324 



burrow. Within the burrow itself the cuts produced by the mandibles are very difficult 

 to see. It is these successive cuts, the outlines of which appear as the above described 

 serrations, which result in the formation of the burrow. 



Hoek (1893) has suggested that the antennae and antennules, because of their 

 haired scales, might be used in boring. We have not observed any behavior to support 

 this view. These appendages are laid back over the head of the animal and do not 

 appear to take any direct part in the excavation of the burrow. 



A peculiarity of the boring habit is the steadiness and continuity with which the 

 work is carried on. Disturbance of the organism may cause it to suspend boring 

 momentarily, but it quickly resumes its activity and continues steadily at work. At 

 intervals the boring is interrupted and comb-like processes on the first pair of legs 

 are used to clean the mouth parts. 



Frequent observations were made on indi\iduals in the early stages of making the 

 burrow, and whenever observed they were found to be at work. One organism which 

 had started its burrow on Saturday, April 26, was still boring at noon the following 

 Monday, by which time the burrow was two-thirds the length of the body of the 

 animal. When the light was turned on suddenly at night the animal was found to be 

 boring steadily as during the day. 



When disturbed the animal may react in one of several ways. An initial disturb- 

 ance will cause it to crawl into the burrow as far as it can go. Continued mechanical 

 stimulation will cause it to back out. It may then swim or crawl away, or if a more 

 severe stimulus be gi\en it will roll itself up into a ball as do man\- of the isopods, 

 particularly the pill-bugs. It remains rolled up for only a few seconds. Animals which 

 have become located in deeper burrows are slower to leave than those in new or 

 incomplete burrows. 



As the boring progresses, a considerable collection of oblong creamy white fecal 

 pellets collect near the entrance of the burrow. They are discharged by the anus and 

 forced away by the currents produced by the beating pleopods. They have a fibrous 

 structure and are composed of the residue after passage of the wood through the 

 digestive tract. These fecal pellets are ver>- quickly attacked by bacteria, turning 

 black and disintegrating within a few hours after their discharge. In this respect they 

 differ from those discharged by Teredo. 



So far as we have observed, no wood accumulates near the region of the head. All 

 of the particles apparently pass through the digestive tract and are discharged through 

 the anus, as in the case of the moUuscan borers. 



The passage of the wood through the digestive tract is rapid. The casts are quite 

 regular in size and approximate one-tenth the length of the entire digestive tract. 

 The average frequency of discharge is one in eight minutes. It is estimated, accordingly, 

 that a given particle remains in the digestive tract about eighty minutes, and that 

 there are eighteen complete passages in twenty-four hours. These observations are 

 based on animals constructing new burrows, and therefore presumably working at a 

 maximum rapidity. 



The burrow gradually approximates the contour and size of the individual con- 

 structing it. It is at first a simple graded excavation, deeper at the forward end. By 

 the end of twenty-four hours, according to our observations, the burrow is considerably 

 deepened and the head entirely covered when the animal is in the burrow as far as 

 it can go. From the rate of progress here and in other cases, we have noted that the 

 time necessary to construct a burrow of such a size as to contain the animal is from 

 four to six days. During this period the exposed position of the organism greatly 

 increases the possibilities of its being washed away, or crushed. The arrangement of 



