168 
FOREST AND STREAM. 
[March 2, 1901. 
narrow stretch of water in the same northwesterly direc- 
tion for thirty miles before the open sea is reached. 
Paper Mill Creek is formed by the junction of the San 
Geronimo and Lagunitas, each of which is quite small, 
the water cold and swift, running over a rough rocky 
bottom, and everywhere overhung wit-h trees and bushes. 
After uniting to form the main stream, the water fre- 
quently runs over quiet stretches of gravelly bottom, but 
for the most part the bed is rough and rocky and there 
are many cascades and deep rocky pools. The stream 
for nearly its. entire length is well shaded with over- 
hanging trees. The width of the stream in its normal 
state is about fifty feet, and its depth averages about two 
feet, but its size is increased enormously after each heavy 
rain. The water in the upper end of Paper Mill Creek 
during the six months after 'planting the fry had an 
average maximum temperature of 64 degrees, while at 
the lower end in tide water the average was 71 degrees. 
Nicasio Creek rises to the east of Paper Mill Creek, 
and flows into it about five miles above tide water. 
Nicasio Creek is very much smaller than the Paper Mill, 
being only about twenty feet wide, with an average 
depth of six inches. Throughout its length it flows over 
flat, gravelly beds, with few trees or bushes to shade the 
water. The sun has full sweep at it and heats the water 
each day during the summer to a temperature of 75 
or 80 degrees. 
Olema and Hatchery creeks rise to the west of the 
main Paper Mill, and flow almost parallel to it, emptying 
into it near together about one mile below upper tide- 
water mark. Olema Creek is about the size of Nicasio, 
but carries more water during the dry season. The 
water, too, is much cooler, and the upper stream has more 
of the character of the upper Paper Mill. The banks 
are thickly grovra with brush and trees. The last two 
miles of the creek runs through low swampy land, with 
its banks most of the way heavily lined with willows. 
The average maximum temperature for lower Olema was 
63 degrees, and the upper stream averaged but one de- 
gree colder. 
Hatchecy Creek is very much smaller than Olema 
Creek, and te not nearly so long. At first fed by springs 
and running through cool, shady woods, the water is 
qtiite cold, but on gaining the open valley and running 
akrough two miles of marshy lowlands, with but little 
shade, it becomes, on reaching Paper Mill Creek, even 
warmer than the water in Olema Creek. The average 
maximum temperature of the upper Hatchery Creek 
was 54 degrees, while at the lower end the average was 
64 degrees. 
The species of fish found in these streams are limited 
to four. The most important is the steelhead {Salmo 
gairdneri), which runs in from salt water each winter and 
spawns in the streams, keeping them well stocked with 
young trout About the most abundant fish is the "chub" 
(RuHlus symmetricus). The "stickleback" {Gasterosteiis 
cataphr actus) and "blob" (Coitus gulosas) are fairly 
abundant. Occasionally the dog salmon {Oncorhynchus 
ketd) enters these streams and spawns. 
Tomales Bay. nowhere deep, grows very shallow at 
its upper end, where it receives Paper Mill Creek. No 
bar ever forms across the mouth of the creek, as is the 
case in most of the small coast streams, but is always 
open for fish to run either in or out. Temperature of 
Tomales Bay at its upper end averaged 70 degrees. 
The Planting of the Fry. 
The young salmon at the time they were turned loose 
had an average length of 1.35 inches. On being liberated 
from the cans they scattered in all directions, the swift 
current at first carrying them some distance down stream. 
But they soon recovered, and turned their heads against 
the strong current, and found their way immediately to 
the quieter waters along the shallow edges, in eddies, 
quiet pools, or among the stones at the bottom of the 
stream. Some few would even move a few yards up 
stream, and demonstrated that they were able to hold 
their own even against a very swift current. On gain- 
ing quieter waters they rested themselves, moving only 
enough to keep from drifting down stream. As soon 
as they gained a position of this kind they immediately 
began feeding on any particles of food that floated Within 
their sight; often snapping viciously at flies half as large 
as themselves. Each individual acts for himself. They 
do not run in schools, and no more than two or three 
are found together unless the nature of the stream com- 
pels it. In eddies and pools they are thrown together, 
but in a long, shallow, quiet stretch of water they scatter 
out — each one taking a position in the eddy of a small 
stone or other object, where they have to barely move 
a fin to keep their position, while the water above them 
is moving swiftly. Lying in such places they watch for 
insects or any particles of food that may be floating past. 
They feed principally upon objects floating on the sur- 
face, but often they are seen to strike at objects Ijeneath. 
After striking at an object they almost invariably return 
to their old position. Frequently one would be seen 
to move a few yards down stream and take up another 
similar position. The young salmon undoubtedly move 
down stream in this manner; moving down individually 
and by dropping a few feet at a time. 
After two weeks in the stream but very few of them 
were to be found in the shallow water; nearly all had 
moved into deeper holes. Her,e they would be nearer 
the center of the current, where there is more floating 
food and where they could also hide, for by this time they 
are very shy. and dart into the deeper water or under a 
rock at one's approach. The young steelheads which 
hatched out shortly after this time were not nearly so 
shy at the same age. 
Four weeks after planting, practically all had moved 
into the deeper holes, and as long as they remained in 
the streams they occupied these places. In the deep 
waters of the lower Paper Mill, where the water was 
four and five feet deep, they would hang stationary in 
the center of the current a foot or two below the surface, 
darting now and then at objects on the surface. 
Movement Down Stream. 
As before stated, the only noticeable movement down 
stream was the occasional movement of individuals drop- 
ping from one resting place to another. During the first 
three weeks after planting, the water in the streams was 
|]ierh ancl jriudd.v, and except in Hatcl]e|-v C^re^k, ^qq^, 
observations could not be made. It was only a matter 
of a couple of days until most of the salmon had dropped 
down out of the upper Hatchery Creek. If there was a 
similar movement of the young salmon in the other 
streams during the first few days I do not know; but 
when the first seining was done, a month after planting, 
the fry were found more abundant in tidewater than in 
the upper streams where they were liberated. There 
can be no doubt that there was a decided movement 
down stream during the first month, which probably 
began immediately after planting. 
Forty-five days after planting, tlie fry were found in 
considerable numbers in brackish water. As this was 
the first time seining was done in brackish water, we 
do not know how much sooner they reached this point. 
We subsequently learned that the fishermen near Mar- 
shall, fifteen miles down the bay, caught the young sal- 
mon in considerable numbers as early as the last of 
April, or fifty days after they were liberated. At this 
time for about a week they caught fifteen or twenty at 
a haul in their 300-foot nets. After a week they caught 
PLATE I, — SALMON FRY. 
Description of Plates. — ^The three accompanying plates are to 
illustrate the rate of growth of quinnat salmon try from the 
streams of Marin county, Cal. 
In Plate 1. are four outline figures of the fry at progressive ages. 
Fig. 1 reprsesents the young salmon with yolk sac just absorbed; 
Fig. 2, 25 days after absorption of yolk sac; Fig. 3, 46 days after 
absorption of yolk sac; Fig. 4, 62 days after absorption of yolk 
sac. Figs. 1, 2, 3 and 4 are from specimens 1.35, 1.7, 2.51 and 
2.98 inches, respectively. 
Plate II. represents a young quinnat salmon taken in brackish 
water, 107 days after absorption of yolk sac — showing loss of mark- 
ings below lateral line and the faded out parr marks, caused 
by the brackish water. _ Drawn from an average specimen 3.2 
inches long, This drawing should be compared with Plate III., 
which figures a young salmon from fresh water. 
Plate III. represents a young quinnat salmon taken in fresh 
water 162 days after absorption of yolk sac. Drawn from an 
average specimen 3.55 inches long^. 
only two or three in a haul, and then, as the fish grew 
scarcer, they caught only two orjhree in a day's seining. 
After the middle of June, three months after planting, 
they caught no more. 
It is unfortunate that we have to rely solely on the 
fishermen for information as to when the salmon reached 
salt water. Information obtained in this way cannot 
usually be relied upon, but the young salmon was a new 
"fish in the bay and the fishermen could not help^ noticing 
it. The stories of the different fishermen) agree so well 
that in this case it is safe to rely on them. 
The first seining in the bay was on April 29, fifty days 
after planting, on which day we caught one of the young 
salmon. This was at a place about two miles from the 
mouth of Paper Mill Creek. The seining was done with 
a thirty-foot net. and although we seined almost daily 
in the bay after this and with a fifty-foot net, not another 
salmon was caught. 
, The young salmon were found in brackish water, near 
the mouth of Paper Mill Creek, up to May 11, two 
months from' the time of planting. This date probably 
marks the end of the rim. Those that remained con- 
tinued to go out a few at a time, until by the sixth 
month but very few were left in the streams. After the 
sixth month no seining was done until the eleventh 
month, which was after the winter rains. After the most 
thorough seining at this time only three of the frj' could 
be found, showing that practically all had run out before 
the end of the year. 
By the middle of the second month, when the water 
had become clear enough to watch the young salmon 
from_ the bank, it was found that they retained their 
positions in the current for hours without moving down 
stream even for a few feet. In one instance an albino (one 
without color) kept one position for three daj^s, after 
which it could not be found either above or below its 
old position. These facts all tend to show that there 
was a decided movement of the salmon down stream 
during: the first mpnth, gp^ ^-^^'^ ^^^^ ^'^^^ ^^'^^^ moved 
out miich slower. It probably took the rains of the 
next winter to drive them all out. 
We tried another experiment. By stretching a net 
across a narrow place near the mouth of Paper Mill 
Creek as the tide was coming in, it was demonstrated 
that the salmon play back and forth with the tide before 
passing into salt water, as they repeatedly ran into the 
net ahead of the tide when it was fixed in this way. 
Movement Up Stream. 
In all of the streams except Hatchery Creek the fry 
were planted where they could easily move up stream 
two or three miles if they so desired. The streams were 
carefully seined above the planting places to determine 
to what extent they moved up stream. In Nicasio Creek 
only a very few had moved up, and they had gone but 
a fourth of a mile above where they were liberated. 
We found a few in Olema Creek a fourth of a mile above 
where they were planted. In Paper Mill Creek I found 
that none had moved up stream. All of the side streams 
were seined to see if the fry had entered any of them, 
but it was found that they had entered only one — a very 
small stream fed by springs flowing into lower Hatchery 
Creek. The water in this small stream was much colder 
than the water in the creek and was grown full of water 
cress. Nearly a hundred of the fry had entered this and 
were found as far up as they could get. Some had even 
found their way through a perfect mat of grass and 
drifted rubbish. The current in this stream was very 
slow, and it entered the creek at such an angle that it 
was a sort of trap, and I have no doubt the fry got in 
there in trying to get down stream. Although some of 
the fry do work up stream for a short distance, they 
are the exception. Why they move down stream we do 
not know, and we will have to credit it to instinct. A 
young steelhead or other trout when frightened prefers 
to dart up stream rather than down, and if one wades 
down stream and attempts to drive them before him he 
will find that they will not drive, but will dart between 
his legs or around him in their frantic efforts to get tip 
stream. Knowing this trait of the fish, in seining for 
them the net is always hauled down stream. With the 
salmon fry it is different — when frightened they will run 
even more readily down than up stream. By walking 
filong a small stream they tein be driven either up or 
down._ In seining for them they are caught as readily 
by seining against the current as with it. 
Growth. 
The most important part of the work was finding the rate 
of growth of the young salmon while remaining in the 
stream. This was determined by capturing series of the 
fry at intervals and making careful measurements of 
their length. 
The young salmon were put in the stream on or about 
March 10, thirty days after hatching, at which time their 
yolk sacs were absorbed. At this time their average 
length was 1.35 inches, and the rate of growth is figured 
from this time. Although the rate of growth could have 
been determined more accurately had larger series of fry 
been saved, it is accurate enough to demonstrate a re- 
markable growth and to enable us to make comparisons 
between the different streams and different parts of the 
same stream. It also gives us some idea what to expect 
in larger rivers. 
The most rapid growth was during the first month, at 
the end of which time they had, in the lower stream, 
reached an average length of 2.28 inches; a gain for 
'■he month of .93 of an inch. A few individuals reached 
a length of 2.40 inches. 
Through the second month the growth was not quite 
so great, and during the third and fourth months corre- 
spondingly less. At the end of the second month the 
salmon in the lower end of the creeks averaged 2.9S 
inches in length, and a few had reached a length of 3.1 
inches. 
During these first two months the fish in the lower 
streams averaged considerably larger than those above. 
At the end of the first month those above averaged 1.71 
mches, against 2.28 inches for those below. At the end 
of the second month those above averaged 2.75 inches, 
against 2.98 below. 
After two months there was no marked difference 
between the salmon in the lower streams and those 
above, except in Hatchei^y Creek, where those- aboVe 
remained very small. 
At the end of four months the average length of the 
salmon was 3.35 inches. The rate of growth during the 
four months, taken from an average of all the salmon 
retained, was 0.52 of an inch per month. After four 
months this rate of growth fell off very considerably, 
and at the end of the fifth month they had reached an 
average length of only 3.46 inches. After the middle of 
the sixth month no more seining was done in the streams 
until the following February, or eleventh month, Avhen 
one salmon was found in Olema Creek measuring 3.64 
inches. On April 14, thirteen months after planting, 
another was found in Olema Creek 4.45 inches long. 
One other was found in Olema Creek, of which I have 
no measurements. 
Nearly all the salmon had run out of the streams by 
the end of the fourth month, and those remaining after 
this time were evidently stunted individuals, as is shown 
by their very slow growth. An instance proving this 
is shown in the following: In March, 5,000 of the fry 
were P;Ut in Bear Valley Creek, which is dammed at its 
lowest end and fixed with a screen so that the fish cannot 
get out. Here, eleven months after planting, two salmon 
were caught, each measuring 5.5 inches. At the same 
place, fifteen months after planting, two specimens were 
caught, one 6 inches^ the other 7.5 inches long. This 
remarkable growth in a stream where they were con- 
fined is another point that leads to the belief that the 
salmon which preferred to remain, after four months, in 
the streams where they were not confined were individ- 
uals stunted or the physical inferiors of those running 
out during the first four months. 
The growth of the salmon was the same in Paper Mill. 
Nicasio and Olema creeks, but in Hatchery Creek the 
growth was not so raoid. The salmon in the lower end 
of the ci'eek greAy rapidly enough, but those above grew 
very slowly, gaining on an average the first three months 
onlv .15 of an inch per month. At the end of the fourth 
||]f^ M descen^^4 ^'^Vi ?K?^- 'W. 
