ay, and as long as a certain proportion go up the stream there's
no great harm done. But if every one of the salmon is caught, as happens
when nets are put all the way across a stream, there will be none to
spawn, and in a few years there will be no fish in that river."

"Do the fish always return, when grown up, to the river in which they
spawned?"

"That is disputed. But the large proportion of such fish do not travel
very far from the mouth of the river in which they were born and the
natural impulse for fresh water at spawning-time leads them naturally to
the nearest stream. So, it is imperative that some fish be allowed to go
up-stream, or in other words, that salmon-catchers allow a certain
proportion to escape their wheels and nets."

"They ought to be willing enough to do that, I should think," said
Colin; "it's for their own good in the long run."

"A lot of them want quick profits now, without any regard for the
future," his host said scornfully. "Of course, there are laws for
fishery regulation in many of the States, but inspectors have their
hands full in preventing violations. In Alaska, which is a territory
still, that supervision is done by the government through the Bureau of
Fisheries."

"It must be a little aggravating to the salmon men, just the same," said
Colin thoughtfully, "when they are trying to keep their canning
factories going full blast, to have to allow half the catch to go on up
the stream. But," he continued, "why don't they catch the salmon coming
down the stream again? I should think that would settle the whole
question."

"It would," said the professor, "if they came down! But they don't.
Every single salmon, male and female, that goes up the river in the
spawning season dies up there. None of them ever comes down alive."

"I don't think they did that way in Newfoundland!" ejaculated Colin in
surprise. "When I was staying with my uncle there I saw lots of salmon,
and it seemed to me that they went down the river again."

"They did," was the reply. "The Atlantic or true salmon does not die
after spawning, but not a single fish of any one of the five different
kinds of Pacific salmon ever spawns twice. Every yard of the shores of
the upper reaches of Pacific coast rivers is covered almost solidly with
dead salmon from September to December!"

"How awful!"

"It makes some places uninhabitable," the professor replied. "Where a
market is near enough, the dead fish are collected