August when water levels were usually low. At Car- 
yle a few spring and fall collections were made and at 
he Sullivan station collections were made throughout 
he year. 
The seine most commonly used was 12 feet long and 
| feet deep, and had 3/16-inch bar mesh. Greatest ef- 
iciency was achieved tn riffle areas of moderate current. 
ools up to 3¥2 feet deep were sampled by dragging the 
eine over the bottom. 
Most collections were made during the day. For food 
tudies, fish were collected both day and night on August 
7, 1966, from the Sullivan station. Another day-night 
ollection was made there on October 10, 1966. 
In collections used in indicating habitat preference, 
he river's current, depth, and bottom materials, and the 
umber of each species, were recorded for each seine 
aul, This permitted a rough quantitative comparison 
f habitats of the four darters. 
Quantitative collections were made during August 
nd early September in both pool and riffle habitats by 
olsoning, using emulsifiable rotenone. Block seines were 
sed at the lower and upper ends of the station. Potas- 
um permanganate was used below the station to oxi- 
ize the rotenone. All small fish were preserved. Upon 
ompletion of a collection the average depth, average 
locity, bottom composition, and total area were re- 
rded. Bottom types were classified as silt, sand, fine 
ravel, course gravel, rubble, or bedrock. Current was 
easured with a Gurley pigmy current meter, set 1 
ch from the bottom unless otherwise stated. Tur- 
dity was measured using a USGS turbidity needle. 
Many specimens were handled in the field when 
aking habitat comparisons and studying coloration. 
me 167 maculata, 130 phoxocephala, 39 caprodes, 
1d 52 shumardi were processed in the laboratory for 
tal length, weight, and sex. Observations were made 
1 gonad size, egg size, and egg counts. The presence 
parasites was recorded. 
Stomachs were analyzed for 274 specimens of Per- 
na. Of these fish 71 maculata, 38 phoxocephala, and 
) caprodes were from Sullivan; 16 maculata, and 20 
oxocephala from Vandalia; 6 maculata, 32 phoxo- 
phala, and 2 caprodes from Carlyle; and 13 phoxo- 
phala, 5 caprodes, and 42 shumardi from the New 
hens-Evansville area. 
All organisms were washed from the stomach, 
ouped into families, and counted. When whole or- 
nisms were not present, heads were counted, since the 
ad was usually the last portion of most organisms to 
digested. When food items had been recently en- 
Ifed, the total wet weight of the stomach contents 
's taken and the percent of total weight was obtained 
‘each group of organisms present. Intestine contents 
re also examined. The location of food in the digestive 
ct was often useful in determining the periods of feed- 
 Stomachs of some predatory fishes from darter 
bitats were examined. 
Several scales were taken from the side of the body 
t above the lateral line at a point below the junction 
5 
of the two dorsal fins. Scales were placed in a drop of 
water on a slide and read on a scale projector with 
magnification of 45X (described by Van Oosten 1923). 
The annuli (represented by a crowding of circuli) were 
found only on the anterior field. Measurements were 
made of the scale radius, and of the distance from the 
focus to each annulus, and were used, along with total 
lengths, in back-calculating the length of the fish at the 
end of each year of life. 
Back-calculations of age at each annulus were an- 
alyzed using a SUPAC program on the IBM 7094 
computer. A number of different regression lines were 
tested to determine which would give the best fit for 
scale radius-body length relationships. Because of greater 
variation in scale radius-body length relationships of 
larger fish, the logarithmic regression gave the best fit 
and was used in back-caculating lengths at each annulus. 
DISTRIBUTION AND HABITAT 
P. caprodes has the most extensive range of the four 
darters, occurring from southeastern New Mexico east 
to northwestern Florida and north to the Hudson Bay 
drainage. In Illinois it is found occasionally throughout 
the state (Smith 1965). In the Kaskaskia River it was 
rare in the headwaters, occasional at Arcola, reached 
its greatest abundance in the Sullivan to Shelbyville 
area, was fairly common through the Vandalia area, 
and then was occasional downstream to the New Athens 
area (Table 2 and Fig. 4). 
In the area of the river from Sullivan to Shelbyville, 
P. caprodes was found in runs with moderate to fast 
current and medium to large gravel and rubble. The 
species has been reported in similar habitats in other 
parts of its range by Greeley (1927, 1928, 1929), 
Metcalf (1959), Blair (1959), and Trautman (1957 \e 
In the 1966 collection made with rotenone at the Eagle 
Creek station, nine individuals were taken in the pool, 
and six in the riffles. In the middle and lower portions 
of the river, most P. caprodes collected in riffles were 
young-of-the-year fish. 
P. maculata also has a wide distribution occurring 
from the lower Mississippi north to upper Minnesota 
and east to the Lake Ontario region. In IlIlinois it is 
widely distributed in the eastern two-thirds of the state 
but is sporadic in western counties (Smith 1965). In 
the Kaskaskia, it was occasional in the headwaters, 
relatively abundant from Arcola to Eagle Creek, com- 
mon through the Vandalia area, and then sporadic 
throughout the rest of the river (Table 2 and Fig. 4). 
This species is migratory, and was more abundant in 
the headwaters in winter and spring than in summer. 
Larimore, Childers, & Heckrotte (1959) found it to be 
one of the first to repopulate Smith’s Branch after a 
drought. 
In the Kaskaskia, P. maculata was usually found 
over a wider range of bottom types than the other 
darters studied. In riffles it was usually found in the 
deeper places where the current decreased. In the 1966 
