52 AMERICAN HYDROIDS. 



It will readily be seen that this table flatly contradicts the preceding one as well as the pre- 

 vailing opinion concerning quantitative bathymetrical distribution, in that it indicates a steady 

 and rapid increase of species from the top to the bottom zone. In the top zone 53 per cent of the 

 stations yielded 50 per cent of the species, while in the bottom zone 1 per cent of the stations 

 yielded 2 per cent of the species. Were the distribution of species equal throughout, the second 

 and fourth columns would obviously agree. It will be seen, however, that in all zones below the 

 first the percentages in the fourth column are materially greater than in the second, and that this 

 disparity increases, in a general way, as we descend to the last zone, where the relative proportion 

 of species is more than twice as great as in the first zone. Several notable variations occur, 

 especially in the eighth and ninth zones, where there is a great increase in species, which may be 

 due to a few especially fortunate dredge hauls. This variation does not necessarily invalidate the 

 indication that there is an increase of life as we go downward. 



There is still, however, a serious source of error in our computation, and this lies in the fact 

 that there is not a direct ratio between the number of stations in a given zone and the number of 

 species. An illustration will make this clear. Suppose an entomologist were to go out collecting 

 in a new locality for ten successive days. The first day he secures 10 species and the second day 

 he also secures 10 species, but some of them would probably be identical with those secured the 

 nrst day. Each day, although he may be equally successful in the number of species secured, 

 he will find a less number of novelties. In ten days he will not collect 100 species, and the 

 longer he works the slower will be the increase in his list until the insect fauna is completely 

 explored, when there will be no further additions at all. In this way it can be seen that the fewer 

 the number of dredging stations in the given zone the greater will be the proportion of species 

 secured. 



This consideration is fatal to our confidence in accepting the actual percentages as shown in 

 the table, but I do not regard it as sufficient to invalidate the general induction derived from the 

 table regarding an increase in the number of species along with au increase of the depth. My 

 reasons for this position are: 



First. None of these zones can be regarded as having been explored with any thoroughness, 

 excepting, perhaps, the first. In other words, the percentage of error is not likely to be very great, 

 because there is in no case any approach to a complete list of the species contained in a given zone. 



Second. The territory covered by these zones is often very great, reaching from near the South 

 American coast to Newfoundland, and through many degrees of longitude. This diminishes 

 greatly the likelihood of repeatedly securing the same species. 



Third. As a matter of fact, there are few species that are recorded from many different stations. 

 While examining the Albatroxx material at the Smithsonian Institution I found only four species 

 recorded from more than ten stations, and from other sources I have secured a similar record for 

 only three more. At the most, not more than 10 species of the 121 can be regarded as having been 

 secured with sufficient frequency to figure materially in the result of our computation. 



Fourth. The difficulty in dredging at considerable depths and the chances of specimens being 

 lost on their way to ihe surface would offset to an appreciable extent the repeated finding of species 

 in the shallower zones. 



It seems practical, therefore, to claim for this investigation that it indicates very clearly, if it 

 does not prove, that plumularian life increases in species down to a depth of 500 fathoms. Below 

 that depth the data are insufficient to warrant any deductions. It can not, however, be too 

 strongly impressed upon the reader that this result is merely an indication to be confirmed or 

 denied in the light of prolonged and careful investigations, which will doubtless be uudei taken in 

 the future. 



It is interesting to note in this connection that Professor W. K. Brooks gives good reason 

 for supposing that the original bottom life of the ocean establishes itself neither near the shore 

 line nor in the abyssal regions, but between the two. 1 



'The Journal of Geology, II, No. !">, p. 170. Tin- nrticle on The Origin of the Oldest Fossils, and the Discovery 

 of the Bottom of the Sea, is of very great interest in its relation to the problem of the bathymetrical distribution 

 of life. 



