Development Of Oxytetracycline Resistance In Rainbow Trout (Oncorhynchus  mykiss) Gut Flora And Sediment In A Recirculating Culture System     W. Bane Schill1 and Graham L. Bullock2   1USGS, National Fish Health Research Laboratory, 11700 Leetown Road, Kearneysville, WV; 2Conservation fund, Freshwater Institute, P.O. Box 1889, Shepherdstown, WV     Last year we reported that rainbow trout cultured in a recycle system and fed oxytetracycline (OTC) at 2.5 gm/100 lbs of fish for 14 d developed resistant gut flora, but similar resistance did not develop among heterotrophic bacteria found in fluidized sand biofilters.  Studies were conducted this year with feces and sediment (feces and uneaten food). Heterotrophic counts were performed on R2A and Mueller Hinton media with and without 100 ug/ml OTC on individual trout feces (50-55 trout each sample period), on 12 fecal pools (5 trout/pool), and on sediment.  Feces and sediment were sampled before OTC feeding; on days 3, 7, 10 and 14 OTC feeding; and 3, 7, 14, 21, 28, 35, and 42 d post OTC feeding.  During the first cycle of OTC feeding, resistance (average % of total heterotrophic counts on both media) in feces of single and pooled trout increased from 11% before OTC feeding to 73% by day 14.  Upon completion of OTC feeding resistance dropped to 35.9% by day three, increased slightly on day 14 post OTC feeding and dropped to 1.26% by day 21.  The second cycle of OTC feeding was begun immediately after completion of the first cycle.  The same general pattern of resistance was seen using fecal samples of individual fish as had been observed with pooled samples, and sediment resistance generally reflected that found in feces.  Surprisingly, sediment resistance was no higher than that found in feces despite the fact that OTC levels in sediment were determined to be as high as 1900 ppm.  A definite bimodal pattern of resistance was seen with the first peak on day 14 of the second OTC feeding cycle and the second about 14 days after OTC was withdrawn. A second population of rainbow trout was obtained to determine if these patterns of OTC resistance in feces were universal or population-specific.  Two cycles of OTC feeding were given and the same sampling days of pooled feces and sediment were used.  Results from two  trials showed that resistance patterns during the first feeding cycle were similar to those observed during the two trials of the first fish population. We did not, however, observe the bimodal response during the second OTC feedings that we had detected with the first fish population. Additionally, sediment resistance levels did not reflect those obtained with feces in these trials.