No. 1.] 7HEZ DEVELOPMENT OF BALANOGLOSSUS. Ly 
Stream. The dorsal side of the larva is turned towards the 
observer. Larvae of this age, kept in a dish, now went into 
the sand, also those newly caught buried themselves as soon as 
supplied with sand. Unfortunately I have not sufficient data 
to show definitely under what conditions the larva went 
into the sand. A fewwere seen to go into the sand when 
only slightly smaller than the preceding stage (Fig. 5); some 
of them would again come out of the sand and not go in per- 
manently until a later stage had been reached, others seemed 
to remain buried. Whether the latter continued to decrease 
in size until they were as small as the larvae of Fig. 6 is 
uncertain, but from a few observations I am inclined to think 
that they did not continue to shrink. 
The process by which the larvae went into the sand was 
easily seen. The tip of the proboscis was thrust obliquely or 
even vertically downwards between the coarse grains of sand. 
Its end then expanded and the body drew up nearer to the 
point of anchorage. The proboscis was then thrust downwards 
again and the process repeated. As soon as about half the 
proboscis was buried the whole body was lifted up, from its 
previous horizontal position on the sand, into a vertical position 
and then was rapidly drawn into the sand. 
It was very noticeable that so soon as the proboscis was 
thrust into the sand a thick mucous was thrown out from the 
surface of the proboscis and collar region to which the sand 
granules stuck, forming an irregular tube around the animal. 
Often the larva disappeared in the sand in less than half a 
minute; at other times the body was not drawn out of sight 
for several minutes. 
An examination of the larva shows that the most striking 
change, if we except the diminution in size, is the increased 
size of the proboscis, as compared with the rest of the body. 
The collar is more distinctly marked than in the preceding 
stage, and it will be noticed that the posterior boundary groove 
is not continuous across the mid-dorsal line but turns forward 
on each side of the mid-line, leaving a plate of ectoderm belong- 
ing to the extra-oral area between the two sides of the collar. 
We will return again to the description of this region. The 
