164 W- S. DEMBOWSKA. 



my own observations, this statement is fairly correct. I might 

 add that Dromia can do the same with any kind of sponge, 

 fastened to any immovable object. But in this case the hy- 

 pothesis of Vosmaer becomes not quite adequate as the sponge, 

 torn off from a large stone, may be perfectly flat and there will 

 be no conformity in shape. As far as the writer is aware the 

 process of cutting and tearing off of the sponge by Dromia has 

 not yet been described. The following observations and experi- 

 ments pour some light on the whole question and they demon- 

 strate at the same time the complication and the variability of 

 the behavior of the crab. 



THE PROCESS OF MANUFACTURING A CASE. 



The process of manufacturing a case by Dromia may be easily 

 observed if we put in the aquarium, which contains a crab, a 

 comparatively large piece of ordinary writing paper. Before 

 putting in the paper we remove the sponge from the crab's 

 back. The paper must be boiled for a few minutes in water, 

 otherwise it would float. 



As a rule very soon Dromia grasps the edge of the paper with 

 its claws and begins the work. The crab enters under the paper, 

 holds the edge tightly and turns with its back to the ground. 

 Now by means of rhythmical movements of both chelae the crab 

 cuts out of the paper a piece which is fairly similar in size and 

 shape to the removed sponge. Nearly always Dromia starts its 

 work from the edge. To state it more exactly, the crab does not 

 cut the paper, but it tears it, and therefore the edges of the new 

 case are always very uneven. Both chelae grasp the paper until 

 their points nearly touch one another, at the same time their 

 planes are perpendicular to the edge of the future case. At the 

 next moment both points are directed forward (upward if the 

 crab lies on the back) and the paper between them tears. At 

 once Dromia grasps the paper a little farther and tears it again 

 in the same way. The animal pursues its work systematically 

 and it keeps working always in a definite direction. As the 

 border of the resulting case is approximatively elliptic, the mutual 

 position of the chelae cannot be always the same, but it varies 

 according to a definite rule. It often happens that the crab 



