192 
BULLETIN OF THE UNITED STATES FISII COMMISSION. 
bright blue areas of the carapace of the adult lobster are clearly seen. The fine trans- 
verse hair line, just above the cervical groove, behind the second antenna, can also be 
detected. The cross-shaped figure seen on the upper surface is formed by the wide 
median absorption area and the cervical groove with which it is continuous on either 
side. Besides the fringing hairs, the whole surface is covered with short set* which 
correspond in some measure at least to the hair pores of the adult carapace. In 
certain cases some of these changes, as in the relative lengths of the rostrum and 
anterior lateral spines, are far less marked. The anterior median carina is not promi- 
nent after the third stage. 
The carapace of the fifth stage is illustrated in fig. 114. The characteristic tendon 
marks, which have been described, are very conspicuous. The rostrum is narrower 
and in the sixth stage is about one-fourth the length of the entire carapace. 
In a young lobster 35 mm. long, which had molted about twelve times, the shape 
and areolation of the carapace were very similar to that of the adult. 
The abdominal segments . — In the first larva (plate 20) the abdomen has its full 
number of segments. The first is unarmed and partially covered by the carapace. 
The second, third, fourth, and fifth somites bear early a prominent median spine, which 
projects upward and backward from the posterior end of the tergum. Each of these 
segments also bears upon either side a long, sharply pointed pleural spine, which 
projects backward. 
The median spines, of which the last two are the longest, are gradually reduced 
during subsequent stages and finally disappear at the fourth molt. Meantime the 
pleural spines become shorter, shift their position slightly, and in the fourth larva 
point downward. 
The disappearance of the median dorsal spines is, however, not uniform, but 
subject to considerable variation, as shown by the following observations upon eleven 
larvae in the second and third stages. 
Table 36. — Variation in time of disappearance of the median tergal spines of the larval abdomen. 
No. of 
larva. 
Stage of development. 
No. of abdominal somite. 
.... I 
2 
3 
4 
5 
1 
Second larva 
Rud. 
1 
1 
1 
2 
do 
Rud. 
1 
1 
1 
3 
do 
0 
1 
1 
1 
4 
do 
0 j 
1 
1 
1 
5 
Third larva 
0 
0 
1 
1 
G 
do 
1 
1 
1 
1 
7 
do 
Rud. 
1 
1 
1 
8 
do 
0 
1 
1 
1 
9 
. - . .do 
0 
1 
1 
1 
10 
do 
0 
0 
1 
1 
11 
do 
Rud. | 
1 
1 
1 
rud. -= spine rudimentary. 1 = spine present. 0 = spine absent. 
We see by the preceding table that the median spine of the second abdominal 
segment may entirely disappear at the second molt or persist in either a rudimentary 
or perfect condition even after the third ecdysis, while the spine of the third somite, 
though usually present, is sometimes absent in the third stage. (Nos. 5, 10.) 
The sixth abdominal somite bears at its posterior margin a pair of dorsal spines 
on either side of the middle line. These curve backward over the telson, and are 
much smaller than the median spines (fig. 33, plate 20). They disappear at the fourth 
molt. 
