1897.] NEW YORK ACADEMY OF SCIENCES. 383 



have been much wider, and would in fact liave been more nearly 

 like that represented for the Peridinium of column I. If now 

 they were to die suddenly out at the surface and the debris were 

 to gradually settle into deeper layers, an irregularity would there 

 be caused, such as is here evident at the three-quarters depth. 

 Such irregular periods of growth are doubtless less common in 

 the deeper layers of water, since the conditions must be more 

 stable in these depths, and this striking increase of Goscinodis- 

 cus at the 84-fathom depth, therefore, probably represents a pre- 

 vious period of active growth in the upper strata, and if one 

 were to restore to life and the former conditions of growth all 

 the dead and crushed debris of the three-quarters level here 

 brought out it would doubtless form such a column as is given in 

 the one just described for Peridinium. And conversely, if con- 

 ditions of active reproduction were suddenly to become unfavor- 

 able at the surface for the two species of Peridininm here plotted 

 (column I), and they were provided with a resistant test like that 

 of the large diatom, their vertical distribution would tend to as- 

 sume, by cessation of growth and settling of debris, that which 

 has now been described for Goscinodiscus. 



The dividing line between the two type species — i. e., between 

 the quantity of Goscinodiscus occulus iridis a,nd that of the large 

 and beautiful G. asteromphalus — in this second column is shown 

 for each level by a small white vertical bar placed in the black 

 areas. All to the left of these bars is G. occulus iridis (with a 

 few individuals of similar size from other species or genera), 

 while all to the right of these white bars is G. asteromphalus. 



In the third column of Plate II, portraying the distribution of 

 the diatom Melosira, is shown a very uniform increase in the 

 numbers of that organism from surface to bottom with a slight 

 falling off at the one-quarter depth. The individuals of this 

 species here enumerated were laid in short chains and each short 

 chain was counted as a unit in the determination, but the\' are so 

 small and transparent that it did not seem advisable to attempt 

 the separation of those living at the time of the collection from 

 the dead, and the whole, therefore, is plotted as living material. 

 It is certain, however, that the diatom can live at this depth, as 

 many healthy chains can be found at the bottom stratum ; and we 

 here conclude also that, under other conditions, the adundance of 

 this form would be much greater in the overlying strata of water, 

 and that the more rapid change of conditions above has left a 

 larger quota alive in the slower-changing depths. That is to say, 

 the increase is not so much due to conditions of activ^e growth 

 below as to the disappearance of a large proportion of those in 

 the superficial laj-ers, together with a settling of the debris into 



