12 Natural History Bulletin. 



the r^piditv with which they were worn away by scouring 

 over the sandy bottom. 



In dredging, there is a tendency on the part of beginners to 

 use too Httle rope. It is best in the long run to be generous in 

 paving out the line, twice the amount needed to go straight 

 to the bottom being none too much as a general thing. The 

 ovster men have a saying to the effect that •• the man with a 

 long line has the biggest pile by night." 



Perhaps three-fourths of our specimens from deep water 

 were brought up with the tangles. We found the bottom rocky 

 almost everywhere at depths of from sixty to two hundred 

 and sixt\- fathoms. Large patches of smooth bottom would 

 be encountered, but the peculiar jerking of the line, which is 

 soon recognized as the danger signal, indicating rocks and 

 trouble with trawl or dredge, was a frequent occurrence, so 

 that we were aiwa\'s uneasv until the dredge left the bottom. 

 The tangles are by far the most efficient instrument for such 

 moderate depths, and we finally came to rely almost entirely 

 upon them. These tangles were made after a pattern 

 suggested by Mr. James E. Benedict, of the Smithsonian 

 Institution. A four foot length of one bv two inch iron bar is 

 bent in the middle at nearly a right angle. Five iron rings 

 are bolted at regular intervals to the inner side of this bar. 

 The ends of five two foot leni£ths of chain are fastened to 

 these rings, and through each link is passed a six foot strand 

 of two and three-fourths inch Italian hemp rope. Each strand 

 is tied to the link at the middle, and then carefully unravelled 

 throughout its entire ]enc{th on each side of the knot. There 

 are six such strands to each six foot length of the rope. The 

 dredging cable is attached bv a hook to a ring bolted to the 

 outer side of angle bar. •• Mousing " should be placed over 

 the hook to keep it from slipping out of the ring when in use. 

 This tangle differs from that previouslv used, in the fact that 

 the bar is bent and not straight, and in the use of the chains 

 instead of fastening the ropes directlv to the bar. This latter 

 feature we found to be an excellent one. as the chain weighted 

 the swabs sufficiently to insure their dragging closely to the 



