Carolina Madtom 
71 
Extreme lower Neuse River, Craven County. — Probably the most 
surprising record of the Carolina madtom from the Neuse is an individual 
that was trawled from the main river near New Bern in 2 to 3 m of 
water over sand in 1981. Extremely heavy rain in December (the highest 
in 15 years) preceded the capture of this fish in January; it had 
apparently been swept downstream with the heavy flooding that followed 
the rain. The collection should probably be considered a stray from 
typical areas of occurrence upstream. 
Tar Drainage 
Tar River proper. — Both historically and recently the Tar River 
proper has harbored the most substantial populations of N. furiosus 
known anywhere. Six collections on the mainstream have yielded 45 
specimens since 1877, and three large populations are known to be 
extant. Searches by us, some U.S. Fish and Wildlife Service personnel, 
and North Carolina State Museum employees in October 1984 and May 
1985 revealed four new sites for the Carolina madtom in the Tar River. 
In addition, several old sites were re-collected (e.g., Tarboro and Rocky 
Mount). All sites had an abundance of madtom habitat, and numerous 
additional specimens were observed but not preserved. Numerous mussel 
shells, cans, bottles, jars, and other litter in the Tar River provide an 
abundant variety of nesting sites and undoubtedly contribute to successful 
reproduction in most summers. It should be noted that the recent 
successful collections in the Tar River were made under ideal conditions 
of extremely low, clear water and during peak recruitment or spawning 
periods. The Tar River remains one of the finest large streams in North 
Carolina and will probably continue to be the stronghold for the 
species. 
Swift and Fishing Creek systems.— Both of these streams offer 
abundant habitat for N. furiosus and are physically similar to the Little 
River near Goldsboro. The largest series ever taken of this species (67 
individuals) was collected in 1959 from Fishing Creek during a combined 
day and night sampling period. Re-collection of these sites between 1982 
and 1985 yielded 14 specimens (all from Swift Creek), but sampling was 
impossible in Fishing Creek because of extremely high water. Re- 
collection of Fishing Creek under optimum conditions is especially 
warranted because this stream may harbor one of the largest extant 
populations of N. furiosus. Several large collections of N. furiosus were 
made by NCWRC personnel in 1963 in Swift and Fishing creeks (Fig. 4) 
after application of rotenone. To our knowledge, no voucher specimen 
is extant for any of these collections, but the large number of specimens 
taken (Fig. 4) is evidence that these two streams support, or did 
support, large populations of N. furiosus. 
