716 
BULLETIN OF THE BUREAU OF FISHERIES 
A spiller is so placed that the fish, which enter the trap with the tide and then turn 
and swim against it, are led into the spiller through a narrow web tunnel which can 
easily be closed when the current is running in the opposite direction. Two spillers 
thus have a big advantage over one in that each one can be filled in turn, unless the 
trap is in an eddy where the current does not reverse itself with the tide. The pot 
aids in the fishing as the fish would not readily pass from as large a chamber as the 
heart directly through the narrow tunnel leading to the spiller, but the salmon are 
removed only from the spiller. 
The construction of the earlier traps was modified to some extent when certain 
regulations were put into effect. In 1897, the length of a trap lead was restricted to 
2,500 feet, and it was further provided by law that there should be an end passageway 
of at least 600 feet, and a minimum lateral passageway of 2,400 feet, between all traps. 
These regulations had the effect of preventing a complete blockade of a whole 
area. For instance, in 1895 a string of three traps, each one connected with its neigh- 
bor, extended in a southeasterly direction off Cannery Point, the southeast tip of Point 
Roberts, for a mile. Two other connected traps near the international boundary 
extended for four-fifths of a mile. Such long strings of traps were not uncommon, 
and the law advanced conservation by breaking them up. 
Another law, passed in 1897, prohibited traps from operating in water over 65 feet 
in depth. However, this law was not observed for several years. In 1913, soundings 
by the State Fish Commissioner (Washington State, 24th and 25th reports, 1916, 
p. 36) revealed 11 traps operating in water exceeding the legal depth by l%-27 feet, 
The owners admitted having driven these traps in the same locations for 12 years, but 
changed them to conform with the law. 
NUMBER IN OPERATION 
The total number of traps operated each year in Puget Sound has been rather 
difficult to obtain owing to the fact that a trap need be driven only once in 4 years in 
order to hold a location. Furthermore, where the driving of one trap would tend to 
lead fish away from another it has been the general practice among companies to 
drive the one location for fishing and to hold the other by driving a “dummy” trap 
there at least once every 4 years. A dummy trap was very poorly constructed, and 
hung chiefly with old, worn-out gear. The object was merely to comply with the law, 
the dummy not being expected to catch more than a few dozen fish. 
In addition to these dummy traps there have always been some traps of an 
experimental nature, especially in years of abundant runs and good prices. Many of 
these locations have been driven but once, others have been tried from time to time. 
The efficiency of the traps has not varied as much as the number in operation 
from year to year might seem to indicate, since the best locations are practically always 
fished, and many of the extra traps, added during years of abundant runs or high 
prices, are driven in inferior locations. 
The number of traps in operation, exclusive of dummies, is given in table 6. 
Between 1895 and 1900 the traps doubled in number three times, reaching a peak of 
163 in 1900. During this first great expansion many inferior locations were tried and 
later abandoned, as shown by the lessened number in all years except for those of the 
