BULLETIN OF THE UNITED STATES FISH COMMISSION. 
26 
which are almost continually in motion, the sand is loosened and the grains more or 
less separated. By the contraction of the longitudinal muscles the rear of the body 
is brought up nearer to the head, and then the circular muscles contract and extend 
the body again. It is prevented from slipping back by the anchors, which are elevated 
by the contraction of the circular muscles and hold against the sand. Since the 
contraction begins next to the rear end and moves forward, the head end is pushed 
onward, the anchors there lying flat in the skin. 
This process of alternate contractions of the two sets of muscles is very obvious 
to an observer, and takes place very continuously, though not rapidly. In this way a 
synapta can move through the sand from 2 to 3 centimeters a minute, and an inhcerens 
of average size can get entirely out of sight in 5 or 6 minutes. One of the most 
remarkable provisions for the use of the anchors in locomotion is their much greater 
abundance and their considerably greater length in the posterior part of the body. The 
use of this is clear when one realizes how the rear of the body acts as the resisting 
base against which the muscles work in pushing the anterior end forward. This dif- 
ference between the anchors of the anterior and posterior ends of the body seems to 
have been overlooked hitherto, so great an authority as Ludwig (’98) saying of inhce- 
rens “Anker und anker platten differieren in vorderen und hinteren korperabschnitt 
nicht merklich von einander.” In the same paper, however, he calls attention to the 
fact that in 8. digitata the anchors of the posterior end of the body are about 50 per cent 
longer than those in the anterior end. While the difference in inhcerens is not quite so 
great as that, yet it is very noticeable, being above 33 per cent. In both species there 
is an increase also in the length of the plates, but, as would naturally be expected from 
the passive part they play, it is not nearly as great. The following figures will bring 
out the condition in inhcerens very plainly. A specimen 14 cm. long was chosen at 
random and cut into seven approximately equal pieces, and in each piece 10 anchors 
and plates, selected entirely at random, were measured with these results, No. 1 being 
the most anterior, No. 7 the farthest back; measurements are all in microns: 
No. 1. No. 2. 
No. 3. 
No. 4. 
No. 5. 
No. 6. 
No. 7 
Average length 
of anchor. 
Average length 
of plate. 
Average length 
of anchor. 
Average length 
of plate. 
Average length 
of anchor. 
Average length 
of plate. 
Average length 
of anchor. 
Average length 
of plate. 
Average length 
of anchor. 
Average length 
of plate. 
Average length 
of anchor. 
Average length 
of plate. 
1 / 
Average length 
of anchor. 
Average length 
of plate 
Number 
141 
124 153 
134.5 
162.5 
135.6 
164 138 
202 
157 
204 
145 
197 
135 
88 
. 87 
.83 
84 
. 75 
. 71 
.68 
Per cent i ncrease in length . 
' .09 
.08 
.06 
.008 
.009 .017 
.23 
. 14 
.009 
— . 077 
—.03 
—.07 
Maximum length of an- 
- 
chor in each section 
155 171 
178 
197 
220 
220 
224 
The most striking feature of this table is the abrupt and marked increase in the 
size of both anchors aud plates in the fifth section. The average length of anchors 
for the first four sections is 155/<; of plates, 133/1; ratio = 86 per cent. For the last 
three, 201 it is the average length of the anchors, 144 of the plates, and the ratio - 71 
per cent. That is, the anchors increase 29 per cent and the plates 8 per cent. The 
slight decrease in the last section is probably due to the fact that comparatively little 
strain could be brought upon the anchors situated so very near the tip of the body, 
but it is noticeable. that the longest anchor of all was found in that section. The 
