chlor epoxide, a-chlordane, and y-chlordane were found 
in soils in this study. Although small unidentified chro- 
matographic peaks occurred infrequently, they were not 
|-hydroxychlordene but may have been nonachlor or 
chlordene. 
Earthworms appeared to metabolize heptachlor rapidly 
to its epoxide. The decline and disappearance of hepta- 
chlor residues in earthworms paralleled the decline of 
heptachlor in soils, the total process taking longer at 
higher rates of application. Earthworms contained no 
quantities of a-chlordane but seemed to ,accumulate 
y-chlordane and store it without conversion. However, 
there are indications that on the DC-200 column (the main 
column used in this study) a metabolite of a- or y-chlor- 
dane, oxychlordane, may appear as heptachlor epoxide 
(Polen et al. 1971). Oxychlordane was found in the fat of 
animals but not in soil or plants (Polen et al. 1971). 
Heptachlor levels in earthworms declined in our study, 
whereas those of heptachlor epoxide increased. The total 
of heptachlor and heptachlor epoxide declined very little 
during the 2 years, confirming similar results of a study 
covering an 18-month period (Smith and Glasgow 1965). 
Heptachlor + heptachlor epoxide in earthworms from the 
0.56 kg ai/ha plots was present in quantities similar to 
those reported by others for fields treated with similar 
amounts (Smith and Glasgow 1965; Stickel et al. 1965). 
Reports of y-chlordane in earthworms were made by 
investigators using gas chromatographic methods (Gish 
1970), but not from those using colorimetric methods 
(Smith and Glasgow 1965; Stickel et al. 1965). 
Influence of Environmental Factors 
Determinations of moisture, organic content, and pH 
- were made in all soil samples, as were moisture and lipids 
in earthworms (Table 6). These factors were compared 
with the total residues in respective soil and earthworm 
samples to study possible relationships. The associations, 
by chemicals, were discussed earlier. 
Organic content was consistently correlated with soil 
residue levels for dieldrin, total DDT compounds, and 
total heptachlors and chlordanes (r = 0.28, n = 72), and 
to the amount of chemical applied. Levels of organic 
matter were greater at the 2.24-kg (4.1%) and 8.97-kg 
(4.1%) plots than at the 0.56 kg ai/ha plot (3.7%); per- 
haps this indicates that microorganisms that attack 
organic matter were inhibited by higher levels of applica- 
tions. 
The influence of organic content of soil on residue levels 
has been extensively reviewed (Bailey and White 1964; 
Wolcott 1970) and generally this factor has been more 
useful than others for predicting pesticide behavior in soils 
(Wolcott 1970). We observed with Wolcott (1970) that the 
addition of other factors to the multiple regression equa- 
tion did very little to improve the predictive value. 
Percent lipids were significantly lower in earthworms 
from plots which were mulched and rototilled (0.78%) 
than from those only mulched (0.90%). Moisture in earth- 
worms from the 0.56 kg ai/ha plots (72.3%) averaged sig- 
13 
nificantly less than that from the 2.24-kg plots (73.0%); 
and that from the 2.24-kg plots averaged significantly less 
than that from the 8.97-kg plots (73.7%). Moisture in 
earthworms from the DDT plots (72.5%) averaged 
significantly less than from the dieldrin (73.2%) or hepta- 
chlor plots (73.3%). Dieldrin, chlordane, and heptachlor 
have been demonstrated to be more poisonous to 
earthworms than DDT (Davey 1963; Van Der Drift 
1963); heavier doses are more devastating than lighter 
ones (Davey 1963). Earthworms were noticeably more 
difficult to obtain in the heptachlor and dieldrin plots, 
especially those plots treated at 8.97 kg ai/ha, and it is 
possible that these insecticides may have interfered with 
water regulatory mechanisms. 
All of the above factors fluctuated with time (P < 0.01) 
and were suspected to influence residue levels in earth- 
worms. The influences of the individual factors on diel- 
drin, total DDT compounds, and total heptachlors and 
chlordanes were discussed under the individual chemicals. 
As would be expected, the amount of chemicals in the soils 
was of primary importance in determining quantities of 
residue in earthworms. The influence of the remaining 
factors varied slightly with the chemical. 
A composite, predictive equation accounting for 80.0 % 
of the variability in earthworm residues was based on the 
totals of all chemicals 
Y = -1.61711 + 0.66087 X, - 0.02909 X, + 0.03817 X, 
+ 0.07432 X, + 0.35864 X, - 0.01540 X, 
where ¥ is the predicted logarithm of ppm total residues in 
earthworms, X, is the logarithm of ppm in soils, X, is 
months after treatment, X, is percent moisture in earth- 
worm, X, is inches of rainfall during the 2 weeks before 
sampling, X, is percent lipids in earthworms, and Xg is 
percent moisture in soil. The descriptive equation, like 
those for the totals of the individual test chemicals, may be 
useful for making general predictions of insecticide resi- 
dues in earthworms at a specific time following applica- 
tion. The accuracy of such predictions has not been deter- 
mined. The value of this equation may be limited by the 
methods used to estimate the other variables and may vary 
with chemicals other than the ones used in our study. If 
the other variables are to be determined by measurement, 
then it would be less complicated simply to measure the 
residues in earthworms. 
Seasonal Fluctuations of Residues in Earthworms 
Residues in earthworms were higher, usually signifi- 
cantly so, at certain times of the year than at others. Those 
maximums generally were evident in late spring and 
occurred for all individual chemicals and the totals. The 
seasonal fluctuations in earthworm residues appear to 
coincide with seasonal behavioral phenomena of earth- 
worms. Populations of Allolobophora caliginosa and A. 
chlorotica are almost wholly inactive during hot dry 
periods, partially active during January, and generally are 
closer to the surface in spring and fall than in middle to 
