HABITS, ETC., OF OEREBRATULUS LACTEUS. 105 



all observations indicate that the worm roams about con- 

 stantly in the mud after food, and that it is able to move 

 quite rapidly, even when forming a new burrow. The enor- 

 mous amount of slime secreted by the worm^s ectoderm may 

 be inferred from the fact that it keeps the walls of its burrow 

 constantly coated. 



This slime often oxidises the iron elements in the mud, with 

 the result that the latter is tinged a dull rusty red in the im- 

 mediate vicinity of the burrow. 



Eepeated attempts were made to ascertain just how the 

 burrowing was done, by digging out specimens and throwing 

 them on a fresh surface. It was noticed that the worm 

 thrust its pointed head a little way into the mud and then 

 seemed to brace itself for a further effort, the head at the 

 same time being withdrawn slightly. The significance of 

 this fact was in doubt for some time, but on placing a few 

 worms in a glass aquarium with five or six inches of mud at 

 the bottom, they immediately began to burrow near the glass. 

 It was then seen that the bracing action was due to the fact 

 that the worm was driving its proboscis into the mud ahead 

 of itself. The proboscis was protruded for six or eight inches, 

 and then enlarged slightly and crooked a little at the end. 

 Aided by the hold thus obtained the worm was able to thrust 

 its head rapidly into the mud, keeping it narrow and pointed 

 the while. 



The extreme tip of the head was then contracted into a 

 broad rounded form, and the wave of contraction thus started 

 passed slowly backward along the anterior portion of the 

 body, thereby moving the latter forward about an inch. As 

 soon as this wave was fairly started in its backward motion 

 the head became pointed, the proboscis was thrust forward 

 again, and the whole process was repeated. The tip of the 

 head was then contracted, and a second wave was started 

 backward before the first had reached the centre of the body. 

 In this way one wave followed another so quickly that the 

 resultant motion was nearly a steady advance instead of being 

 jerky. This rapid burrowing by means of both proboscis and 



