and Littoral Manganese Nodules 171 



extra cable which had been paid out, I sounded when the cable 

 was up and down, and found exactly 104 fathoms, so that the 

 anchor had held and remained in the deep place while the ship 

 >wung into water that was 44 fathoms shallower. When the 

 anchor was brought up, a large mass of clay and dead shells, 

 principally Pccten, was sticking to one of the flukes; and this 

 was the specimen which contained the manganese nodules, 

 which at that time were not supposed to exist anywhere out 

 of the deepest oceans. As the anchor held firmly in the first 

 instance in 60 fathoms, and was not examined before being 

 lowered into the iO4~fathom spot, it might be held to have 

 picked up the sample in 60 fathoms, and kept it all the time it 

 was holding the ship in 104 fathoms; but this is not likely. 

 On 24th September 1878 the anchor was again dropped in 

 apparently exactly the same spot, and it brought a quantity of 

 clay and shells exactly like those of the 2ist, but containing 

 no nodules. On ist October a few nodules were got in the deep 

 trough ; and dredging in 50 fathoms on the north side and in 

 40 fathoms on the south side none were found. On the south 

 side the ground was very rough and rocky in 40 fathoms. 



Although everything went to show that the abundance of 

 nodules really occurred in the deep trough, it was disappointing 

 that I could never get complete confirmation. On I5th July 

 1881, however, I was again successful, for the anchor-dredge 

 brought up a large bagful of mud, containing abundance of 

 nodules. On this occasion the position of the ship was care- 

 iullv watched, showing that she dragged over about half a 

 mile of ground, beginning in 100 fathoms and gradually shoaling 

 to 85 fathoms, when the anchor was brought up. There can, 

 fore, be no longer any doubt that it is really in the deep 

 trough that the nodules occur, and <m\\ in a very limited area 

 of it. 



The mud of 2ist September 1878 was divided in two equal 



portions, one of which was preserved and the other was analysed 



mechanically. It was separated into tlm< ( ..n-titn. -UN. the 



nodules, the shrlN. and tin- residue, a s.mdy day. The number 



nl.t.mi.d was 83, and they weighed 142-7 



whence the mean weight of a nodule was 1-7 gri nun i, 



