62 
depressions. Anal shield triangular, irregularly marked with paler, its margins con- 
vex. First and second pairs of prolegs one-third shorter than the third and fourth; 
anal prolegs clasper-like. In locomotion the hump at the base of the abdomen is 
scarcely noticeable. 
Tubercles much as in Stage Y ; i and ii on segment IV and iii and iv on all segments 
shining black, the others inconspicuous. On segment I the second pair are separated, 
the posterior one much smaller, and the pair at the base of the legs are separate and 
of equal si7e. The additional ventral tubercles present in stage Y are absent. On 
segments II and III, tubercle iii is three times as large as i and ii; iv is minute and the 
ventral tubercles are absent. On segments lY to IX, iii is largest, i and ii are equal, 
and iv is farther forward. On segment lY, vi is closer to vii, and v and vi are equal. 
On segment Y, v is smaller than vi, while iii and vii are equal. In the group, i is 
smallest, iii a third larger, iv twice as large, and ii four times as large. On segment 
YI, iii is just above and in front of the spiracle, and v and vi are subequal. On seg- 
ment IX, iv is smaller and less distinct, v is farther down, near the base of the proleg, 
and vi is just above the base of the proleg. On segment X, i, ii, v, and vi are equal, 
iii is smaller and closer to the spiracle, iv is distant from the spiracle, and vii is four 
times as large as viii. On segment XI, i, ii, and iii are equal, as are also iv, v, and vi; 
iv is just behind the spiracle, with vi beneath it, and viii far down on the side of the 
segment. On segment XII, iii is minute and more depressed, directly below i, and iv 
is in the stigmatal stripe at the posterior edge of the segment. On segment XIII, 
vii consists of 8 inconspicuous tubercles on the outer side of the anal proleg, viii 
midway between the base and apex of the proleg on the inner side. 
POSSIBLE CAUSES OF COLOK. TAEIATION. 
In connection with the color variation above referred to, Mr. Girault 
has prepared a table showing- the growth and color differentiation of 
three larvae which proved to represent the three well-marked color 
varieties; all hatched from eggs laid by the same female (var. 
ochracea), and were reared during September under similar conditions, 
on cotton. 
Table XYII. — Variation in three Jarrx reared unJer similar conditions. 
In star. 
I. 
II. 
III. 
IV. 
VI. 
Dark variety. 
Rose varietv. 
Green varietv. 
Normal, marked distinctly 
with deep orange. 
Dark, marked with dull orange. 
Like the type, hut lighter and 
greenish. 
Dark, greenish, marked with 
dull orange. 
The usual brown type: oliva- 
ceous above with olivaceous 
side stripes and yellow stig- 
matals. Twenty-four hours 
later, dorsal region lighter 
and stripes very conspicuous. 
Just before ecdysis. larva 
green. 
Grayish; gray above with whit- 
ish and yellow lines; dark, 
the side stripes dark and the 
stigmatals yellow. 
Normal, lemon yellow, 
marked with dull orange. 
Pale, marked Avith dull 
orange. 
Olivaceous, marked with 
dull orange, becomingpaler 
later. 
Ground color pale ochreous 
marked with rust red; later 
with greenish lines above 
and conspicuous side 
stripes. 
The usual brown type dorsal 
region greenish with yel- 
lowish lines; side stripes 
with crimson spots, stig- 
matals pale, margined. La- 
ter with the crimson spots 
more distinct. 
Normal, yellowish, marked 
with deep orange. 
Marked with dull orange. 
Grayish, marked with dull 
orange. 
Typical at first, but after 24 
hours changed to green 
with dusky side stripes 
and whitisii stigmatals. 
Typical; side stripes brown; 
"later pale green with pale 
yellow lines above and 
darker green stripes. 
Stigmatals pale yellow. 
The Tisual pink type; pink Theu.sualgreen type; body 
above the pale lines; side greenish, the stripes 
stripes broad, velvety black; j darker green and the 
stigmatals reddish yellow. " lines above pale yellow. 
