

 about 100 feet wide was treated on each side of 

 the stream . These strips served as a buffer 

 zone to prevent insecticide from reaching the 

 stream during the remainder of the spraying op- 

 eration. The entire treatment of the stream 

 was completed in approximately fifteen minutes . 

 Both streams were treated in similar manner, 

 except that special care was taken to keep the 

 insecticide out of the Conasauga Lake stream; 

 no care was taken to keep it out of the Brasstown 

 Bald stream. 



METHODS OF STUDY 



Prior to the DDT treatment, four bottom 

 samples were taken at each collection station 

 with a Surber square-foot sampler. Three sta- 

 tions (1, 2, and 3) were established on the Cona- 

 sauga Lake stream above its point of entry into 

 the lake, and one (4) in the stream below the dam. 

 No stations were established in the lake. At the 

 same time, two stations (5 and 6) were estab- 

 lished and sampled in a control stream . These 

 control stations were comparable to the experi- 

 mental stream stations 1 and 2 but no compara- 

 ble stations could be found for stations 3 and 4 



Collecting of drift samples was accom- 

 plished by placing a Surber sampler in the stream 

 for periods o£ five minutes . The nature of these 

 streams was such that one could screen approx- 

 imately 90 percent of the water flowing in the 

 stream. Drift sampling was initiated one-half 

 hour after the spraying was begun, and samples 

 were taken at one -half hour intervals for the 

 first hour and then at hourly intervals until dark . 

 After that, they were taken at irregular inter- 

 vals up to 72 hours after treatment. All spec- 

 imens were placed in containers with 10 percent 

 formalin, taken to the laboratory, sorted into 

 taxonomic groups, and tabulated. 



The amount of insecticide which reached 

 the creek surface was determined by the use of 

 4 1/2- by 5 -inch oil -sensitive cards. The vari- 

 ety of cover conditions suggested classifying the 

 cover into three groups: (1) low canopy, in 

 which the laurel, rhododendron, hemlock, etc . 

 formed an almost solid canopy from immediate- 

 ly above the water surface to a height of 10 - 15 

 feet, (2) high canopy, in which the canopy was 

 from 20 - 50 feet from the water surface, and 

 (3) open, in which the sky was visible from the 

 creek . Three spray cards were placed under 

 each cover type. 



Interpretation of the cards was made by 

 the author and Mr. Dale Vandenburg, U.S. For- 

 est Service, the forest entomologist working on 

 the project. 



EFFECTS OF TREATMENT 



Measurements of deposition of DDT on 

 the water surface of the two streams are shown 

 in table 1 . In the Conasauga Lake area, where 

 care was taken to keep away from the stream, 



