ON EOCK REMAINS IN THE BED OF THE ENGLISH CHANNEL. 103 



with the small mean dimeusion of 4'8 cm., and a distribution of 1 

 stone to 247 square yards. The dredge had liere passed through 

 a large and flourishing settlement of Pallasia mnrata, Allen, and a 

 vast number of the tubes, together with several living specimens of 

 this valuable Polychiete, were brought up in it with little else. As 

 it was evident from the next haul, (18), that we had passed outside the 

 limit of the Pallasia settlement, we steamed back to the ground of 

 Position (17) in the hope of securing some more specimens, setting 

 this run at half a mile. But the Pallasia ground was missed, and 

 a very heavy haul of 34 stones was brought up at (19) with a mean 

 greatest dimension of 10 cm., which was one of the highest averages 

 obtained during the work. These two positions, (17) and (19), cannot 

 have been separated by very many yards from one another, though 

 the results obtained were totally different. 



Again, the two blank hauls with the conical dredge previously 

 referred to, (71) and (73), at 19 and 29 miles respectively, were 

 followed in each case immediately afterwards by a haul with the 3' 6" 

 dredge at (72) and (74). The first of these latter gave 213 

 stones with a mean distribution of 0"8 per square yard. In the 

 second, the safety-stop of the dredge was broken, owing to the rough- 

 ness of the ground passed, and no more definite conclusion was there- 

 fore obtainable than that very heavy stones had been encountered 

 within about a quarter-mile of (73), where the conical dredge revealed 

 only coarse shell-sand. 



(3) Size and Weight. — A single greatest measurement of each of the 

 stones was taken. Ultimately all stones of less than 4 cm. were left 

 out of consideration, and a mean was obtained for each haul, derived 

 from the products of dimension x number in each case, at intervals of 

 1 cm. Similarly, the samples were weighed, and a mean was worked 

 out in pounds per stone, for each haul. 



A cursory glance at Table I (p. 114) does not convey any very definite 

 impression as to the relation between size or weight and distance of 

 position from the Eddystone. A certain rate of increase occurs with 

 distance, but it is very irregular. This is evidently due to the fact 

 that the main bearing of the work does not lead directly out towards 

 mid-Channel, but nearly four points to the westward. If a line be 

 drawn due E. and W., Mag., through the Eddystone, a distinct 

 increase is obtained, both in size and weight, in a direction perpen- 

 dicular to this line, i.e. due S., Mag. Parallel lines being accordingly 

 drawn E. and W., Mag., at 5-mile intervals outwards, and a mean being 

 taken for all the samples falling within these several intervals, a fairly 

 regular curve is obtained for both size and weight (pp. 105, 106). 



NEW SERIES. — VOL. VIII. NO. 2. H 



