4 Erwin H. Barbour, 



Several were attached to the familiar transverse portion, 

 as in the first type, — a fact not known till the third expedi- 

 tion was sent out and more exhaustive search was made, 

 revealing two examples. One spiral secured on the third or 

 Morrill geological expedition by Mr. Thomas H. Marsland, 

 when dug out, instead of ending abruptly as if cut off, as we 

 might have expected from other experiences, or instead of 

 ending in a transverse portion as some do, terminated below 

 in three massive spherical enlargements. (See Fig. 17.) 





Fig. 2. — Eagle Crag, seen from the north, showing the DaimoneHx or Devil's cork- 

 screws in place. C, see Plate IV., also Fig. 4. 



Can it be, then, that these great "twisters," instead of 

 being fucoids or sponges growing from below upward, are 

 roots, boring their way from above downward, and becom- 

 ing so completely modified as to lose their identity as 

 roots } 



The invariable perpendicularity of the Devil's corkscrew 

 suggests the possibility that sedimentation was going on at 

 a far livelier rate than supposed, otherwise these specimens 

 must have rotted away or toppled over in spite of any pre- 

 servative quality of the water. 



304 



