236 WM. S. MARSHALL AND C. T. VORHIES. 



anterior end and more sand added, making a total of 2.5 mm. 

 Fourth, cut again. Fifth, had cemented at anterior part; cut 

 again. Sixth, cemented as before ; cut again. Seventh, eighth, 

 ninth and tenth, cementing and cutting repeated each day. 

 Eleventh, cut edges at anterior end cemented as before and a 

 little more new sand added to this end. 



7. Six cases were cut, two dorsally, two ventrally, and two 

 laterally. The time during which these experiments were carried 

 on was at least two weeks in each instance, giving sufficient time 

 for full action on the part of each larva. 



Of the two cut dorsally one ignored the cut entirely but built 

 en a ring of new sand at the anterior end after eight days had 

 elapsed. The other at once cemented the cut edges for a dis- 

 tance of 2 mm. at the anterior end, then, for a period of nine 

 days, did nothing further. Repetition of the cutting had no 

 other effect than to cause, within a short time, the larva to glue 

 the edges for a short distance at the anterior end. 



The two cut ventrally took much more notice of the injury to 

 the case than the preceding. One of these closed the cut half 

 way down from the anterior end and at the end of two days had 

 completely glued the cut edges together. Repeated cutting 

 caused, with one exception, a partial repair within twenty-four 

 hours. After five such repetitions an additional day was allowed 

 for repair, at the end of which time, the cut edges were com- 

 pletely glued. The second had at the end of twenty-four hours 

 cemented the cut at the anterior end for a distance of 2 mm.; then 

 for eight days it did no further repairing. The cut was then 

 reopened and at the end of twenty-four hours was again partially 

 closed. The next day the larva had turned in its case and was 

 working at the posterior end and on the following day had com- 

 pletely cemented the cut edges. 



The first of the two laterally cut specimens would always close 

 the cut for about one third the distance at the anterior end, and 

 once, when allowed four days, completely repaired the injury. 

 Twice a new ring was added at the anterior end instead of the 

 usual repairing. The second of these larvae did nothing for four 

 days. It then closed the cut for a distance of 2 mm. at the 

 anterior end and for four additional days made no further repair. 



