882 TRANSACTIONS OF THE [dEC. 13, 



From the data obtained by us, and represented b}- the plotting 

 on Plate II., it will now be seen that the dinoflagellate PeriiH- 

 nium represented in the first column, is very abundant at the 

 surface, also that, of the two species represented, the one, P.fusus, 

 is much more numerous than the other, P. divergens, the latter 

 being onl}^ about one tenth as numerous as the former, as is here 

 represented by the stippled portion on the right side of the 

 column. At the one-quarter depth (28 fathoms) the decrease in 

 this organism is very marked, while the falling off continues 

 thence gradually to the bottom. It is plain, however, that P. 

 fusus can maintain itself alive at the bottom and that the larger 

 species, P. divergens, shows a more uniform distribution through- 

 out the upper half of the vertical, but runs rapidly out in the 

 lower half, being absent at the last two levels. 



The distribution of the diatoms included under the heading 

 Goscinodiscus is much more varied, as is shown in the second 

 column of the plate ; taking all the elements together, there is 

 shown a uniform increase from surface to middepth, then a rapid 

 increase to the three-quarters depth, with a final falling off in the 

 last quarter distance of the vertical. It will be seen, however, 

 that there is a large element of dead and broken debris here in- 

 cluded (represented by the oblique-lined sides of the column), 

 which we assume to have settled from surface strata, and which 

 would thus naturally increase toward the bottom. In the living 

 and uninjured part of the material one sees the same plan, except 

 that the proportion of living to dead organisms is much 

 greater at the surface than at any other level. The living dia- 

 toms of this group are over 82% of the whole number at surface, 

 as against 23% at the three-quarters depth, and as against 29^ 

 at the bottom. 



It may be that some currents at superficial levels are at work 

 in shifting off the material, so causing it to be settled or eddied 

 into belts and strata below, and it may be that there is also a 

 higher rate of disintegration in the bottom strata to account for 

 the lessening quantity there, but this seems hardly possible 

 within these short limits, and we believe that this difference in 

 quantity at different levels is rather due to those circumstances 

 which cause a sudden loss in the rate of reproduction in the 

 upper strata, and from which there is a subsequent settling of 

 the debris into the lower. As will appear upon a following page, 

 the diatoms here considered are very variable in quantity at the 

 surface, being at some times very abundant and at others (as on 

 the day when this vertical Avas taken) hardl}^ perceptible. From 

 this it is certain that the column in black as here plotted would 

 on another day, during a period of rapid growth at the surface, 



