71 
Injury to ripe peaches still attached to the trees is recorded by 
Webster" in Ohio. BoUworms attack peaches about as they do 
tomatoes (see PI. XI, fig. 3). 
Peas, beans, and okra are all bored into and the green seeds eaten, 
the bollworms often being very destructive to these plants (see PI. XI, 
figs. 4 and 5). 
From eggs laid on rosebuds early in the spring a lot of larvae feed- 
ing upon the unopened buds were reared. 
CHOICE OF FOOD BY LARV^ 
Aside from field observations, a series of experiments were made in 
the laboratory to determine the choice of food by larvae when plenti- 
fully supplied with several kinds. When given corn, cotton squares, 
green tomatoes, and tobacco buds, the corn is almost always chosen, 
often after first tasting the others; but with difi'erent parts of the corn 
plant selection is not always so uniform. Apparently the tender bud 
and ear are about equally attractive. In the case of cotton the flowers 
and squares were liked best, as is shown in the appended table. 
Table XXIII. — Choice of differeiit parts of King cotton bij holhwrms. 
Number of larvae. 
First series. 
Second series. 
Total. 
Part of plant. 
>> 
>. 
>. 
ol 
>> 
>> 
>i 
Remarks. 
a 
>> 
cj 
^• 
!>. 
<a 
k1 
c3 
>. 
'C 
cd 
'O 
•^ 
r^ 
^ 
>. 
-o 
aj 
^3 
r? 
rrt 
^ 
-C 
"C 
w 
03 
-a 
-d 
-a 
-a 
t3 
-d 
A 
Tli 
t3 
A 
a> 
xi 
2 
i 
5 
X 
> 
£ 
O 
o 
2 
1 
^ 
.^3 
c 
> 
.^3 
o 
o 
^ 
& 
H 
;=H 
U^ 
ID 
w 
^ 
OQ 
H 
1^ 
t^ 
m 
CO 
W 
p!H 
m 
Leaves 
f> 
8 
?, 
1 

n 

^ 

n 

1 



8 
9 
Fifth choice. 
Tender growth 
2 
2 
1 
4 
2 


1 
4 
2 

1 



n 
8 
Fourth choice. 
Flowers 
(a] 
(n) 
('-') 
(a) 
(a) 
<? 
<? 
4 
fi 
'?. 
4 
R 
8 
4 
8 
(n) 
99 
First choice. 
Squares 
4 
8 
S 
■8' 
■5- 
9. 

1 
4 
?, 
'.^ 
? 
1 
9^ 
14 
Second choice. 
Bolls 
1 
1 
3 


2 
4 



2 
3 
2 
1 
3 
11 
11 
Third choice. 
« Absent. 
The experiments from which these data were obtained were made 
by placing leaves, squares, etc., together under a large bell jar with 
the larv?e. Then every day an examination was made and the part 
chosen b}^ each larva recorded. No doubt if the experiments had been 
tried in the open sunlight, the exposed tender growth would not have 
been chosen so frequently. 
RET.ATIVE ATTRACTIVENESS OP UPLAND AND EGYPTIAN COTTONS. 
It was noticed quite frequently that Egyptian cotton was more or 
less free from attack by the bollworm, while Upland cotton growing 
adjacent was badl}^ infested. To get accurate data on this point a 
«Bul. No. 8, n. s., Div. Ent., U. S. Dept. Agr., p. 41. 
