PLANKTONIC STUDIES. 631 
OCEANIC POPULATION—STATISTICS. 
Statistics in general is known to be a very dangerous science, be- 
cause itis commonly employed to find from a number of incomplete 
observations the approximate average of a great many. Since the 
results are given in numbers, they arouse the deceptive appearance 
of mathematical accuracy. This is especially true of the complicated 
biological and sociological conditions, whose total phenomenon is con- 
ditioned by the codperation of numerous different factors, and is, 
- therefore, very variable according to time and place. Such a highly 
complicated condition, as I believe I have shown, is the composition of 
the plankton. If, as Hensen actually wishes, this were to be sufficiently 
analyzed by counting the individuals, and oceanic population statistics 
were thereby to be made, then this would only be possible by the forma- 
tion of numerous statistical tables, which should give results in figures 
of the plankton fishery quantitatively in at least a hundred different 
parts of the ocean, and in each of these at least during ten different 
periods of the year. ; 
A single “reconnoitering voyage” on the ocean, a single “trial 
trip,” limited in time and place, like the three-months Atlantic voyage 
of the National expedition, can furnish only a single contribution to 
this subject. Butit can in no way, as Brandt thinks, offer ‘ firm foun- 
dations” for the solution of this and that “thorough analysis” (23, p. 
525). If, also, after six years the 120 catches should actually be counted 
through (after a labor of more than 17,000 hours), if by statistical 
arrangement of this numerical protocol, by rational reckoning of their 
results, a serviceable conception of the quantity of individuals of the 
oceanic region investigated should be obtained, then at best this one 
computation would give us an approximate conception of the conditions 
of population of a very small part of the ocean; but from it by no means : 
can we, as the investigator of Kiel wishes, arrive at couclasions bear- 
ing upon the whole ocean; for that purpose hundreds of similar com- 
putations must be made, including the most diverse regions and based 
upon continuous series of observations during whole years. The Z00log- 
ical stations would be the best observatories to carry out complete series 
of observations of this character, not such trial trips as the three-months 
voyage of the National.* 
*In my opinion the results of the National expedition of Kiel would have been 
quite different if it had been carried out in the three months from January to March, 
instead of from July to October. On the whole, the volume of planktonic catch, at 
least in the North Atlantic Ocean, would have more than doubled; in some places 1 
would have been increased many fold. Its constitution would have been entirely 
different. If the expedition had by accident fallen in with a zoécurrent, and its 
voyage had continued init fora few miles, the contents of the nets would have 
certainly been a hundredfold, possibly a thousandfold, greater. 
