•r.'S ■ ? . • " 
• 5 • 
In order to be able to compare the relative abundance of fishes in 
CN ) 
s imilar habitats there was established tj. an index which we are calling 
the INDEX OF RELATIVE ABUKDAHCE. This index represents the number of fish 
per square yard recovered in any given area* It is calculated by dividing 
the number of fishes picked up at eaeh station by the approximate number of 
square yards effectively poisoned and searohed. Since the poison kills 
every fish in the area where it was effectively used the chief errors 
involved in the application of poisoning methods relate to 1038 of dead 
fish by waves, currents, and inability of workers to find and recover all 
the dead fishes which sink to the bottom eventually* Those fish that swim 
away or are oarried out into 
% 
the value of the index* 
water too deep for revovery influence 
Since not all of the fishes killed at each station can be 
recovered the values of the index are lower than the actual fauna, but 
since the percentage of non-reoovered fishes probably remains more or 
less the same for all stations, the index is of -some value in comparing 
the relative abundance of fishes between ^stations. 
Sufficient rotetone is used to kill the fishes in the area 
effectively fished* Since the eels are the last affected among all the 
s pecies observed, it is fairly certain that ample rotenone was used if 
the ei-ls are dead after s/d to an hour after the cloud of rotenone*laden 
, mJL 
waster has past an area* At all stations eels were killed* The amount 
of powdered cube root containing the rotenone varied from 5 to 35 pounds 
per station depending on conditions* All stations where the rotenone was 
not effectively used have been excluded from the determination of the 
index of relative abundanoe as recorded in the accompanying table* 
It may be observed from the accompanying table that the average 
index mt for the northern marshall islands for 10 stations is ftstli 0*119 
