222 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
One of these experiments was made at Baird station by using fyke nets, set in 
McCloud Eiver. In a small stream without too strong a current this method might be 
employed advantageously, but in the rapid current of the McCloud, which, though 
not a wide stream, carries a large volume of water, the fyke-net exiieriment proved a 
complete failure. 
One or more large wooden traps have almost every year been built into the rack 
which extends across the river, and at times, especially during a rain storm accom- 
panied by a marked rise in the river, large numbers of salmon are taken, but at other 
times only a few, and at all times only a small x>ercentage of spawners are captured 
in the trail. The trap is quite a valuable auxiliary to the seine, but it would be a poor 
dependence if relied inion exclusively, because, although it will secure a great many 
unrixie fish, the ripe ones, which are the ones that are wanted, finding an obstruction 
in their way, settle back to the spawning-grounds below and remain there. 
Large dip nets have been occasionally used at the Clackamas station, in Oregon, 
the fishermen standing on the rack at night and dipping below it. Toward the end 
of the season this method secures a considerable number of spawners, but it involves 
labor and exi^ense, and after all it is an open question whether most of the s^iawners 
taken with the dip nets would not have been cajitured in the regular course of fishing 
The following xilan deserves a brief descrixition, as it is, I think, unique among 
methods emxdoyed by fish-culturists for caxituring salmon: 
There not being any entirely satisfactory seining-grounds at the Clackamas station, 
and the river just below the rack being shallow, we resorted to the Indian method of 
fishing. The aversion of the salmon to heading downstream is well known, but when 
they are very much frightened (stamxieded) they will turn around and rush down- 
stream at their utmost speed. The Indians take advantage of this and build a clam 
of rock or wickerwork, or anything that will xireseut an obstruction to the frightened 
fish. This dam is shaxied like the letter V, with the angle downstream, and at the 
angle, of course, is a large trax>, which they can easily enter but can not escaxie from. 
This method of capturing the breeding salmon was the xirincixial one employed at the 
Clackamas, and it worked very satisfactorily. 
At Baird station, before it became customary to put a rack every year across the 
river, the seine fishing was exclusively done after dark, and was usually kept ux) all 
night. Since the rack has been used the seine has been hauled more or less in the 
daytime, with x>Prfectly satisfactory results. We generally begin fishing now about 
4.30 a. m., and keep it up as long as the fishing warrants it. We begin again about 
5 o’clock in the afternoon and continue as we do iii the morning. 
The seines used at Baird station are from 120 to 170 feet long, made of about 
28-thread twine, with a 4-inch mesh and a 20-foot bag, taxiering down to about 6 feet 
at the ends. The seines have to be double- leaded, on account of the xiowerful current 
of the McCloud. 
METHODS OF SPAM^NING THE SALMON. 
All methods of spawning salmon are in general the same, as of course they 
must necessarily be. There are, however, some slight differences in details, chiefly 
in holding the xiarent fish and in the manner of inixireguating the eggs. 
Where there is plenty of helxi and the salmon of medium size, the most expeditious 
way of holding the fish seems to be for the man who spawns the female salmon to hold 
