Tennessee River Fish Surveys 
101 
From 1979 to 1981 TV A conducted extensive fish surveys on nine 
streams in the Tennessee River drainage. A total of 118 samples were 
taken at 32 localities using various collecting techniques. This paper 
reports the findings of these samples, including new occurrences, range 
extensions, and relative abundance estimates for some species. Parts of 
this report are based on literature searches and may not reflect all the 
efforts by many other collectors whose data have not been published. 
Previous studies and surveys have been reported on all of our nine 
study streams except Copper Creek and the Paint Rock River. The pub- 
lications and data considered in this report are: Duck River — Whitney 
(1957), Anonymous (1975), Nieland (unpubl. data); Powell River — 
Woolman (1892), Anonymous (1970), Masnik (1975); Clinch River — 
Masnik (1975); Nolichucky River — Ward (1960); North Fork Holston 
River — Cope (1868), Jordan (1889), Ross and Carico (1963), Feeman 
(1980), Feeman (1986); Buffalo River — Anonymous (1973); Elk 
River — Jandebeur (1972). 
DESCRIPTION OF STUDY AREA 
The general study area of this project was the Tennessee River 
drainage (Fig. 1). Five of the study streams are in the Great Valley 
physiographic region of southwest Virginia and east Tennessee: Clinch, 
Powell, North Fork Holston, and Nolichucky rivers; Copper Creek. The 
other four streams are in the Highland Rim physiographic region of 
middle Tennessee and northern Alabama: Duck, Buffalo, Paint Rock, 
and Elk rivers. Table 1 gives specific locality information with river 
mile, county, state, and map coordinates. 
METHODS AND MATERIALS 
The sampling scheme of this study consisted of two stages. Da.ta 
collected in the first stage were used to determine the host species; three 
streams with known populations of Conradilla caelata (Duck, Clinch, 
and Powell rivers) were sampled once in 1979 and three times in 1980 
(Table 1). Data collected in the second stage were used to determine 
which streams had suitable populations of the host species and could be 
considered as transplant sites; nine streams were sampled three times in 
1981 (Table 1). 
Three collecting techniques were used: (1) seine-snorkeling, (2) 
backpack electrofishing (day and night), and (3) seine hauling. Both 
quantitative and qualitative data were obtained. Transects were run 
over a known area, using seine-snorkel and backpack electrofishing 
techniques to collect quantitative data. Six transects were taken in each 
habitat type present (e.g., sand and gravel, rubble, vegetation) to get a 
representative sample of the site. Relative abundance estimates were 
calculated from these data. 
Sites were also sampled at night for nocturnal species and others 
difficult to collect in the day. Electrofishing transects were repeated in 
