t I

Studies on the
PACIFIC PILCHARD or
SARDINE (Sardinops Caeuleo)

SPECIAL SCIENTIFIC REPORT: FISHERIES Na15

UHfTED STUTES OEPUIMENI OF THE INTEmOII
FISH MD WIIOLIFE SERVICE

Studies on the
PACIFIC PILCHARD or
SARDINE (Sordinops Caeulea)

SPECIAL SCIENTIFIC REPORT: FISHERIES Na15

UNITED SUITES DEPtRTMEIIT OF THE IKTERIOR
FISH HMD WIIOIIFE SERVICE

Explanatory Note

The series embodies results of investigations, usually of
restricted scope, intended to aid or direct management or
utilization practices and as guides for administrative or
legislative action. It is issued in limited quantities for the
official use of Federal, State or cooperating agencies and in
processed form 'for economy and to avoid delay in publication.

Washington, D. C*
May 1950

United States Department of the Interior
Oscar Lo Chapman^ Secretary
Fish and Wildlife Service
Albert Mo Day, Director

Special Scientific Report ~ Fisheries
NOo 15

STUDIES ON THE PACIFIC PILCHARD OR SARDINE
(SARDINOPS CAERTTLEA)

CONTENTS

Page
lo structure of a Research Program to Determine

How Fishing Affects the Resources by Oscar Eo Setteo •oooooo 1

2o Determination of the Age of Juveniles by Scales and

Otoliths 5 by Lionel Ao Walford and Kenneth He Mosher • o » ooo«31

So Determination of Age of Adults by Scales, and Effect
of Environment on First Year's Growth as it Bears
on Age Determination, by Lionel A<, Walford and Kenneth Eo Mo:£b3fo ^6

4o Influence of Temperature on the Rate of Development of Pilchard

Eggs in Naturej, by Elbert Ho Ahlstrom ooeooa».eoooo o-^

5o A Method of Computing Mortalities and Replacements,

uy Ra±ph ro oio-ximan oe««ooeo ooooooooooooo ©-OO

?o Thermal and Diurnal Changes in the Vertical Distribution of Eggs

and Larvae J by Ralph Po Silliman ooooooo o aeoeooo J81

Notes This is a reissue of Special Scientific Reports 19-24,
■ issued in 19438

1» structure of a Research Program to Determine How Fishing Affects

the Resources

By

Oscar Eo Sette l/

CONTENTS

O O O O O O

o o o o

o • «

Preface

Introduction

Vital Statistics . » .

Source data » « •

Analysis oo

Reliability
Tagging research o • • • o • «

Source data o <> ••><>« •

Analysis eoo«ooo*o
Refiruitment research o • o . •

Source data <> o o > » » <>

Analysis.. . . . o
Stanmary.
Literature cited

ee*ooooooiioe9

o o • e • e

o « o • e o e

0 0 0 4

oooooovoo*

o o • 0 o o

• eoooooo»9

• •ooeevoeoAV

0*0090fl0«04e

00400000

ooooo«ee«e«

oeoooeoooo*

• 0000*»**90

o«ooev«»«

9 e e o o o o

oveoooooooo

o»«oooooeo«o

oooooooooooo

Page

1

6

6

7

15

16

16

17

19

21

28
28

l/ Now Director, Pacific Oceanic Fishery Investigations, Fish and
Wildlife Service, Honolulu, T. H.,

INTRODUCTION

The fishery for Sardinops caerulea, known as "sardine" in California
and "pilchard** in the Pacific Northwest, became important during the war
of 1914-1918 and has since grown to be the largest in North Americao It
started in California and reached its greatest development hereo As early
as 1919;, research on the fishery was under way by the California State
Fisheries Laboratory,, an agency of the California Division of Fish and
Gamso More recentlyj as the fishing spread into northern waters, Oregon,
Washington, and British Columbia have engaged in research and, since 1937,
the Uo So Bureau of Fisheries, later becoming a part of the Uo So Fish
and Wildlife Service, has also participatedo With five agencies working
together, this has become a cooperative research program, in the best
senseo In the spring of 1936 and annually since then, the biologists
have met to discuss the problems of this fishery euod to coordinate their
efforts to solve themo

Out of the research activities and from these meetings there came
evidence of the complexity of the problem of research on the sardine and
of the need for clarifying relations between various phases of the re-
searcho These have been discussed periodically and at length within the
staff of the South Pacific Investigations of the Uo So Fish and Wildlife
Service o At one of these staff meetings was drafted a statement of ob-
jectives and of the information required to attain themo As the time for
the 1942 conference approached, Dr<. Richard Van Cl«ve, Chief of the Bureau
of Marine Fisheries of the California Division of Fish and Game, suggested
the desirability of a formulated outline as a guide for discussiono In
collaboration with Vernon Brock of the Oregon Fish Commission, such an
outline was drawn, following a procedural diagram which I constructed at
the same time, and embodjdng results from our Service staff meetings and
the suggestions of Dro Van Cleve and Mro Brooke The diagram mentioned
( opposite ) thus represents the ideas of a nxjmber of personso It
was distributed and discussed at the 1942 meeting of the five agencies
•without eliciting demands for important revisions o While the program
is thus the product 'of a number of persons, the exposition given herewith
is that of the Fish and Wildlife Service laboratory at Stanford University
and includes argument on what may be considered controversial topics.
It is appropriate, then^, that the responsibility for the exposition be
assumed by the author who hastens to add that the program is subject to
constant revisiono

This discussion is mainly an elaboration and justification of the
diagramg which by itself is somewhat cryptic, owing to the compression
of complicated ideas into short phrases « In the outline the titles of
procedures or projects either under way or rather fully planned are en-
closed in solid-line rectangles and connected with the others by solid
lines o Those recognized as desirabl® or essential but not yet integrated
into the working program are indicated by broken lines o Heavier lines
follow the main procedural path and lighter ones connect tributary adjust-
ments with the main procedural lines „ The titles at the left margin.

o
It:

g
^

on
p

I

<

u

3

E

b

lu

o

^^

!<

z

a3^

^

1%

not enclosed by rectangles, refer to material to the right of eacho
To preserve the closest possible connection between text and diagram,
the exact phraseology used in the diagram is repeated as headings in
the textj, even thoughj, in some instances, it results in rather awkward
context o

Objective

Underlying the investigations of the Uo So Fish and Wildlife Service
on the Pacific sardine or pilchard are certain ideas or theories, the
discussion of which will clarify that of objectives » In the first place,
it is recognized that before fishing took place the sardine population
had filled its ecological niche and that, apart from fluctuations such
as occur among all organisms, the deaths, on the average, equaled the
births and the population was in equilibrium, ioeo neither increasing
nor decreasing in total numbers. Moreover, the deaths under these con-
ditions were all due to "natural" causes. Predators were, of course,
responsible for a substantial portion of the natural mortalityi but the
basic influence tending to keep the population in check must have been
competition within the populationo This competition may have been of
many different forms, all, however, connected with the density of the
population itself « It might be failure in reproduction which could arise
from the overabundance of spawners, the debilitating effects of which
might affect the viability of their eggs j or if the viability of eggs
were unaffected, simple sprea^ into marginal spawning grounds with con-
sequent failure of eggs to hatch would keep the population in checko
Or the competition might simply increase mortality through division of
a limited food supply among more mouths o Or the pilchard population,
grown to maximal size, might so well supply predators with food that
they would multiply to an extent which would check further increase of
the pilchards o Whatever the mechanism, the existence of a depressing
effect of population size on population increase is implicit in the idea
of natural population equilibrium. It also involves negation of the
idea that a fishery can ultimately destroy a populationo

What, then, happens when an important amoimt of fishing takes place? '
There is an added cause for deaths* These deaths are largely among adults,
and to the extent that catch mortality goes beyond replacing natural mor=
tality, that is, to the extent that fishermen take fish that would not
be eliminated by sharks and other predators, the number of adults in the
population is reduced and the size composition is changed in the direction
of fewer large individ'oels » But the fact that the population grows lesser
in numbers and its individuals smaller in size, is no proof that the fishery
is taking more than it ought to take, for another thing is happening at
the same time o The depressing effect of population-size on population-
increase is diminishingo If internal competition had been strongest among
the commercial sizes, catch mortality would largely replace natural mor-
tality and the population decrease would be slighto If it had been strong-
est among pre -commercial sizes or between commercial sizes on the one hand
and pre-commercial sizes on the other hand, then replenishment of the com=
mercial population would increase somewhat in proportion to the amount of
"thinning oute"

So far the argument has been speculative <> But it is not necessary
to rest on thiSo There have been many experiments in which populations
have been reared under laboratory'- conditionSo These have involved such
diverse forms as yeast,, diatoms^ protozoans, insects, and even fishes o
In all of them the rate of increase was rapid while the population was
small and slowed down as it grew larger, finally reaching a size beyond
which there was no increase o Initiallyp when the population is small,
even though the reproductive rate is high, relatively few individuals
are added because there are only a few producers. Finally, when the
population is at its maximum and there are many reproducers, there are
also relatively few individuals added, either because the reproductive
rate has become depressed or because mortality has become so high as
to offset the additions from reproductiono Usually both a depressed birth
rate and incre' sed mortality operate together to keep the net increase
in numbers of individuals at a minimurrio In contrast to the small increase
in population~size at both low and high population densities, is the large
increase that takes place when the population is about midway between
these extremes. Then, the population is large enough to have manyre =
producers and small enough so that the reproductive rate has not been
greatly lowered and mortality has not yet been intensified. In this state,
the increase in numbers of individuals per unit of time is large*

The lesson is obvious o A population has its maximum increase when
it is neither at maximum, nor minimum, size, but when middling in sizej
and that is the level at which the most individuals can be regularly
removed and still be fully replenished by the population's inherent ten=
dency to growo

Of course, with a population of fishes in their natural habitat,
things are not as simple as in these experiments. With the pilchard
it is almost beyond question that predation would be a large influence.
Fortunately, there have been population experiments that included the
element of predationo In these it has been shown that the predator popu-
lation increases as the prey increases - with a time lag, to be sure.
Therefore, holding a prey population at a middling size would similarly
limit its predators^ especially with a species like the pilchard which
is in a dominant position among prey fishes. So the factor of predation
should not greatly alter the fact that replacement will be maximal in
a medium-size population. The time lag in the growth curves of composite
predator-prey populations does introduce fluctuations and may disturb
not only the general level but also the precise level at which the prey^
population would otherwise have its maximal replacement. This, however,
does not affect the general concept.

According to this idea, the optimal amount of fishing would be that
which lowers the population numbers to a level at which maximum increase
takes place. Such a fishery will remove only as many fish as will be re=
placed and the number removed will be the largest number that can be regu-
larly removed without further disturbing the population-size.

Howeverp there are things to be desired in a commercial fishery
beyond merely a continuing yield of maximal members of fisho The stock
mijiS'c be S'off ioiently abundant to be economically fishable and the sizes
of fish in the catch should be economically desirable o

At the present stage of research on the pilchard population, we
do not have any notion of the level at which maximixm replacement would
take places nor as to whether the abundance at that level would be an
economio one or whether the size composition would be commercially de=
sirableo We can, however^ be confident that the replacements the abund-
dance,, and the size composition will be affected by the fishing intensityi
and we can be sure also that we shall want the replacement to be equal
to the catch so that no continuous decrement will ensue o

Therefore„ for the time being, the objective is "to determine at
various levels of fishing intensity the quantity and quality (sizes of
fish) in the average annual catch." Choice of the optimum intensity can
then be made among the consequences of the various fishing intensities
according to what is economically and socially desirable o That will
undoubtedly involve other than biological considerations o

It should be noted particularly that it is not an objective to de-
termine whether abundance is decreasingo The population can be expected
to decrease long before maximal yield is reached. It is not along an
objective to determine whrther the average size of the fish is decreasing.
This probably will be the finding but is not of itself a sufficient in-
dicator of a desirable or undesirable situation. No mention is madi^ of
protection of s pawners or of spawning grounds, or of size limits, or of
closed seasons, or of wasteful practices in utilization, or of other topics
that often are the concern of conservationists. These items may or may
not be involved eventually. At this time, they are secondary to the de-
termination of what consequences attend different amounts of fishing and
cannot be intelligently considered until these are known. In fact,, the
method of conserving a commercial fishery involves social, economic,
and political considerations and lies in the field of political economy.
Biologists need only determine the consequences of certain actions and
make them known to the law-makers and administrators.

Characteristics of Catch and Stock

The objectives are stated in terms of v/hat will be caught at dif-
ferent levels of fishing intensity. With sufficient information, the
characteristics of the catch could be defined each season, and over a
long time, empirical curves could be made that would show the relationship
of intensity of fishing v/ith quantity and quality of catch. Indeed,
save for data on fishing intensity, ready collection of which was pre-
cluded by certain practices of the fishery, this job of biological ao-=
coxaating has been done by the California State Fisheries Laboratory over
a period of 23 years during which the fishing intensity has varied widely.

Had it been possible, to include fishing intensity or its complementp
relative abundancej^, today's task would be enormously facilitated«
Eren so, it would be necessaty to relate or transform the characterise
tics of catch to those of the stock, for replenishment depends on the
fish that are left in the seao

Of course, if the fishery took a random sample of what is in the
sea, the characteristics of catch would be the same as those of stocko
But if we can b e sure of only one things it is that the fishery does
not take a random sampleo In the Pacific Northwest (Oregon^ Washington,
and British Columbia) the fishery over=samples the older fish? in so'jithern
California, the youngo In Central California, it perhaps over-^samples
those of intermediate af;eSo In certain areas between ports, notably
off northern California and southern Oregon,, there is no fishingo At
a single port, the -size composition changes significantly from d ay to
day, week to week, and month to month, and there is no assurance that
the catch of the various sizes is proportionell to their relative abundance
in the seao In fact, it is obvious that the various components of the
stock are represented in the catch in proportion to the length of time
each is subject to fishing in a particular area and not at all in pro=
portion to its abxindance in the sea©

Thus the major problem of research is how to deduce the characteris=
tics of the stock from the available samples in the catcho Once those
are known, total mortality can be estimated by computational processes
developed by Bayaa&v, Thompson and Bell, Ricker, and others© Also under
certain conditions, it will be possible to separate natural from catch
mortality! and both being known, to compute the recruitment (the equiva=
lent of the actuary's birth-rate, taken as of the attainment of commercial
size) o

Analysis

In the diagram, the central part^analysis is taken up with the
processes concerned with adjustments, conversions, and combinations in=
tended to transform data from the catch into attributes of the stocko
These processes leading to *vital statistics* are so compleg that simple
appraisals of their reliability by conventional statistical methods are
unavailing* Confirmation must, therefore, be sought in results attained
through independent approaches » The only other approach that is currently
being followed is taggingo The processes involved in this branch of re=
search are noted in the left=hand portion of the diagram and will be dis=
cussed under the heading of "T&gging researoho* Even after determinations
of mortality and recruitment have been made by two methods and one found
to varify the other, it is still necessary to investigate recruitment be^
cause changes in it are of differing significance, depending on whether
they be due to the changes in the stock itself or to fluctuations in the
environment o The fBiaasesable means of dealing with this problem are listed
in the right-hand part of the diagrams, and will be discussed under the
heading, "Recruitment researcho"

=^ At uniform availability and with no competition between units of gears
Relative abundance = catch=per=unit=of =eff ort.? and
number of units of effort = Catch

Catch-per-unit-of -effort

VITA.L smTisncs

Source Data
Total catch. - The hasic data for this central branch of research

are the experiences of the fishermen themselves. Their "'Total catch"
is needed indirectly to derive "Intensity of fishing." It is used also
to compute "Vital statistics," a connectijon which is not shown in the
diagram. These statistics are collected by the several Pacific Coast
States and the Province of British Columbia as a matter of administrative
routine.

Boat operation. - Records of activity, particularly of the time spent
fishing, would be invaluable for computing "Catch=per-unit=of -effort,"
as it would permit by-passing two adjustments and would avoid an addi-
tional difficulty inherent in statistics of "Catch"=per-boat-week:,*' Un-
fortunately, records of boat operation are lacking except for a very small
sample resulting from an observer's interviews with fisherraeno Kftiether
or not these interview-data may be pressed into service in lieu of more
complete operational records remains to be seen.

Boat-catch-records » - For the time being, the records of individual
daily landings by fishing boats, are the main reliance in deducing changes
in abundance or in availability of the pilchards from time to time and
from place to place. These records collected by the State and Provincial
Governments, give only the quantity of sardines in each delivery to the
processing plants. In California, each delivery represents one night or
a fraction of one night of fishing, in the Pacific Northwest, from a frac-
tion of a day to several days of fishing. In neither region is there any
record of nights or days spent in fruitless effort.

Length samples . - "Length samples" cover the measurements and desig-
natlons of sex of a sample, usually 50, of individuals, from a delivery
of sardines. In California, prior to the season of 1941=42, semi -weekly
samples from 5 deliveries were taken by the California State Fisheries
Laboratoryo Sine© thenj, daily samples up to ten in number were taken by
the California State Fisheries Laboratory at San Pedro and Monterey and
by the Fish and Wildlife Service at San Francisco. Supplementing these
are special series of 100-fish samples collected by che Fish and Wildlife
Service at San Francisco and Monterey daily over certain periods during
the three seasons prior to 1941-42. In the Pacific Northwest, sampling
has been somewhat less systematic. In British Columbia, the Fisheries
Research Board of Canada has taken one 100-fish sample as nearly daily
as feasible. In Washingtonj, the Fish and Wildlife Service in some seasons
and the Washington Fisheries Department in others hav6 collected one to
several 50- to 100-fish samples daily. In Oregon, the Oregon Fish Commis-
sion has taken one to several 50- to 100-fish samples sometimes daily
and sometimes intermittently.

Scale samples o - Beginning in 1941-42j scale samples have been sys-
tematically collected from ten fishes in each length sample by all of
the agencies engaged m sampling,. During three prior seasons similar
samples were less regularly obtained by the Fish and Wildlife Service
and in these and earlier seasons some scales and otoliths were gathered
by the California State Fisheries Laboratory in California and by the
Fisheries Research Board in Canadao The utilitj'- of these earlier collec-
tions is dubious but is now being surveyed by the Fish and Wildlife Service©

Analysis

Catch-per-boat-weeko - The processes described under this heading
have been devis'eTTor handling the California statistics for the seasons,
1932'!-33-1940-41o Modifications will be necessary for use with data for
the Pacific 'Worthwest and for other periods in California. The accepted
manner of compiiting the catch-per-boat eliminates two sources of variability
that are extraneous to fluctuations in abundance or availability,, namely;
variations in number and in the type of size of boats fishing. This is
accomplished by comparing a boat's performance during two separate periods.
The resulting pairs of relative catches are combined by averaging to repre-
sent fleet performance J, and combined by linking to fonn a time series.
Because the opportunity to fish is variable, depending on the duration
of the moonless portion of the night, the statistical series have been
divided into lunar periods and each lunar period into four lunar weeks.
Year-to-year ratios are then computed between pairs of homologous lunar
weeks. However, this statistic still contains several extraneous sources
of variability. One is the varying amount of idleness of boats, assumed
to be fishing, but actually in port on account of repairs to machinery
or gear and of miscellaneous causes. It has been assumed that this is
random in occurrence, and has no trend effecti and lacking any record,
no account of it is taken in the computations. Tno other sources of vari-
ability are treated below.

Adjust for efficiency. - There may be changes in efficiency connected
with the skill of crews and modifications of gear. As yet no means has
been devised of dealing with such variations.

Adjust for weather. - By computing regressions of average daily catch
per boat according to weather conditions a relation between the two is
derived and used as an adjustment factor. The effect of variation of
available fishing time each night is also incorporated in the regression
systemSo

Extensive experimental analysis of wind movement as recorded by
anemometer at such meteorological stations as were available, has led to
the discarding of this source of weather data. Instead, the wind movement
is deduced from barometric gradients in a manner which gives a representa-
tion of conditions over an extensive sea area adjacent to each fishing
port. Direction as well as force is correlated with catch, and both enter
into the adjustment. Other elements such as fog and precipitation appear
to have no statistically consistent effects on the catch, probably because
conditions at the obserx'ation points differ too much from those on the
fishing grounds.

8

The weather adjustments already computed have measured the effect
on daily catch=per=boat as an 8=-season average effect and, while reliable
over a long period^ are not accurate for short periods of time or for
specific daySo

Adjus t for limits „ <= Processing plants^ at times, have limited the
amount of f i^TPreceivable from a boato Under these conditions, compliant
boats stop fishing when the limit is approximately attained and the av^
erage catch must be less than if there had been no limito

The adjustment for the effect of limits j however, is based on the
assumption that their principal effect is to shorten the period of fishing
per nighto When a limit catch is attained by a boat early in the night,
that boat's fishing effort should be figured as only a fraction of a nighto
This may be computed as the ratio of the actual, to the total available,
fishing timeo

Accoimt is taken of boat^capacity, as well as imposed, limits* The
time of attaining limits is recorded only for the catches of boats whose
skippers are interviewed when a sample of fish is takeno This is only
a small fraction of the whole number but it is assumed that the average
time applies to the entire fleeto Due to the smallness of the sample^
the adjustment is computed as an average for an entire lunar weeko It
is probably reliable for long, but not for short, periods o Complete
operational records for the boats would be better if these were available o

Convert weight to numbero - Catch statistics are always in terras of
weight, but data must be in ntanbers of fish for computation by the methods
of vital statistics o As the relationship of weight to number fluctuates
in accordance with size-composition, the conversion is made on a daily
basis o From a predetermined length=weight relationship, the average weight
of fish in the day's length=f requency distributions is computed and this
average weight divided into the pounds -caught-per-boat-day, coaverting
that statistic into ntjmbers-caught^per-boat-dayo For samples taken in
1941 ■=42 and subsequent seasons, the average weight is determined by direct
weighing of each sample o

If now, we were to weigh each day's frequency by each day's numbers-
caught-per°boat=day and sum the frequencies for all ports and all days
for each season^ the resulting curves might be taken as representing the
relative abundance of each size of fish for the series of seasons and the
task of converting catch statistics into stock statistics be considered
complete o

Indeed, Hodgson (1939) regarded this very process, less some of the
adjustments, as giving the relative abundance of each year^class in the
East Anglian herring fisheiyo To use an analagous method for the Pacific
sardine would be to assume either that? (l). The average boat's fishing
capacity was equal in the fleet of each port, or that an equal proportion
of the fish population was drawn upon by each ports and (2), the various
sizes of fish were available randomly from each port, or that the length
of time a complex of sizes of fish was available at each port was propor-
tional to the popuiationo Howeverj these assumptiors are either known or
suspected to be erroneous o Various further devices of weighting, ®©lec=
tion, and combination need be empioyedo

Ad jus t for aggregation habits » ■= If a series of samples taken at a
particular fishing port be 'compared as to size-composition, it is apparent
that among them there are samples that resemble each other sufficiently
to have come from the same population but differ enough from certain other
samples to indicets that the latter must have come from different popula-
tions o Since ths various samples come from different schools it appears
that certain groups of schools have a uniform pattern of size-composition
and, if so^ t'ney must have been associated for a significant period of time
during which their reaction pattern was the sane within;, but different be =
t"vveenj the groups of schools o This corresponds to views expressed by wT~Fo
Thomps on (1926 po 163)? "Catches of a certain type^ or 'runs" having cer-
tain characteristics, prevail for varying periods of time, and each of
these periods is a variable unit itselfo'* Our view differs only in sup-
posing that there may be simultaneous ocoxirrence of different types of
"'rvmSp'* -- a possibility doubtless appreciated by Thompson or one that
certainly would have been,, if the fishery of a given port had spread over
as large an area_„ then as nowo It will be convenient to refer to what we
have called ''groups of schools'* and what Thompson has called "runs" as
school-groups <,

Obviously the population fished from a given port is not randomly
distributed as to si2,e<, Unless the period of time each school-group is
"available" to the fishery is either g (1) Random or (2) proportional to
the abundance of each school-groupj our sampling system, even though ^strati
fied" as to time, and even though weighted according to catch-per-boat
in the time strata, cannot faithfully represent the population of school
groups fished from that port during a fishing seasono

The summation of such weighted samples would weight the sizes in
accordance with the frequency or length of time the school-groups domi-
nated by those si^.es were fished by the flee to Since the length of time
the school-group is present and available on a ground is more likely to
be associated with its reaction pattern and the fluctuations in oceano-
graphic conditions tlian with the relative numbers of fish represented
by each school-group, such weighting would represent largely availability
rather than abundanos of f isho

At the present time (spring of 1942) means of distinguishing school-
groups are being developed » Methods of employing the school-group idea in
adjusting samples will have to await analysis of the occurrence and persis-
tence of school-groupE o It may lead to a method of weighting in school-
groups or of combining samples in certain manners or may merely serve as
back-ground information for appraising the validity of methods based on
other principles o Since methods are still uncertain the rectangle labeled
"'Adjust for aggregation habits'" has been related to the main line of pro-
cedure by a broken lineo

Compute seasonal weights o = Apart from the departure from random
distribution of s'cEool"^roups within short periods of time, there is an
even greater difference between the fall and winter populations in Califor-
nia, especially in the seasons prior to 1938-39o The fall fish are pre-
dominantly young adultSj, 3 to 5 or 6 years of age 5 the winter fish, old

adults over 5 or 6 years of apeo Usually the fall season has been longer
than the winter season and direct suiranation of samples would heavily weight
the young adults. Judging from preliminary computations, which indicate
greater abundance -» more marked a decade or more ago than at present - in
the winter when the old adults are dominant, weighting in the opposite
direction would be more appropriate. Obviously, these phenomena need be
taken into account in combining the samples for the two parts of the fish-
ing season.

Compute regional weights. - The sardine at different ages tends to
frequent different places along the Pacific coast (F. W. Clark 1940 pp.
44-46). In general, smaller individuals prevail in the south and larger
ones in the north, with many exceptions in detail. Hence, the fishery
at any one port does not afford a true cross -section of the entire popu-
lation. To get that, samples from the different ports must be combined
in proper ratios. Logically they should be weighted in proportion to the
fraction of the total population available from each port. As these frac-
tions differ from one part of the season to another, and in any case are
difficult to determine, the computation of regional weights is no simple
task.

In general, the basis for such weighting might reasonably be the
estimate of relative abundance provided by the catch-per-boat-week. But
the method by which this statistic has been computed uses a different yard-
stick at each port. That standard must be calibrated by some kind of
detennination of the relative fishing capacities of the boats in the fleet
of each port. This project is under development but has not been
advanced to the point where its usability is assured.

Another measure for determining the weighting for each port might
be derived from the statistical analysis of tagged-fish returns. It would
depend on computing the proportional amount that tagged fish are diluted
by release in the population of each fishing region. Certain difficulties
common to many tagging computations have so far prevented its use; but
its desirability is indicated hj the broken line running from "Statistical
analysis'* of the tagging procedural line to ** Compute regional weights."

Combine samples. - At present, alternate ways of combining samples
are under consideration. The first and simplest plan assumes that the
weightings have fully discounted variations in availability so that a
simple summation of weighted samples will suffice to give seasonal fre-
quency distributions that will portray relative abimdance of each size
class and shovr the relative abundance of all commercial sizes from one
season to another. As before stated, however, there is reason to distrust
this assumption and unless farther considerations dictate otherwise, an
alternative will be adopted.

The second plan does not require assumptions as to whether the weighted
samples represent relative availability or abundance. It is designed to
provide an adjustment converting availability to abundance that will not
produce erroneous results even if availability is not afactor to be dis-
counted. In essence, it is a method of selecting among the weighted samples

11

for all sampling days at all ports the ordinates for each size-category
which most nearly represent homologous levels of availability© The method
i^akes its form from the idea that each size^category is present in varying
degrees of availability to the fishery from time to time and from place
to place throughout the extent of the fishing grounds,, If then, the weighted
sample-day freatiencies correctly portray the availability of each sizs^
category;, then^ for each size-category a frequency distribution of ordinate
heights will describe the number of times this size=category occurs at
various levels of avai lability o

The problem then is one of selecting from these ordinate-height fre-
(fancies the ordinate classes representing homologous levels of availa-
bilityo Theoretically, it would be desirable to have a selection that
would represent maximum ( or 100 percent) availability of each size-cate-
goryo This would require selection of the maximum ordinate among the
whole collection of ordinates for a given size-categoryo But the ordinate
heights contain two important sources of variability apart from availability g
errors introduced by weighting factors and those introduced by the random
departures of mumbers in each size class from the true number in the popu=
lation from which drawno Therefore the maximum ordinate in many instances
wouldj by chance, be above rather than at the maximum level of true avails
abilityo To avoid this it has been decTded tentatively to take the ordinate
class at the ninth decile positiono In other words, we would derive a
frequency curve which represents the 90 percent level of apparent avail-
ability of each sizeo For the reasons just stated, this leered of apparent
availability may be very close to the 100 percent level of real availabilityo

The first plan requires that availability of each size-category have
the same pattern from year to year, the second requires merely that at
one time or another during the season the levels of availability of the
several size classes approach their maximao It appears that the latter
involves a more conservative approach. It is to be expected that knowledge
of aggregation habits, leading to the classification of school-groups,
will provide important modifications in, or appraisals of, the proposed
methods o

Determine length frequencies of year-classes » - Assuming that one
plan or another has produced a properly weighted combination of samples
to represent the stock in the sea, there remains the problem of converting
each season's length-frequency, into an age-frequency, dis tributiono
Its solution, for the years prior to the beginning of routine scale sampl-
ing in 1941, involves the use of certain statistical constants derived
from scale readings,

i

By this is meant determination of the paremeters of the length-fre=
quencies of single age-groups or year classes that will define the shape
of the frequency curves in sufficiently general terms to be applicable
to years other than that in which the scales are collectedo So far the
only parameter we have worked with is the standard deviation, as a fanc"
tion of age, assigning each age group to have a normal distribution of
lengths. While apparently reliable results are obtained, there is e'.7ldence

12

0

that skewness may be introduced by selective availability of the fish
or by fisherman selaotion in favor of largest m'smbers of young year-
classes, or, in certain localities, converse selection of favored small
fisho It may be necessary to include a skew factor to some extento
There also are indications of multimodality in the length-frequencies
of individual year classes. Whether this is sufficiently pronounced
to require statistical recognition remains to be seeno

De te rmi ne growth rateo A growth curve or curves is needed to give
another parameter, tHe mean or modal size, of the frequency distributions
of year classes. Having age readings made in any one year it is simple
to establish a growth curve, but such a curve would be a true one for
only that year. There is evidence that growth differs from year to year,
from year-class to year-class, and, more disturbing still, it appears
to differ by fishing areas « For instance, a given year"=class has a lower
modal length in the San PedrOj than in the Monterey area. We are faced
by the dilemma of using either a general growth curve fitting no specific
situation, or a variety of curves to fit situations that are largely un-
knowTio It remains to be seen whether there is sufficient regularity in
geographical or annual variations to permit age determinations of recent
years to be applied to the length data of former years,

Conve r t length-composition to age-composition, - The translation
of a length-f re quency curve to an age-frequency curve has proved unex-
pectedly difficult. As may have been anticipated from the previous dis-
cussion of parameters, the only method so far found to be applicable is
that of discovering what combination of year=class length-frequencies
will fit a season's length-frequency. The fixing of two of the parameters,
standard deviation and mean, leaves only the third parameter, relative
area, to be varied in the curve fitting. Even so, it is a laborious pro-
cess by the trial and error method, but as yet we know no other. Further-
more, error in either of the two fixed parameters very seriously affects
the results. In other words, the problem should be approached with extreme
caution.

An expedient that greatly facilitates the process may be employed
in certain situations v/here one is justified in the assumptions! (1) That
the year-classes were of uniform initial numerical strength and (2) that
they experienced uniform annual mortality since becoming of commercial
size. These assumptions are approximately justified in dealing with a
length-frequency distribution of fish pooled from a considerable nijmber
(enough to "average out*' inequalities in initial year-class strength)
of contiguous seasons during which the fishing intensity was nearly enough
constant to have produced nearly uniform mortality. This severely limits
the combinations of frequency-curve areas and a unique fit is soon dis-
covered. The goodness of fit serves as an indication of how nearly the
assumptions were fulfilled, except that uniform trends through the series,
either upward or downward and in either year=class strength or intensity
of fishing, would not be revealed by poorness of fit.

IS

Abundanise and aj^e composition of stock. - When the weighted, combined,
lerigt1i-'f'"requency™iTtrlbution has been converted to age=f requency we have
a series of season's curres representing the relative abimdance of each
age of fish in the stooko These are useful in deducing the relations be-=
tween abundance and spawning and between abundance and competition effects.
This is indicated by broken lines in the diagram and will be discussed
later. At the moment we shall pass on to detennination of mortalities
and recruitments.

Compute *•' Vital statistics." - By vital statistics we refer to re-
crui 'tifien t~"and mortaiityT"^-^ fisheries counterpart of the actuary's
births and deaths. One might also use the term '*population dynamics .**
The computations involved are adaptated from methods previously developed
or employed by BaranoVj, 1918 1 Thompson and Bell, 1934.; and Rieker, 1940.

Total mortality is readily computed in situations where recruitment
and m-ortality can b e reasonably assumed to have been uniform during as
many years (prior to the year in question) as there are ages present in
the stock. Then, it is necessary to have only one season's age distribu-
tion, for each age class will be a constant percentage less numerous than
the next younger. The constant percentage then is the annual mortality
rate.

Such simple situations are rare, but, fortunately, they are not the
only ones amenable to analysis for mortality. That statistic may be readily
deduced from the age frequencies of two successive seasons, providing
triat they truly represent the stock in the sea. Here the decline would
be measured by the relative numbers of a given year=class in the two
frequencies and the percentage reduction of a year=class from the first
to the second season would represent the annual mortality rate. If all
year classes have been fished with equal intensity, hence have suffered
equal satoh mortality, and also have experienced equal natural mortality,
the indicated total mortality will be sensibly equal for all but the
''entering'* year classes o Inequalities, therefore, may be ascribed to
ago^selective, natural mortality or age -selective fishing intensity.
Correlation of the absolute quantities of each year=class in the catch
with their respective mortalities should distinguish between these two
age-selective processes.

Recraitmento = As used herein, recruitment means the nvmber of sar-
dine 3~TeaGlain g~c'ommei ciai age each yearo Oiven ag-: -distributions that
reliably represent the abundance and age composition of the stock, the
recruitment is directly proportional to the numbers of individuals in
the year-class that has, for the first time, appeared in the commercial
catch. The only complicating feature is that there has been nothiiig in
the previous adjustments to insure that the entering year-class will be
fully represented if, in its first season, it is only partially available
to the fishery. Hence, in this respect, we cannot consider the weighted
cui-Ties as reliably representing the stocko However, from a series of
age-distributions it should be possible, without great effort, to dis-
tlnguish the age at which a year^'class becomes fully available, deduce

14

further its percentage availability in previous years, and so arrive
at a nvimerical evaluation of its relative strength at commercial age*
Indeed, trial computations have practically proved the feasibility of
this calculationo

Natural mortalityo ~ In computations so far undertaken we have as-
sume d~~thi~"probRbility that mortality from natural causes is constant.
This greatly simplifies the calculation of catch mortality. However,
this assumption should be examined critically and perhaps revised if it
is discovered that predators (other than man), and therefore predation,
has increased or decreased to an important extent in any two periods
under consideration.2/

Also the probability of death from natural causes may vary with
age, and the formulations may need appropriate revision. These eventu-
alities cannot be appraised until a series of age-distributions, properly
representing the sardine stock, become available.

Catch mortality. - While remaining aware of the possible necessity
for revising the assumption of constant natural mortality, it is con-
venient to employ this assumption to simplify the computation of catch
mortality. Given two periods, in one of which the intensity of fishing
was greater than in the other, there is a unique combination of differ-
ing catch-mortalities in the first and second periods and of constant
natural mortality for both periods that will account for the total mor-
talities and the total catches in the two periods, respectively. The
mathematical formulation must take into consideration that the natural
and catch causes for death are operating simultaneously and, therefore,
complete with each other for the lives of the fish. It also involves
consideration of whether an increase in fishing intensity, i.e. the em-
ployment of more effort or gear, brings in competition between fishing
units. That is difficult of appraisal for the sardine fishery because
addition of boats may as easily produce an effect the opposite of compe-
tition, i.e. cooperation, through extending the scouting area and com-
municating information on the location of schools. The effect of coop-
eration has been neglected in the computations so far contemplated.
It should be taken into account, and probably will be, as soon as a
method of measuring it can be devised.

"Tf It may be possible in such a case to regard the sardine as having
a number of predators, including man, and base the formulation on the
effects of the simultaneous probabilities of death from the several
categories of predation. The mathematics would be complicated and prob-
ably require data on more than two seasons, especially if a multiple
correlation method were employed, but should be feasible, given an ade-
quate quantity of data of reasonable reliability.

15

On ths other handj, it will probably be safe to rep;ard competition
betv/een gear as negiigiblej, a point of view that is tenable if cooperation
is the effective result of increasing the nianber of boats and also if
the schools of sardines are so widely distributed that instances are
relatively rare wherein a boat is prevented from making a catch by the
pre~emption of space by another boat»3/ We do not consider in the cate-
gory of competition the situation where the probability of catching
a school of sardines has been reduced by the number of schools taken
out of that locality previcuslyo This is a phenomenon of reduction of
the stock by catch mortalityo It is taken into account by appreciat-
ing the distinction between average abundance and initial abundance
in any particular period of timeo (See Ricker 1940 po 45) o

RELIABILITY

Having computed one set of values on natural mortality, catch mor=
tality, and recruitment <, one gains a general idea of the magnitude of
eachj, but the result cannot by any means be considered a determination.
The various computations and adjustments have contributed errors <> Also
the original data had variability which perhaps was increased rather
than diminished by the subsequent treatmento The mathematical result,
therefore, has a probability of differing from the true value by an
amount which cannot be known from one set of values alone.

In a series of sets of values, however^, from which one might com-
pute a regression,, savj, of catch mortality on fishing intensity, the
spread of values around the regression line would give a measure of
variability and one could be said to have really made a determination
if that spread is small in relation to the magnitude of the observed
change in catch mortalityo

While this would appraise the random error, it might not detect
even a large amount of error consistently above or below the true value
for the entire series o Such an error might easily be introduced by
simplifying assumptions or by failure to take into account certain sources
of consistent bias or selsctiono It could be detected only by comput-
ing the same statistics by an entirely different method and from a dis-
tinct source of data. The tagging technique affords such a method and
'atilizes a considerably different set of original data, although, to
the extent that the same commercial catch is involved, errors in fisher=
men's sampling majr be in the same direction as in the other methodo

3/ Persons familiar with the fishery might say that such instances are
common rather than r».r6o It is true that there often seems to be a race
between two or more boats to set their seines around the same school.
This may actually take place at times, but more often a boat, preparing
to make a set, betrays its intention by its actions, and so leads other
boats to come to the same spoto They do sOj, not necessarily in the hope
of beating the original boat to the school it has sighted^, but in the
hope, often realised, that other schools are to be found in the same vicin-
ityo All such instances should be classified as cooperational rather than
competitive. TTnder the latter category, there should be included only the
cases where two boats independently locate and try to set upon the same
school. Such cases oi'' true competition probably are not more numerous
than would serve to offset the contrary element of cooperation.

TAGGING RESEARCH

Tagging has been extensively practiced by the California State
Fisheries Laboratory, and the Fisheries Research Board of Canadaj also
to a lesser extent, by the Oregon Fish Commission and the Department
of Fisheries of Washington,, Having been an interested spectator rather
than a direct participant, my discussion may siiffer from lack of famil=
iarity with this technique, which is simple and direct in principle
but difficult and complex in its applicationo

SOURCE DATA.

Tag fisho - The tagging operation itself needs little discussion
but various aspects of the process require attention in subsequent adjust-
ments of the datao It may be remarked here, however, that the tags are
of metal and inserted into the body of the fish following in general
the method first developed by Rounsefell and Dahlgren (1933) « They are
recovered by electro-magnets installed in the fish-meal line of the reduc=
tion plants o The fish to he tagged are obtained, for the most part, from
commercial fishermen and are subject to the same selections that influence
the raw materials used in the vital statistics methodo Since the numbers
of "samples'* tagged are fewer, the danger of non-random selectivity is
greater©

Tag mortalityo - It is known that tagged sardines suffer high mor-
tal! tjrTr'^~"the~operati on itself or from the attendant handlingo The
only sources of data on the magnitude and variability of tagging mortality
ares (l) A limited nianber of experiments by the California State Fisheries
Laboratory in which tagged fish were held in live cars for direct cbser-
vation, and (2) a correlation establishing the relation between length
of time of confinement before tagging and the subsequent returns o

The live-car experiments proved that mortalitj'- is high and variable,
that small sardines siiffer greater mortality than large ones, and that
the size of the tag affects the rate of mortalityo It is, of course,
difficult to perform a sufficient number of such experiments to establish
a reliable mean mortalityo Moreover^ there is uncertaintj' as to whether
the difference in treatment received when the fish are released in the
live-car rather than directly into the sea (as in the regular tagging
operations) has caused the live-car mortality to differ from that in the
sea releases. This uncertainty is especially acute with respect to libera-
tions in the Pacific Northwest, where tags have been applied with a dif=
ferent instrxjment (tagging-gum) from the scalpel and forceps used in
California and where the fish were tagged directly from the fishermen's
seine instead of being brailed into a holding net for tagging; and, also,
where the roughness of the sea generally imposed more difficult condi-=
tions for the handling than in Californiao

Recover tags. - Since sardines are handled in bulk and mostly by
machine, few tags would come to light without a specific recovery system^
That now in use consists of magnets installed in the meal-lines at can-
neries and reduction plants o Difficulties arise in achieving installation

17

of magnets in all plants, and in connection with their effectiveness when
installed in different mechanisms and under different operational methods «

Test magnets o => This requires that each installation be tested to
determine its recovery efficiencyo Periodic tests of the same installations
ars made to discover fluctuations in performance and detect any changes
that may occur as plant operations are alteredo The test consists of
mixing a definite number of tagged dead fish with an ordinary lot destined
to go through the reduction system and recording the percentage of tags
recovered from the magneto A possibility that the efficiency of recovery
of test-tags may differ from that of tags in sea-released fish arises
from the fact that the former lie within the body, unattached to the
tissues, and not far from the open, insertion incision; whereas the iat=-
ter are encysted by tissue and the incision healed shuto It is likely,
therefore, that the efficiency may be somewhat over-rated by the testSo
On the other hand, a few recovery experiments indicate a lessened effi--
ciency of recovery from fish destined for canning than from those des"
tined to go directly to the reduction machinerj^c This is probably due
to the loss of tags from the body cavity in the cutting and eviscerating
process. The performance of testing experiments in connection with canning
operations has met with obstacles preventing the accimulation of suffi-
cient experience to gauge the reliability of efficiency ratingso Their
value for adjustments is also lessened by the custom of mixing whole fish
with cutting offal in the ordinary operating procedure o In all, the ef-
ficiency of recover^'- varies widely and while important for our calculations
is difficult to determine o

Analysis

Adjust for tagging mortality o =■ Two basiis adjustments are made to
deduce the actual number of viable tagged fish releasedo The first util-
izes the live-car experience in adjusting release records downward by
the indicated amount of mortality induced by the tagging process, with
due regard to the size of the sardines o The second takes into account
the additional mortality brought about by the increased time of holding
to which the late-tagged members of a given batch of fish were subjectedo
It involves establishing a regression of returns on serial order of tagging
in blocks of 100 or other suitable number of individuals for each batch
or group of batches of fish taggedo These regression values adjust not
only for the time -connected differential mortality within a batch but
also for the differences in mortality between small batches tagged in a
short time and large batches requiring more time for tagging^

These adjustments have been employed in the analysis of retur^ns from
California-tagged fish, but no comparable methods are available for these
tagged in the Pacific Northwest, though it would appear that the second
adjustment at least would be feasible for releases in that areao

Adjust for recovery efficiency. - This adjustment applies the magnet
efficiency records and also incorporates a calculation taking into account
any amounts of fish run through plants when or where magnets were not in
operationo To some extent a differential adjustment for reooveries from
reduction of whole fish and reduction of canning offal is feasible from
a comparison of returns from the two categories wherever plant records per-
mit segregationo Since canning is not practiced in T/ashington and Oregon,
this feature is not a problem there, but it exists in British Columbia as
■well as in California.

Adjust for intensity of fishing. - One essential statistic sought
through the tagging technique is the catch-mortali tvo If it were possible
to release all the tagged fish immediately prior to the fishing season
and if they were immediately distributed at random through the commercially-
fished population, this statistic would be the ratio of first-season re-
turns to the number tagged. Another essential statistic, total stock,
could then be computed simply by the proportionality?

number of fish tagged = nximber of returns
' toTal stocTE totaT~catch

but tagging can be done only at intervals during the fishing season and
the tagged fish probably diffuse only gradually through the general stocko
Hence, first season's returns are almost useless and it is necessary to
deduce from subseauent years' returns what the first year's returns would
have been under the simple conditions described aboveo That deduction
is made by extrapolating back to the first year a line representing the
ar-nual rate of decline of tag returns. Since the annual returns depend
in part on the amount of fishing done, some adjustment must be made when
the amount of fishing changes during the series of years included in the
tagging experimento

The adjustment used for the California statistics has been to 'oom-
pute for each season the number of returns per unit number of fish caughto
An identical adjustment would be feasible also for the Pacific Northwest
tagging returns. While this adjustment may suffice for useful approxima=
tions, it obviously gives identical treatment to a fluctuation in catch
whether due to a change in the amount of fishing or in abundance of the
sardines. Yet these two phenomena have different effects on the returns.
Furthermore, the fishing in a current year^ by removing a certain number
of tagged fish, influences the returns of subseouent years. Except
therefore, in situations where it may be safely assumed that the fluctua-
tions in catch have arisen only from moderate and random changes either
in intensity of fishing or in abundance, this adjustment may requir® re^-
vision. Where a trend exists in fishing intensity or in abundance there
would be particular likelihood of erroneous results. Such cases would
require additional adjustments appropriate to the particular circumstances
and possibly materials from the vital statistics methods could be drawn
upon for such analyses o This has not been indicated in the diagrsimo

}

statistical analysis » <- As explained in the previous section, the
statistical analysis of tagging experiments depends mainly on a line
'^or .curve) representing the trend in annual adjusted returnso The inter-
cept of the line at year one (an extrapolation) is used to compute catch
mortalityo By another computation^ it also results in estimate of total
jommercial stock. The rate of decrease in annual returns is taken as
the total annual mortality. The remaining statistic, natural mortality,
is derived algebraicly from catch mortality and total mortalityo

Since the trend curve is based on returns during a series of years,

the statistics resulting therefrom, represent the conditions, not in any

one year^ but are an average (not necessarily an arithmetic mean) for
the series of years.

It has already been pointed out that the adjustments, particularly
that involving the size of the catch, may have affected the data in a way
that calls for modification of the above -de scribed computations. Instead
of that, determination for a nxxmber of periods of overlapping years might
point the way to an interpretation or correction of the statistics for
individual periods or even individual years. Thus far, however, records
are available for only two periods and these possibilities cannot yet be
examined.

A further aspect of the computations so far made is that when Pacific
Northwest catch and Pacifio Northwest returns of fish released in California
are excluded different rates are indicated than when they are included.
This points either to the non-availability to the California fishery of
at least a portion of the stock that migrates seasonally to northern waters
or to some discrepancy of experiment or analysis so far unrecognized.

Fortunately, there is reason to believe that many of the difficulties
of interpreting tagging restiits may disappear or be resolvable as more
data accumulate.

The value of the tagging method as a largely independent method of
dstermining mortalities cannot be over=emphasized and it is particularly
encouraging that the computations have led to determinations practically
identical with those resulting from a preliminary application of the
method of vital statisticso

At the same time it should be emphasized that both the tagging,
and the vital statistics^ methods depend, in the last analysis, on the
representative nature of the sample dealt iwi the At its present stage
of development, the method of vital statistics includes a much larger
sample, better distributed in time and space and more thorbughly adjusted
to exclude the effects of extrinsic influences than does the tagging
me thod u

RECRUnMENT RESEARCH

Determination of the amount of recruitment would be an outstanding
achievement, but would be, of itself, of only limited usefulness so long

as nothing is known as to which of two major influences determine its
vaiueo These influences are those i (l) Not connected (let us say^, ex-
trinsic influences) J and (2) connected (ioeo, intrinsic influences)
with the size of the sardine population itself o The first category would
include hydrographic and oceanographic conditions that may influence
the amount of spawning and the survival of youngo It would also era--
brace competition with, or predation byj other marine animals o The
second category would include the direct influence of population numbers
on the amount of spawn produced and the competitive effects of popula-
tion numbers on the number surviving. The amount of spawn would be
directly, and the amount of survival inversely, proportional to popu=
lation numbers „ The survival rates might be different for the various
stages of youngo

So far as intrinsic influences are concernedj, one would expect re =
cruitment to be small even at low levels of fishing intensity where there
is a very large spawning population and also a very crowded condition
which could impose a high mortal! tyo At intermediate levels of fishing
intensity there would be maximal recruitment resulting from a still large
spawning population and the low mortality associated with an unorowded
conditiono At extremely high levels of fishing intensity there should
again be low recruitment due to a very small population producing insuf-
ficient spawn for maximal recruitment even in a very much thinned-cjit
condition where competition within the population would result in neg-
ligible mortalityo

If the intrinsic influences were operating alone,, it would be e
simple matter to construct a curve of recruitment according to levels
of fishing intensity by merely observing what happens to recruitment
over a range of fishing intensities. But the extrinsic influences,
operating simultaneously also affect the recruitment o For instance,
preliminary analysis of vital statistics for two periods of the sardine
fishery showed that quadrupling of fishing intensity was acoompaaiod
by doubling of recr-uitmento The earlier period with the low intensity
of fishing covered 8 seasons^ and the later one of high intensity, 4o
Both periods were thus of sufficient duration to have had a variety of
extrinsic influences and their effects on mean recruitment should to some
extent ''average out," Yet one would be bold to conclude that they had
indeed averaged out and that the increase in recruitment was in fact
due to the increased intensity of fishing. On the contrary, one could
almost as easily argue that through the later period^, there had been
4 years of favorable extrinsic conditions that were responsible for
the increase.

Yet a choice must be made between the alternatives. If the increased
recruitment were in fact due to the thinning oxit of the population by
intensive fishings then this high rate of utilization^ with the attendant
large annual catches could go on with no ill conseouences other than a
not intolerable reduction in average size of fish. But if it were due
to a lucky run of good survival years^ the present rate of fishing could

SI

not go on without serious diminution of the population, resulting in both
markedly lower annual catches and a marked and perhaps intolerable de=
crease in average size of fishT

At the moment there is no basis for making the choice, but the pro-
gram charted in the right-hand portion of the diagram, facing page 1
which is now to be discussed, is designed to distinguish between the
effects of the extrinsic and intrinsic influences »

Source Data

Egg and larvae sampleso - By a plankton=coliecting program designed
to sample the egg and larval pilchard population in waters of the spawning
region, it is planned to obtain material for two sets of data on each sea=
song 1) The numbers of eggs spawned and 2) the survival of the larvae to
the post-planktonic stage o

To serve these purposes, quantitative collections of eggs and larva®
must be madeo Features of quantitative technique so far employed in this
branch of the program are use of; The oblique method of tmiring to sampi*
all egg-and-larva-bearing stratai current meters in the nets to measure
the quantity of water strained in each haulj regular pattern of stations?
and regular periodicity in making oollectiono Additionally, there have
been special collections or special methods employed in the regular collec=
tions to test the reliability of the tow-net method for sampling eggs and
larvae o These special inquiries have convinced us that quantitative work
is feasible, if also difficulto Some of the techniques still have to be
improve d»

Hydro graphic and meteorological da tap = Observations on hydrographic
conditions taken simultaneously with the collection of eggs and larvaej,
are intended to provide a record of conditions to be correlated with the
time, place, and volume of spawning and with the survival of larvae « It
is already known that the dominant features of oceanic circulation along
the west coast, and particularly the maintenance of fertility of the wa=
ters through the upwelling process, are connected with the winds. With
the establishment of the relationships of the oceanographic features to
the meteorological influences, on the one hand, and to the amount of spawn=
ing or to the survival of larvae, on the other, it may be possible to es=
tablish directly the relationship of amount of spawning or survival of
larvae, or both, to the meteorological conditions » Hence simultaneous
observations on hydrography, meteorology, and the young stages of sardines
promise to elucidate the influences on recruitmento

The particular oceanographic observations so far programmed, include?
Those on temperature and salinity down to 500 meters regularly^ and pilot
observations to greater depths j determination of oxygen and phosphate
content for the same stratai and counts of the diatom population for each
tem=meter level to 60 meters in deptho All of these observations and

22

their subsequent analysis are being undertaken by the Scripps Oceano-
graphic Institution in cooperation with the Fish and Wildlife Service.
Acvjessory experimental work on rates of biological processes in the
survey area is also under way at that Institution,

Analysis

Compute total numbers of each stage, - With reliable quantitative
collections of eggs and larvae at each station, it is further necessary
to integrate the total numbers over both time and spaceo The time ele-
ment involves proper weighting to account for the time interval between
successive observations over the station pattern. This in turn involves
certain assumptions, or preferably a determination of a curve of the
volume of spawning as a time function. The integration over space de-
pends upon a curve of distribution of the organisms in the sea area
sampled. Both phases of integration are still in the developmental
stageso An additional element in the computation is an adjustment
necessitated by the smaller catches of larvae by day than by nighto

For the time being, work will be directed toward determining total
n'ombers of each stage in the particular area included within the egg
and larval survey. This covers only a part of the spawning region.
Extension to other areas must await either better boat facilities or
a fortunate discovery of relationships with meteorological conditions
which may provide a certain reliability in extrapolation.

In any even, the integration must be proceeded or followed by
certain adjustments.

Adjust for drift, - Since the survey area is of limited extent and
contiguous to other possible areas of spawning, some allowance must be
made for gains in numbers by organisms drifting into the area and losses
from their drifting out. This aspect is not serious with respect to eggs,
which hatch iii three days and^ therefore, cannot drift farj but it is
important in the case of larvae which may spend weeks or even months in
the drifting phase© With the pattern of circulation determined from
hydrographic observations, an allowance should be feasible. The prin-
cipal difficulty now foreseen is the amount or rate of lateral diffusion,
which would not be apparent from the circulation pattern alone. This
adjustment is therefore still in the problem stage.

Adjust for rate of developmento - The numbers of eggs or larvae found
in each stage will vary inversely with the time occupied in passing through
the stage. One occupying a short interval of time will afford a smaller
accumulation of individuals than one occupying a longer period. Adjusting
nianbers to allow for the '^accumulation effect" involves knowledge of the
rate of development or of growth. This has already been worked out for
the eggs, but the rate of growth of larvae has yet to be determined,

28

Infant survivalo - Having fully adjusted the data on numbers at each
stage of infant survival, its rate should be quite simply described by
a frequency distribution of the successive stages. The simplest result
to be expected would be a J=shaped ciirve transformable into a straight
line by suitable mathematical procedureo Differences from one season
to another in the slope of this line should register the changes in survival
rate. If, however, bhere are variations in that rate during the larval
existfSncej a more ccmples: curve would result and its interpretation would
be more difficulto

Amount of spa-wningo -■ There are two approaches to the measurement
of amount of'~s"pimirng~through i) Computations from data produced by the
vital-statistics method that would effectively enumerate the spawning
stocks and 2) sampling the sea water for eggs. Both have serious obstacleso
The approach through spawning stock involves the appraisal of fecundity
of females by sizes or ages^ an appraisal rendered difficult for lack
of samples of spaTmers which resort largely to grounds farther offshore
than the area in which commercial fishing takes placeo4/ Therefore,
this approach has been indicated by a broken line in tTie diagram 'facing
page lo The approach through sampling for eggs is far simpler in principle,
and in terms of the diagram would be directly from the previous computa=
tion of total numbers. Howeverj, it involves surveying larger areas than
has so far been feasible with the single vessel available for this research.
Satisfactory enijmeration by either method depends on substantial augmenta-
tion of the sea work.

Correlations with hydrographic and meteorological conditionso - So
far as recruitment'Ys~ determined by the amount of spawning and modified
only by the rate of larval survival, determination of the former and cor-
relation of the latter with hydrographic and meteorological conditions
would give the relationships necessary to the interpretation and predic-
tion of changes in recruitment rate. This correlation must relate to
the conditions in the one area of survey, but if it were found very exact
and involved hydrographic or meteorological features of wide-spread na-
ture, the results would possibly apply to the entire range of species.
It is more likely, however, that ttie more complex and extended procedure
discussed in the following paragraphs will be necessary.

Multiple corr€<lation to distinguish effects of population density
from o ce anogra phi c irf laences . - Thus far a rather simplified view has
been taken - one that stresses the laxn^al stages as the only factor crit-
ically modifying the relationship between amount of spawning and recruit-
mento A more comprehensive viewpoint would recognize that there are other
influences and that they may be exerted at any stage of life. To illustrate
this, the recrutiment diagram facing page 25 is given. Instead of a pro-
cedural diagram as in the one facing page 1, this is one of mathematical
equivalents and correlations.

~^ Frances N. Clark, 1934, has studied fecundity in material taken from
the commercial catch. Her work proved the paucity of spawning-ripe
individuals in the catch and also the difficulties in determining the
numbers of eggs spavmed per female per season.

24

Structiire of the recrtiitment diagramo - The diagram reads from bottom
to top, beginning with the production~of~eggs within the female and iead=
ing through successive stages to net recruitment at the topo At the left
are variables pertaining to population densities (intrinsic inflaences)^
and in the middle are the survival rates (dependent variables) upon which
the two categories of influences (independent variables) operate to affect
the recruitmento Algebraic notations relate equivalents o Thus the frae-=
tion "number of eggs hatched** over "number of eggs spawned" equals the
"egg survival rate", and the ''number of eggs hatched" also equals the "den=
sity of larvae", at least initiallye Correlation notations are given as
arrows pointing from tha independent to the dependent Tariabie with a sign
to indicate whether the correlation should be negative or positive.? and
where inde terminates, ioe,, depending on the way the independent variable
is expressed;, the plus=minus sign (=) is usedo

Discussion of the recraitment diagram,, - It is not intended that the
recruTTment diagram should be an outline of precedure. Rather it indicates
a complex of relationships too extensive and inclusive of too many subjects
■apon which data are, for the most partj non-sxistent at present and diffi =
cult to obtain in the fut'ureo It illustrates tha manifold phases of the
recruitment problem and should serve as a reminder of the fragmentary nature
of less comprehensive treatanentSo

This commentj however^ should not discourage effort to solve the prob-
lem of recruitment by some less comprehensive or abbreviated process,, If
the survival rate in certain stages of life is relatively constant from
year to year these stages can, without risk, be ignored » If this be done
and high correlation is found, much time will have been saved. If, on the
other handj the results turn out to be indeterminate, it would merely mean
that the stages that were ignored did, infact, have variability and should
have been consideredo Since that requires data for which there are ccw no
collection facilities, we will merely have to decide between surveying
the significance of available data or neglecting themo It is clear, how^
ever, that a program for determining additional aspects of certain reproduc=
tire attributes of the pilchard population and also of oceanographic con=
ditions, including study of that portion of the marine fauna which may
exert predation on any stage of the pilchard, should be started at the
earliest opportunity.)

Discussion of an abbreviated recruitment analysis » - It is interest-
ing to speculate on how limited a view would have to be taken if one were
confined to data already available or in process of becoming available o
The diagram opposite gives a scheme of correlation analysis on what
appears to be fairly reasonable grounds » It ignores intra-ovarian, and
eggs, mortalityo Such sampling of eggs on the spaTming grounds as has been
made is impressive in the relative constancy of the numbers found and fails
to suggest important variability in survivaio

The scheme includes only the oceanographic extrinsic variable affect=
ing larval survival because it hardly seems that competition between

RtGKUITMElNT

//

Number of juvenilts
reaching
commerci&t 5i£&

Number of larvae
reaching post-
pl&nktonlc stage

Juvenile
survival
rate

Number of larv&e
reaching post-
pUnktonio stage

Number of
eggs hatched

Larval

survival

rate

Density of
larvae

//

Number of
egfg^ hatched

Number of
ejgs spawned

surviVAl
rate

Number of
eggs spawned

Potential spawn-
Ing capacity

intra
ovar
surv
rate

lan
ivai

s

Abundance
and age com-
joosUlon of
commercial
stocK

Density of

commercial-
sized fisfi

J

Certaiin oceanogrdphic
conditions

Planktonic food ?

Abundance
of competing
or predatory
forms

Certain

oce&nogaphic

conditions:

Drift?

Planktonic

food?

Certain

oceanographic

conditions:

Drift?

Temperature?

Turbulence?

-o

c '*
o "^

o^

^t
o -^

— o

on
_o o

O Z

-t

_iJ —
0,0

e

-o

C C

O

Certain oceanogfraphic

conditions:
temperature J
Planktonic food?

26

■a

o

E

<

z

o

<

-»
X
<

o
Z
o

<

-J
<

z.

Z
o

<

V.

"V

^

v^

e

c I
a) ..

I
<1

o

1.
3
O

« I

c
o
«>

id

ol

5^

1

o--;;

id

-2g

-o

c

Q> O

> «-

Q>.C

tf

OO

VJ

27

larvae can be important as they form such a very small fraction of the
plankton communityo Owing to the lack of data, it ignores competition
from other f onus » This is perhaps the greatest weaknesso

It includes only the intrinsic influence on juvenile survival and
considers that the competition takes place between juveniles and post=
juveniles rather than within the juvenile category itself o (This particu=
lar selection has a back-ground of observation too extensive to describe
hereo It was suggested by the virtual disappearance of dominance of year
classes since intensification of sardine fishing took place.,) o T/ith ju"
veniles, alsOj the competition from other forms is ignored for the same
reason that it was among larvae. Howeverj utilization of data on predator
species of commercial fishes invites attention.,

The diagram opposite places the elements on a time~scale hori-
zontally, showing the approximate season and year from which the source
data are drawn and the approximate season and year in which the independent
would be expected to influence the dependent, variable^ recruitment.
Thus recruitment of the current season (n) would be estimated from the
(fully adjusted) sample of the current commercial fishing season. The
numerical value of this recruitment would be correlated withs (1) The
density of the post-juvenile population during the previous season (n-l)
as estimated from the (fully adjusted) sample of the previous commercial
fishing seasons I (2) the strength |or persistence) of northwesterly winds
during the spawning and developcaental portion of the second previous sea-
son (n-2) as estimated from weather records of that portion of the second
previous seasons and (3) the amount of spavining during the second previous
season (n-2) as estimated from the (fully adjusted) sample of the commer-
cial fishing season with, of course, due allowance for the percentage of
fish that are mature and the proportionality between s ize of spawner and
number of eggs spavmed.S/ The curved arrows connect source data with
the equivalent derived Trom ito The straight arrows with plus and minus
signs point from dependent variables to the independent one. It is as-
sumed that correlation between northwest winds and recruitment is positive
through its influence on upwelling and the attendant "fertilizing* of the
sea water. It could be negative (or curvilinear, ioe. positive in some
parts of its strength or persistence range and negative in others) through
its effect on drifting larvae away from the more favorable nursery grounds.
Which of these actually occurs would be revealed by the eorreiation analysis©

The multiple correlation process indicates in the diagram. oppO"
site, three independent variables. If the correlations or regressions
are rectilinear, 6 degrees of freedom would be absorbed by the multiple
correlation process. It appears that data on the two intrinsic independent
variables will be derivable for eight seasons when the portion of the
program now in progress is completed. By that time, however^ two more
seasons will have elapsed and if they can be added, a series of ten seasons
will be available. The meteorological data^, if regular Weather Bureau

"5/ Using approximate determinations published by Frances N. Clark, 1934.

28

observation will suffice, would, of course, be available for the same
years. With ten seasons and the loss of six degrees of freedom there
would be left four degrees of freedom on which to base judgment as to
significance of results. This nvunber is perhaps too low to do nore than
indicate whether the general hypothesis is on the right track, 'Certainly
it will not be enough to conclude that the hypothesis is disproved.
(Of course, correlation analysis can only disprove: it can never prove
by hypothesis.) ~~~~'

SUMMARY

As the pilchard program now stands, tv/o lines of evidence when the
work on them is completed, will give the rates of catch and natural mor-
tality, and the rates of recruitment over an 8-season period, 1932-3
to 1939-40, with the prospect of adding two subsequent seasons. By ap-
propriate computat' on, estimates of the total size of the population
and its size-composition will be available on an annual basis from the
"vital statistics** approach, and also available as a mean for a group
of seasons through tagging studies. This will afford estimates of the
effects of fishing at different levels of intensity on the quantity and
quality of the catch. The estimates will be true only for the seasons
covered by the analysis. To extrapolate them so as to predict what con-
sequences will follow any particular level of fishing intensity, a third
line of evidence is needed for determining v.'hether the recruitment has
been conditioned by the size of the stock itself (and hence predictable
from the "vital statistics" evidence) or v/hether it was affected to an
important degree by oceanographic conditions (and therefore unpredictable
except in teni;s of range of variations about a certain mean condition).
This matter is being investigated. Great difficulties have been met
and still others may be anticipated, but in spite of these possibilities
of solution exist.

An important phase of the entire problem is whether the range of
intensities of fishing, which happened to have been included in the 8
or 10 seasons under study, is sufficient to cover reasonably well the
range to be anticipated in the future. The range could be extended to
lower intensities by going back to the records of still earlier periods
of fishing. To extend it to higher ranges would depend on developments
in the fishery. To permit development to higher ranges of intensity
could involve the risk of reducing the stock so far below a desirable
level that recovery might be slow. Appraisal of the degree of risk in-
volved and of the desirability of undergoing that risk will depend on
the nature of the results that flow from the current program, and may
well be deferred until they become known.

LITERATURE CITED

Baranov, F. I. 1918, Nauchnyi issledovatelskii iktiologisheskii
institut. On the question of the biological basis
of fisheries, Izvestia I (l), pp. 81-128, 12 figs,

Clark, Frances No 1934 Maturity- of the California sardine (Sardina

caerulea), determined by ova diameter measurements .
California Division of Fish and Game. Fish Bulletin
Wo. 42, 49 pafres, 19 figs.

1940. The application of sardine life-history to the
industry. California Fish and Game, Vol. 26, Mo. 1,
pp. 39 - 48 1 1 table, 2 figs.

Hodgson, W. C

Ricker, W. E.

1939. An improved method of sampling herring shoals.
Cons. Perm. Int. pour I'Expl. de la Mer, Rapports et
Proces - Verbaux des Reunions. Vol. CX, pp. 31-38,

5 figs ., 4 tables.

1940. Relation of "Catch per unit effort" to abundance
and rate of exploitation. Jour. Fish. Res. Bd. of
Canada, Vol, 5, No. 1, pp. 43 - TO, 4 tables, 3 figs.

Rousefell, Goarge A., and Edwin H. Dahlgren.

1933. Tagging experiments on the Pacific Herring,

Jour, du Cons., Cons. Perm. Int. Po\ir L'Expl. de la

Mer, Vol. VIII, No. 3, pp. 371 - 384, 6 figs. Copenhague.

Thompson, W. Fi

1926. Errors in the method of sampling used in the
study of the California Sardine, in The California
sardine. California Div. of Fish'Tnd Game. Fish
Bull. No. 11, pp. 159 - 189, 13 figs.

Thompson, W. F., and F. H. Bell.

1934. Biological statistics of the Pacific halibut
fishery. Int. Fish. Conim., Report No. 8, 49 pp«,
18 figs., 15 tables.

30

£o Determination of the Age of Juveniles by Scales and Otoliths

i7

Lionel Ao ?falford and Kenneth Ho Mosher

CONTENTS
Preface Page

The Pr'OOXelU oooooooooooo«eoe9«ooooe 32»

Collection of Material ..oooo.o.o.oo.oo 34

Age and Growth of Juvenile Pilchards as Judged from

Length Frequency Curves » ooooo.o.o.o.oo 35

Age and Growth of Juvenile Pilchards as Judged from

Scales and Otoliths. ao<i<>>a«»oo«<>o<>o<> 38

The Use of Scales for Determining Ageo . . » » o o « 38

The Use of Otoliths for Determining Ageo . ., « o o <> 45

Growth of Juvenile Pilchards ..0000,00000 46

Difference in Size Between Localities. . » « o o o o 48

Discussion and Conclusion .„ ,oooc.o48

?„/ Now Chief J, Branch of Fishery Biology, Fish and Wildlife Service;,
Washingtonj, D, C»

'/ Now Fishery Research Biologist, Alaska Fishery Investigations^
"° Seattle,, Washington

THE PR03LELI

Up to the present, no method has been developed for determining
the age of individual specimens of the Pacific pilchard, or sardine.
Lacking a satisfactory substitute, it has thus not been possible to
isolate year classes of fish taken in samples of the commercial catch,
so as to trace, by serial sampling, their growth in length, or to fol-
low their passage through the fishery.

Such available knov;ledge as bears on the growth and age of pilchards
results from the occurrence of dominant year classes in the population
from time to time, which can be readily recognized by their prominence
in systematically collected samples of the commercial catch; By this
method it has been shown (Clark, 1936, pp. 18-2I|.) that year classes
have entered the fishery when their mode i/as between 190 and 200 ram. ,
that their approximate growth could be followed through as many as five
years, when their mode was in the neighborhood of 23P mm., and c"buld
be traced virith somewhat less certainty, if they persisted in the fishery,
for as many as five additional years, when their mode was near 260 mm.
Thereafter, according to Clark's work, the hitherto dominant groups have
completely lost their identity.

This statistical method is useful only for studying those year
classes abundant enough to produce persistently prominent modes in the length
frequency curves. However, the method does not permit determining whether
a dominant group is composed of one or of several year classes, nor can
it be used for studying individually or collectively the more numerous,
adoLiinant age classes. Furthermore, as Thompson (1926, p. 18?) says,
"....our vie\T of the course of a given year-class is probably much dis-
torted by the conditions of selection and by the errors of sampling."

Neglecting, for the purpose of this study, the problem of sampling,
an ideal method of analyzing frequency curves would expose and recognize
individual year classes from the youngest to the oldest, and of all
degrees of abimdance. To be accomplished this necessitates age deter-
mination of individual fish.

Thg desirability of such a method was emphasized by Thompson (op.
cit. p. S2) who says, "There are two ways in which the abundance of
fish may be affected; first by the variations in mortality rate at
various stages of life preceding capture, and second by the direct
effect pf the enrivonment upon the movements or habits of the fish.
In the first of these the analysis for size or age must be of supreme
utility, being the sole means of determining the relative proportion
each age forms of the catch..."

Since the ages of individual fish are determined usually from scales
or otoliths, Thompson made a preliminary examination of pilchard scales
in the course of his investigation on that species (Thompson, op. cit).
He used for this purpose a collection of 19h specimens, ranging from lUU
to 228 mm. in length, taken from October 28 to December 2. These were

32

then read for age, v;ithout knowledge of the size of fish, and v/ith fre-
quent changes in magnification to avoid being influenced by the size
of the scales. It was found that 20 percent of the scales were illegible,
and 5I4 percent of the read'ings questionable. In a subsequent examina-
tion, with reference to the size of the fish, 61 percent of the readings
were questionable, and 25 percent of the scales illegible. A comparison'
of the two readings indicated that, what vdth disagreements and illegi-
bility, only about 25 percent of the total age determinations were com-
pletely acceptable, \fhen the length of the fish at each age was calcu-
lated from the scales by Lea's method (Lea, 1913), Thompson found such
discrepancies between the actual lengths of fish at the several estimated
ages and the lengths at those ages as calculated from older specimens
as further to discredit the dependability of reading pilchard scales.
He says, then (Thompson, 1925, p. 53), "So important is the analysis
by age regarded at present that it is with the greatest regret that it
is temporarily omitted. Vfe — the dii-ector and the successive assistants —
have vainly attempted to read the age marks on the sca.les and otoliths.
This does not mean that the attempt to utilize thera will be abandoned. .
... we may be able to develop a method of accurately reading the scales
according to the age their lengths should indicate. This attempt is al-
ready in progress." And, on p. 57, 'HVith this simple presentation of
results the use of scale readings is concluded for the time being."

No further studies were carried to completion following Dr. Thompson's
preliminary report, however, and no other publications have appeared
to date on this subject. Nevertheless, the possibility of determining
the age of pilchards from scgles and otoliths was not closed, and the
belief was expressed from time to time by those studying the species
that the subject must be further studied.

Accordingly, by agreement vath the California Division of Fish and
Game, the staff of the United States Fish and v;ildlife Service engaged
on pilchard research took as one phase of its program a study of the
problem.!/ This paper, the first report on the resiJ.ts of that study,
inquires critically into the utilization of scales and otoliths for de-
termining ages of pre-adult sizes of pilchards; that is, those ranging
from the smallest taken by bait fishermen up to those around 185 wra.
in length, which is approximately the size at which 50 percent of the
females become mature (Clark, 193U). In addition, somewhat less critical
results are given for fish up to around 220 mm., at which size 100 per-
cent of the females are mature.

i'This study has been made possible only by the cooperation of
members of the fis.hing industry, who aided in our collection of material.
The authors are grateful to the following fishermen or dealers, who
took an active interest in securing samples of the bait fisheries:
Messrs. J. L. Sullivan, Eugene Sullivan, Dominic San Phillipo, and
Leonard Schipper at San Dlegoj Joe Dixon and George I^son at Newport
Beach} S. Carmen and Frank Pugleise at San Pedro; lH. Martinelli at
San Francisco} and Thomas A. I^lartin, Jr., at Seattle, Washington. The

38

Collection of I^terlal

So as to determine the characteristics of an age mark and to observe
the time Vichen it forms, scales of young fish of known age, tl^at is to
say, young of the year, were studied during an l8-raonth period beginning
with March, 1938. l.fe.terial v^ras obtained by sampling bait fisheries in
southern California, which is the center of this industry, and which is
probably the region of raaxijTial spavming (Scofield, 193U). '.'eekly samples
at San Diego, Newport Beach, and San Pedro vrere taken wherever they were
available, i.e., in the fresh fish markets, from fishermen with whom
special arrangements to save young pilchards had been made, from can-
neries, from bait receivers, or from the live-bait tanks of commercial
pleasure boats v Enough fish v;ere taken to total, for all localities,
about 1500 specimens each week, and to average about !?00 from each lo-
cality. This vfork was carried on regularly from Ivlarch, 1938, to May,
1939. Lleanwhile, to have a representation from other parts of the pil-
chard range, fish were obtained elsev/here by various means.

Beginning November, 1938, and extending thi^ough October, 1939,
weekly samples vrere taken of the bait fisheries at San Francisco and
Monterey. In addition, several samples from Mexico were supplied by
tuna boat captains j and in July, 1938, Captain C. W. Thomas of the coast-
giiard cutter, Hermes , invited the senior author to accompany the Hermes
on a voyage to Magdalena Bay, Lovrer California, Virhere, for one week,
daily samples were taken, totaling altogether 1500 specimens.

Kerckhoff Marine Laboratory of the California Institute of Technology
provided laboratory facilities for the southern California station. All
the staff members of the United States Fish and > midlife Service (formerly
Burea.u of Fisheries) engaged on pilchard research have contributed some-
thing toward the progress of this study. Mr. Robert Luckhardt was en-
gaged in sampling the bait fisheries in southern California and San Fran-
ciscoj Ltr. Ralph Silliraan canvassed the possibilities of obtaining young
pilchards in Washington and Oregon. ISr. 0. E. Sette, in charge of the
pilchard investigation, gave advice and criticism throughout the entire
course of the study,

IJr. \\'illiara, Bowen, microscopic technician of Stanford UniveiTsity,
supervised sectioning and staining of the scales. Preparation of the
scales and otoliths for microscopic examination, subsequent clerical work
on the statistical processes, and the preparation of graphs were done
by !/«■. P. A. Projects 702-3-1 and 10917, respectively, Microtechnical
and clerical work was done by N. Y. A. assistants, furnished through
the Stanford Student Employment Office,* and Stanford University has gen-
erously provided the working quarters of the South Pacific Investigation,
giving the staff free access to its libraries and, other useful facilities.

Finally, we acknowledge with gratitude thQ critical reading of the
manuscript by Dr. 1". F. Thompson of the International Pacific Salmon
Fisheries Commission, and Dr. Frances N. Clark of the California State
Fisheries Laboratory,

S4

Meanwhile, since it had been reported that young pilchards are
taken occasionally along the V;ashington coast by fishermen, and appear
from time to time in the stomachs of salmon caught there, one of the
Service's staff, Mr. Ralph Silliman, was stationed at Seattle to secure
any specimens that might be obtainable. By examining a large number
of salmon stomachs and by making special seine hauls, Mr. Stilliraan
obtained a san^sle of young fish at IVestport in April, 1938, and others
at Tokeland, V/illapa Bay, in September and October. (Silliman, unpub-
lished manuscript. )

Altogether, 175 samples were taken, totaling 71,800 fish. For
measuring, the fish were laid on a board especially designed to obviate
numerical bias on the part of the operator, (Sette, 19U1) and the body
length (i.e., tip of head to end of fleshy part of caudal peduncle)
read to the nearest millimeter. For each sample, scales were removed
from the first $0 to 75 specimens having any suitable ones still adher-
ing. Ordinarily they were taken only in an area of about 1 1/2 centi-
meters square centered by the tip of the pectoral fin, or, in a few
exceptions, as close to that area as scales were available. They were
removed with forceps, dipped in water, wiped with the fingers Y»'hen still
fresh to remove adhering slime and tissue, and preserved in envelopes .
Otoliths were removed from most of the specimens from which scale samples
were taken, 'fhese were washed in water and preserved dry in envelopes.
Scales were taken from 11,500 fish; otoliths from over 10,000. Not all
these were read; only enough to secure reasonably adequate representation
from each month,

AGE AND GR01,TH OF JUVENILE PILCHARDS AS
JUDGED FROM LENGTH FREQUENCY CURVES

Because the first two year classes are each usually recognizable
from older fish by their distinctive size, length frequency curves are
useful for identifying the age of young fish up to two years. They there-
fore provide a starting point for associating age with number of rings
on the scales and otoliths.

Under simple conditions such an extensive sampling as was carried
on should provide a series of frequency curves truly representing the
total population of young pilchards in the sea. Unfortunately, for
various reasons, conditions are not simple. Judging from the protracted
spawning season (Clark-, 193U) and the extensive spawning range (Scofield,
193bj Silliman, unpublished manuscript), the year's brood along the coast
probably consists of a nvmiber of groups of fish, each representing an in-
dividual wave of spavming in a particular locality. Since pilchards,
like other species v^rith similar habits, appear to school more or less
according to size, it is conceivable that, so long as the several groups
differ in size, each will travel more or less independently, entering
and leaving the field of the bait fishery at irregular intervals, thus
offering the fishermen only a limited availability. This effect, must
be enhanced by the fact that bait fisheries do not cover the entire ground

35

traversed or occupied by pilchards — there are no such fisheries of im-
portance north of San Francisco, or betvfeen tlonterey and Los Angeles
Counties, or between Turtle Bay and Magdalena Bay, nor do the bait boats
operate farther than about l5 miles offshore. In addition, the method
of fishing is such that only those fish schooling near the surface are
caught, while those schooling deeper are missed. Ivloreover, the fisher-
men tend to seek certain sizes; in some places the smaller fish, in
others the larger, depending on local demand. For these reasons only
a highly and variably selected part of the population could be available
to us.

The effect of this selectivity might be partly eliminated by weight-
ing the samples according to the catch, or according to relative abun-
dance at the several localitiesf but statistics bearing on the bait fish-
ery or on regional abundance are not available, and such weighting is
therefore not at present practicable. Because the samples were of vari-
ous sizes they were all weighted equally (to 1000 fish); likev/ise, the
monthly totals of the vreighted samples at the several ports (to 1000 fish);
and the totals of all California ports thus weighted vrere summed by months.
The original data were divided into two parts, as indicated by vertical
lines in figure 1, the one comprising what was judged to be approximately
all the youngest year class, the other the older ones. Each part v/as
weighted separately, to 1000 fish, thus emphasizing the height of the
several modes vathout regard to their relative abundance.

Judging from figure 1, a nevf complex of groups of fish first ap-
peared in appreciable numbers in the California samples in June, in 1938,
ranging in body length from about iiO to 95 mm., and having several modes,
the dominant one at 70 mm. Save for a few specimens in Ivlay, fish of that
size-range were not available earlier, in spite of every effort to obtain
them, and were not available again through May of the following year.
A fairlj'- similar group appeared in the bait fisher^' of San Pedro in 1922
(fig. 2; Higgins, unpublished manuscript), showing that such occurrence
is probably normal. It is probable that this group represents the young-
est year class. This conclusion is based on the following considerations:
In the most sinistral group of modes for June, 1938, in figure 1, there
is a range of about 55 mm. This leaves only UO mm. to the left in which
to include a possible younger year class. It wovild be extremely unusual
for the zero year class to have a narrower range than the I's; or to be
less widely separated from the I's than the I's are from the II' sj or
for the first year of growth to be less than the second. 3ut, neverthe-
less, even if there were an additional year class, then the fish in the
above mentioned UO-mm, range should have grown almost, or quite past
the hO-mm. point by the following Ii/lay. Before that time, they should
have been taken by the nets of the bait fishery, which, judging from
figure 1, could hold fish as small as UO mm. That no such fish were
taken in our samples is strong evidence that they did not exist in the
waters of the area studied. It seems beyond question, therefore, that
the most sinistral group of modes for June in figure 1 does represent
fish produced in the spawning season of 1938.

36

Subsequent grov/th can be traced by the monthly progression in the
position of that group from June, 1938, through Jfey, 1939, when the
dominant mode was near 120 ram. It is evident that the year class was
composed, at first, of several subgroups, the availability of vrhich
fluctuated more or less independently during the year. This accounts
for Irregularities in the progression.

As for the ages of fish older than 0 year class, it may be supposed
that in curves to the right of those belonging to year class 0, the small-
est fish, those around 8U to 100 mra. , in March, 1933, v/ere the smallest
one-year olds, and may be homologous with those ranging from 130 to
about 170 mm., in October, 1933, and to those of an undeterminable range
above 150 mm., in Ilay, 1939. From figure 3 it is sho^m that fish as
small as 1^0-l60 mm. were taken in the f5.ll comnercial fishery for "adults"
at Monterey in October, 1939- If this October, 1939, curve of figure 3
be comparable vrlth the October, 1938, curve of figure 1, it could be con-
cluded that at Monterey in 1939 some pilchards entered the fall commercial
fishery when in their second year. Nothing can be told by inspection
of frequency curves in figures 1 and 3 about, the, ages of the fish larger
than these sizes; they may be all one-year olds, or there may be an ad-
mixture of older fish. A knowledge of the age composition of these fish
is possible only by determining the ages of the individual specimens.

The modes of curves in figure 1 do not coincide in position with
analogous modes indicated for 1921 and 1922 (fig. 2). These differences
may mean that the grovrbh rate of the comparable year classes was dif-
ferent, or that survival T^as different; or it may mean that selection
was different in 1938-39 than in 1921-22. In any event, it appears from
these differences that one cannot generalize, as to size and growth, from
one year class to another, or from the sampling of one year to that of
another.

Young pilchards collected in Lower California in July and August,
1938, from ViTashington in April, September, and October, 1938, and from
Oregon in I;lay, 1939, do not correspond in size to those taken in California
during those months. This indicates that the California bait fishery
does not dravr from the entire juvenile pilchard population of the Pacific
coast. Lacking data for Lower California, ITashington, and Oregon in
other months, little can be judged from these data as to the age of fish
represented in the available frequency curves. Are the fish taken in
Lower California in July and August, 1938, fish in their first year, or
their second? The same question may be asked about fish taken in Vj'ashing-
ton in April, September, and October, I938, or at Coos Bay in Ilay, 1939.
These and other questions bearing on the age composition might be answered
by studjr of the scales and otoliths.

37

AG:: AND GRO'T'I OF JUVSMILE PILCHARDS AS-
JUDGED FROI'! SCALES AND OTOLITHS

The Use of Scales for Determining ^ige

The scales of pilchards are typical of fishes in the fanily Clupeidae.
Like herring scales, they are sciilptured on the upper surface of the
unexposed part with low transverse folds, (Figure 12) which are in most
specimens variously interrupted ^.Tith rather coarse radii; and they are,
with fevf exceptions, alnost entirely unsculpttired on -their exposed part.
Like herring scales, they differ from scales of many other species that
are used for age determination, such as cod, haddock, or salmon, in
that the surface folds — "circuli" — are tjiDically not circular or concen-
tric, but transect the scale from the dorsal to the ventral margins.
Careful study has shown no irregularity in spacing between these folds,
extending clear along their length, that can be related to age; there
is no indication of winter narrowing of spaces, or of spring vifidening;
nor any periodic change in texture that involves entire folds.

Close inspection of an adult pilchard scale, however, reveals other
marks, concentric v^ith the margin, along the line of which the trans-
verse folds are more or less distorted and irregularly spaced. These
are similar to the year marks on the scales of European herring, and
are what Dr. Thompson counted as annuli in his study (1926) reviewed
above, and also what the authors of this paper have considered as such.

If it could be demonstrated that they are really annual marks,
the problem of determining the age of pilchards might then become a
mere matter of counting them. However, they are not distinct in ordinary
preparations, and special mounting is necessary to bring them out.

It Tfas found during the present study that unless a pilchard scale
be perfectly clean it is almost invariably useless fbr age determination.
Because of the tenacity of the dried mucous, blood, guanin, etc., to the
delicate surface structure, dirty pilchard scales cannot be cleaned,
once they have dried, by washing in water and rubbing. Consequently,
the scales must be cleaned at the time they are removed from the fish,
before they are preserved in envelopes. For" dirty, dried scales, a
moderately satisfactory method of cleaning is the folloviingj Soak the
scale until thoroughly moistened in a 10 percent solution of sodium hy-
droxide, then rinse in water, rub vj'ith fingers to remove offal, rinse
in 10 percent acetic acid to neutralize, and then rinse in clean water.

For mounting, many liquid media were tried; glycerine, water, mix-
tures of the two, and various kinds of oils and glues. These substances
all rendered the scale too transparent for use, seriously decreasing the
visibility of the surface structure, an effect found to be more or less
proportional to the refractive index of the mounting medium (see figures
25 and 26).

38

It v;a5 found that pilchard scales appear to excellent advantage
and the surface structure is brought out in bold relief in a medium of
air. Accordingly, the specimens were mounted dry, — 6 or 8 for each fish,-
between two glass slides stuck together at the ends with glue and clamped
together tightly, until the latter dried, so as to flatten the scales
and insure their remaining in place. £/ To mount scales from 100 fish
required about 6 hours .

The specimens thus prepared were examined with the aid of a project-
ing microscope so arranged that the projected image appeared on the table
beside the microscope. It is the impression of the authors, perhaps
purely personal, but probably not, that the year marks appear much more
clearly in the projected image than when viewed directly tlirough the
microscope.

For each specimen a punch card designed for the ready tabulating
and sorting of data was laid on to the image so that a millimeter ruler
printed on the card extended along the mid-longitudinal axis of the
sculptured part of the scale, its zero line coinciding with the base
of the latter. Lines were then drawn through the ruler where the lat-
ter was crossed by "annuli." Subsequently there were recorded on the
card such pertinent data as sample number, length of fish, locality and
date of collection, dimensions of scale and position of marks, etc.
Thus a permanent quantitative record was obtained for every examination.

To learn the characteristics of an annual mark, scales of young
pilchards were examined over the course of a year. A concentric mark
was observed to have formed consistently by early spring. As a result
of these preliminary examinations, vre recognized as annuli those marks
having the following characteristics t

An annulus is concentric with the margin of the scale. It
is not, always a sharp or unbroken line; nor are the segments
of an interrupted annulus alvfays perfectly cocircular (if the
shape of a scale may be called circular in this discussion).
But the course of an annulus, continuous or broken as it may
be, can usually be traced, by careful scrutiny if necessary,
entirely around the sculptured part of the scale from left-
hand to right-hand margins. Sometimes they can even be fol-
lowed around the unsculptured part. Annuli are clearly sepa-
rated from each other and do not ordinarily meet at any point.
If an annulus has formed, it is present in all the normal scales
of an individual. ' ~

Tt'e have recognized as adventitious, or otherwise unperiodic ("false
rings") those marks distinguished by the following characteristics:

-AiThile this paper was in manuscript, a publication by Aikawa (19U0)
was received. It de'^cribes the same mounting technique as that used
in our study.

39

They are merely short or unassociated. arCs; or If- they

completely circle the sculptured part, they are usually not
concentric v,rith the margin; furthermore, they frequently
join an annulus at the base. In any case, they may be less
distinct than an annulus, being vague and indefinite, or
they may be much more distinct, having a scarlike appearance;
that is, with very pronounced irregularities of pattern,
the folds being broken or otherwise obviously abnormal (See
figures 21;, 27). They rarely appear in all the scales of
an individual

For the first annulus, which is sometimes difficult to distinguish
from a scar, especially if it be nearer the center than normal, the
following rule v/as applied: A mark conforming to the above description
of an annulus V\ras counted as an annulus, provided it appeared in all
the scale specimens examined; but if in one or more specimens said mark
had an obviously scarlike appearance, it was considered a false annulus,
and was not counted.

Although a false annulus may appear at any time during the year,
a true annulus is formed only once annually, and during a certain sea-
son. The distance between annulus and the margin of the scale — the
marginal increment — represents the grov/th since the annulus was formed.
For fish in their first year, the marginal increment is the distance
betv/een the base of the sculptured part and the margin of the scale.
If one were choosing unperiodic marks to determine age in samples of
fish, then frequency curves of the marginal increments should show no
progress in width during the year, for there should be no consistency
in the time of appearance or position of the marks. But, on the other
hand, if one were choosing truly annual marks, there should be one time
of year when the marginal increments are narro^;^est, and another time
when they are i\i.dest, with intermediate widths betvreen. This argument
is the basis for the follovj-ing test as to v;hether or not the annulus
is formed once yearly.

A stratified random sample of 3,000 mounted scales was dravm (by
Mosher) so as to represent about equally all the months of the period
studied, and, where possible, to represent equally northern and southern
California material. The labels of the slides v/ere then masked, the
slides mixed, and dra\m at random from a box. The scales were then
examined (by walford), and measurements recorded as described above-
Since the annulus is not apparent at the time it is fonning on the very
edge of the scale, but only sometime afterward v^rhen enough marginal
growth has occurred to set it off clearly, no zero marginal increments
were recorded. Figure h shov/s the marginal increments thus measured
for scales with 0, 1, and 2 rings.

Obviously in the specimens of these year classes examined, the
marginal increments vrere relatively narrow in April, 1938, and increased
in width from then until the folloviring fall, when narrow marginal in-
crements began to appear again. In other words, the marks called an-
nuli in this study must be really annual in occurrence, and may be used
to determine age, at least for the first two years of life.

40

It has been argued (p. 5) from figure 1 that the group of small
fish taken In June, 1938, 'with a dominant mode near 70 ram. represented
fish spavmed in 1938. According to the scale studies, the scales of
these fish were consistently v:ithout annuli. It may therefore be sup-
posed that fish v^ith one and two rings on their scales at that time
of year were of year classes 1937 and 1936, respectively. In November,
one annulus became evident near the margin of some specimens identified
from figure 1 as belonging to year class 1938 • By March, all the mem-
bers of this class had one annulus, while those having tv;o and three
were now identified with year classes 1937 and 1936, respectively.

If viTe choose an arbitrary boundary to separate "narrow" from "wide"
marginal increments, say the 32.5 mm. point in the data used for fig-
ure h, it is possible to define a time when the annulus became evident
during 1938-39. The propor.tion of fish having "narrow" marginal in-
crements, in percentage, by months, is given in figure $. In southern
California, according to figure 5, an annulus became evident, on the
average, in pilchards of year classes 1937 and 1938 early in January,
when 50 percent of the specimens examined showed a "narrow" marginal
increment, and 50 percent a "vride" one. In central California the an-
nulus did not appear until late February for year class 1938, early
March for year class 1937.

This difference between central and southern California held con-
sistently during the months for which comparable data are available
(fig. 6). Since the scales vrere read without knowledge of the catch
localities , this is further evidence that the marks called annuli had
not been formed sporadically.

The annulus, being a very narrow zone, practically a line, became
evident only when enough growth had occurred to expose it. Therefore
it must have formed some time previous to the late fall and winter,
probably in the late summer and fall, when, Judging from figure 5,
growth of the scales is slight.

If the evidence be accepted that the marks counted as annuli are
really year marks, it remains to be shown whether the counts can be
made -virith reasonable consistency. To test this, a stratified random
sample of 2k2 scales that had already been read was drawn (by Mosher)
from among some 3,000 scales, so as to represent with approximate equality
the first four year classes. The labels of the slides were then masked,
and the scales read for the second time (by V/alford). The results,
given in Table I, indicate that one person can count what he defines
as year marks consistently.

To test the similarity of readings by two persons working sepa-
rately, a sample of 367 scales from the fall fishery of 1939-UO that
had been read by VJalford was read independently by Mosher, with no
previous experience at scale reading, and recorded as described on
page 6. The results, given in table II, show a high percentage agree-
ment between the two sets of readings, and indicate that at least tv^o

4l

persons can identify the same marks as annuli, and coimt the same number
of rings in a significantly large proportion of the cases. The results
summarized .in both tables I and II give evidence that a year mark has
been defined on page 8 in simple enough terms to serve as a usable guide.
They do not necessarily indicate the percentage error in the counts of
age rings, or in the relation betv;een the number of rings present Jind
the number of years of life, for it is v/ell-nigh impracticable to obtain
such a measure. For the present study, however, it is likely that this
error is not disproportionately large j otherwise, figure I4 v/ould have
been a chaotic picture.

TABLE I. -Comparison of Original Scale Readings V.'ith Those Ifede
at a Later Time by the Same Person.

Number of
annul i in
second reading

Number of
readings
ing sec

first •
match-
ond

Number of first

readings not
matching second

Percentage

agreement

0

6U

0

100

1

50

U

93

2

70

3

96

3

h9
233

2
9

96

Ta3LE II. -Comparison of Scale Readings by Tvro Different Persons

Number of
annul i in
second reading

Number of first
readings match-
ing second

Number of first

readings not
matching second

Percentage
agreement

1

iiU

7

9U

2

123

lU

90

3

63

12

8U

U

30
330

37

88

Although in this study the labels of the slides vrere masked, so as
to obviate any possible bias caused by knovirledge of the size of the fish,
there was no way of concealing the size of the scales. The authors may
thus have been influenced by scale size, tending to assign older ages
to large scales and younger ages to smaller ones. If this influence were
serious enouf;h to dominate the determinations, the fact might become evi-
dent from a correlation study on age, body length, and scale length.

42

The latter tv/o variables are rather imperfectly correlated, in
the present material, with considerable variatior) of body length for
each scale length, and also of scale length for each body length.
(The coefficient of correlation betvreen body length and scale length
was .78, with P of less than .01. The coefficient of regression for
scale length on body length was .8U; that for body length on scale
length was .73.) There is, of course, no ivay of knowing the true
magnitude of correlation between age and either of these two variables.
There are these possibilities for consideration, however:

(A.) Age might be more highly correlated' with scale length than
with body length. Then, on measuring scales from a number of fish
with identical (or nearly so) body length, we should expect some degree
of positive correlation between age and scale length. In addition,
if, in our age determinations we had been influenced by the size of
the scales, rather than the number of age marks, we should expect the
above correlation to be enhanced. To examine this possibility, a col-
lection of 987 scale measurements ranging from 190-219 mm. of projected
image, and compromising four year classes, was divided into groups of
ten-millimeter classes according to body length; and for each of these
body-length classes the coefficient of correlation betvfeen age and scale
length was calculated. The results, given in table III, show no sig-
nificant correlation between these two quantities. The probability
(P) is high, averaging .Ul6, that if there were no correlation what-
ever, the coefficient could, by random causes alone, be as high as
indicated. This lack of correlation is evidence that the age deter-
minations w6re not influenced by scale length to any appreciable de-
gree. Incidentally, it is also evidence that there is no intrinsic
correlation betv/een scale length and agej i.e., no correlation other
than what results from the dependence of scale length on body length.

TABLE III. -Correlation (r) Between Age and Scale Length
for Fish of Identical Classes of Standard Length.

Body length of fish ^
in millimeters i
date

i

<

r

P

190-199
Nov. 1939

82

1
.21

>

.05

190-199
Dec. 1939

91

.19

>

.0^

200-209
Nov. 1939

236

-.oU

>

.^

200-209
Dec. 1939

193

.002

>

.9

210-219
Nov. 1939

17U

.12

>

.1

210-219
Dec. 1939

! 211

: .003

>

.9

—'Number of ipeciraens

49

(B. ) If, on the other hand, age is nore highly correlated v;ith
body length than v.lth scale length, we shovild expect some degree of
positive correlation between age and body length among fish v:ith
identical (or nearly so) scale length. To examine this possibility,
scales of SS9 fish taken in the fall commercial fishery for adults,
comprising four year classes, and ranging from 190-219 ram. in body
length, v/ere divided into 10 mm. classes according to scale length.
The coefficient of correlation betv/een body length and age v.':'. thin
each of these scale-length classes was then calculated. These cal-
culations (Table IV) show significant correlations betv.'een these
two quantities, with a low probability (P), averaging less than one
in one huncired, of arriving at such high values by chance alone, if
there were no correlation. These results could not have been modi-
fied by any influence of scale length, for the design of the analysis
precluded that.

TABLE IV. -Correlation (r) Between Age and Body Length for
Fish Having Identical Scale Length Classification.

Scale length
(mm. of projected
image)
Date

Nl/ '

r

P

190-199
Nov. 1939

s$

.70

<.01

190-199

Dec. 1939 :

so :

.U3

<.01

200-209

Nov. 1939 :

81 !

.76

<.01

200-209
Dec. 1939

8U

.68

<.01

2IO-2I9
Nov. 1939

152

.52

<.01

210-219
Dec. 1939

137

.U6

<.01

i/ Number of specimens

(C.) The results reached under A and 3 are conclusive; but,
nevertheless, consideration might be given to the possibility that
age is correlated in like degree with both bo(3y length and scale length.
Then the coefficient of correlation for age on body length with fish
having identical (or nearly so) scale length, and that for age on
scale length with fish having identical (or nearly so) body length
should be about equal. Then, if in our age determination, the authors
had been influenced by the a., .e of the scales instead of the number
of age marks, the correlation coefficient for age on body length should

44

tend to be the lo^.^er, and that for a^e on scale length the higher value
of the tv/o. Obviously, judging from tables III and IV, just the oppo-
site of this result is deraonstrated. Thus, so far as can be deduced
from this evidence, there is no confirnation of a supposition that scale
length significantly influenced the age determinations in the present
study. Indeed, it appears unlikely that it vras of any influence what-
ever.

The Use of Otoliths for Determining Age

Another line of evidence by which to judge the correctness of scale
readings is given by a study of otolitlis. This was carried on simul-
taneously by Mosher, independently of the scale v/ork, and without refer-
ence to it.

Otoliths were collected from 6000 fish of the same series, de-
scribed above, as represented by scales. These were washed and pre-
served in envelopes in the same manner as the scales. They were sub-
sequently mounted as follovra ; A piece of thin cardboard was cut in
the shape of a standard microscope slide, but about l/8 of an inch
shorter and narrovfer. This was punched with eight small round holes,
and laid over a drop of Canada Balsaia on a microscope slide. Into
each hole v;as placed an otolith, then a drop of Canada balsam; and
any air bubbles were burned out with a hot needle. Another microscope
slide was laid on the mount as a cover glass, and the preparation put
in a warm place to harden.

The mounted otoliths were examined in a micro-projection apparatus
and the dimensions recorded on punch cards, as described on page 8,
v/ith the base of the ruler placed at the center, as indicated in Figure
28. The inner edge of each translucent zone was taken as representing
the year mark. These readings were later checked by examination with
a binocular microscope and reflected light, vrtiich brought out the year
marks sharply as blue-black against the white opaque intervening zones.

Preliminary examination of otoliths from fish known to belong to
0 year class and collected over the course of 12 months established
the following criteria of a year mark:

An annulus is a more or less translucent band concentric
with the margin of the otolith, the intervening spaces
being opaque. It can usually be traced entirely around
the otolith, although it is more easily observed at the
blunter anterior end than at the sides or posterior end.
Annul! tend to be zones rather than lines.

Marks having the following characteristics were considered to be
adventitious, or otherwise unperiodic;

Though translucent, they tend to be lines rather than zones j
they frequently meet an annulus at some point.

46

Although the first two year-marks are generally well defined on
otoliths, the third is less so; and the subsequent ones are so closely
crowded and so easily confused with false year marks that they cannot
be counted with a very high degree of confidence.

To determine whether or not the "annulus" is formed once annually
in otoliths, a stratified random sample of 2600 specimens, representing'
about equally all the months of a year, was examined, and the marginal
increments, -vrfiich included the band comprising the annnlus, were measured
as described on page 8. To avoid possible confusion between true and false
year marks in otoliths, no translucent zone was considered until it was
wide enough to preclude its being a false annulus.

TJie marginal increments for otholiths, when plotted in the same way
as had been done for scales in figure 4, proved also to be narrowest during
one part of the year, and to increase progressively thereafter. In figure
7, age classes 1937, 1936, and 1938 have been assigned to fish having,
at their first appearance in our samples, one, two, and zero rings on
their otoliths, respectively. It is evident, in figure 7, that the annulus
is a true year mark in otoliths as well as in scales, and became visible-
earlier in otoliths than-in scales in 1939, showing as early as June in
some specimens, and by October in all. This is consistent with the con-
clusion reached above that in scales the year mark forms during the summer
and fall.

To define more exactly a time when the annulus became evident in
otoliths, we can choose, from the data used for figure 7, arbitrary boun-
daries to separate "narrow" from "wide" marginal increments, as was done
for scales. Thus the 35-mm. point was selected for year class 1938, the
20-mm. point for year class 1937, and the 15-mm» point for year class 1936.
The proportion of fish having "narrow" marginal increments, in percentage,
by months, is given in figure 8,

Parallel age determinations, by scales and by otoliths, were made
in 1036 specimens. The results, given in table V. indicate a high agree-
ment for the first two year classes, and a moderate agreement for the thirdi
Doubtless a large portion of the disagreements among the three-year-olds
may be attributed to the difficulty of counting the rings on otoliths
■with more than two marks.

Growth of Juvenile Piloharde

The age composition of samples from the 1938-39 bait fisheries of
southern and central California, as determined by scales, is indicated
in figure 9 where the growth of fish comprising year classes 1938 and

46

1937 can be traced from early youth to the size at entrance into the
fall commercial fishery for adults; and though data for year class
1936, also included, are rather scanty, they are sufficient to give
some indication of size composition and growth. A similar figure made
from otolith readings was essentially identical with figure 9, and
is therefore not here reproduced (Table 6 ).

TABLE v.- Comparison of Age Readings by Scales and by Otoliths

Number of Number of otolith Number of otolith Percen-
annuli by readings matching readings not match- tage agree-
scale reading scale readings ing scale readings nent

1 355 6 98

2 UI9 36 92

3 166 5U 75

9hO 96

Precise knowledge of the growth of the pilchard is seriously
impeded by the difficulties of sampling, discussed on page U , Growth
is illustrated in figure 10 (Table 8 ) where the average lengths of
each year class among fish taken in California are plotted by months.
Data are from the bait fisheries and from the fall commercial fishery
for adults.

The differences from year to year bet'reen the average length
of fish of corresponding ages in figure 10 are no doubt due in part
to actual differences betvreen the year classes. In large part also,
however, they are due to differential selection. I.ioreover, valuations
from a smooth growth curve for individual year classes are probably
due to varying selection.

Samples of fish of year class 1937, taken from April, 1938, to
May, 1939, were mostly from the southern California bait fisheries,
where there is a tendency for bait fishe"rmen to select against the larger
fish, resulting in over-representation of the smaller members of that
year class. Furthermore, judging from figure 9, there was a tendency
for the larger fish of year class 1937 to leave, and for smaller ones
to enter, the field of' the fishery at intervals during the year. The
consequent selective availability undoubtedly contributes to the irregu-
larities of figure 10. From June to October, j'-ear class 1937, as well
as 1936, was sampled from the San Francisco bait fisheries, which seemed
to select against the smaller fish, producing an effect opposite to
that obtained in southern California. At the same time, beginning in
September, samples were taken from the fall fishery for adults at Monterey.
Here a selection results from a rather conplex differential availability.
Fish of year class 1938 taken in the fall coniuercial fisher;/- for adults
appear to represent the larger membersof their class, perhaps partly

47

because the latter were schooling with older fish, and partly because
the fishermen consciously selected against very small fish, which have
low economic value for reduction or canning purposes. In general the
fish available during most of the fall are sjnaller and yoimger than
those available during the winter. This results in a selection, by
differential availability, against the larger fish of year classes 1937,
1936, and 1935.

Because of these differences in demand and availability from place
to place, from fishery to fishery, and from month to month, and the
consequent imperfections in our sampling of the population, figure 10
shows a very irregular series of .points to which such a curve as might
be fitted could only rather roughly represent the average growth of the
pilchard in California during its first four years. Nevertheless, in
all essential respects, the points, as far as they go, are in harmony
with the points (designated by large asterisks) obtained from the modes
of frequency curves in figure 2, and from data published by the CaLLifornia
State Fisheries Laboratory (Clark, table 2, I936).

Difference in Size Betv/een Localities

The foregoing discussion has referred only to fish collected in
California. The statistics given in figure 10 and table 8, therefore,
describe only a portion of the total Pacific coast population of those
year classes. That this is true is shown by figure 11, which gives the
age composition of such samples as were obtainable in Washington, Oregon,
and Lower California. Obviously, if it were possible to sample properly
the entire coast, the range of each year class would probably be broader
than shown in figure 9. Though the data from northern and southern
grounds are too scanty to deserve very extensive interpretation, they
show that fish found north of California in 193Q-39 averaged smaller
by the end of their first year than those of the same age from Californiaj
and fish taken from Lovrer California averaged larger than those of the
same age taken in California. These differences are no doubt reflected
in the scale grov/th increments, and should furnish a tiseful tool for
studying migrations.

DISCUSSION AND CONCLUSIONS

The scales and otoliths of most fishes are characterized by folds
or sculpturing in which variations of pattern may be related to age.
If they are so related, for a given species, they need not correspond
in number vrlth years of life; or, if they do correspond, they may not
be recognized, for they may be so obscure as to be counted incorrectly.
It is no wonder, then, that scale and otolith reading is a process
peculiarly open to subjective influences, and requires, for serious work,
a stringent test and proof of validity.

48

In the foregoing pages it has been demonstrated that, if they
be properly prepared, scales and otoliths of juvenile pilchards
(Sardinops caerulea), show ringg that are associated in number with
agfT It has been shown that these rings are formed once yearly, at
a definite ticie of year, namely, the summer and fall. They therefore
correspond in number with years of life. Though they are sometimes
relatively obscure, the rings present may be counted v/ith what appears
to be significant accuracy, say 85-95 percent for the first three year
classes; and results, which a later report will cover, give promise
of extending the use of scales to much older ages. Of the scales ex-
amined during this study, over 99 percent were considered legible.
Those npt legible were generally specimens that had been improperly
cleaned, or that had regenerated centers. The age detenninations were
in all cases made without reference to the size of the fish, and there
is evidence that the operators were not unduly influenced by the size
of the scales.

The counts of annuli on otoliths v^ere found, by an independent
study, to agree with those on scales. It is the opinion of the writers,
however, that otoliths are less useful than scales for determining age
of pilchards beyond the third year.

Judging from scales smd otoliths of juvenile pilchards taken from
the bait fisheries in California during 1938 and 1939, and from the
fall commercial fishery for adults in 1939, fish of the year class 1938
had a modal length around 70 mm. when they first appeared in the bait
fishery in June, 1938, growing to near 120 mm. by the following April,
by which time the first annual mark has formed; and fish of year class
1937 had a modal length near l60 mm. v/hen their second mark had formed.
Because of the rather complex selection resulting from the requirements
of the fisheries, as well as from variations in availability, the sampling
was highly imperfect, and these figures are mere approximations. Fur-
thermore, they apply only to fish taken from California fisheries.

Young specimens collected in Lovrer California were larger than
those of corresponding age taken in California; and those taken in Oregon,
Washington, and British Columbia were smaller. These differences suggest
further applications of the study of pilchard scales, for they are, no
doubt, reflected in differences in corresponding scale growth increments
among the populations of the different regions. Thus the scales may prove
useful "for studying the migrations of the pilchard, as well as for deter-
mining their ages.

LITERATURE CITED

AIKAV/A, HIROAKI

I9U0. On the age and race of the Japanese sardine, Sardinia

melanosticta (Temminck and Schlegel). Records of Oceano-
graphlcTorks in- Japan, Vol. 11 (2), pp. 81-112, 10 figs.

49

CLaRK, FRANCES N.

19314.. Maturity of the California sardine (Sardina caerulea)
determined by ova diameter measurements. Div. Fish
and Game of California, Fish Bull. U2, h9 pp., 19 figs.

1936. Interseasonal and intraseasonal changes in size of the
California sardine, (Sardinops caerulea. ) Div. Fish
and Game of California, Bureau of Comnercial Fisheries,
Fish Bull, hi, 1936, pp. 28, 11 figs*

HIGGINS, ELIIEK

1922 A study of the age and rate of growth- of the California
sardine, Sardina caerulea, (Girard). Thesis presented
to the Department of Zoology, University of Southern
California, in partial fulfillment of requirements for
Degree of Master of Arts, 61 pp., 9 figs. (Unpublished.)

LEA, EINAR

1913. Further studies concerning the methods of calculating the
growth of herrings. Conseil Perm. International pour
1 ' Exploration de la Mer. Publications de Circonstance,
No. 66.

SCOFIELD, EUGENE C.

193U- Early life history of the California sardine (Sardina

caerulea), with special reference to the distribution
of the eggs and larvae. California Div. Fish and Game,
Fish Bull. Ul.

SETTE, OSCAR E.

Digit bias in measuring and -a device to overcome it. Copeia,
July 8, 19U1 (No. 2.)

SILLIMAN, RALPH

Seasonal occurrence of the Pacific pilchard (Sardinops
caerulea), off the coast of Washington as indicated by
its presence in salmon food, (Unpublished manuscript)

THOIIPSON, WILL F.

1926. The California sardine and the studj' of the available supply.
In 'the California Sardine, by the staff of the California
^ate Fisheries Laboratory; *Fish and Game Commission of
California, Fish Bull. 11, pp. 5-66, 18 figs., 3 tables.

60

TABJjE VI -Frequency Record of Juvenile Pilchards Saii5>led in California, by Ages,

MARCH, 1938

APRIL, 1938

Uiy, 1938

Fish
length

Total
fish
measured

Fish i
From
scales

aged
From
otoliths

Total
•fish
measured

Fish i
From
scales

iged
Fl-om
otoliths

Total
fish
measured

37

fear c
36

lasses

37 36

37

fear classes

36 37 36

36

-

-

-

-

-

-

-

-

3

1*0
hh

•"

_

_

~ —

"*

~

""

^ _

1
1

U8
52

—

*"

~

^ _

—

~

«

_ _

-■

56
60
Sh
68
72

■■

-

:

: :

-

-

-

: :

-

-

-

:

: :

-

:

:

-

:

Ik

80
8U
88
92

-

-

-

-

-

-

-

-

-

1

U9

3U

17U

-

-

-

3

-.

-

-

-

96
100
10(t
108
112

17
11
23
23

2

h
5
2

-

5

8
10

32
15U
365
577
90U

-

-

-

10

58

116
120
12U
128
132

3b

la

us

25

7
8
7
6
9

-

12

11
8
k

10

6^1

507
U37
UI4O
377

1
1

7

5

-

3

10
7

6 1

266
hi?.
35U
355
383

136
1I4O
iWi
lli8
rl52

17

10

9

8

11

7
1
2
2

i

1
5

3
1

/■

5
1
1 2

30I1 ■
139

66

U3

22

7

u

3
9
9

5

1
2

5
lU 1

5 2
11

9 2

360 ■■

357

306

255

188

■156
160
16U
168
172

9
6

1
5
2

1

2

1
1

1

19
10

1
2

1

11
5
2

2

5

h
9
5
5

5 2

1 3
1 6

5
3

116
1x2
30
33
31

1?6
180
18U
188
192

1
1
h

-

-

-

1

1

■ 2

3
1

1
2
1

16

16

U

h

2

I96
200
201i

1

-

-

-

-

-

1

~ T

3
2
2

208
212

~

:

"■

~ —

1

~

—

^ _

1

216
220
22U
228

_

-

„

_ _

-

-

-

- -

i

232

-

-

*

_

_

-

.

. -

.

236
2U0

2Uk
2U8

252

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

.

.

_

_ _

_

_

_

_ _

.

256

—

-

—

- -

-

-

-

-

1 ■■

TOTALS

939

72

15

83 . 2

5096

71

liO

77 29

36lh

51

TABU Vl-^Cont'd, )'-ft"«<J'-'««=y Rscord of JUvanlle Pilcharda Sampled In California, by Ages

JUNE,

, 193£

!

JULY,

, 1938

Plah aged

Total

Fish aged

Total

nsh

ftXB

Tron

fish

From

Froa

fish

length

scales

otoliths

measured

scales

otoliths

measured

fear classes

• Year classes

38

37

36

38

37

36

38

37

36

38

37

^

36

«,

^

•

_

-

.

-

-

-

-

-

-

.

-

UO

-

-

-

1

-

-

6

-

-

-

-

-

-

-

liU

-

-

-

3

-

-

65

-

-

-

-

-

-

-

Ii8

_

-

-

3

-

-

31

-

-

-

-

-

-

-

52

-

-

-

1

-

-

10

-

-

-

-

-

-

-

56

_

-

.

2

-

-

17

-

-

-

-

-

-

5

60

1

.

.

3

-

-

28

2

-

-

1

-

-

2U

6U

2

.

•

6

.

-

38

u

-

_

h

-

-

103

68

2

_

.

9

-

_

36

u

-

-

6

-

-

156

72

1

-

-

9

-

-

27

u

-

-

6

-

-

151

76

_

-

-

10

-

-

31

7

-

-

7

-

-

lU.

60

.

.

.

7

_

.

103

9

-

-

12

-

-

18U

6U

.

•

_

U

-

-

92

u

-

-

lU

-

-

13U

88

.

_

.

5

-

-

29

5

-

-

17

-

-

156

92

-

-

-

3,

-

-

5

1

-

-

8

-

-

105

96

-

-

-

z

-

-

2

1

-

-

b

-

-

bh

100

•

_

•

-

-

-

2

2

-

-

1

-

-

35

lOli

_

•

•

2

-

-

1

-

-

-

1

-

-

11

108

-

-

-

-

-

-

-

1

-

-

-

-

-

3

112

-

-

.

-

-

-

-

-

-

-

1

-

-

8

116

-

-T

-

-

-

-

1

-

-

-

1

-

-

2

120

_

_

»

.

U

-

lU

-

-

-

-

-

-

5

12U

.

2

•

_

5

-

72

-

-

-

-

-

-

7

128

«

9

.

-

7

-

272

-

-

-

-

-

-

U7

132

-

10

-

-

U

-

556

-

2

-

-

h

-

197

136

-

— B-

-

-

W

-

m

-

1»

-

-

h

-

337

11,0

.

12

_

_

11

-

791

_

2

-

-

5

-

578

lljli

•

10

.

•

8

1

651

-

15

1

-

12

-

517

lb8

•

10

.

.

U

_

U77

-

6

-

-

10

-

li03

152

-

12

1

.

u

2

362

-

11

-

-

6

-

270

ss

-

1^

"6

-

3

3

2U0

-

6

-

-

2

-

21b

«

12

U

.

1

1

180

-

8

1

-

h

2

1U9

16U

»

6

12

.

«

3

87

-

9

-

-

h

h

111

168

.

W

IX

_

.

^

ti6

-

6

2

-

-

2

6U

172

-

1

u»

-

-

.

28

-

1

3

-

-

\l

176

-

-

E

-

.

-

16

-

^

1*

-

-

30

180

.

.

5

_

.

1

11

_

-

-

-

-

lU

18U

_

-

U

,.

•

1

8

-

-

1

-

-

11

188

.

.

1

•

_

.

3

_

-

-

-

-

8

192

-

-

1

-

_

_

3

-

-

-

-

-

2

196

-

-

-

-

-

-

2

-

-

1

-

1

5

200

_

-

-

-

-

_

2

.

-

1

-

-

2

1

20U

.

»

-

_

_

•

2

_

.

-

-

-

-

3

208

-

-

-

-

-

-

3

-

-

-

-

-

-

2

212

_

-

-

-

-

-

3

_

-

-

-

-

-

2

216

-

-

-

-

-

.

3

-

-

-

-

-

-

1

220

'-

-

-

-

-

-

1

-

-

-

-

-

-

-

22U

-

-

-

-

-

-

1

-

-

-

-

-

-

-

228

_

-

-

.

-

-

1

•

•

.

.

-

-

-

232

-

-

-

-

-

-

-

.

-

-

-

-

-

-

236

-

-

-

-

-

-

2

.

-

-

-

-

-

-

2it0

-

_

-

_

-

-

-

_

-

-

-

-

-

-

2hU

-

-

.

-

-

-

1

-

_

-

-

-

-

-

2U8

_

•

_

_

-

-

.

-

•

-

-

-

-

-

2S2

-

-

-

_

-

-

1

-

-

-

-

-

-

-

256

-

—

-

—

"

—

~

~

"

~

"•

■"

"■

■■

TOTALS

6

108

65

70

68

12

51U1

, ^

70

lU

8U

52

21

U293

52

TABLE VI {Cont'd.). -frequency Record of Juvenile Pilchards Sampled In California, by Ages.

AUGUST, 1938

SEPTEMlfflH, 1938

Fish

Fi-om

From

fish

From

From

fish

length

scales

otoliths

measured

scales

otolith!

1

measured

Year c;

lasses

Tear c

lasses

38

37

^

38

31

^

38

37

36

38

37

i6

^'f

1

-

-

-

-

-

1

-

-

-

-

-

-

_

5i.

-

-

-

-

-

-

1

-

-

-

-

-

-

-

6Q

1

.

-

-

-

-

5

-

-

-

-

-

-

-

6U

h

-

-

h

-

-

U7

-

-

-

-

-

-

-

68

7

-

-

6

-

-

202

-

-

-

1

-

-

16

72

19

-

-

16

-

-

30U

-

-

-

2

-

-

75

76

11.

-

-

29

-

-

2BU

3

-

-

i

-

-

2U2

80

10

_

-

20

-

-

118

5

-

-

12

-

.

287

8U

3

.

-

2

-

-

3U

10

-

—

13

-

-

2U2

88

3

-

-

1

-

-

10

8

-

-

10

.

-

187

92

3

-

-

-

-

-

9

11

-

-

Hi

-

-

210

96

1

-

-

1

-

-

'8

'5 ■

-

-

111

-

-

211

100

6

-

-

-

-

-

7

Hi

-

-

15

.

-

76

lOU

1

-

-

1

-

-

9

7

-

-

7

-

-

18

108

-

-

-

-

-

-

1

1

-

-

2

-

-

5

112

1

-

-

-

-

-

3

-

-

-

-

-

-

3

116

-

1

-

-

-

-

2

-

-

-

-

-

-

U

120

_

-

_

-

.

-

2

1

-

-

1

-

-

2

12U

-

_

-

-

_

-

3

-

-

_

-

-

-

u

128

_

2

_

-

-

-

9

_

_

_

_

-

-

2U

132

-

6

-

-

1

-

67

-

-

-

-

-

-

101*

136

-

16

—1

-

V

-

23«

-

2

-

-

1

-

288

lUO

-

2U

1

-

13

-

U76

-

h

_

-

6

-

b89

Ihh

_

23

2

-

20

-

630

.

9

_

_

10

.

686

1U6

-

25

-

-

Ifl

-

678

-

m

1

-

13

-

671

152

-

11

3

-

16

1

5U9

-

16

2

-

19

-

517

156

-

10

-

-

la

1

3bi

-

7

1

-

11*

-

327

160

-

5

-

-

10

2

17U

_

6

_

-

11

.

175

16U

-

2

1

-

2

1

107

_

-

_

_

3

.

7U

168

-

3

-

-

1

1

70

.

_

_

_

1

2

91

172

-

1

1

-

1

1

U9

-

-

3

-

1

2

68

176

-

1

1

-

-

2

^0

-

-

_

-

-

-

59

180

-

-

2

-

-

1

U7

_

»

_

_

_

-

U9

18U

-

-

2

-

-

2

33

_

_

.

.

_

_

19

188

-

_

-

_

-

1

28

_

.

.

«

.

_

16

192

-

-

1

-

-

-

29

-

-

-

-

-

-

8

"196

-

-

-

-

-

1

16

-

-

-

-

-

-

2

200

-

-

-

-

-

1

11

•

..

.

_

-

-

1

20U

-

_

-

-

-

_

18

_

«

_

..

«

_

2

208

-

-

-

-

-

-

5

_

•

•

_

_

.

1

212

-

-

-

-

-

-

11

-

-

-

-

-

-

1

216

-

-

-

-

-

-

?

-

-

-

-

-

-

1

220

-

-

•

_

.

.

3

_

_

.

•

_

_

•

22U

•

»

•

.

_

_

2

_

_

_

_

_

_

_

228

_

.

_

*

_

.

»

_

_

_

_

.

.

1

232

-

-

-

-

-

-

-

-

-

-

-

-

-

1

536

■■

~

—

~

—

"

1

-

-

-

-

-

-

-

TOTALS

7U

129

lit

80

103

15

U703

65

58

7

96

79

1*

5257

53

TABLB TL (Cont'd. )--F*«quancy Itocord of Juvenile PllehardB Saopled In Califoml*, by Ages.

Fish
length

OCTOBBt. 1938

Piah aged Total

Tram Troa fish

scales otolitha neaaured

Tear classes
38 37 36 38 37 36

__---- 1

- U.

3U

17
-r — :: :: 2 = = 55

8 - - 5 - - 212

lit - - 5 - - 350

13 - - 9 - - 262

5 - - 6 222

-1 = 7 = = 311 —

2 - - 6 - - 277
7 - - 5 - - 202
5 - - 6 - - 95
3 85

-7 = = 1 = = 911

3 - - 2 - - 95
2 - - - - - 72

2 1 - 1 - - 62

- 1 - - - - 5$

—I 5 = = = 155

1 10 - - 1 - 363

- I4 - - 2 - 1*9U

- 11 - - U - U69

- 22 1 1 17 - U73

— = — 15 1 = — ?2 = 555 —

- 16 3 - 20 1 218

- 6 1 - 12 1 125
31-31 65
1 3 - 2 - 37

— :: 1 1 = 1 : 55 —

- _ - _ 1 _ flo

- - - - - 2 m

12 86

- - - - - 2 70

— = — = — I — = — :^ — n 55 —

2 35

----- 1 20

----- 17

- 6

—z — :: — : — = — = — :: 5 —

6

------ 1

— :: — :: — = — = — = — :: 1 —

sovgMBEa. 1938

Fish aged Total

feoa From fish

scales otoliths measured
Tear classes
38 37 36 38 37 36

13

2 - - 3 - - 1*0

3 - - U - - 112
5 - - 9 - - 180

-5 = = 1 = 555

1 - - U - - 15U

3 - - 111

2 - - 2 - - 11*3
U - - 3 - - 127

-5 = = C = = 157

5 - - 5 - - 55

6 1 - 5 - - 37

31
1 1 - 2 - - 2U

-l — I — :: — = — : — = 15 —

1 - - - - - 22

- 2 - - - - 78
6 - - - - 198

1 7 - - 2 - la6
— = B 1 = = = 755

- 6 - - 13 - 800

- U - - 13 - 670

- 8 1 - 13 1 5U8

- 2 1 - 6 - U80
-■= 7 5 = 5 5 357

2 h - 3 U 283

- 2 U - - 7 211

6 - - 6 lli7

- - 3 - - 2 62

—z — r — 5 = — = — 5 511 —

2 - - 1 59

- - 1 - - 2 U6

- - 1 - - - 38

: : i : : : i

- - 2 - . - 16

8
------ 2

----- 2

— = — :: — I — :r—z — : 2 —

------ 1

------ 1

------ 1

...... 1

60
6b

68

72

-75-
80
8U
88
92

-55-
100

lou
108

112

-115-
120
12U
128
132

-I5ir

lltO
Uilt

1148
152

T5r

160
16U
168
172

T7r

180
18b
188
192

155-
200
20b
208
212-

-215"
220
22b
228
232

TJ5-

2bO
2bb

2be

252

■?5r

260
26b

TOTALS

69 98 11 56 86 16 5862

b2 63 35 53 57 31 6812

54

TABLE VI (Cont'd. ).-Froquoncy Record of Juvenila Plleharda Sampled in Califoi la, by Ages.

DECEUBER

, 193«

1

JANUARY. 1939

1

Fish aged

Total

Fish aged

Total

Fish

Fi-om

From

fish

From

Fi"om

fish

length

scales

otoliths

I

measured

3Ci

iles

otoliths

measured

Y.

sar c.

lasses

Y.

ear classes

38

37

36

38

37

36

38

37

36

38

37

36

68

_

_

.

_

.

_

^

1

_

.

_

_

«

^

72

-

-

-

-

-

-

-

-

-

-

-

-

-

1

76

-

-

-

-

-

-

-

-

-

-

-

-

-

1

80

•

.

.

—

—

-

—

—

»

—

•

*

-

-

8U

-

-

-

-

-

-

3

1

-

-

-

-

-

6

88

.

_

-

-

-

-

27

1*

-

-

1

-

-

3U

92

h

-

-

2

-

-

115

6

-

-

2

-

-

121

^

1?

-

-

6

-

-

200

5

-

-

5

-

-

H5

100

16

-

-

5

-

-

166

3

-

-

6

-

-

129

lOh

h

-

-

1

-

-

U3

2

-

-

10

-

-

16U

108

h

-

-

-

-

-

IIU

3

-

-

9

-

-

158

112

3

-

-

2

-

-

lou

7

-

-

1

-

-

199

116

-

-

-

5

-

-

89

7

-

-

li

-

-

333

120

1

-

-

1

-

-

75

8

-

-

9

-

-

390

121.

1

-

-

6

-

-

53

U

-

-

2

-

-

273

128

-

-

-

2

-

-

33

2

1

-

2

-

-

160

132

-

-

-

-

-

-

19

-

-

-

1

-

-

81

136

-

-

-

-

-

-

lU

U

-

-

-

-

-

53

lUO

-

-

-

1

-

-

17

-

1

-

-

1

-

39

UJ4

-

-

-

-

1

-

30

1

1

-

-

1

-

96

ma

-

1

-

-

-

-

65

-

3

-

-

2

-

187

152

-

3

-

-

3

-

1U8

1

19

-

-

2

-

350

156

-

5

-

-

1*

-

528

-

19

2

-

7

-

556

160

-

11

-

-

16

-

W8

-

28

-

-

10

1

68U

16U

-

6

3

-

8

1

U90

_

18

1

_

11

.

639

168

-

6

2

_

12

-

39U

-

2

1

-

13

7

392

172

-

9

2

-

lU

1,

3I46

-

3

-

-

10

3

230

■1?6

-

U

a

-

b

b

21,7

-

-

-

-

6

1

129

leo

_

8

7

_

3

5

157

-

-

2

_

-

U

88

16U

-

3

1

-

3

1

UU

-

-

1

_

-

U

67

188

_

1

2

-

-

2

96

-

3

1

-

1

1

72

192

-

1

1

-

-

1

93

-

1

-

-

h

6h

196

-

-

1

-

-

-

77

-

1

2

-

-

-

Uh

200

-

-

1

-

-

-

61

-

_

•

_

-

_

33

20li

-

-

-

-

-

-

U8

•

•

2

_

.

1

28

208

-

_

_

-

_

_

U6

»

..

.

.

_

«

21

212

-

-

-

-

-

-

36

-

-

-

-

-

-

10

216

-

-

-

-

-

-

22

-

-

-

-

-

-

lU

220

-

-

_

-

-

-

8

_

_

_

_

_

..

6

221,

-

-

-

-

-

-

15

_

-

.

«

-

-

7

228

-

-

-

-

-

_

6

_

_

_

_

_

.

6

232

-

-

-

-

-

-

5

-

-

-

-

-

-

2

236

-

-

-

-

-

-

5

-

_

-

-

-

-

1

21jO

•

•

.

.

.

«

1

«,

^

»

.

_

_

_

2U,
21,8

-

-

-

-

-

-

1

-

-

-

-

-

-

-

252

-

-

-

-

-

-

-

_

_

_

-

-

-

-

256
260

~

■"

^

—

_

^

_

~

^^

^

^^

_

^

^

261,
268

—

•

_

~

^

~

•

^

—

:

—

_

~

—

272

•-

—

—

—

-

—

1

-

-

—

—

-

—

"

TOTALS

50

62

28

31

69

22

UhlQ

?9

99

13

50

61,

26

6017

55

TABLE VI (Cont'd. )>-Fre<l>i8ncy Record of Juvenile PUcharda Sampled in California, by Ages.

FEBRUARY, 1939

UARCH.

, 1939

Fish ased

Total

Fish

aged

Total

Fish

From

From

fish

From

From

fish

length

SI

sales

otoliths

lasasured

scales

otoliths

measured

fear (

:lass<

as

Y

ear •

classes

38_

37

36

38

37

J6

38

37

36

38

37

36

76

_

.

_

_

_

_

1

_

_

_

_

_

_

_

80

-

-

-

-

-

-

7

-

-

-

-

-

-

7

8li

-

-

-

-

-

-

-7

1

-

-

1

-

-

39

68

1

-

-

1

-

-

16

h

-

-

1*

-

-

103

92

3

-

-

h

-

-

96

U

-

-

5

-

-

2U»

96

19

-

-

la

-

-

279

7

-

-

6

-

-

U96

100

U

-

-

13

-

-

35U

2

-

-

7

-

-

<u

lOU

5

-

-

2

-

-

209

2

-

-

h

-

-

1*27

108

6

-

-

1

-

-

125

6

-

-

5

-

-

U27

112

h

-

-

-

-

-

70

9

-

-

9

-

-

5I46

116

-

-

-

-

-

-

63

12

-

-

h

-

-

571i

120

5

-

-

tt

-

-

80

29

-

-

15

-

-

575

12U

5

-

-

3

-

-

66

32

-

-

12

-

-

539

128

6

_

-

3

-

-

62

22

-

-

7

-

_

U96

132

3

1

-

-

-

86

9

-

-

6

-

-

3U0

:i35

1

-

-

1

1

-

89

5

-

-

5

-

-

271

iwt

3

U

.

3

3

_

105

7

_

_

7

1

_

201

lUU,

3

2

-

2

_

109

1

-

-

h

1

_

18b

1U«>

8

1

-

li

-

108

_

2

•

9

.

-

125

152

1

10

-

-

6

-

136

-

1

-

5

-

-

76

156

-

12

-

1

8

-

213

-

1

-

i

1

-

U9

160

-

21

1

-

10

-

282

-

3

_

•

1

-

50

l6ti

-

17

U

-

8

-

236

-

7

1

-

1

-

69

168

-

19

3

-

3

-

132

-

13

1

-

3

1

129

172

-

12

3

-

2

-

96

-

27

5

-

7

-

179

176

-

8

2

-

-

-

93

-

35

1*

-

5

2

223

180

-

7

5

-

1

-

122

-

Uo

11

-

12

7

2U6

lea

-

9

3

-

1

2

136

-

29

12

-

u

Hi

228

188

-

1

2

-

-

u

117

_

15

18

_

5

U

170

192

-

1

6

-

-

u

106

-

u

12

-

10

95

196

-

-

1

-

-

h

06

-

3

8

-

-

6

tK

200

-

-

2

-

-

2

51

-

6

-

-

h

35

20U

-

-

-

.

-

1

35

-

-

5

_

_

2

17

208

-

-

-

-

-

2

30

•

-

1

-

-

2

6

212

-

-

-

-

-

-

18

-

-

-

-

-

-

2

216

-

-

-

..

-

1

15

•

.

•

-

•

-

1

220

_

_

_

_

—

.

6

_

_

_

_

_

_

_

22U

_

-

_

_

-

_

3

_

«

_

_

.

•

_

228

-

-

-

-

-

-

3

-

•

-

_

-

-

-

232

-

-

-

-

-

-

-

-

-

-

-

-

-

236

-

-

-

-

-

-

1

-

-

-

-

-

-

-

2U0

-

_

_

_

-

_

_

_

•

«.

_

_

_

_

2lth

_

_

_

_

_

_

1

.

•

_

_

_

•

•

2U8

-

_

-

-

-

-

-

-

.

-

-

-

.

1

252

-

-

-

-

-

-

-

-

-

-

-

-

-

-

"256

~

~

—

"

"

•

~

"

"

•

-

"

—

~

TOTALS

76

132

33

5U

U9

20

3850

152

180

66

U7

Ui

59

7879

56

TABI.E VI (Cont'd, j.-rrequencjr Record of Juvsnlls Pilchards Sai^pled In California, by Ages.

nsh
length

100
lOU
108
112

TTS"
120
12U
128
132

APRIL, 1939

Fish aged Total

Froa PVom fish

scales otoliths measured

1 - - - - - lU

U - - - - - 70

8 - - 8 - - 185

19 - 12 - - 321

-J? — - — =~~n — = — : — jrr

26 - - 12 - - 2U0

36 - - 16 - - 191*

2U - - 8 - - DJ*

16 6 - - 8tt

Ti — I — = — 3 — - — - — wr

13 1 - U - - 56

6 - - 1 - - 28

2 1 - - - - 26
1 2 - 1 2 - 36

-I 5 = 1 1 = W

6 - - 6 - 80

- 13 1 - 2 1 1U7

- 18 - - 8 - 177

- 23 - - 9 - 213

— : — 15 — 3 — = — Ta — 5 — mr

- Hi 3 - 5 - 123

- 9 1 - ii - 61

- 1 - - - - 20

- 1 - - - - 11,

— 1 1 : = r ir

6
_ . _ _ _ _ 5

----- 1 12

3

—z : :: = : :: 5-

______ 1

_____ 2

-= r z = = z r-

_____ 1

-z z = z z z 1"

m, 1939

Fish aged Total

From From fish

scales otoliths measured

38

Year classes
37 36 38 37

36

38

Year classes
37 36 38 37

36

92

_

_ - _ -

-

1

•

- - _ -

_

_

9i>

1

_ _ - -

-

5

-

- _ _ _

-

.

1

"liT

28

19

7

3

1

~T

13

7

2

2
lU
32

81

277
233

177
111

136
lUO
IUI4
lli8
152

"15^-
160
16U
168
172

2

-T

1

U

U

7

39
29
22
33

IT
3li
liO
51
61

180
18U
188
192

1

10

7

18
7
9
1*

155"
200
20li
208
212

IT
20
111
12
6

lOlT

179

115

99

51

"515-
220
22U
228
232

19

17

7

6

21*0
2UU

2U8
252

"25r

TOTALS

201 108 9 83 55 I* 2957

77 95 30 35 76 35 2106

57

TABLE VI (Cont'd. )'-f*«<J^8ncy Record of Juvenile Pilchards Sampled in California, by Ages.

JULY, 1939

Fish
length

TOTALS

U6 UP 30 17 3U 36 780

82 88

- Ul 56 355

58

TABia VI (Cont'd. )'-Frequency Record of Juvenile Pilchards Sampled In California, by Ages.

AU

CrU.ST,

1939

SEPTEMBEK,

1939

Fish

aged

Total

Fish

aged

Total

Fish

Fi-

ora

From

fish

From

From

fish

length

scales

otoliths

neasured

s

cales

otoliths

measured

Year classes

irear

classes

38

37

36

38

37

36

38

37

36

38

37

36

156

_

_

_

_

.

^

_

_

_

_

_

_

_

5

160

-

-

-

-

-

_

_

_

.

_

_

_

.

3

16U

-

-

-

-

-

-

-

-

_

_

_

-

_

5

168

3

-

-

5

-

-

6

-

1

_

_

_

_

10

172

1

-

-

1

1

-

5

-

1

-

-

-

-

15

176

6

1

-

b

2

-

11

-

1

-

-

.

_

?

180

3

1

-

K

h

1

13

-

2

_

_

_

-

7

18U

-

3

-

i

2

-

7

-

1

.

-

-

_

8

188

-

1

-

-

3

-

lU

-

1

-

_

_

.

5

192

-

9

3

-

6

1

38

-

3

1

-

-

-

19

196

1

111

1

-

7

3

-

6

2

-

-

-

w

200

.

16

12

-

U

9

82

-

18

5

.

.

.

59

20U

-

38

29

-

9

13

75

-

27

36

.

-

-

78

208

-

36

51

-

3

15

100

-

17

27

-

_

.

60

212

-

2U

56

-

h

lU

75

-

10

19

-

-

-

36

216

-

5

I4I

-

-

9

U3

-

-

11

-

-

-

13

220

_

1

21

-

-

U

27

_

1

3

_

_

_

6

22U

-

■ -

6

-

-

1

6

-

1

_

-

_

2

228

-

-

1

-

-

-

3

_

-

-

-

-

-

1

232

-

-

-

-

-

-

1

-

-

-

. -

-

-

-

236

2hO
2Uk
2U8

-

-

-

-

-

-

1

-

-

-

-

-

-

-

-

-

-

-

-

-

1

-

-

-

-

-

-

-

252

-

-

-

-

-

-

-

-

-

-

-

-

-

-

256

~

~

"

■"

"■

~

~

—

~

~

~

~

~

-

TOTAUS

m

1U9

221

17

hS

70

567

.

90

lou

_

_

^

379

OCTOqER, 1939

Fish
leagth

From From
scales otoliths

fish
measured

172

Tear classes
38 37 36 38 37 36

3

176

■ ■■ 2 '

130
18U
188
192

200
20U
208
212'
"215-
220

8
23

h

2
9

9-
7

■7-

5
3

8

1;

-?r

38
51
23
19
"IT

TOTALS

U3 3U

198

TABLE VII. -Frequency Record of Juvenile Pilchards Sampled Outside California,

by Ports and by Ages.

JULY. 1933

AUGUST, 193?

— —

~ ■

Southern

Lower California

(

Central Loner California

riah aged

Total

Fish aged

Total

Fish

Trca

Ttm

fish

From

From

fish

length

scales

it

otolith!

p

naasured

scales

otoliths

measured

iar classes

Year classes

38

37

36

38

37

2k

3L

37

36

38

37

36

6U

_

^

_

_

_

_

-

1

-

-

-

-

-

1

68

_

_

.

-

-

-

1

2

-

-

1

-

-

10

72

-

-

-

-

-

-

25

3

-

-

h

-

-

15

76

•

-

-

-

-

-

68

2

-

-

la

-

-

55

80

«

•

_

-

-

271

7

-

-

I.

-

-

la

8U

_

.

.

-

-

U0I4

6

•

-

8

-

-

32

88

1

_

.

-

-

2hh

5

-

-

1

-

-

10

92 .

1

-

-

-

-

lUO

1

-

-

-

-

-

3

?6

3

-

-

-

-

BU

-

-

-

i

-

-

u

100

U

.

«

-

-

57

U

-

-

h

-

-

lU

lOU

3

«

•

-

-

58

5

-

-

b

-

-

28

108

5

.

_

-

-

63

7

-

-

11

-

-

26

112

10

-

-

-

-

77

6

-

-

6

-

-

11*

US

... ^^

-

-

-

-

73

3

-

-

2

-

-

B

120

10

-

-

-

-

62

1

1

-

2

-

-

3

12U

9

1

-

-

-

35

-

-

-

-

-

-

-

128

2

-

-

-

-

-

20

-

-

-

-

-

-

-

132

3

-

-

-

-

-

8

-

-

-

-

-

-

-

-I3*'

-

-

-

-

-

2

-

-

-

-

-

-

-

lljO

-

-

-

-

-

-

2

-

-

—

"

—

—

"

lliJ*

-

-

-

-

-

-

-

-

-

-

-

-

-

-

UtS

-

-

-

-

-

-

-

-

-

-

-

-

-

-

152

-

.

-

-

-

-

-

-

-

-

-

-

-

-

156

-

-

-

-

-

-

-

-

-

-

-

-

-

-

160

-

-

-

-

-

-

-

-

-

-

-

-

-

-

I61t

-

-

-

-

-

-

-

-

-

-

-

-

-

-

168

-

-

-

-

-

-

-

-

-

-

-

-

-

-

172

-

-

-

-

-

-

-

-

-

-

-

-

-

-

176

_

>

-

-

-

1

-

-

-

-

-

-

-

-

180

_

-

-

-

-

-

-

-

-

-

-

-

-

-

181,

-

-

-

-

-

-

-

-

-

-

-

-

-

-

188

-

-

-

-

-

-

-

-

-

-

-

-

-

-

192

-

-

-

-

-

-

-

-

-

-

-

-

-

-

1^6

"

"

1

~

~

~

~

"

~

■■

"

■■

~

—

TOTALS

62

1

1

U6

-

1

169a

53

1

-

62

-

-

2614

60

llLBrx vn ^Cflnt'd. ).-nraquenc7 Record of Juvenile Pllcharda Samplod Outsljde C&lli'oml&,

b7 Ports and by Agea .

aiPTEIIBER. 1938

(XT03ER. 1938

Vtohlngton

Washington

Fish
length

Froa
scales

nsh aged

Ftxm
otoUths

Total

fish
■sasured

Fi-on
scales

Fish aged

F*«m
otoliths

Total

fish

□eaaured

38

37

ear c
36

lASses
38 37

36

3L

37

Tear
_36

classes
38 37

Jk

60
6k
68
72

^

~

-

-

~

-

5
8
9

u

"■

-

-

"

"

7

5
u

6

76
80
6U
86

92

-

-

-

-

-

-

1

-

-

-

-

1*
U

2

100
lOli
108
112

-

2

-

-

-

1
2
5

1
1

2

2

1

2

-

-

-

-

5

2

1
1
2

116
120
12U
128
132

-

3

5
2

, 3

-

-

-

3

10

15

U

17

-

3

-

-

-

-

1

1

"136 ■
lUo

152

-

1
3

1
2

1

-

-

y

10
6
7

3

-

-

-

-

-

-

1

156
160
16b
168
172

-

-

-

-

-

1

1
2

-

-

-

-

-

-

-

TOTALS

-

22

1

-

-

95

38

10

.

_

•

•

50

61

TABLE VII (Cont'd. )--Fre<JU9ncy Record of Juvenile Pilchai-ds Sampled Outside California,

by Ports and by Ages.

Fish
length

MAI, 1939

dragon

FlBh aged Total

Trca From fish

scales otoliths measured

Tear classes
36 37 36 38 37 36

__--_- 1

___.-- 1
------ 1

----- 2

3

_-.-_- 12

-1 :: = = : = 15

- . - - - - 2U

- - - - - - U5

- - - - - - U3

- - - - - - 56
— z : : 1 1 57

- 5U

1 - - - - 30

19

3 - - - - 7

— : (5 1 : : r 3

- 3 1 - - - 2
7 - - - -

U - . _ .
k - - ~ .

loU

108
112

TTS"
120
12U
128
I3S

lUh
1U8
152

"I5r
160
16U
168
1/72

180
18U
188
192
-19^
200
20I4
208
212

TOTALS

UO

375

62

Table Villi Statistics of length frequency curves for California pilchards
found from scale studies to belon:': to year classes 1936 to 1^38,

Nuniber

Mean

standard

Standard

Standar'd

Date

of

body

3rror;

deviation

error;

s

pecinens

lenfth

mean

standard

Year Class 1933

deviation

1938

July

UU

79.27

1.59

10.52

1.12

August

7U

70.69

1.30

11.22

0.92

September

65

91.07

1.11

8.97

0.79

October

69

9U.66

1.81

15.06

1.28

hovember

U2

108.59

2.51

16.26

1.77

December

SO

98. 9U

0.90

6.35

0.63

1939

January

59

108.96

2.11

16.21

1.U9

February

76

109.05

1.36

16. 2U

1.32

ilarch

152

118.10

1.01

12. hO

0.71

April

201

122.25

0.73

10.35

0.52

May

77

123. 2I4

1.05

9.17

O.lh

Year Class 1937

1938

April

72

126.39

1.56

13.23

1.10

May

71

1U3.UU

1.59

13.37

1.12

June

108

1U6.U8

1.13

11.78

0.80

July

70

153.01

1.18

9.87

0.83

Au^f^st

129

1146.15

0.79

9.03

0.56

September

58

IU8.33

0.76

5.78

0.5U

October

98

153.23

0.36

3.53

0.61

Hovember

63

161.66

1.U5

11.53

1.03

December

62

167.08

1.26

9.39

0.39

1939

January

99

157.62

0.89

3.90

0.63

Febr'oary

132

162.59

1.01

11.60

0.71

March

180

176.01

0.62

8.35

O.UU

April

108

169.37

0.9U

9.78

0.66

May

95

177.60

1.10

10.7li

0.78

July

183

19U.17

0.93

s.uu

0.66

Aucust

150

202.55

0.60

7.35

0.U2

September

90

199.83

0.95

9.06

0.68

October

U3

20ii.l5

0.67

U.hl

0.U3

Year Class I936

1938

April

15

1U9.63

2.21

8.56

1.56

May

UO

165.20

1.90

12.00

1.31

June

65

168.02

1.05

0.U6

O.7I4

July

lU

172. 6U

3.50

13.09

2. 17

August

lU

165.93

i;.57

17.09

3.23

September

7

158.93

3.51

9.30

2.U9

October

11

163.32

1.98

6.95

i.ue

November

35

186.93

2.27

13. U3

1.60

December

28

178.78

1.69

8.92

1.19

1939

Janiia,ry

13

130.38

\x.h3

15.99

3.IU

February

33

177.72

2.21

12.70

1.56

I.iarch

36

186. U5

1.00

9.26

0.71

April

9

177.52

2.32

8.h7

2.00

May

30

190.50

2.6U

1U.U8

1.87

J'lly

91

205. Ul

0.73

6.93

0.52

August

220

209.35

0.U2

6.25

0.30

September

102

206.28

0.53

5.33

0.37

October

33

207.68

0.95

$.hS

0.67

63

DO 120 rSO 140 190

BODY LENGTH, MILLIMETERS

— CAUFOI^IA

— COOS SAY, OREGON

— LOWCT CALIFORHIfl
ACTUAL NUMBCRS.

WASHINGTON

Figure !• n^equency curves shoiring hody lengths of pilchards sampled from
the Pacific coast bait fisheries during 1938 and 1939 ♦ The saiples are
summarized by months*

64

•0 flo M n to

I I I I I I 'I M I I r I I I i I I I 1 I I I M ■■ I I I I I I I I

iooioK)io«cso>o ipaowioe io ro

BODY LENGTH, MiLLMETER3
SMIfUS TWCM AT SM HOMO lUII tS, l«I1 -FCm IX.ItU CMJf FISMCMCS UU .

Figure 2. Frequency curves showing body lengths of pilchards sampled
from bait fisheries of San Pedro from March, 1921, to February, 1922,
The samples are summarized by lunar months. Original data furnished
by California State Fisheries Laboratory, and discussed by Higgins
(unpublished manuscript).

65

u

O

U.

o

ac
III

ISO . 30 40 50 60 70 SO 90 200 10 20

T I r i I I I I I I I I I I I I I I I I I I ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I

1939

JULY

I I ' I I I ' ' ' ' I ' '

JUNE

SEPT.

■ ■■■''■' ' -."i^i-ii^ 1 1 t.^t-^*->^j/r\i

- 5

0
5

0
5
0
10
5

120 30 40 SO 60 70 80 90 200 10 20 30 40 9C 60

BODY LENGTH, MILLIMETERS.

— MONTEREY AND SAN
FRANCISCO BAIT CATCH.

— MONTEREY COMMERCIAL

Figure 3* Rrequency omrves ahoirlng body lengths of pilchards sampled
from central California bait fisheries, and, in October, ftrom the
Ubnterey conmercial fishery for adults. The samples are summarised
by mon^is.

66

-""'••f7';Tr7"i'y?7~'"'*"rrTT?7

II3S '

i I 1 1 1 1 1 1 1 1 1 I I I I I I L_

_1 I I U

- «

• ae

ti I* n St •> «• tS M •> U TS T« » ■■ as •• I

■ 0« 11) III It) m I)) ■)• hi) Ml IS) IB* M

MARGINAL INCREMENT , MM

Figure 4* Rrequencies of marginal increment widths in scales having
0, 1, and 2 rings. Dimensions are in millimeters of projected image;
magnification is 431 diameters. The 32,5 mm, point is arbitrarily
chosen as separating the narrcvr from the vide increments*

67

4UG

SEPT OCT NOV

DEC. JAN. FEB. MAR APR HAY JUNE

100

' 1 1 1 1 1 1 1 1 1 1 1

100

0 ^^"^

"

«o

0)

^ .0

—

SOUTHERN CALIFORNIA

y^

:

90
SO

S TO

Z <o

_

^

/^

-

70

-

nsa-^

/

-

eo

d 50

-

// 0

-

50

S 40
* 30

^

0

_

40
30

1 20

Z '°
^ 100

* 90

«>.

^^^^

-

20
10
100
90
SO

-

f /

1 1

g *

i"

-

CENTRAL CALIFORNIA

//

/ /
/ /
/ /
/ /

/ /
/ /
/ y

TO

<" <0

11.

° 50

1-

U 30

o
oc

III 20

a. .

-

SO
30
40

JO
20

•38- a

10

1

1 1 1

1- 1

10

1 1 1 1 1

" *U6

SEPT OCT NOV

DEC JAN FEB MAR APR «AY .ONE |

•1

Figure 5« Percentage of scales having narrow marginal increments •

68

1 — I — I — I — I — r

1 — I — \ — r

"1 — I — r

939

MAR.

APR.

I I I

8 13 18 23 28 33 38 43 48 S3 58 63 60 73 78 63 88 93 98 103 108

MARGINAL INCREMENT, MM.

SOUTHERN CALIF.

CENTRAL CALIF.

C.V.R

Figvire 6. Comparing the widths of neirginal increments in scales of
fish from southern California with those ftrom central California.
Dimensions as in figure 5»

69

1 — I — r 1 '

.'.'■.' A 1 ' 1 J. J. ' ■ J. ■ ■

- .-■■■»■■

«r*-

WIDTH OF MARGINAL ZONE ID HLLMETCRS OF mOJCCTCO IHAM

■lm> ap<vf-n*a ci-Aii mf

Figure 7» Ptequencies of marginal increment iridtha in otoliths assigned
to year classes 1936, 1937, 1938. Dimensions are in millimeters of
projected image j the magnification is 125 diameters. Die boundary
between narrow and -wide increments is taken as 35 mnu of projected
image for specimens of year cHass 1938 j 20 nm. for those of 1937 j
and 15 mm. for those of 1936*

70

100

90

>-

O 80

: // :

3 70

o

y / /

^60

- *// /

u.

// /

50
LU

*^~\ / • h

< 40

S 30
O

£ 20

a.

- \ V '/ '-

10
n

V

1 1 1 IS 1 1

APR. MAY JUNE JULY AUG. SEPT. OCT.

OTOLITH READINGS ^ ®'® ^^'^ ^"-^^S

O 1937 YEAR CLASS

+ 1936 YEAR CLASS

9

Figure 8. Percentage of otoliths irith narrcw marginal increments.

71

40 to »0 TO »Q »0 K>0 IP

40 90 M

»o »o aoo >0 to 10

I-

|::

i H. ll . ■ ■ .1 ■ ■ .. 1 I I ■ . I I I I . I I ■ ■ ■

40 M to TO

Figtire 9« Shoiring the age composition (first three year classes only
are given) of 2,600 ^)ecijnens taken f^om California ssnQxLes
represented in figure 1* Ibe ages irere determined from scales.

72

s -

200 ■

T

I I I I I I I I I I I I I I I I I I I I I I I I I I I I I ' I I I I

0 »^

FREQUENCY DATA

I I I I I I I I I I

SAMPLES FROM
e YEAR
4 YEAR
• YEAR

CL^

SAMPLES FROM
FISHERY FOR

■ YEAR
i YEAfl

■ YEAR
B YEAR

I t I I I I I I I I I

BAIT FISHERIES
199 1938
t99 I9S7
199 1936

COMMERCIAL
<|DULTS

99 1938

1937

1939

1939
■■■■■■''■■■

I I I I I I I

_1

lll^ll§iSlriSaliig5iSlriliSSII§ig;~ISiliigSI£|:ililiiggis;s;Sil3l

Figure 10. Maan body lengths of year classes 1935 to 1938 pLottal
according to age, and to time of capture* Ages irere determined by
scale s| asBiples irere from the California bait fisheries, or from
ccxninercial fishery for adults.

73

U 0

z

^-7W\ , A^-^,

SEPT. ivsa

CALIF.

^^^/V^""^

100 no ItO 130 MO ISO

BODY LENGTH, MILLIMETERS

MO ITO in ito too

OPEN CURVE - I93B YEAR CCASS
auUX CURVE- lUT YEAII CLA39

Figure U* Comparing the body lengths of fish collected In California
with thcBe taken during the same months in Washington (Wash.),
Oregon (Ore*), northern Loner California (N.L.C.) ox* southern Loirer
CaUfomia (S.L.C.).

74

Figure 12, Above, pilchard scale with no annuli; body length of fish
63 mm. J caught June 22, 1938« Below, photomicrograph of a portion
of a longitudinal section through a pilchard scale to show the
surface folds.

75

Figure 13. A pilchard scale with no annuli. Body length of fish

1^ mm. J caught July 9, 1938*

76

Figure 1U» A. pilchard scale with one anniilvis. Body length of fish
137 nmi.j caught May 26, 1938.

77

Figure 15. A pilchard scale with one annulus, taken from under the
lower margin of the pectorAl fin. Body lehgth of fish 123 nra.j
caught April 17, 1939.

78

Figure 16* A pilchard scale with one annulus, which Is rather obscure*
Body length of fish 134 nun.; caught January 26, 1939.

79

Figure 17 • Pilchard, scale ?rith one annulus.

80

Figure 18. Pilchard scale with tiro amiuli. Body length of fish

180 mm.,} caught May 9, 1939.

81

Figure 19. Pilchard scale -with three annul! . Body length of fish 199 ran.

82

Figure 20. Another scale from the same fish as indicated in figure 19,
taken from \inder the lower margin of the pectoral fin»

83

Plgur« 21. Another scale from the same fish as indicated In figure 19 >

taken from the caudal peduncle*

84

■Figure 22. Pilchard scale with four annuli, and with one scar near the
center (F). Body length of fish 207 mm.j caught August 3, 1939.

85

Figure 23. Example of a poorly marked pilchard scale, having three
annuli, a false ring (F), and a scar (S) near the center, which
might be confused with an annulus (but see figure 24) . Body length
of fish 215 nim.j caught September 21, 1939.

86

Figure 2A» Another seals from the same fish as indicated in figure 23.
Note the false ring (F), and scar-like appearance of the center, where
a mark somenrhat resei±>ling an annulus appears*

87

lilgure 25» Pilchard scale having three annuli, mounted in glycerine.
THIhen mounted in any liquid medium, particularly those with high
refractive index, pilchard scales become transparent and the surface
structure loses clarity of detail. This effect increases with time,
and after a few weeks, a scale so mounted becomes so transparent as
to be almost completely invisible , and can then be detected only by
dose scrutiny at a sharp angle. Body length of fish, 215 nmi.j
caught August 3, 1939.

88

Figure 26. The same scale as shown in figixre 25, here mounted dry.

89

Figure 27. Two scales from a fish 111 mm, in body length, caught
March 15, 1939. In the top specimen are seen clearly one annulus
and two false rings, (F and S) one of which is easily confused
with an annulus. In the lower specimen, only traces of the latter
two marks persist, while the true annulus remains evident.

90

CENTER

Figure 28. A pilchard otolith with no annulus. Body length of fish

70 mm.; caught June 20, 1939.

91

Figure 29. Left, a pilchard otolith with no annulus and a false mark (F)
in the growth lone of the first year. Body length of fish 76 ram. 5
cstught June 30, 1938» Right, a pilchard otolith with one annulus
some distance from the margin, and with a false mark (F) in the
growth zone of the first year. Body length of fish 121 mm. Caught
Majjr 24, 1938.

92

Figure 30, Left, a pilchard otolith with two annuli and a false mark (F)
in the growth zone of the first year. Body length of fish I65 mm.;
caught January 18, 1939. Right, a pilchard otolith with two well-
defined annuli. Body length of fish 151 nmi.j caught February 7, 1939 •

93

Figure 31, Pilchard otolith with three annuli. Body length of fish

178 mm. J caught January 18, 1939.

94

Figure 32. Pilchard otolith with four annuli. Body length of fish
213 ram, J cau^t August 10, 1939.

95

3o Determination of Age of Adults by Scales, and Effect of
Eavlronment on First Year's Growth as it Bears on Age Determinationo

By

Lionel Ao Tlfalford and Kenneth Ho Mosher

CONTENTS
Preface Page

Int^^OdUCtiOno oooaouoooooooooooooooooooo y*

The Material = Its collectionj, preparation and examinationo o o o 98
ErrCiT of interpreting the annulus oooooooooooooooo 100
Recognition of the new anrmlias ooooooooooooo
Evidence bearing on the validity of scales for determining

o o . o 105

the age of adult pilchards ooooooooo

107

Effect of environment on growth of the pilchard, and its bearing

on age de termination oooooooooooooooooooo 112

Sranmary and conclusions oooooo.oooooooooooooo 116

i^i te^ature citei_io ooouoouoooooooooooooeooo xxo

96

INTRODUCTION

The objective of conservation research on the Pacific pilchard, or
sardine, has been stated by Sette (19'i-3) thus: "To determine for different
levels of fishing intensitj'' (i.e., effort), the quantity and quality (sizes
of fish) in the average annual catch." The statistical coiaputations in-
volved in this determination require a current measurement of the age
composition of samples of the comraercial catch. Owing to the close over-
lapping in lengths of fish comprising the several ages, it is not feasible
to deduce age composition or even to discover with certaintj?- the growth
rate by a study of length frequency curves. To attain these ends, it is
necessary to employ some other method of determining the age of individual
fishes.

The scales and otoliths of the pilchard, as isi many fishes, are marked
with rings Y/hich appear to be related in number to age. As usual ajnong
fishes, these marks are not diagramatically clear in the pilchard, being
of various degrees of obscurity, and recognition of those that are formed
only once annually (the age marks) can be easily confused by the presence
of those that are formed fortuitously, the adventitious marks. In a pre-
vious reporti' , the authors established criteria for distinguishing between
the two types. These were critically tested on .i^^venile material sampled
from the bait fisheries of California, and found to serve satisfactorily
for determining ages up to the third year.

It is now proposed to test the application of these criteria to the
scales 'Of adult fish as taken in samples of the commercial catch, and to
answer this question: lifith what order of accuracy can the age of a pilchard
be determined hy recognizing and counting the age marks on its scales?
The results of this test and the ansv;er to this question will furnish a
basis for judging whether age reading from scales nay confidently be used
for age composition records needed in the various studies on this species.

Thanks are due to our colleagues for their critical attention .ind
interest given during the development of this study; in particular, to
0. E. Sette, v.rho has suggested a considerable part of the methods that
are new. We are grateful also to Dr. H. U. Sverdrup and Dr. Vj. E. Allen
of the Scripps Institution of Oceanography, the former for his critical
reading of that part of the manuscript touching on oceanographic matters,
the latter for making available a large volume of unpublished data on
diatom abundance. Vfe are indebted, too, to Julius Phillips, of the Califor-
nia State fisheries Laboratory, for his suggestions, cooperation, and
criticism. The mounting of scales and a large part of the routine clerical
work were done by employees from works Progress Administration projects:
765-08-3-22, 65-2-08-286, and 265-2 -08-3U.

17

Walford and liosher, 19U3.

97

TH3 :.&T::RIAL: I?3 COLLEGTIOI.', PREPAiLiTIOIJ iUJD SXAiaNATION

The material for this study ■'.vas collected from the commercial fishery
serving reduction plants and canneries, as detailed in Table 1.

Table 1. -

Source, date, and number

• of ;

specimens

in scale

collections used for

this

study

No

. specimens Agency

Locality

Period

aged

collecting

California:

U. 3i4nd Calif J'

San Pedro-iKi-

1/10/UO

- 1/16 /UO

686

San Pedro-x-;:-

1/21/m

- I/27AI

761

U. S. and Calif.

Monterey •-;!■

9/20/39

-12/21/39

3072

U. S.

Monterey -;k;-

10/21/UO

-10/29/UO

998

U. S. and Calif.

Monterey -"-"-

I/2U/I4I

- I/30/I1I

2U1

Calif.

San Franc is co-;:-

9/lO/hO

- 2/Ul

35U6

U. 3.

San Franc is co^'^s-

8/UO

181

U. S.

Eureka -;w

9/UO

158

U. 3.

Ore con:

,

Astoria-;;-

3/7/UO

- 8/lo/Uo

liiO

U, 3.

YiTashington:

Grays Harbor -"-

7/19/39

- 3/17/39

789

U. 3.

Grays Harbor -;:-

8/11/UO

10

I: hJ

British Columbia

7/22/l;0

- 9/21/UO

832

The entire fishin<:^ seasons were sampled systematically by the Fish and
Wildlife Service at Grays Harbor, IVashington, in 1939 and 19H0; at Monterey,
California, in 1939j and at San Erancisco,. California, in I9U0. In I9U0,
the Fisheries Research Board of Canada collected scales at Barkley Sound;
these the board, through the courtesy of Dr. John Hart, contributed for the
present study. In addition, special sanples were gathered in collaboration
with the California State Fisheries Laboratory at lionterey in 19U0 and at
San Pedro in I9U0 and I9I4I. Other special samples were taken, as shovm in
table 1, f or various purposes. The specimens collected at Eureka, California,
were from pilchards from the stomachs of whales landed by the ViTialing Station
of the San Francisco Sea Products Company, and through the courtesy of the
Station, sent periodically to our laboratorj"-. In addition, scales collected
in I9UI - U2 in collaboration v.'ith the California State Fisheries Laboratory,
are drawn upon in this paper for illustrative or corroborative purposes.

In the regular collections a sample consisted of 30 fish, taken at, random,
without regard to size, from one boat's catch. As many samples were taken
daily as to occupy a man's full time in collecting, measuring, and tabulating.

~' Collections made in conjunction vfith routine sampling.
-;h^ Collections made for special purposes.
±.1 United States Fish and 'lildlifo Service.
2/ California State Fisheries Laboratory.
-^/ Fisheries Research Board of Canada.

98

Body length of the fish was measured as described by Sette (I9UI). For each
sample, scales were obtained from the first 10 specimens having any still at-
tached. Generally about 10 or 12 scales were taken from the side of the body
near the region touched by the tip of the pectoral fin. As necessity, required,
however, fewer were taken, sometimes only one, and from v/herevcr they could
be found.

As the scales were collected from each fish they were placed in a small
vial of water to which had been added about two drops of 2 percent phenol
for preservative. The vials were kept in covered trays, 50 to a tray, and
held in numbered places by round slots bored in the bottom. The collections
v;ere sent daily to the laboratory at Stanford University, where they were
mounted.

Generally the field man assigned to the sampling work was able to take
five samples daily, that is scales from 50 fishi'. In Washington, whenever
the landings were too few to permit collection of this number of samples,
scales were taken from enough spociraens to brin;^, up to that time, the daily
average to ^0.

In the laboratory'-, the scales vrerc cleaned and dried by rubbing between
the fingers, and mounted between tv/o dry, clean slides which viere then bound
together at the ends ^irith cellixLose tape. The mounted scales were examined
vfith the aid of a projecting microscope. For each specimen a paper strip
printed vj-ith a millimeter me.asure v:as laid on the image along the midlongi-
tudinal axis of the sculptured part of the scale with the zero line at the
base of the sculptured part. Locations of annuli v/erc traced on the paper
strip and later recorded as dimensions after reading the millimeter measure.

These as vrell as such pertinent data as sample number, length, and sex
of fish, locality and date of collection, name of reader, dimensions of scale,
position of marks, calculated length of fish at past ages, etc., were recorded
on a punch card, providing a permanent detailed record for every scale reading.

It required tv/o hours for one person to mount 50 scale samples^ two hours
for one person to read ^0, and about four hours for one person to tabulate
the data, calculate growth increments, pvinch the cards, and check these steps
for $0 readings. Thus the routine processes in the laboratory required about
eight man hours for one day's field collection from one port.

The readings Virerc all done by the authors, 'j/ith certain exceptions, to
be discussed below, each examined alternate samples of scales, Pfelford the
odd-numbered ones, Ifosher the even. 3y this method a check could be furnished
on each man's reading, since the two sets, on theoretical grounds, should be
satistically homogeneous if the operators used identical standards for inter-
preting annuli. The results of this check will be disciissed below.

Occasionally the scale collection from a specimen was obviously contam-
inated with one or more scales from other fish. V/hcn that occurred, the ma-
terial was discarded unless at least three quarters of the scales were obvi-
ously from the same fish, in v/hich case these were read.

1/ Beginning with the 19U1-U2 season, an off ort vras made to sample approxi-
mately one-fifth of the landings.

99

The criteria of an age mark as set forth previously (^Yalford and llosher,
I9U3) remains unchanged, namely;

"An annulus is concentric with the narf^in of the scale. It
is not always a sharp or unbroken line; nor are the segments
of an interrupted annulus always perfectly co-circular (if
the shape of a scale may be called circular in this discus-
sion). But the course of an axinulus, continuous or broken
as it may be, can usually be traced, by careful scrutiny if
necessary, entirely around the sculptured part of the scale
from left-hand to right-hand margins. Sometimes they can be
followed even around the unsculptured part. Annuli are
clearly separated from each other and do not ordinarily
meet at any point. If an annulus has formed, it is present
in all the normal scales of an individual."

In the scales of fish older than six years, the peripheral annuli are
crowded and frequently irregular, so that they cannot be made out equally
well entirely around their course, particularly in very old specimens. Oi
such scales, the annuli are generally most distinct at the antei'ior "cor-
ners" of the soulptured part of the scale (see figure 1). There they are
most easily located and counted, and thence they can be traced, even though
with difficulty, entirely around the sculptured part of the scale. The
criteria of a false year ring, given previously (Walford and Mosher, 19h3)
apply for older fish as well as for yoimger ones. These are:

"They are merely short or unassociated arcs; or if they com-
pletely circle the sculptured part, they are usually not
concentric with the margin; furthermore, they frequently
join an annulus at the base. In any case, they may be lesg
distinct than an annulus, being va2\ie and indefinite, or
they may be much more distinct, having a scar-like Appear-
ance; that is, v/ith very pronounced irregularities of pattern,
the folds being brbken or othein/ise obviously abnormal.
They rarely appear in all the scales of an individual."

ERROR OF INTERPRETING THE ANNULUS

Having established certain definitions of an annulus and of an adventi-
tious, mark, it is pertinent to learn how accurately they can be employed.
Expected errors are of two typea: those caused by faults of the scale and
those caused by faults of tht3 reader.

0f the first type are those cases in vitiich a scale fails to record a
year mark; also those in which an adventitious mark majr look enough like
an annulus easily to be mistaken for one. These cases of scale faults pro-
vide exceptions to our criteria of true and false annuli, and they are as
liable to the same interpretation in a second examination as are normal
scales. The percentage of errors arising from these causes probably can
never bo directly measured .

100

Of the second typo ar^ those cases in vfhich a reader simply mis-counts
the rings on a clearly marked scale; or those in which he overlookc one or
more unusually obscure marks j or those in wluch, by faulty judgment, he
miscalls an adventitious mark a true annulus. Th'jse errors are, in gen-
eral, the fault of the reader, and are relatively ■'insusceptible to dupli-
cation in a second reading, particularly by a diff'^rent reader. Therefore,
the results of parallel readings may be expected to provide a measure of
the personal error — the reader fault — in interpreting our criteria of year
marks and in counting them.

Scales of 1,187 specimens from the San Francisco fall fishery of 19hO
were read by Walford and by Ifoshcr, each \vorking separately and independently
Subsequent comparison sho^-ved that l,Oli? or" 88. percent of the pairs of read-
ings were identical. The 138 specimens for vj-hich readings were in disagree-
ment vrere then subjected to a third examination, this time by the two opera-
tors working jointly, and without reference to either of the two previous
readings. On this occasion, k specimens v/ere discarded as unreadablei/ ;
123 of the joint readings agreed with one or the other of the tvro previous
readings ("single disagreements'") J ^^^ 1^ were different from either
("double disagreements"), Jn 31 instances the "single disagreements"
concerned the innermost ring, in h9, the marginal ring, and in U3, one
of the intermediate rings. In t^^.rms of percentage these quantities are
25, ho, and 35, respectively. Since the average number of rings present
in specimens of this group is 11.2, those percentages should be close to
25, 25 and 50 respectively, if all rings WGro equally difficult to recog-
nize. Obviously, the marginal ring was more obscure than the others. In-
deed, during the joint readings, it v^as found that uncertainty about the
criteria for recognizing a newly formed ring at the margin during January
and February was the cause of several of the disagreements, and a special
study v/as indicated (see page 9 ).

A similar test was made by three workers, Tfelford and Mosher; and Phillips
of the Calif oniia St9.te Fisheries Laboratory. Scales from li83 specimens,
read first by Phillips, Vrcre subsequently divided into t\fo lots, of xvhich
one vras read by ITalford, the other by Mosher. The two sets of readings,
i.e., by Phillips and by Walford-Hosher, wore identical for U31 or 39 per-
cent of the scales. The 52 specimens for which readiogs were in disagree-
ment wore then subjected to a third reading, this time by the three workers
together. IVhen these joint readings vrerc compared with the original read-
ings, it v>ras found that throe disagreed vj-ith both Phillips and with IValford-
Mosher ("double disagreements"), and I4.9 agreed v;ith one or the other ("single
disagreements"). Disagreements concerned the first annulus in 13 of the cases
the ■■marginal annulus in 15, and an intermediate ' one in 21: or, expressed as
percentages: 26, 31, and U3, respectively. Thus there was no marked tendency
for any one annulus to cause more disagreement than iuiy other.

A third test v/^ith parallel independent readings was made by Walford
and Mosher on 973 specimens taken from lYashington. This sample contained
a higher percentagd of fish over throe years old than those described above.
For that reason, and also because the scales had not been viell cleaned vfhen
collected, this test v;as more severe than the others. Nevertheless the two
readers agreed in 709 or 73 percent of the cases. Of the remainder, 235
were "single disagreemerfts", and 29 were "double disagreements."

±/ Less than 0.2 of 1 percent of the scales had been discarded as unreadable
on first reading. These were specimens with regenerated centers.

101

Thus in three tests involving parallel readings of 2,6U3 specimens,
there was disagreement on U5U of the cases — "disagreements" — and agreement
in 2,1;^3 — "agreements." ViThen the former were re-examined by the two par-
ticipating workers, the final decision agreed with one or the other of
the two original readings in i;08 cases and disagreed with both in only U6.

In close to 90 percent of the cases, disagreement involved a choice,
of two adjacent ages, and, therefore, uncertainty as to the existence of
only one of the rings ; and only 10 percent involved uncertainty as to more
than one ring. Save for the first of the three tests, no one annulus was
the subject of disagreement more frequently than any of the others. For
the first test, disagreement over the marginal annulus, which was somewhat
in excess of expectation, was later corrected as a result of a special study
(see page 9 ).

A reader's decision as to which of two adjacent ages a "disagreement"
should be referred, might possibly be determined by chance. Assuming this
to be true, half the specimens should be assigned to the younger age, half
to the older. Consequently the size composition of fish comprising the
two halves should be statistically identical, and should agree equally well
or badly with the size composition of the "agreements" .of the two adjacent
ages in question. To examine this hypothesis, a series of chi-square com-
parisons was made, the results of which are summarized in table 2. The
tabulated figures give the probability of obtaining, in a second series
of samples, differences in size composition as great or greater, by random
error .

Table 2.- P values obtained on chi-square comparison oetween size
composition of agreements and revised disagreements.

* Final determination on revised disagreements

Agreement^ ,

t

Nurat

)er of rings

number of

I 3

! h .

5

! 6 :

7

8

! 9-

rings

3

.714(2)1/

.001(2)'

.

'

U i

.00U(2)

.26(3)

.001(5)

i

5 !

•

.ooi(U)

.16(7)

.001(5).

6

.0015(6)

.08(5) .

.ooi(U).

1

7

!

!.001(U):

.90(5) i

.ooi(U)

8

; :

.001(3)

.72(2)

.00(1)

9

•

1

.01(2) .

.60(1)

—' Figures in parenthesis indicate degrees of freedom.

102

According to table 2, the size distribution of the fish yielding
"agreements" on a given age, when compared v.'ith that of fish assigned
to that same age by revision of readings that had originally disagreed
("revised disagreements"), gave P values high enough to indicate that
the tviTo distributions had been of the same population in the statistical
sense. On the other hand, vfhen a parallel comparison was made v/ith ad-
jacent year classes, the P values were all so low that there could be
no question that the fish v/ere of different populations in the statistical
sonse. It appears,' then, that the final decisions on the disagreements
v;ere not determined by chance, but must have been, on the whole, correc-
tions of errors in the first reading.

It is, of course, not possible to Imov/ the absolute error of the
final determinations. Among the "agreements" can be erroneous readings
that agree by chance; but for scales of ordinary difficulty the number
of these is probably lov/, being something less than one percent of the
casesl/. In addition, there must be errors among the joint readings.

There may be more of these than are found among "agreements," since
only the more difficult scales tend to be the subjtct of disagreement.
P\irtherraore , the final decisions in the joint readings may be determined
often by the domination of one of the participants. Finally, there are
the errors resulting from scale faults, referred to above.

Errors from all these causes are not measurable by any knoinm methods.
The material and methods at hand provide merely this incomplete measure:
If B disagrees with A, A is not held to be in error if a joint reading
concurs vfith him rather than with B. Hence, an index of a reader's error
is the number of age determinations v/ith v;^hich there is disagreement in
both a parallel reading, and in a subsequent joint reading, expressed as
percentage of the number of specimens ^aged. For the three tests discussed
above, this index of error, by ages and by three readers, A, B, and C,
totalled as given in table 3.

Table- 3. - Index of

error

Number

! Number of

i Number of

t Index of error

of rings

! specimens read

t errors by

! by reader'

: by> readdr

! reader

A :

'B i C

! A : B

C

A i

B

! c

1

66:

66 : 26

1 1:2

6

1.5

3.0

! 23.0

2

272-;

225 :318 : 5:5

13

1.8 ,

' 2.2 .

U.i

3

862:

853 :112 : 22 : 39

8

2.6 ,

li.6 .

7.1

h

! 68U:

683 : 20 : U9 : 62 .

h

7.2 : 9.1 i

20.0

5

253:

235 J 7 : 28 : 29

1

11.1 : 12.3 !

11;. 3

6

16U:

181 : 0 : 12 : 33

-

7.3 : 18.2 J

-

7

62:

66 : 0 : lU : 20

-

22.6 : 30.3 1

-

8 i

. 31:

UO J 0' : li : 7

-

12.9 : 17.5 .

-

9 .

! 33:

27 : 0 : 9 : 10 .

-

27.2 ; 37.0

•2,U27:2

,376.:iiB3 i

lUU : 207

'J''^

S.9

a. 7

b.b

■=' This is based on the average knovm error of each man being close to
seven percent. The probability of both misreading any one scale on the
average, assuming purely ramdom error, ,is (0.07)^ = .OOI4.9.

103

Table 3 gives evidence that the index of error varies from age to
age, being higher for the older tlian for the yo^unger classes; and, there-
fore, the error on age composition must vary according 'to the number of
old fish present'. The index of error varies from reader to reader; also
for any given reader, from time to time, being more or less proportional
td the degree of his fatigue. It has been our experience that rreaders some-
times unconsciously adopt faulty reading techniques irhich result in their
making some types of errors consistently over a peribd of time. Thus,
Reader C (table 3) tended during the test to overlook the marginal ring
on scales with two annuli, and -consequently he recorded an excess of ' year-
lings. Faults of this kind ape.; detected and eliminated during the joint
readings .

To avoid such errors, frequent tests should be made. Indeed, it would
be desirable to make parallel readings of all scales wore it not impractical
in routine work involving several hundred specimens weekly. Fortunately,
at least 95 percent of the California .catch is composed of fish younger
than 5 years, for v/hich the index of error is low enough to justify sub-
stituting for parallel readings the procedure described below. For the.
remainder, the relatively high error necessitates making parallel readings
on all specimens.

In the light of the foregoing results, the following procedure was
adopted for future routine v/ork: The slides will be dealt among the several
readers like playing cards, so thas-^ all have equal-sizedj randomly distrib-
uted portions that should prove to be satistically identical as to body
length composition. That being true, the age compositions of the several
portions should also be statistically identical; and significant differences
among them should be detectable by the chi-square test, as described below.
IVherover such differences occur, the scales v/^ill -be re-read and the cause
of the differences found and cori'octcd. Mcanvvliile, specimens aged as five,
years or older iid.ll be given parallel readings; and joint readings will then
be made of the disagreements.

Follovfing thia procediire, the scale collections taken during the 19liO-
Ul season at San Francisco,. 3i210 in number, v^ere dealt between Vifalford and
Mosher. Tha1^ the t^TO portions were statistically similar in size composition
is indicated by a chi-square comparison, ivhich gave a probability of 0.53
that a second pair of samples dravm from the same population vro\ild differ
in size composition as much or more by chance alone.

For the same reason, if these tv;o random portions mqto alike in size
composition, they should be alike in age composition. A chi-square coiiparison
was therefore made between tho totals of each year class in Ifalford's portion
•and those in Lloshcr's. This test gave a probability of 0.13 that a sucond
pair of samples dravm from the same population would differ in ago composi-
tion as much or more by chance (chi-squaro vj-as 8,U Vi'lth 5 degrees of freedom).
Though high enough to be acceptable as evidence th^it Walford and Mosher had,
in the main-, read their respective lots of scales alike, it is at the same
time low enough to suggest the presence of sc«nc non-random differences between
the tvra scries of age distributions, due, possibly to some difference in the
interpretations of the two readers.

104

To learn then whether such a difference laight be centered about fish
of certain size categories, a chi^square comparison of age composition
V\ras made for each size class.

For this purpose, size classes vrere made as small as possible — usually
two millimeters — but lar<];e enough to contain, for each member, a minimum of
ten specimens. The results of this analysis indicated no significant dif-
ferences betvreen the two readers — all P values virere well above .05 — except
for eight classes in the size range 211-226 millimeters. For each of these
was indicated a probability of less than 0.001 that a second pair of samples
from the sajiie population could differ as much or more by chance. Accordingly,
scales of the 1,131 fish in this size range vrere reexamined, the two readers
exchanging their respective portions so as to check each other's first inter-
pretations. It transpired from the subsequent joint readings of the disagree-
ments, that the chief cause of disagreement had been confusion as to the
criteria for distinguishing betvreen old and neiJ- rings .at. the margin, par-
ticularly during January and February, v/hen accelerated grovrth is exposing
the nev;-ly formed annulus.

This difficulty vfas at length resolved by a special. study as described
on pages 9 to 11, which provided an objective method of making the neces-
sary distinction. After the method had been applied, the two original por-
tions, Vfelford's and Mosher's, >vere again compared as to age composition,
by the chi-square test. On this occasion the probability v^as O.kh that the
two distributions would differ as much or more by chance in a second pair
of samples drawn from tho same population.

Thus an equivalent of parallel reading has been devised, v/hich saves
considerable labor, yet serves to detect and correct errors in the inter-
pretation of tho critv.ria of age marks. The method was applied during the
19U1-U2 season, when three persons were employed. The average P value ob-
tained on first readings was .U8,. indicating no significant difTerence in
results among the three readers^;/.

RECOGiJITION'OF THE NST Alff^LUS

In a previous paper (L'alford and Mosher, 19h3), it was shown that the
nevf annulus appears on the scales of young pilchards, up to three years of
age, from late fall to early spring. Judging from the present data, it ap-
pears at the same season in the scales of adults. Because commercial fishing

1/ It has been suggested by those critically reading the manuscript of this
paper that the above discussion tends to over-simplify age determination of
pilchard scales. It must be emphasized that a considerable degree of' judgment
must be exercised in recognizing age marks. This reauires skill obtained
from experience illuminated by Imowledge of the biology of groirth. It also
requires much patience and a certain sympathetic attitude tbvrard scale reading.
Nevertheless, the evidence described on pages llff v/as obtained despite the
subjective nature of these imponderables. It appears, however, that age
reading cannot be practiced equally successfully by all investigators.

106

for adults is cari'ied on only from late July or Aunjust to February or March,
and fish in sufficient quantity for study are not available during the inter-
vening time, it is not feasible to follow an entire year's cycle for adults
as it was for the young. In July, however, v;hen specimens were first taken,
the last ring was found to be fully formed in all exarained, and was completely
surrounded by a narrow marginal growth zone; -and no specimens were found
with developing new rings. During succeeding months the margin beyond the
las f ring seemed to increase in width, and by late fall nev; rings began to
appear in a few specimens. Iiiliile these were in the process of formation,
they could be easily distinguished from old rings by the follo'v.'ing criteria:

an annulus is recognized as forming v/hen it does not appear entirely
around the sculptured part at once, but generally only on the most distal
parts of the edge (see figure 3)j or, when it extends entirely around the
sculptured part, it touches the edge at one or more points.

These criteria seemed to serve for distinguishing new rings -until about
early January. From then on, however, an increasing number of specimens
occurred vfith the last (subraarg:'jial) ring remarkably close to the edge of
the scale, but complete, and entirely bounded by a marginal zone. For these
the above criteria could not be applied. Instead, at the time of first
reading, the operators tried to judge whether they were new by v/hether or
not they v;ere at any point of their course obviously closer to the edge
than could be accounted for by a year's growth. Unfortunately, there are
intermediate cases v«-hich could be attributed equally vrell to a year of ab-
normally slow growth or to a fe-v; weeks of abnormally rapid grovrth. Conse-
quently, the decisions on these cases rested more or less with chance, and
as shorm by the studies on parallel readings, this was one of the chief
causes of disagreemem; among the readers.

In an effort to devise an objective standard for distinguishing new
from old rings, therefore, a statistical study v/as made on the width of
the marginal growth zone. One "vvould expect the latter to be narrowest shortly
after a new ring has formed | an^ to increase thereafter in width during
the year, reaching its maximum just before the follovdjig new ring appears
(Walford and Mosher, 19U3)« Accordingly, tho distance between the last
annulus' and the edge of the scale — the marginal zone — ^was measured to the
nearest 0.2 millimeter of projected image (equivalent to 0.0066? millimeters
in actual width), on all scales taken from July I9J4O through February I9UI.

These- data wjre tabulated by number of rings (not by assigned year,
classes), according to width of margins, in the form of a frequency distri-
bution. For example, all scales having fojir rings wore tabulated together.
It ivould be expected that early in the season, they would all be of year
class 1936; later in the season some of them (those with new rings) v/ould
bo of year class 1937 • The tv^ groups vrauld be expected to have the same
relative variation about their respective moans, and to put this on an
absolute basis for purposes of graphical study, sizes v/ere converted into
logarithms, which were tabulated into classes of equal intervals (Schrek
and Lipson, 19Ul). In figure ^> the results of this transformation are given
for scales with four rings, as an example of a case of average difficulty.
A dominant mode to the right, v/hich can be follo^/cd through the season, is
taKen to represent the margins attributable to the past season's growth;
and a nuw mode to the left, appearing in November, probably represents margins
attributable to the next season's grovfth.

106

These data have been fitted with nomal pi'obability curves, according
to a graphical method devised by 0. E. Sette (unpublished manuscript). A
single normal curve fits the data for July-August and for September-October.
This group is assi;^ned to year class 1936. It obviously persists thi'ough
the season, retaining certain consistencies, namely, first, a mode v/hich
advances slightlj'', indicating slov; autumnal scale grov/th; second a standard
deviation v/hich remains constant. For November-December, and for January-
February, it V7as necessary to add a second, overlapping, c^irve on the left.
This group is assigned to year class 1937. As in the right-hand group,
this too -has the characteristics of a mode v;-hich advances, as is to be ex-
pected, from scale growth, and a constant standard deviation.

To measure the degree to vrtiich the selection and placing of these hy-
pothetical curves fit the empirical data, comparison was made by the chi-
square method. A P value of .06^/ was obtained, (chi-square = 30. U; degrees
of freedom, 19), of borderline significance.

Chief contribution to the chi-square comt^: from the sixth to ..jjith
classes. It appears, then, that the low P valu i first obtained was due
to minor irregularities of the data, rather than to selection and placing
of the hypothetical curves. If these be combined, the seventh class with
the sixth, the ninth with the eighth, P becomes .^6 chi-square 1$.^9; de-
grees of freedom 17).

This, then, is an objective means of allocating proportions of new
to old rinf3;s v/here the marginal J.ncrements overlap. In the four-ring
series (Fig. 1) allocations were made as follows for^ November-December:
To the older agu, l/6 the specimens in size-class 7, 3A i" size-class 8,
3U/35 in size class 9. The remainder in these classes were assigned to
the younger age. In January -February, to the older age were allocated
9/26"^ of the specimens in size-class 8, 31/33 in size-class 9. The remainder
in these classes wore assigned to the younger age.

The assignment of ages to individuals in these critical cases by the
method just described, precludes using those indi'/iduals for other studies
involving sizes, ages or growth, and the statistical cards vrere marked
accordingly.

This method of solving the problem of distinguishing between^ new and
old rings has been incorporated into the routine of age analysis in pilchard
research.

evidence bearing on the validity .of scales for
deTjiRI-HNING the age of adult Pilchards

The forugoing section has dealt exclusively with the question of how
correctly one or more scale readers can assign a given age to a given fish,
or to the measurement of the "reader error". la the following discussion
the effect of this error plus that of the scale error (cf. page U ) will
bo examined.

-^ I.e., the probability is .06 that a second series of empirical data
vrould differ as much or more from this theoretical series of curves, by
chance .

107

The annual occurrence of age marks in pilchards of juvenile age had
been proved in the follo"King wayl/: A stratified random samgle of scales
was prepared to represent equally all the months of the year. The labels
on the slides vrcro then masked, the slides mixed, and the scales examined
and measured vifithout Icno'i/ledge of the size' of th.i fish or the time of year
collected. ViTien the measurements thus obtained vere collated, it uas found
that the marginal increment vras narrov.'er at one season — the apring—than
at any other, and increased in width during the remainder of the year. Its
groi/rt-h was rapid in spring and early summer follo;/ing the formation of the
annulus in the i.'inter, and slight from late sumracr to v/inter.

The foregoing demonstration was possible because young pilchards are
available near shore in California throughout the year and aru taken by
the bait fisheries. "Adult'' pilchards, on the other hand,, arc migratory,
departing for thuir spai/ming pilace at the time of maturity, in the spring,
and not reappearing until raid-summer. Hence, the commercial fishery for
adults is seasonal, and the scale material collected for this study does
not include an important period of the year's groirth. Despite this hiatus
in data, thu results aiscussed on pages 9ff and given gr^iphically in figure S,
indicate, as far as they go, that in adults, as in the young, the annulus
appears during one season of the year — the winter, and that the marginal
increment increases in width until the next annulus is formed.

For critical proof of the valid:j.ty of scale reading, the annual groivth
increments on the scales were studied in search of peculiarities in dimen-
sions, marks or other- Irregularities associated consistently with certain
year classes such as Lea had discovered for th^ Nor^vegian herring (Lea 1919).
To provide homogeneous data, an effort v;as made to select scales consistently
from the same part of the body, namely the side, near the tip of the pectoral
fin. Unfortunately such selection vras unfeasible -viti inever the fish sampled
were in poor condition. Under these circumstances, scales were taken v/-herever
they could be found, often from the back or near the caudal peduncle, where
pilchard scales are characteristically smaller than elsewhere » For adjust-
ment to a common basis to permit comparative studies, therefore, scale mea-
surements were translated into terms of body length of the fish, by means
of Lea's formulae:

^n ^n

3- = pT- , and (1)

^n - ^n-1 = "^n (2)

Tifhere s^ is the distance from the center of the scale to a particular annulus,
n; S the distance from the center of tht; scal^j to the edge of the scale;
fn the length of the fish at the time the annulus n was formed; F the length
of the fish at the time of the observation; t^ the growth increment in the
year preceding the time the annulus n was formed. Hence, t =• f, ; ^2 = ^2~^\>
etc. The symbol "f" is the same as Lea's "1"; it abbreviates the term "fish
length" as "s" does "scale length".

T7~~

-' I'.alford and Uoshcr, I9I43.

108

This formulation assumes simple, perfect correlation between body length
and scale length, with a regression having its origin at zero. It was found,
however, that the assumption is no more valid for pilchards than for many-
other fishes studied. The scales are not formed until the fish has reached
a length of 3 to U centimeters. For a time thereafter, until the body be-
comes fully scaled, the gi'owth rate of the scales is rapid in relation to
that of the body; then it decelerates, until at length the slope of the re-
gression, b, becomes constant. This. occurs nelov* the minimum length of
fish found in our samples. Hence, for the material used in this study,
the relation of scale length to body length may be expressed by a simple
straight line regression of the type y = a + bx, and Lea' 3 formula (1) must
be modified to:

fn ^ in

S F + a

(3)

For calculating a, the regression of scale length on body length was
plotted separately for each of four year classes, those of 1936 to 1939,
to Tihich specimens had been allocated by scale studies.

The values of a and b were as follows :

Year Class a b

1936 -2.2378 .7326

1937 0.8U6U .7285

1938 -20.3317 .8U05

1939 2.081U .72UO

Significance of differences, in terms, of P= (according to Fisher's t
test) is as follows:

Year Class

Compare

id v/lth Year

Class

1939

1938

1937

1936

a

.30-.U0

.01

.li0-.50

b

.50

.01

.50

1937

a

.50

.01

b

.50

.01

193"8

a

.01

5 .o5r.io

In regression of scale length on body length, year classes 1936, 1937,
and 1939 did not vary from each other beyond the range of random errorj but
year class I938 differed significantly from each of the other three, both
as to a and b.

y That is, the probability that if an additional pair of samples be drawn
from the same population, they would differ as much by chance as the two
under comparison, or more.

109

Thus the fish assigned to one year class had an intrinsic quality in
coCTnon in addition to the number of rings on their scales, peculiar to their
class, vifhich persisted during three years of this study. It is fortunate
that this peculiarity applies to year class 1938, for io is that class on
which the most complete collection of material is at hand, going back to
first appearance in the bait fisheries, when identification of its age as
fish of the year was virtually beyond question. And it is that class from
which the material vras drawn on which the validity of age determination of
juvenile pilchards was proved (V/alford and Mosher, 19U3). Thus fish iden-
tified with. certainty as belonging to year class 1938 have been marked by
Nature with a characteristic that may continue to serve for distiiiguishing
them as a group for several additional years.

Since the relation of scale length to body length can vary among year
classes, it is necessary, for approximating past grovifth as accurately as
possible, to determine a for each year class separately. To do this requires
measurements over a sufficient range of sizes - therefore years of growth -
to provide a significant measure of the regression. ViTith present material,
this vfas possible only for the four year classes mentioned above. Consequently,
in the folloviring discussion, averages concerning those year classes alone
have been adjusted as of formula (3) (page 13); and with the appropriate
a values (page 13 ). For the rest, formula (1) has had to suffice.

Summarizing: To dei^ect important anomalies in the past growth of the
several year classes, the lengths at the past 'ages (f ) were calculated by
Lea's formula (1). These were averaged, and averages for year classes 1936
to 1939 v;^Gre adjusted to take into account departures from exact proportion.-
ality of scale' length to body length; in other words for the failure of the
regression of scale length on body length to intersect with coordinates
0,0.i/ Valu-J3 of X^ Y/ere obtained by subtracting Tj^ from Tj^-1.

The results of thuse calculations for the first year's growth increment
(^ ), given in Table h, are remarkably variable. Among year classes 1932
to I93O collected in 1939-i;0, for example, the average first year's growth
increment (lq_) panged from 98 to 115 mm. Such apparent inconsistency could
be caused by changes virith age and size in the scale length-body length re-
lationship. It could also occur if increasingly larger portions of the size
range of each year class' became available to the fishery as the year class
became older. Or it might result from systematic errors in marking the lo-
cation of the annuli, vihich were somehov; associated •yv'ith age.

-' These adjustments -v^ere made graphically as follovirs: The average total
length of a year class under consideration v/as located on the scale length-
body length regression line, and a line dravm from there to the origin of
the graph.- Each f viilue for that year class v/as then located on the latter
line; and the intersection betv/een, the ordinate passing through that point,
and the scale length-body length regression line gave the adjusted -value of
that T.

110

Table U, - Average calculatca first grov^th incroni<:;nts , by year classes

Year
class

Averagfe calculated length for fish cnllected In season of -

1939-hO

Unad

HSl

Aa.i«^

2r

19U0-U1

ir.

Vnal1t~^' A-s1t-

TT

Unadi.J=/t Ad,1-.£

-2r

1932
1933
193U
1935
1936
1937
1938
1939
19U0

111.83
llU.02
102.22
10U.U9
97.95
113.91
115. lU

99. U
llU.U
123.6

108.97
115.20
107. OU
107.62
101.98
113.76
108.93
102.66

103.0 '•

llh.l «

118.5 «

102.0 *

110.00
111.26
106. U8
103. 81|
102. 7U
112.03
99.89
92.91
116.73

10U.3

113.6

113.0

92.0

i' Unadjusted figures.

£/ Adjusted for deviations from exact proportionality of scale length to
body length.

It is reasonable to expect that all thoso causes •vJould be progressive
and rather regular in their effect, resulting In constant increase, or decrease
in comparable values of t with increase of age. Also, this trend should be
continued in the collections of later years.

,It is apparent at once from table U, hovfever, tnat far from progress-
ing regularly, the comparable values of t fluctuateci' with no evident order
in each of the three years of data studied. Thus it seems improbable that
the fluctuations could be associated vdth age. On the other hand, it does
appear that they are associated v.-ith year classes. If so, any given growth
increment (^) of a given year class should deviate from a norm of that in-
crement in the same direction every yoar, as long as that class is wholly
and significantly represented in the samples v;:th respect to the increment
in question. The meaning of this will become clear presently.

To establish a norm for "^j^, the 1- values for year classes 1932 to
1938, inclusive, v;ere averaged, using €ho material collected during the
season of 1939-UO, and these grand averages vrere taken as norms for that
season. Deviations of the 1^ values for individual year classes were then
plotted as curve number one in figrore 6. The same procedure was follovrcd
VTith the material collected in the next sodson (I9UO-UI), furnishing a
second curve in figiu^e 6. Although an additional year class, that of 1939,
was present, it did not enter into that season's norm; but the deviation
of its ^1 value from the norm is indicated by a dotted line. Next the ma-
terial collected in 19U1-U2 was similarly treated and the results plotted
in the third curve of figure 6. Year class 1939 was not included in the
norm, but deviations of its Tj value is indicated in figure 6 by dotted lines.

Ill

TheSG thrive curves arp closely simil.-ir. In all tnroe, the first year's
grov/th increment for year classes 1933, 1937, and 1938 deviates above the
norm. In six out of seven points all throe curves are in agre-ement. The
probabilitj'- that such agreement could occur by chance is slight. i/ In the
absence of random occurrence, then, the agreement must mean that specimi'^ns
having certain growth characteristics in common were allocated consistently
to the same year classes tliree years in succession. It remains to be shown
iThether the allocation vfas correct.

EFFECT OF ENVIROMLIEOT ON THE GRa'/TH 0? THE PILCHARD,
AND ITS BE.IRING ON AGE DETER!IINATION

If the anomalies described in the preceding section are a characteristic
of the year classes, they must reflect some varying environmental condition
that influences grovrbh of body and scales. The most obvious- such element
is temperature. Surface temperatures of the sea at several points along the
coast have been compiled by the Scripps Institution of Oceanography, and have
kindly been made available for this study by the Director, Dr. H. U. Sverdrup.

Mean temperatures, at these points, for calendar years and for various
combinations of months for years 1932 to 193B inclusive, were averaged, and
deviations computed in the same v;ay as had been done with the growth incre-
ments. In all combinations of months tried, the deviations from the norm
are negatively correlated with the anomalies in the. calculated first year's
growth. The means of temperatures measured at Scripps Institution Pier for
the period June to August, however, had a higher negative correlation than
those for other localities and other combinations of months; and deviations
from their norm are shown inverted as the fifth curve of the Series in fig-
ure 6. The months represented are those in which growth of young pilchards
is most rapid (cf. lui'alford and Mosher, 19U3).

Of more fundamental influence than the temperature on the marine environ-
ment, hooce on biology of the pilchard, is wind. For it is .the northvrest
winds prevailing along the California coast during spring arid summer that
produce the phenomenon of upwelling. This results in the transport of nu-
trients to the surface v;aters, making them available to the plankton organ-
isms» Thus, Tiltimately, it results in the production of food which the pil-
chard eats .3,/

i/ The deviations given in the data collected in 1939-UO (top curve o:f fig. 6)
are correlated with those corresponding in the data collected in 19iiO-Iil
(second curve of fig. 6), with an r of .93 J and P of less than .01, according
to Fisher's t test. ""

_/ Temperature data were studied for the following localities: Scripps In-
stitution Pier at La Jolla, Lat, 32° ^2' N.j the Pier at Balboa, Lat. 33° 36' Nf
Huenemo, Lat. 3U° 9' N.; Pacific Grove, Lat. 36° 33' N.; North Farallon Is-
land, Lat. 37° U2» N.J Blunt «s Reef, Lat. U0° 2?' N.

3/

- For a discussion of the relation between wind and the circulation along

the Pacific coast, see Sverdrup, 1938.2, 1938.3; Sverdrup and Allen, 1939.

112

Lacking available records of offshore wind force, a series was derived
from barometric pressure data given on daily weather maps published by the
United States ''.leather Bureau (cf. Sverdrup, 1936.2). First were tabulated
the daily barometric pressures at each of the four corners of a 300-mile
square placed off the coast of southern California (cf. table 6). From
these data were obtained two series of differeu^'.es representing pressure
gradients which vfould be expected to produce winds parallel and at right
angles to the coast line, respectively; and their differences were- summed
by periods of time as described below.

The resiolting two series of sums wer«j treated, with due regard to the
effect of the earth's rotation, as if they were conponents, the resultant
of which is proportional to direction and extent of wind movement. The
direction proved to be northwesterly for each period of the entire series
studied.

Sums w^ere made, as indicated above, for calendar years, and averaged
for the period 1932-38, inclusive, to furnish a norm from which deviations
v/ere determined, as had been done with .growth and temperature data, discussed
above. The same pr.ocodurc was followed using various periods of time and
various combinations of months. Without exception the yearly deviations
from normal indicated -v/ind force were found to be correlated positively with
anomalies in the first year's growth of pilchard. Hov,'ever, the period April
through the following torch gave a somewhat better correlation than other
periods, and has been shovm in the fourth curve of figure 6. This period
corresponds with the first y<iar's grov/th as registered on the scales: i.e.,
between the time of spavming and the time the first annulus is formed.

Other data bearing on the environment of the pilchard,- made available
by the Scripps Institution of Oceanography, are average monthly svirface sa-
linities for the sane localities as cited in footnote 2, page l6. Those
taken at the Scripps Institution pier v/ere treated In the same way as de-
scribed for the average monthly temperature records (page l6). Deviations
from a norm based on the twelve-^onth period April to the following March -
the same as that used for v;ind force - were plotted as the sixth curve of
figure 6. This curve bears a strong similarity to the other five.

Thus is indicated a positive correlation between the force of northirest
vdnds, the salinity of the surface v^ater, and the first year's gro^rth of
pilchards; and a negative correlation between the latter and surface tem-
peratures. To have demonstrated this required that the year classes, as
determined from scales, be allocated correctly to the years in which they
had been produced. It required also that a high enough order of accuracy
be maintained in the scale reading to preserve the distinctive character
of the curves.

A further conclusion to be drawn from these curves is that the fish
of a given year class normally have a common history during their first year.
They are subject to the same vicissitudes of marine climate and therefore
probably also of food supply.

113

According 'to Lgvo-S (1929), pilchards feed on diatoms and dinoflagel-
lates, as v/ell as upon copepods and othdr zooplankton. Moreover, the amount
of food ingested is positively correltited with the abundance of surface
plankton in the surrounding water. This in turn is controlled by tempera-
tures j "low temperatures associated virith upvelling of subsurface v/aters
favor the grovrth of diatoms, which in turn attract the sardine" (lewis,
op. cit., p. 179). This being true, it is reasonable to expect grovrth
in abundance of diatoms to favor growth of pilchards.

For study of this question. Dr. \f. E. Allen has kindly made available
statistics on relative diatom abundance, which he has observed for several , /
years at the end of the Scripps Institution pier at La Jolla and at Heuneme.-'
Numbers of all species of diatoms per liter per vreek were svimned by months;
and the monthly totals siiramed in various groupings of months. No significant
correlation bet^reen anomalies in these totals and anomalies in growth was
found, no matter what grouping of. months v^as used. Hov;^ever, the niunber of
weeks in the year that the count of diatoms per liter exceeded 1,000, or
2,000 or 5,000, gave measures 6f diatom persistence, all of which shov/ed a
positive correlation v/ith growth. As with the temperature and salinity data-,
these measures were summed by various combinations of months. The most satis-
factory colrrelation was given by the niomber of weeks between June and the
following May, inclusive, that the count exceeded 2,000 per liter. Devi-
ations from the norm of that statistic taken over the period 1932-1938 are
given in the seventh curve of figurc'6. The period used, June through the
following May, lags two monthSv^behind that used for \7ind data, and may in-
dicate the time required for the actich set up by the northiifest wljids to
affect the supply of diatoms.

Although the abundance of diatoms thus appears to affect grovfth directly,
it is a sustained abimdance of these plants above a certain minimal level
that is most effective in promoting growth, rather than sporadic flowerings
of great abundance interspersed with periods of scarcity.

The relations betvfeen grcftvth and other factors shovm in figiire 6 might;
not necessarily mean that pilchards normal]^ spend the grov^ing part of their
first year of life near the locality where these data were collected. How-
ever, second and later growth increments v/ere found not to be correlated
with wind force, surface salinities, and temperatures measured at La Jolla.
This may be explained in part by the fact that pilchards migrate successively
greater distances as they grow larger, the extent of their migrations prob-
ably being a function of size more than of age. Consequently, the history
of members of the same year class becomes less uniform as the fish grow older,
though they probably never leave the Pacific coast save for soavirard migrations
of not more than 300 miles. This is all consistent with the conclusion that
the bulk of pilchards in a year class do normally spend the first year of
life in or neaf southern California, and that the subsequent history of the
fish becomes so diverse that year-round conditions in any one locality can
no longer reflect growth among the whole population of that year class.

-' For the method of sampling, see Allen, 1936.

114

The most striking inconsistency in figure 6 is the lack oi' correlation
for 1939 between growth on the one hand and diatom persistence and surface
salinity on the other; also between surface temperature and surface salinity.
Although the surface watjr at La Jolla was abnormally warm in 1939, it vras
also of abnormally high salinity; and though the diatom count vras high, the
average growth of fish of the year was below nornKil.

The breakdovm of correlation between temperature and salinity suggests
an important anomaly in the current pattern in 1939 in the vicinity of La Jolla.
The relatively i/eak northvtrest -winds prevalent that year are consistent v/ith
relatively weak upwclling; and the increase in salinity and temperature can
indicate an increase in the admixture of "southern water" (cf . Sverdrup,
1938). The biological evidence of a change in hj^'drograp hie conditions is
equally suggestive. Certain species of diatoms normally occurring several
weeks of the year at La Jolla were absent in Dr. Allen's samples of 1939,
notably, Chaetoceros curvisetus, Chaetoseros laciniosus, Licmophora lyngbyei,
and Lithodesmium undulaturit Among species occurring more frequently in 1939
than normally, those most noteworthy v/erc Rhizosolenia alata, Rhizosolenia
fragilissima and Rhizosolenia irabricata. The ecology of these species is
too little known to permit of interpreting their occurrence in terms of physi-
cal oceanography. It may be of significance, however, that among species
of the first group mentioned above, all have been listed by Cupp and Allen
(1938) as being south temperate neritic . forms , except Licmophora lyngbyei,
which is "tychopelagic". Among those listed Ik the second group ,"Trhizosolenia
alata is an oceanic temperate form? and the other two are north temperate
forms .

More direct evidence that a change in the environment of the pilchards
took place in 1939 is -the fact that an abnormally high production of yoiaig
pilchards occurred that year in v^raters far to the north of southern California,
v^hich is regarded as the usual chief spavming ground. This may have been
the result of abnormally high survival of young hatching from the number -of
eggs normally spavmad north of California. It may also have been the result
of abnormally heavy egg production in northern \/aters, permitted by a north-
ward extension of oceanographic conditions favoring spavming. In support of
the latter hypothesis is the fact that adult pilchards taken by fishermen in
mid-summer off Oregon and Vfashington were full of spawn, in contrast to their
usual immature condition at that time. That spavming actually occurred north
of California is proved by the fact that the Fish and vaidlife Service that
year collected pilchard eggs and larvae in tow nets off the Oregon coast
(Walfbrd and Moshcr, 19i|l).

ViThatevur may account for it, an extraordinary rtumber of young pilchards
of year class 1939 were subsequently observed and caught from Oregon to
Alaska, as well as along the California coast. It has been previously ob-
served that the first year's grovfth decreases nortbvard (Walford and Mosher
I9U3). Thus the northv^ard extension of the range of year class 1939 could
alone account for its abnormally low first yearns growth in spite of the
apparently favorable food supply in southern California.

115

suiaaRY km coiiclusions

Annual rings appear on the scales of adult pilchards, as in the young,
and at the sane tirac of year, namely, the v>rinter and early spring. If the
scales be properly prepared, those rings can bo recognized and distinguished
from false age marks by means of criteria set up in this study. The ac-
curacy of following these criteria differs among readers and for different
ages of fish being lower for older ages than for younger ones, but for the
age distribution prevailing from 1939 to 19U2, and for the persons reading
the scales in this study, it averaged near to 93 percent.

The most important evidence supporting the validity of age determina-
tion by scales is that the fish allocated to each year class had growth char-
acteristics In comraon, v/-hich appeared consistently in three seasons of sam->
pling. These characteristics were expressed as the average first year's
growth increment, calculated from scales. Departures of these averages
from a norm v;ero significantly correlated with anomalies in certain" elements
of the environment, thereby proving that the identification of the year
classes had been correct.

The correlations indicated that, in general, grovfth during the first
year is favored by a sustained presence of diatoms v at optimum abundance.
In turn, the latter is favored by upwelling, v/hich is induced by northwest
T/inds, and accompanied by low surface temperatures and high surface salinity,
at least at the locality of observation.

In addition to an abnormally high first year of grov.'th, year class 1938
was marked by having scales v;hich averaged smaller than normal in relation
to the length of the body.

No mtthod has yet been devised of measuring the error of scale reading
occasioned by the scales being irregularly marked; i.e., having too many or
too feviT age rings. If this error were unduly high, however, specimens as-
signed to a given year-class would actually be composed of several ages;
and if such nixing be complete, the average growth of the several year-classes
should be uniform. In that event, the correlations referred to above would
have become obliterated. The demonstration of the correlations, and their
persistence for three seasons is evidence that the error vras reasonably lov/,
we judge, for the age composition prevailing, something of the order of ten
percent or less.. Thus, the relatively high accuracy of age determination
of pilchards-'by scale reading — betvreen QO and 90 percent — and the fact that
results "are reproducablc by different persons, warrants the confident utiliza-
tion of this technique in the program of pilchard research.

LITERATURE CITED

Allen, Winfred Emory

1936. Occurrence of marine plankton diatoms in a ten-year scries of

daily catches in southern California. American Journal of Botany,
vol. 23 (1), pp. 60-63.

116

Cupp, E. E. and, 17. E. Allen

1'938. Plankton diatoms of the Gulf of California obtained by Allan
Hancock Pacific Expedition of 1937. University of Southbrn
California, Allan Hancock Pacific Expeditions, vol. 3 (5),
pp. 61-95, 15 plates.

Jacobs, Yfoodrow C.

1939. Sea level departures on the California coast as related to the

djTiamics of the atmosphere over the north Pacific Ocean. Journal
of Marine Research, vol. 2 (3), pp. 181-1 9U.

La Fond, Eugene C.

1939. Variations of sea level on the Pacific Coast of the United States.
Journal of Isiarine Research, vol. 2 (l), pp. 17-29.

Lea, Einar,

1919. Report on "age and growth of the herring in Canadian waters".
Canadian Fisheries Expedition, I9II4.-I5. Department of Naval
Service, Ottawa, pp. 77-l61i.

Lev/is, Ralph C.

1929. The food habits of the California sardine in relation to the

seasonal distribution of microplankton. Bull. Scripps Institu-
tion of Oceanography, La Jolla, California. Technical Series,
vol. 2 (3) pp. 155-180, 2 figs.

Schrek, Robert, and Henry I. Lipson

.19^1. Logarithmic frequency distributions. Human Biology, vol. 13 (1),
pp. 1-22, figs. 1 - 5A.

Sette, Oscar E.

I9UI. Digit bias in measuring and a device to overcome it. Copeia,
19i;l (2), pp. 77-80, 2 figs.

19U3. Studies on the Pacific pfilchard of sardine. (Sardinops caerulea).
1. - Structure of a i-esearch program to determine how fishing
affects the resource. United States Department of the Interior,
Fish and V/ildlife Service, Special Scientific Report No. 19.
Mimeographed, 27 pages, 3 figures.

Sverdrup, H. U.

1938. .Research within physical opeanography and submarine geology at
the Scripps Institution of Oceanography during April 1937 to
April 1938. Transactions of the American Geophysical Union,
Nineteenth Annual Meeting, pp. 238-2U2, h figs.

1938. On the process of upvelling. Soars Foundation, Journal of
Marine Research, vol. 1 (2), po." l55-l6U.

1938. Oceanic circulation. Proceedings of the Fifth International
Congress of Applied Mechanics, 1938, pp. 279-^93, I6 figs.

117

Dvordrup, H. U. and M. Ji. Allen.

1939. Distribution of diatoms in relation to the character of v/ater
nasses and currents of southern California in 1938. Sears
Foundation, Journal of Marine Research, vol, 2 (2), 1939.

ITalford, Lionel A. and Kenneth Moshor.

19hl. Extension of pilchard spavming to nortn Pacific waters indi-
cated. Pacific Fishoman, February, 19U1, p. U7-

I9I43. Studies on the Pacific pilchard or sardine (Sardinops caerulea).,
2.- The determination of the age of juveniles by scales and
otoliths. U. 3. Department of the Interior, Fish and Wildlife
Service, Special Scientific Report No. 20. 'Mimeographed, 3U pp.,
32 figs.

118

.' ♦■»

Figure 1. Left - A scale with U rings, the fourth close to the ed^-^e
and touching it at the base of the sculptured oortion.

Right - A scale with \\, rings, the fourth well wj.thin the
edge.

Owing to their large size, irregular surface, and fine structure of
the external folds, sardine scales do not photograph well.

In reading scales, it is necessary to focus continually, and sor.ie-
times to alter the lighting in order to see all parts of all rings. The
following plates show one what can be seen at a particular focus and
lighting.

119

^4^-s^ /<

Figure 2. A pilchard scale with three rint^s, with a v;ide mar)3in beyond
the third ring.

120

Figure 3. A pilchard scale with 5 rings.

121

Figure U. A pilchard scale with six rings.

122

Figure 5. - Frequencies of widths of nargln beyond last ring In scales having
four rings. The data are fitted with nomal curves to distinguish
specimens having a recently formed annulus.

12 3

IT

O

o
o

>

UJ
Q

<

II 1 1

1 1 >i r-

V-

12

<*

Q

^r

O

8

-

^r

*

^r

1

4

^^,.M^

y^

r

+

S"-""^ N.

/

—

>v

/

4

\^ ^

/ ^^^-^^

8

8

^.^^v >--"■■

^''*"*«.^^^^ / ^^■''^'^ '^^

4

2_^

^^^^^^ \>^

x^ ^^

+

^•'''^ ^V

J^ \

2^ ^

^""^^ X ^^

4

<

^^^^ ^r \

1- ,

^^""^"-^ /^ V

4

'

y — -x ^

e

<^ .

+

3.- ■■ •-v^

/

t '

'"^n,^^

/

'

^*^*^

/ \

*

4

4V^^

^ »

01

»

l'

8

^.^^

12

1-

^^^^

\

*

o

12

^s^

>
\

8

en

^^^

\

o

16

>s„^

^ \

4

ro

>v

+

^v^

^^^^^^^

^^^^^ \

UJ

4
2

.---^''^^

.^f^^^ ^N

4
8

l-

\

\

<

. •

a:

~

Ny

UJ '

2

+

x^^,^,^.^-

UJ

t-

■

2

■

^ ^

lO

4

•

y/

J2

* *

^.

y /

;38

1

^^\^

— ^-— — '' '

8

^^ \.

.04

6

y^ Ny

y/^ y^

+

6 >y^

y^^ y^

_

4

X-.^

/^ /

X)4

^

<

2

^

^

J08

o

_+

^ ^s^

/

•'2.

r^

—

X.

y^

2

^^ — ^.^^^

jT

4

^""^^

^^^.^.^^y^

'

1 1 1

1 1 1 1

1932

1933

1934

1935

1936

1937

1938

1939

Figure 6. - Deviations from seven-year norm of growth In body length as
compuced from scales collected during thres seasons, compared
with similar deviations for various elements of the environment.

124

in

1

01

Cm J- ^

1

ct

Ti O - -

o

0)

CO

>JD

■tj c:

i^\

O O 1

e

CTn

<X> UI O r-i

1

o

f-'

i-H CO ^ J-

^.

rH (U - -

Vh

O m
O

Ti

C 1

•D

•H CTn o

1

*^

ir*-^

CO

.H

D

O

rH

CO

1

u

C\J OJ
tM -:t ^

1

m

t:! o - -

CO

0)

LPv

-►J c!

(h

1^

O O 1

CD

cn

QJ CO O rH

1

0)

rH

rH CD ^ ^

>=

r^ 0) - •-

o tn

+3

o

w

C! 1

»H

■H O^ O

1

.M

r^,^

Vi

- -

Vh

O

1

rH CM

1

tJ

tw ^ ^

c;

CO

Ti o - -

tu

■J^

to

<t! J-

-1^ c3

+j

1-^ rv-^

fj O I

0) CO O rH

CO

O (T^

1

1 rH

r-l CO J- ^

CO

r-H <D - -

ti

O CO

u

1^

O

o

l-H

C! 1
•H CTn O

1

rA

■H

P-

1

<H

•-t CM

1

O

tM J- J-

-d o - -

03

u

^

f^

-^^ C

-U

t^

O O 1

'Ifl

cr\

OJ CO O rH

1

a

t-t

rH CD J- J-

OJ

rH 0) - -

rH

O CO
O

O

1

,o

- ••

tH

o

1

rH (\t

1

ea

VlJd- J-

Q>

Td O - -

•H

(U

o

C\J

*^ a

fl

r^

O O 1

0) CO O rH

(U

cr>

1

g

r-t

r-l CD J- J-

rH CD - -

0)

O CO

^4

o

Pi o^ Q
•n r^^

1

iri

^

. ^

ITv

(D

-■>,*i

r-l

Td AD

1

,Q

O C

t-i

OS

pq ->a)

ir>

&)

rH

rH 1 K>| I I J- VX>a>l^rHCr\0^r-CMVDLnOrH I KO ,-:t rH rH I I I

r-\ r-i r-\ t^\ t<^ I i^^ r^ 1^ CVj OJ rH rH rH

rH(MLr>r— LOcTiOrHJ-r^o r~- J- ovx)i^ojcviC)OrHv.oi^cM i •-* i i

rHrH^lr^CJ0VX)C3^COCJ0 CTv r— ir\ ir* J- rH rH

IViDLOOCMCMOCT^O I — <0 J-J-vXIOjr— QJ"CnCT\tMr^OJrH I I I
CM KMPi I — ao CT\MD I~— VJD CiOV^vx>J--3-rUCM rH

I I I I t rH ro LTNv^XJ ir^VD CVI C^^V£> O-^&OICMCMI I I 1 I I I

I I I I

I 1 rH I rH CM hrvCTN hO O I — *.0 ^Cr>tOr~-0^rH60CMrOrH I I I r-t
rH OJ CM r^ N^-:t CM CM OJ OJ rH

I I OJI^I — 60>^£)l>0,^cr>CMC?O*>£)rHCTNrHO LTN J" CM J" rH rH I I

rHrHOJOJCMJ-r<-NOJrHrHrHrH

1 I I I rH I rH Lr»J- OJI ^rHCr»r— l^rHrHrH IrHh^rH IrH I I

I I IrHrHCMOJ^CboOO l^-d" lOJd" H^ Cr\''0 O (^ l-^ CM I I I rH I
rHrHCMCMCMOJrHCM r-t

I I rH C\J r^ J- Cr\ r— (T\ J- I^OrHt^O-^d-CjMAJ'-Or^-rHCM-d^CVJ I I I
rHrHrHCMCMCMCMCMrHrH rH

I I I I I ICMIrHrHCM^rHrHCM ^<^VX) rH rH I rH I I I rH I I

I I 1 I ICMCMrHrH l^jd-v^^r--OOrHMr>-CMr^| I I I I

I I I ^

I I I I OJ I 1^ r-~U3 IC\0 (T\ r>-^ O60t^f<^C\J<rirHVDVi3rH^ I I

rHrHrHCMCMrHrHrHrH r-^

I I I I I I I I I rH I H I H I rH I I rH I I I 1 I I I '

OJ

I —

hO CD
r-\

CO

o

IX) (0

o

CT^ <M

o

!>j
O -P

ITS .H

o

•H

CTn ^i
CJ^ O
P
o
t<

o ^

CM -P
r^ o

CO
X

1)

C7^ S

CM O

^1

r<^ (0

O P4

CM <U
t:!

in 0

c\j o
o
o
cS

r—

CM o
+j

•H
CM O

a

O -P

rH

CM •«

la)

(0

ir\ 0

-d
CO

I I I i^i^^irvjror^ f>--d- lr^mcM-d■^^l icMCMirHi i

I I I f r-\ i-{ \r^<D J-OOU)CnrHrHlOrH r— VX) CTl f^ rO 60 f<^ rH I I

60 CM>X)OJ-K)CMUD Crt 600JU30^WDCMVJ30^COOJV^ O ^ M C\J

^LTsv^u. r^r^r^tfl 60 a^(T^a^o o rH jH ^ OJ CM ^;^^;^^:;^^^ tn^^^

uK c^ rJ^ ri ^ uk J> r^ ri- J^ lA ck ri> ri. rl, ti> cj^ r^ ri- rl. jA ^ r^ ^ ji^ lA 4.
« tn S^>!d ^^ |C|i-ti-«Dt>ocJ^(7^cr^OOrHrH^HOJOJ^^^O r^-2 -£ S^ ^ ^

o o ^

^D

0) c)

^ (U

O C6 rH

rH CO >a

0) -d

f-i o

Cd -rl O

+3 (k, +5

EH r-i]

125

CO

<u

iH

CD

U

(0

6

o

^

<H

Td

0

ti

'U

QJ

0

4J

a

♦i

J-

0

0

1

CD

en

a)

w

rH OJ

r-l

rH

r-i

CD J- J-

:3

r— !

(U

•- •-

O

0

to

iH

0

td

p!

o

• H

m

a

u

a

(U

>s

tH

+j

tJ

0

1

<n

■D

rH OJ

U

cr>

-U

c! J J-

•H

t~^

y

0

— •-

t(-l

cr

0)

m

rH

rH

cu

«(-l

1— 1

Oj

1

o

0
0

cn

i?5^

•c)

CJ

•- —

ci

■H

tt)

-*»

cd

«o

t:)

w

1

U

f/J

tfH

rH CM

(d

;^

0

J- J-

A3

■d

o

0

(U

1-H

1

60

+j

a

•H

^

ro

0

0

1

p.

cr>

0)

c«

0 rH

w

1— 1

rH

CD

J- J-

«M

H

rH

01

— —

o

0
0

U

U

C

1

ji

•H

CTN 0

■»j

f^J-

no

a

a>

iH

>»

Td

o

,0

<H

rH OJ

«H

0

^ ^

0

0)

" "■

03

t —

4-3

c

0)

r^

0

0

(

•H

a^

1;

0)

0 r-^

0

i-H

rH

(D^ ^

fi

i-H

c

•- —

Q)

0

03

&

0

c

1

(U

•H

■o> 0

u

K>J-

•

Td

^

43

1:3

0

0

-— •

ir\

^

(0

!>s

+=

r-^

Ti

W)

.0

0

c

cd

m

ID

E4

rH

I I I I

I I

CM CM U">UD rHr~-(M60tOivDCrNOr^ LTVCTiVX) r^ rH
rH rH K>, r<^U3 >vD IS\3- t^ OJ

I I I

I rH ^r^o^CT^I^aJLf^CVJC^JC^^O LTN^ vi30^li^060QOO-:i-t-f^(\J I

(\JVO^^VX)JJ-Lr\CriOOrHrHf^l>r^rU rH

<-* OJ r<^ J- m LTN J- ^ ^r, OJ rH

I I rH I C\J vr> r^VJD ^ CTl^ O rH ^ O r<^ ITv J- Lr\ rH rH C\J I I
r-t OJ OJ roJ-r^N^t^^OJ C\J i-\

I I

I I I

I C\J I Ojl — a>^rHJ-bOrHr~-rHv^-Of\|60rHMDC\JrHLPiLr>rHajrH I rH
rH rH J- v^ rr^ CO |v->iVX) r<-^ N^ OJ C— VO OJ CO I — J" ro rH

rHrHOJOJOJOJOJrHrHrH
I I lrHrH>^J-J-| — f^Cr\60COajJ-COr^r^rHeOC\J60B0j- I I I I

rH OJ r<^ LTiUD r — o^(T^l — r— r-i — r^r^rH

I I I I I I I I CM O^ OJ J- ITNJ- vr> r-r^^ VX) OJ cr>>-D I^OM^ I I I I

rH r^ r^ LTvUS C^^ r^ 60 VD IT* OJ r-i

I rHrHrHt^h'>CVJrHVX)ajLrv I — J" GO r<^ CTlVD <TNCOO>^a>N-\OJt-<-N I rHrH
fH rH C\J K^I^U^LTNI-— r^VT) 60U)VO^ r<-\C\J OJ rH

I I rH OJ ^<^ C\J rH

•cjM — c\Jl^^oJJ:^ ir\ I — r^-i — cr\rooo(TN r— -d- J-

r-\ I I

I I t

I rH Lr\ 60 J- t-^ CO r/-^ t^MD >JDCr.r--C\IC\)rHVDViDrH LPi J- I I I
rHOJf^LPvKlOJ- CV-d- r\JOCPir~-rOrH

-:r 00 c\jU3 0-3^ 00 iM'-JJ (-> J^ 00 rvj'>£)<_>.;3^ roc\JUJO-3" eoc\jvi>0^3" oocm
un Lf^vx) vx)i — I — I — 60cocricr\crvOOrHrHrHc\jc\jr^r<>v rr>, J- j=t ir\ir\ \s^\d

t-{ r-t r-< r-i r^ r-i r-\ r-i r^ ,-^ r-i t-t i-t I-* i-\'t-i

I I I I I

tr\ O^ ro I — rH

I 1 I I I

LO CTv K>r~- rH

I I I I I I I I I I

L^^c^^^^^-•r^ L^^(T^^ol — rH

ic\ LTN LOUD VDl — r~-i — eofiO<7^cT^c^^OOrHrHr^t\JC^J

11 I t I

. . u> o^ f^ r — rH

i<^ 1^ r<~\ J- ^ ir\ LTi Lfli
r-ii-ir^r-tr-ir-tt-ir^

CO

CM

O

10
CM

1^

Ko

CM

I —
CO

NO

CM

CM

60

O

CM

(d
o

EH

126

CO

Q)
(-(

CO

o

01

B
o
u

td
o

Tj

u

>»

to
Vi
•H
tH

«H
O

0)

-(->

CO

CO

r-i

cr

CO

o
I

o

CD

tifl

o

Ch
o

0)

o

&

u

-p

c
o
o

in

CO

EH

o .*• -

O I

tn O rH

CO J- ^
(D - -

CO
ii! I

Cm ^ J-
O - -

O to O <H

CO ^

(U - -
(0

d I O

•H CJnJ-

i-H nj
Cm J- J-
o - -

0) rt

+i o I ■ ■

O CO O rH

0) CO ^ ::d-

-t O - -

— I CO

1-^

nd

OJ

1
i-H OJ
J- J-
tH - -
O

a

O I

CO O rH

CO J- J-

O) - -
CO

•rH

1
CTN O

1

i-t OJ
Cm J- J-
O - -

a
o
CO
cO

V

to

?e

^§

(0

rH

CO

o

-H

cS
+>
o
r-1

BO
CVJ

o

CM

(■— CTN
r-l

1

1 U>
J-

"^

J- o

1

r- 1

rH

1

f^ 1
1^

CPi

VD h^ OJ C\J CJN I —

OJ ir»

I I J- 60 r— rH
rH rH 60 O

1^1 I O I

60 rH

rH r-\ C\J t^ O VD
OJ OJ

I I m r~- 60 <TN

l^rH ^ O

1^1 I l-~ 1
J- o

OJ

I I I OJ r^cxj

I 1 ^ f-\^ r-\
OJ rH rH J-

I 60 I I OJ I

I I I I c\j r^

I I r--MD OJ LTN

r-^ rH OJ

I OJ I I 60 I

Ph +> Ph

• O • O
C/2 a W M]

O

o

CT-

O

r!
o I

CO rH OJ

rl C0,:J- J-

-H CD •- -
O CO

o

CT^
CJ^

CO

o

I

pcJ

to

Vi 1

Ti

O rH OJ

a

J- -:J-

+^

iJ - -

o

O

a-

to

rH

(.0 -t

l-l

CD O rH

O

iH ^ ^

o

•- •-

CTN

fO <-^ CTN>.D CTi LfN
r— rH r-\

rH OJ

OJ CNJ LOi rH J- (TN
LTN K-^ aNVD ><0 to
rH VX) '-X) rH OJ r^

I LPv I

OJ

I I

rH OJ

r'^ 1 O rH

rH LTN

.4- ^

VX) ^<-^ CTN m O C\J

f— 1 — Lr\

r-\ r-^

tn

Ti

O

'H

■»J

Pi f

o

O CD rH

eo^ ^'■

1 1 VT) 1

1 1

c>

r—

rH

CO - -

60

r^

■1)

o

CO

o

•H CT> Q

1 r- I 1

1 1

r^^

K-v

UD

r-\ C\j

x) o - -

O I

to CD rH

to J- ^

:j^ — —

•rl CTN O

CO

CO

•H

+3

rt

(0

o

u

rH

•rt

o

cO

>1 f-<

Ch

4J

^ HJ

•H

o

CO

rH

EH

•H

CO

P

o

a\^ ^ rH m CTN

r-: H r-t c^^

I I

60 Q r^ LfN
60 CM J- C\J
Ol

I LTN I
O
C\J

•4J [iH

O ■
t/D ;s; CO

I I I

d

o

-p

W)

fl

c

o

■rl

W)

^

<u

CO

u

CO

o

;s

o
o

CO
•H

o o

tJ a) CD
CD ^1 fn
PL, cu rx(

ra o cti
en S CO

rH) CVT) t^

127

Table 6. - The symbols "a", "b", "c" , "d" refer to the foix.- corners of
the square, mentioned on page 29, for which barometric pressures (in inches
of mercury) A, B, C, D, respectively, nere calculated by interpolation
beti.'een isobars on daily vreather maps. The figures given are monthly sums
of these daily intei'polations . The points are located approximately as
follows:

a, lat. 32^ 26' N.; lonj. 126° 12' -j. ; b, lat. 35° I6' N. , long.

121° 1-:^' ■■;. ; c, lat. 30" 58' N.3 long. 110° 28' \i.; d, lat. 23° I6' N.;

long. 123° 17' 'T.

They are 3o placed that (A+D) - (3+C) will give a pressure gradient in
a direction from v^hich one would infer a wind parallel to the coast line,
assuming deflection to be 73 degrees. Similarly, the difference betvreen
(A+B) and (C+D) would give a pressure gradient from v/hich a wind at right
angles to the coast would be inferred. For the study discussed in this
paper, the aforesaid differences vrere computed from twelve-month sumiTia-
tions of the monthly sums of daily pressures. They were treated as com-
ponents, and converted to their corresponding ■"esultants (R) with the fol-
lowing formula:

R = "V [(A+D) - (B+C)/ + [(A+B) - (C+D)]^

128

O I — t^ O CT\<T\ a>-d- C^^ CM CM O
CM CM I — CTiLnr^LPnH I — I — LTNi-l
h- (^^ U3 I — ^ f^-d- rH f>- CD l^r-v

J-CMr-r^^rHrHiHOCnOf^C-J

a^^o crvcri(j>cr\cr\criC3 cr>cr\CJ^

o

Lf>cr>Ol — rHbOOt~— MJ- Or-^
CMt^iH COr^t^LOiCMOJ- coco

O r~— iH cT^l■f^J:J■ r~-.d- co cm cJ> O

r-r> CM iH 60 00 60 CO to '.jD CT> O r-H
r^_d- r^cr\CMCr\fUCMcnCMOt^

cn6ocr>fOcr>60cr>cnL-ocj^c7^cT\

rq

OJ CO I — ^ I — r— CO r-t ^ cvjJ-VD

COVnJ-lTM — Ot\)OrMI — rHLPl
CTi 50 I^- CM UTi 60 CM CM >.£) SO (T\ (T\

LO CM CM O CJ> 60 CTi CTN*-D O CM t^
t^.rt r^OCMa^CMfOCTNh^Or'^

cr. 60 en crs cr. CO o^ cr> co crv o ■• cf>

en

..__._ O6o<j?m
r-t^r-i Okd ^ 3- o

i-l 60 J- 00 CM O Lr\ J-
r.-OiHrHI — t^r-IO,j^— _ _
UD C\J I — CM O M> ^r^ 60 r<^V£) rH K>i

p J

r^c<~»Or>^rHOOcnCTiOt^rH
c^-* i-^Or^Oi-^CMCnt^>Oi~^

en cj en (J^ o> cn en OA 00 en o> 0"%

OJ CO ;o o r~- lO cm cm r^, (—- O r<^
'JD>^ <-i njO-* r~-r--oc3>cncM
I en en r^ J- 1 — CM 'X) vxi en cm t'-\>x>

1^ 1-1 cTi o r— CO vj:) r--'.D co rH o
M^-:i- cMocMcnrucMcncvJOi<^
cnoo encncnio cna^6o o^cj^cn

m

rH (^ CM m hfA rH J- 60 rH rH CM CM
en rv^ rH O -d- rH 0> 1-^ J- O VX) 60
VX3 rH CO U>C<^CTNr^ir\OVr) rH^

CM CM CTi CM CO
CM O

- - , . _ 'o CO 60 CO cn OJ CM
.■^-d- fMOCMcnrucMcncMOr^i
en6o ^T^cr^cn6oo^cT^eoa^cnen

en I — 60 vi3 LTi ltn I-— md r— o cm rH

60r-~•rHCM60lr^cr^r^OrHCMr^

cnuO 60 C3>0 c^i — iH rH r<~>r^O

r--rH^CMC^CMCMCMrHrHf<-^'ljr> fX^

c^^ f^0l~^0c0t^0c^0r<-\ en

o^6ocncncncT^CT^cnc^^cncncn

CnrHLPvCMJ-'vDrHCnCOr'M-— ir\

m r— ^ r — k-\ r~ en o m ir\v.o c^

N^mrHOt^-r^rHCOCM^LT'LTv

J- QlrHlOCMrHCMrHrHrH^ CM
coJ- f^O c^O c^K^Or^O c^

cnco cT^cnc^tcncno^enlCncno'■.

J- rH h-

CnQUD
V.o3 o

n

CM rO CM
h<M — r^

a> 60 cTi

CM O rH

r^ J- in

rH ^^^L'^

O

CM CM CJ
r<^ h- h^

en 60 cpv

ro 1 — m
CT^r^ J-

w

rH CM rH

(^ r— CO

en 60 en

>X> 1 — CO

.t>

O o O

^

rH CM r^

CTA

ro r— ro

rH

CJ^ 60 C3^

rHcojJ- coencM^ \r\jt r-t ^ rt
r^enrH locncoJ' 6o ^<-^rH cmo
60 Ln CM a^ o I — co-d- m>.D r-^5

rHJ-J' ru-d- r-* ocno^Oc^c^

tv-\ J- r^OmO c<~\CMenr<^Oc^

CT>60 cncJ^<3^cno^enco cncncn

P

r^co r^coinrH soioeni — incu

CM CO CO VD J* en in^ I — 60 rH CM

cno^^^|^-r^ ':^i r-i inrH o cnr<^

rHOOCMrHOAenCnCOOOrH

3^co encncncoa-\cnco cr\encn

ctn I — J" I — vno i^->~D

in r^ 60 CO r>-60U3v£)

r^ r^J-U3 OOJrHI —

^ mj:t CM CMOrHen
r^ J- rv^ O N^O r^c\j

en 60 en en cnocrsen

-p

a

o
o

EH

CT^^- r~-rvJrHj-^ coot^l^rH
CO ru vD I — CO r^ J- ir\ rH cr-, cm O
vx) 60 t — I — r — --i cnvD U5 in CT^ en

o CM rH en o cnvx) i^x) >J3 ' o o cvj
m-:j- c^eni^cnCMOja^rMOcn
encocnooencocncTNCocnencn

rq

coor--06or^mrHrHCMCM
j-LnencMojtrNr^OrHt^rH
■ en 60 60 injd- vx) co invo

en

LncviorHcnr — I — uienrnm

r^-d- r<-\Oi-^CnCMCMCnCMOrn

cnco !T»enenco cnenco cncnen

-c?^ m en en-^ vx) 6o 6o co rH en.:*

en J" 60OCri60rHI — jd->^ovx>

inLnco6ocMen^rHi-HO>vD3

VD r— J- ^LnOJCMrHrHCMLni^

c<^^' i-no mo cnr^o tno c^
enco cnenencno^cnenenencn

-:fOCnOJr>JrHCnrHOr-ir~-in

r^ en cvj 'jn r— vjO en kn rH vx) cj^ f^
cr\^ rH (^o O'JD o cr\r-i ^ in

o o o"> rH r^ I — >x) t— in 60 o o
rr\\ — cMOCMcnojojcriC^JOm
cj^tocncjNcn6oenc7^6ocncn(7>

fn

r~-CM cJ^lneo •aivx> mvenm. rH^X)
rH-* r— c^ocn^vx) rHr--h-ir\
I — UD en CM O r— CT^<X) r^ O J- I —

rvj o CTN cvj en i — md r~-u3 a^t^<-i

r<~\l — CMOCMCTNCMCVlCnCMOm

enco ct\encJ^6o crvcnco crNencTN
hr\t — moo r^r^to mvx) in j^

I — I — ini — l-~\(\J rH CM h^US 50 rH

en J- rH J- o m o J^ in J- cj^ rH

^vjr^^^j^ncMOr^r^o^r^f^^r^
r-r^r— |j>,o cno cnrnent-no en
en60 cncncncncncj". ^CTNcncn

o

U5 r-i<D LnOJI^OCTi
rH >X) J- J- 60OJ-J*

o 50 J" >X) rHencno

r-. ru ru o enr— co co
t-r>,j- r^ o rocncvjCM
en 60 en en enco encn

m

in 60 CM J- irMninr—
en I — r^vD inr^vx) J-
r^ r^ 60 CM I — incovx)

evj ^ CM rH 60 60 CO CO

mJ- CO O cvlcJ^cvJrvJ
en CO ctn cpv cnco cricn

00 <ii

m

en

cr>,<.X) 00 inj-UJUDVX)

O in O CM r-t<D CM CO

60 <X> in ro 50 rH LTNO

• . . . •••.

ixNr— inJ- cMOJCMrH
r^ J- rr>i o mo cnco
en 60 en en (3^cj> encn

P

vD >oD o vX) m ,H rH cr^vx) fu CO en
incM'^x) incneninco r^cor^^

^ CM 60 Q C3^ J- r^ rH r~- O <TM —
inrHOJCVlrHCriOCnCT^OrHCM

c<M — K^o cncnr^cM cnfno en
enco cncncTNCo cna^6o cnencn

ror--LnrOrHCMC060rHj-50CM

incor — cococoi — ror^or^co

rHvX) CMLnCMK>60COCnt — cooo
^ C\l t^ r-t •-( t-l cnVD 60 O CM rH

roj-ry->OK>orurvjenmOcn

enco encncncno"^cT\co encnen

o en en c^ f\r-t t — t —

SX) O inO 60r-r\6060
VX) I — in rH >X>CM I — CO

t^ r-{ CM K^rHCnOCn

cr\ J- ro o mencncu
en 60 en en ct\50 encn

rH in rH CO J- r— J- in ro K^ en rH

50 r-^ cr-j J- I — J- f-l t-\ in J- O r-<

J- rH envx) cMvxi r— in^ r-i j:t in

c^O O cr^ cnvx) r^vx) t — co O rH
N->r~hOcncMcncMCMcj>ajOf^
encocncocncocncncoencncn

o

rr>io coinen^ inrH inr — rrt rH

vXJrHVXJUJenCOVXSmCOOrH^

.jOvx) c^rncM jo cnrH cT-\cnmr—
•

O rH rH en CO r— 1^- inuD 60 rH O
r<-\^ POc7>CMcnCJCM0^CMOcn

enooen'ocncocn. cncoencnc3^

o

en ml — <>x3-d"60vx)in

CO O rH rH r-KD ^CO

^'^J- 50 r-i i^-t~-or~-
rH rH O O COr~-CTM^

rr>,^ cn O ^^JcT^^u^u
en 60 o^ en enco cT%en

J- in^ in ho J- J- in Lr\ en J- t—
tnr — LncomsocMOJcn in'-X) in
c\j in in I — 60 rH envo vo i — -+ in

J- o OJ o en I — t^vx) r~ en. cm m
r^r— r-^O CMcnoj ojcj>oj o en
<n6o enencn5o cT\en5o cncncjA

fp

I — J- C3^ J" c^^ CO CO co in i^ en co
f<^7--^ OJ en in>x) j- co O in co
t^'a coini — cnininrHvxiJ •-{

rvJcArHcr, cor— covx)CoocMCM
ro-d" mcnCMcTiCM OJ cnrno m
encoo>X)cncoencn60cr\cncj>

m

r<-\ ir\ r — ^ inrH cnco
ro t-r\ix) vXJ inx) rH J-

O CnrH rH'X)rOrHrH

j:± o r-i r-i enco cnco
mJ- coo cMcnojoj
en CO en en enco enen

CM

en

(v^rj I — 6ov,o r^enin^ oj-j-
Ovx) in in in J- ir\ rH rH rH in r —

inrH^J r^CMJ-VD OUD in>X)

h-CM Lntv~j|M^rH CMO OJOJr^J-
CTM — hoOrr<,Oc<^fnOmOrn

enco <r>cnenencnenencncncn

inl^

m
en

coojc'^cx5encnfn60rHojJ-Ln
o-NcnO t-ni — r^tni~r>ia^co oj cm
miJ- oj crien>X) oj cm r-! cm oj vxj

rH ^' J* o" f ' CM CM 0> O CM J- OJ

K-\ J- i-r-i o n o m CM o m o m
cn^Q enencncncncnenencncn

VX) en OJ J- rvJr-TM^-MD

r<^i-r^Lnrvj rHOvx)i —

J- Li-NCM >-{ oenoej

CO ^ . . . . ....

hOi U\ Q r-\ ^ t^r-^ OJ l-t

en cn^ mo rr^o r<-MV>,
rH cn 60 en en encn c3>ctn

. . . 0) >j • +3 . < •

cuctiputoppipi'uooa)

l-3p^S'=>Ja»-3'-3'-^'^0!3Q

• . t o -t • -t^ • ' '

^ ^1 fn >j q rH M; P, -1^ t> O

r^ s; -^ S '-3 i-o <^ "J O ^ P>

0) >» •

C^ Jh ^h >.drH'lO

nj (u CO P4capp)^

>-3 fi* a ■=^ ai-a ^-0 =^

129

+3

a
o
o

EH

«

CO o^ r<^0
i-H LPvO CO
O J- rH r<^

CO rH tA^
a^ r»^ O r-a
CO CTi Cr\ CTi

O

CT^ CO rH ,=J-
(M m rH rH

vr) ON O CM
(7^ CM O 1^

CO cr>(3^cr\

m

CO 1 — i-H CO
CM irMTN CM

r^ CO LC^ r--

r— O CM rA
CJ^ r<^ O ro
CO cr\ CTi cTi

en

•^

a\ O O rH

CVJ 0>JDU3

Cr. CM U^ rr>,

C7^ t^ O r<^

CO cricr.cr>

f=i

CM 1 — tr\ t<^

r^ 1 — l-O r^

cn CM J- vo

CO rH fvj CM

CO a\cri cr>

o

CTi J- LfN r~-
O CO r^ CO

CM rH O J-

f— O rH rH

00 C7> CT^ CTi

pq

LO. CM O CM
CM r-t ir>v-D
r^ 00 KMTN

f^ CTi f^ CM
cr\ CM O r^
to CT^ CTi CTv

CO
i-H

<<

O rH VX) K>1

LTN h- rH 1

rH O t^ 1

CTi CM J-' rA
CT^ r^ O r^

CO CJ^ c!>cri

+J • • .

P.-tJ > o
(U O O lU
C/2 O I2i Q

o

rH
rH
O
•-3

a

0)

•rl

d
o

•rt

tn

CO

P
p.

•H
U
O

CO

■tJ

ctf

p
o

CO
0)

o
o
o

■r<

<0

C!
(D
O

CO

CD
(U

Td CO

•H

CO O

CO

o

s

0)
CO

(-1

fe

t

%

•

'-Tl

r—

<u

r-A

Ol

,o

J*

Cfl

I'l

EH

>H

OrH>JDVX)00'Xlr^rH
O CTi J" r— rH f-^ r— J- rH

,-i- r^ Lr\jd- VX) LC^.::f I — KD

60CM^X><T^00I — rvji — t-^
•-{ ^ CJNCM CVI CJ>LOirHO

V.D IfNUD lO r--v£) J- CO VJD

O CM r^a^(■~-^^L^^ro^r\J
J-CMCOrHLfM^OCMvX)

r^r^r — r--co i — h-cneo

CM Vi5 LTiV^O LTn no MD ir\ CM

vncTiLfM^criLfAOJ- CM

00vJ3OCr>OOOC3>CJ^
r^r^CMfHCX^CMCM^^r^

Lr\.d- I — J± CO f<^ CO f-H r<\
I — O J- I — O vx) en r<^>^

C0CT^OOC\iOOrHO

rHrHCMCMCMCMfVJCMCM

ai LfM^ CO ir\M3 I — (J\Jt

^ <o -^ CT\f<^r~-ovx) (^

• ••••••••

c^^coOc7^CMOC3^0oo

rHrHCMrHCVjrjr-tCMt-H

cr^CMOCocT^r^co^*^rH
60 r^cT^co r— C7M — oo 60

t-ii-\r-{r-lt-tr-lr-ir-tr-\

COCTvrHCTiOOrHrH COO

O i-^CMLPiO I — I — '-X)VI3

vx> lJC^ CTi r — 60 I — uD vo co

rHOCVJJ-J-OfMt-^t^
60 CJM^ O CM CTl rH LPl J"

.;:!- ^ I-— IPijit J- LPi LrN<vffl

OlsOj-kD rH lOCTvrH LPv
60.3-1 — CMrHCMir^OJO

rH rH rH rH rH rH rH r-tr-\

ojvn c3^ovo rH ^'-^cT^r^

t^ CM h^ J- J- t^ J-

rH rH rH rH rH rH rH

rH Lf\
rH r-t

CO rH 1 — CTM — f^ rH
OJ O CM r<^ Cn J- O

00 J-
O 60

ro r^ r<~» J- r^ hO LTv J- U^

CM r^ J- ir\vj3 I — CO CTi O

cT\cricr>(T\cj>criCT>cr>CT^

X)

o

o

u

c!
o

t(H

o

rH
O

cd

a
o

•rl
■•J

CO

d

CO

p<
P-.

■H

(h

O
CO

a>
Xi

M
O

d

cfi

0)

o
o

o

d
o

•H
+>

^

to

P
P
•H
U

o

CO
>a

u

cd

CO

d

o

.

Q

g

(U

'-H

W)

cti.

Ii

OJ

>

>

f

^^

ha

CO

Q)

rH

■^^

^

cti

'1

EH

>^

^

r^ CM i^.;:*- CM CMr<>Xi r—

r— 60 CM CTv lOrHI — CM ^

CT^ 00 CM O ir\r-<D60 cm

J- >Xl ITN LPv J-ViDLTIsD ir^

MD V£) rH rH >X)VX)ir\C\J I —

CT^V£) 00 CTi CM LTNTNCM r-<
LO I — LTl tr^ f — I — r^60 I —

60 (^ CO CM lr^fVJ(TkD.=^

^X) I — UD VD I — I — r^60<X)

CTiO CO ir^H-OOCOrHO

UD CO >^ >X) vx>>^r— CO r—

I — Cr\t — O J" 1 — h^lTN r-H

vx> VX) vD <X) VD LTN — r^r^

r\J 00 rH rH 60 (-MD rH 60
W I VX) LPiU) LTv LrNf-<-VX>U3 ir\

cn-d- r^ CM >^ J- tor-- (^

K>, iTN LTN J- ^ OrHirxX)

rH Lf\ (-"^ OO VX> LTXX) CTnO
CMJ* LOCM CM CMU).:^- MD

J- O^J■ O i-i OOCO CM

^o J- ur\^ lOji r--.::t r—

CM r«-^J- l^^>vD r^co(r>o
r^ K^ r^ h<~\ ro r<W^r'^»J"

a■^ ctn CTv CT^ cn oxtnctv ct~.

rHrHrHrH r-i r^ r-t i-f r-i

0)

cd

CO

u

0)

(d

o

o
+>

o
o

•

-d
Id
Hi

O
O

1^

g

0)

u

cd

d

>

•H

W)

CO

0)

s,

•rl

130

0)

^.

w

lU
r-t
P.

a 01

en jj

CO .H

ctl C

- £0

to S

43 ^1

(0 O

tM CO

o o

ui to

§.2

o u

^ o
+3

- f-i
to 0)

^ s-

4->

c! -

o ni

e r-(

rH

>J O

to

M

5 tu

^

o

u
I

CTN

a>

rH

cd

r<^

I ai rj

ol

CHlrH

Pr,

e

I rH OJ

^1'

OJ 1 C\J

I r^ rH I

I I I J-

I I ITi

I I ^

r^ I I r-l

I rH I ^

I t OJ nj

CTv crs <u

o^ cr> o
- e

^1

o

I I

I I 3

o o o

o o o

LTN O O

rH OJ

CM

I 1 .-I J

I I rH r^

I I rH K>

I r\j I K%

I rH I hO

rH nj I rH

I I 1 LfN

I I I J-

I I I ITS

I I rH r^

I I rH r^

to CTv 0> Q)

CO CTi CTi M

<u CT> cr> o

-H - a

lu I I

O I I ob"

a\ O O Q

cr\ o o o

J- LP. O O

CM

cr.

I r-\

r-o.

rvj I rH rH

(\J rH I rH

r^ r-^

r-\ rvi I

111^

I rH I J=f

I I I -d-

f^

1 C\J rH rH

r-{ fT^r-A I

rH rH I C>l

I I CM CM

I (M 1 r^

111^

I rH rM C\J

rH CM I

111^

I rH I 1 I Lr>

I I I J-

1 CM CM

(T^a^ 01
CTsO^ o

- a

I I

I I °8

o o o

o o o

Lr\o o

rH '.M

CJ>

I I rH r^

rH rH I CM

to CTn CTv OJ

0) CJ^ CTv O
rH - a

^1 I

O I

I c«

cr> o o o
a, o o o
^ Lr^ o o

rH OJ

t-^ rH rH I

CM rH ,-H I

rH I N^ I

CM CM I I

I I rH ^

I CM CM rH

OJ rH 1 rH

1 I rH r<-\

rH OJ I aj

I I I ^ rH OJ I rH

I rH I t^

I I

I I rH rv-^

I CM rH

OJ

I I CM OJ

CO cr. CJ^ 0)

to (J> C7> ^

a) CT> CTi o

r-\ - G

U 1 I

o I I a

0-^ O O O

cr^ o o o

^ LO, o o

rH CM

I rH I r^

I CM CM

I I I J

r-^ rH \ t^~\

OJ I I OJ

r-\ <^\

to cTN cr> o)

to CTv cr> *-"

0) cr\ CTv o

H - e

o I I oS

o^ o o o

cr> o o o

J- lr^o o

rH o/

1^-

rt rH

I I I J-

I I r^

111^

I I CM ro

I I J-

1 rH ro I I rH r^

I I ■ I ir^

I I rH rO

CM r^

I ^

rH I OJ rH

to a^
to a-\
1) a>

u I
o I

cr\ O
cj^ O

O^ O

- B

I '-8

o o

CO O

O O

,-{ CM

ro

CT>

I I r^ rH

I 1 I Lr\

I I I J-

I I I Lr^

I I I J-

I I I lr^

I I

OJ

I J-

I r^ OJ rH

to CTi

en cn

U I
O I

cn o
cr> o

CTi 0)

cr\ f-c

cr> o

- S

o o
o o
o o

CM

CTv

111^

I IrH r<-

I I I ir

I I I J-

I I l_-t

I I I Lf

I IrH K-

I |r

I IrH J-

I rH4 N-,

tn
>»

£0

Ti
to

43

o

43

to
o
S

•n

CD
rH

o

e

0)
43

o

0)

C
o

>

o

Si

• rl

to
to
ctf

to

to

43

a
o

s

tocncpv uJ
to o~>cri f-
■j» CJ^o \ c
-H -e

o

A 43

43

^. I

O I

I ^

cr-.OOC
zJ-LPiOO

i-icv

CJ>
rH

■H rH

>

•rt

0)
0) rH

m

^ Oj]

131

4o Influence of Temperature on the Rate of Development of Pilchard

Eggs in Nature.

By

Elbert H. Ahlstromi/
CONTENTS

Preface Page

Introduction ©o»o»oeoo»o»»<>ooo««ooo« 133

Collection of material ...o .....o.o 133

Description of the pilchard egg. «... 134

Time of s pawning .....o....... ..« 135

Duration of developmento oco...s«a..e.o<>.. 138

Water temperatures o.o««....«>«e.o.oo..o 138

Influence of temperature on the rate of

development of pilchard eggs. ....oo...... 139

Summary .« oooooo.oo.oeooooooooo.. l4o

Appendix. «ooooa.o«oa«o..o.o.a..*a 14o

stages of pilchard eggs
Tables

Bibliography 0 ooo.oo«o.««...o.o....e 167

l/ Fishery Research Biologist, South Paolfio Fishery Investigations,
Fish and midlife Service, La Jolla, California.

132

INTRODUCTION

In the first paper of this seriesi' , Sette, 1914-3, has stated that
a major requisite in the program of pilchard research is the measure-
ment of recruitment. This can bo obtained most directly from a quan-
titative sampling of pilchard egf^s and larvae at sea in and around the
region knovm to be the chief spavming grounds. Determinations of the
amount of spavming require the integration (over both time and space)
of the quantitative samples of eggs. To accomplish this necessitates
knowing as exactly as possible the age of any particular stage of de-
velopment, as well as the number of days of spawning represented in each
collection. Since rate of development is strongly influenced by tem-
perature, it is necessarj"- to know the relation between these two vari-
ables to get a precise method of determining the age of eggs in all
samples .

So far it has been unfeasible to procure live spawning pilchards
for artificial fertilization and propagation of eggs. Fishermen rarely
catch ripe, spawning pilchards, and vre have lacked facilities wherewith
we could attempt to catch them. Although eggs newly spawned were col-
lected in the plankton samples, constant temperature apparatus was not
available to permit hatching experiments under controlled conditions .
Hence, it has been necessaiy to approach the problem analytically by cor-
relating temperature data collected in the field vvith the probable length
of time particular stages in the samples had been developing.

The writer is indebted to Oscar Elton Sette, in charge of the South
Pacific Investigations, for continuous aid and advice throughout the
development of this problem, to Dr. L. A. Walford for valuable help in
the preparation of the manuscript, and to Elizabeth Vaughan for assist-
ance and suggestions in the mathematical treatment. The plankton hauls
for the most part were made by the author and other members of the South
Pacific Fisher;^' Investigation staff of the Service, vfhile the occano-
graphic observations, of which the temperatures-data of this paper were
a part, were the responsibility of the Scripps Institution of Oceano-
graphy staff. Sorting of the plankton collections was accomplished with
the assistance of employees from Works Progress Administration projects
Nos. 10917 and 65-2-08-286.

COLLECTION OF UAT^nlAL

The plankton collections on which this paper is based were made
during the pilchard spasming surveys of 19hO and 19U1, conducted by the
South Pacific Invest j.gations of the Fish and Vildlife Service in conjunc-
tion with the Scripps Institution of Oceanography from the research

i/oscar Elton 3ette. Studies on the Pacific Pilchard or Sardine (Sardinops

caerulea). I - Structure of a Program for Pacific Pilchard (or Sardine)
Research. Special Scientific Report No. 19.

133

E. './. Scrlpps. Six cruises, 3ach covcrinf^ a pattern of approxi-
mately forty stations, v/o-ro made during the spring and early suraraor of
each year. The region surveyed is. the area of maxnnum spavming of the
pilchard (Scofield, 193h) • It had as its northern limit the group of
islands of i.'hich Santa Cruz is the largest, and extended in the south
to near the Coronados Islands; the stations ranged fron near shore to
about 150 miles out to sea. The region covered and the pattern of sta-
tions occupied was somewhat different during the second season, but es-
sentially the sane area vras surveyed in both years.

The racthods employed in the collection of plankton- samples contain-
ing pilchard eggs ■vfill be treated in detail in another paper devoted to
the results of the spawning surveys of the 3/cars 19U0 and I9I1.I. For the
purposes of the present article, it is sufficient to state that oblique
hauls were made I'/ith either a one-meter or tvro-meter net from approximately
75 meters deep to the surface, the hauls talcing, on the average, about
2i4. minutes of towing time. Stations v/ere occupied at whatever hour of
the day or night they were reached. Temperatures v/ere taken at each sta-
tion at 10-meter intervals in the upper ^0 motors of livater and at 25-nieter
intervals in the next 50 meters, with reversing thermometers attached
in pairs to Nansen vrater bottles.

DESCRIPTION OF THE. PILCHARD EGG

The pilchard egg can be readily identified by the follov;ing com-
bination of characters: a wide perivitelline space, a single oil globule,
and an irregularly segmented yolk. The eggs are spherical, with the
egg membrane thin, unsculptured, quite transparent, and v^-ith a bluish
or purplish cast. The egg membrane is easily broken, more so than those
of most planktonic eggs; in the plankton saraples studied it vms commonly,
absent from a third to a half of the eggs , probably having been broken
in the net, or in fractioning or sorting the material. Fertilized eggs
average about I.70 mm. in diameter (range 1.35 - 2.05 mm.). This is of
similar size to the fertilized egg of the European pilchard. The peri-
vitelline space is wide, having nearlj'' as great a width as the yolk mass.
Tfe have encountered no other pelagic fish egg of the size of the^pilchard
egg in the area investigated that are so characterized. The yellovfis'h-
brown yolk appears to be made up of a- number of irregular cells,, but it
is composed of a large number pf separate particles of yolk material closely
pressed together. There is a single oil globule, O.I6 mm, in diameter.

In connection •with these studies the development of the pilchard egg
was separated, on the basis of readily observable morphological character,
into eleven stages v;-hich are jiot of equal time duration. A brief descrip-
tion of each stage is given- in the appendix (p. 9 ).

134

TIME OF SPAT/i^v'ING

As there are no actual observations on the time of day pilchards
spawn in nature, this could be determined only by indirect evidence.
The time of day newly-spawned eggs appear in the plankton should roughly
establish the time of spawning. Such newly-spaivned eggs vrould be pre-
cleavage eggs that had not had time to begin cell division, i.e., stage
I (cf. appendix). The times when pre-cleavage eggs occurred in plankton
samples is summarized in tables 1 and 2. According to both tables about
87 percent of pre-Cleavage eggs vrere taken during the four-hour period,
8:00 p.m. to midnight. At that time there were practically no eggs in
the early stages of cleavage. After midnight very many pilchard eggs
undergoing cleavage were in the samples taken, while only an occasional
egg v>ras of precleavage stage.

Eggs of that age taken during the four-hour period before midnight
are considerably smaller in diameter than are those with some embryonic
development. 4Since the yolks are of similar size in both groups, the
difference lies in the width of the perivitelline space^ which is nearly
wanting in pre-cleavage eggs taken during this period; such eggs averaged
only 1.20 mm. in diameter (range 1.02 - I.38 mm.). This is slightly
smaller than the average size of 1.21; mm. that F. N. Clark (193U, p. 3)
reported for mature pilchard . eggs from ripe females. The difference in
size between eggs newly-spawned and those undergoing development (average
size l-,70 mm.) could result from the latter having absorbed water, swell-
ing the perivitelline space up to the size characteristic of planktonic
pilchard eggs.

A few pre-cleavage eggs were taken in the saiiqples after midnight.
The majority of these were much larger than pre-cleavage eggs taken before
midnight (cf. tables 1 and 2), in fact, vfere about the same size as pil-
chard eggs in the same samples undergoing cleavage. Hence, they must have
been in the water for some time, and, consequently, must either have started
development very slowly or had remained unfertilized. Hovrever, a few
pre-cleavage eggs taken as late as mid-day have a very narroiv perivitel-
line space. Either these had been spawned at the normal time but did not
attain the typical planktonic egg size, or they resulted from isolated
instances of deposition during the day. In numbers such eggs were neg-
ligibly fev^.

Further support for the thesis that spavming is rhythmical, occurring
during a brief ,*relatively fixed time each day and not sporadically, is
furnished- by the time distributions of eggs undergoing cleavage. Thus,
eggs in early stages of cleavage v/ore taken immediately after midnight
and not during any other period of the day, while eggs in the last stages
of cleavage were collected in samples around noon. This could occur only
if spawning took place during a brief period each day.

The following account of the length of time of development assumes
that spawning is confined to the period, 8:00 p.m. to midnight, by taking
the midpoint of this period, 10:00 p.m., as the hour when development is
initiated. True, using the time of fertilization rather than of spavvTiing
vTOuld be more precise, but the two processes are believed to occur close
enough together to justify neglecting the time difference betvieen them.

135

1

.^1

iiim.

1

.22

II

1

.19

It

1

.15

II

1

.07

It

1.

31

II

1.35

It

1.

20

It

1

20

II

Table 1. — Samples in which pro-clea.vagc3 uggs were taken
(Regular survey cruises of I9UO-I9UI)

Number N^ombcr Number Av. size
Time of pre- of of pre-
station of cleavage stage II stage III and cleavage
No. collection±/ eggs^ eggs stage IV eggs eggs

11
12
22

U
19
15

6

100^/

1

1

1 30 - 1.60 "

1

1 13 - 1.78

1

1 156 - 1.05 •• , /

U 808 - 1:0U; 1.70 nun.ii/

1

1

3

1 lU - 1.60 mm.

25 - 1.51i "

3

1

2

1 - 60 1.65

1

8

1

1 - 193 i.oU

2

1

1

— The time of collection as used in tables 1 and 2 refers to the time when
the plankton net was going through the 20-10 M. zone.

2/

— ' Number of eggs in the sorted portions of all ntts used at a station,

v/hether of 2 M, 1 M, or l/2 M diameter.
1' Approximate number,
ii' Eggs of two size groups.

136

171.1;

20:16

1752

20:3U

1736

2I1I5

2135

21:20

1751

21:U8

1731

21:53

112

22:17

105

22:30

1952

23:33

2115

23:37

17U5

1:02

1926

2:2U

1915

2:U7

336

U:00

128

5:U5

171U

6:U7

2325

6.58

17U3

7:U9

Ull

8:05

UlU

8:27

1955

8:35

U05

9:30

19U6

10:32

232U

10:36

1715

10:53

IIU

11:00

132

11:15

17U2

12:10

2113

12:27

U09

12:U5

2163

12»U7

330

13:03

I2I1

15:17

1726

17:25

2

—

38

-

30

—

75

-

13

-

U33

— •

156

—

808

_

hS

—

307

-

225

—

lU

-

5

-

t

2

7

1522

—

60

1

80

523

196

103

1

-

193

116

15

-

67

-

13

Tablo 2. — Pre-rloavagc eggs -inc samples of transection-

1/

Number

Number

Number

Average size

Time

of pre-

of

of

pre -cleavage

Station

of

cloavage

stage II

stage III

eggs

No.

collection

eggs

eggs

Rggs

1876
1877
1878
1879
1880
1881
1882
1883
188U
1885
1886

1887
1888

19:00

20:06

21:06

22:08

23:11

00:13

1:12

2:23

3:27

U:3U

5:22

6:30

7:30

26

112

91

68

3

5

h

5

9

10
2
2

2
12

57
330
1198
1088
2157
23U7
1136
156

1.07 mm.
1.13 »
1.28 "
1.32 "

1.68 "

195

306

1/

This table is based on collections from a special series of stations
occupied on the night of April 8-9, 191^1, vrfien stations XTerc taken on
a line extending from off the southern end of San Clomente Island to-
ward the mainland at 2.5- to 5.0-mile intervals, with about an hour's
time elapsing between successive samples. Fortunately, this' series
of samples was obtained from a region where heavy spavming was taking
place the night of collection.

137

DURATION OF DEVELOPMENT

The length of time required by pilchard eggs to develop to the hatch-
ing stage after being spavmed and fertilized was roughly deterrainud by
placing pilchard eggs taken in plankton samples into hatching jars aboard
the vessel while at sea and following their de-velopraent. This was done
during several cruises in I9UO. In some sanplds the development of eggs
v;as followed from the blastodermal cap stage to hatching. From these
experiments, it was concluded that the pilchard egg ordinarily required
about three days to develop to the hatching stage from the time of spawn-
ing. Because the temperature of the hatching- jars could not be kept
constant vifith the facilities available at sea, it fluctuated several
degrees during each twenty-four hour period. Hence, the data derived
from these experiments is not of direct use in the study of the influence
of temperature on the rate of development.

Since the pilchard egg is spawned only during a fevr hours of the
twenty-four (8:00 p.m. to midnight), and since it requires several days
for the process of development, a plankton sample should contain eggs
in various stages of development j furthermore, there should be a sharp
Separation of the stages among eggs derived from each day's spawning.
A tabulation of the stages of devjlcpment present in the samples (table 6)
reveals that a sample typically contains stages in various degrees of
development separated by gaps of one or two stages, which are taken to
represent the time gaps between successive spavmings. Thus, \rc have a
basis for separating the stages present in a sample into their respective
days of spawning^/; with the least advanced stage (or stages) taken to
represent the most recent spawning, and so on. In some samples one or
more days of spawning may not bo represented, but this can usually be
ascertained by a comparison with other samples. As a consequence, there
are often insixfficient landmarks to indicate the age of such stages as
are present.

Having effected a separation of the eggs in a plankton sample into
their respective days of spawning, the actual time of development of each
stage present in a sample can be determined. The midpoint of spawning,
10:00 p.m., is taken as the hour when development commenced. The time
of preservation of each plankton sample marks the endpoint of development
of the eggs present in the sample, and this time was recorded as a part
of the routine observations during collection of the material. From these
data it is possible to derive the probable length of time, in hours,
that each of the several stages in most samples had been developing.

WATHK TEI-IPERATURE3

In correlating the development of pilchard eggs with the temperature
of the water at the station of collection, it is necessary to express the

£/ Separation into days is indicated for each sample listed in table 6.

138

temperature at each station by a single-;' value. For that .purpose the av-
erage of the temperatures at 10 and 20 meters was used. 2' Ti.j.s choice
v;as based on the results of a series of horizontal hauls made with closing
nets (cf. SillLTian, 19U3), which showed that although a few pilchard eggs
are occasionally found deeper than 50 meters, nearly all occur above 30
meters, and by far the greatest concentration- is usually between 10 and
20 meters or even nearer to the surface. Surface temperatures, however,
were not included in the average, because of their rather marked fluctua-
tions .

To bo sure, the tcnperature readings may not represent exactly the
average temperature to which the eggs had been subjected from spavming
to the time of collection. HoTrever, we can assume, relatively safely,
that the temperature of the sea water usually had not changed very much
during this period of time, which, on the average, would have boon only
about a day and a half.

Pilchard eggs vrcre found developing at teir^eratures ranging from
12.5° C. to 17.6 C. as measured by the mean of the 10 - 20 meter levels;
this is an over-all range of slightly more than 5.0° G. Hovrever, the
majority of samples containing pilchard eggs wore taken at temperatures
betvraen 13.5° C. and l6.Q° C, or v;ithin a range of ^,5° C.

INFLUENCE OF TE].P:mA.tURS ON RATE OF DEVELOPLLiNT jF PILCHARD EGGS

The priBiary information upon which the analysis of the relation be-
tween temperature and rate of development of pilchard eggs in nature can
be made arc the following: (1) The estimated age, in hours, of most pil-
chard eggs from the time of spawning to the time of collection and (2)
the temperature of the water at the depths where the majority of eggs,
were developing, taken at the time of collection of the eggs.

An analysis could not be made of the influence of ten^Jerature on
the length of time required by the pilchard egg to develop completely
to hatching, because of the difficulty of determining, from the preserved
material, when hatching would have occurred. However, a correlation was
possible between temperature and length of time of development of the
pilchard egg to the stage immediately preceding hatching (stage XI, cf.
appendix).

Not all records of stage XI could be used in the correlation, since,
in s'ome samples, only a small percentage of the eggs referred to a common
day of spawning had reached this stage, the majority still being in the
preceding gtage; also, in a few samples the age of the eggs could' not be
determined vdth certainty. Therefore, the folloi/ing criteria were set

- Tomporaturos at 10 and 20 meters wore usually closely similar, and both
depth usually occurred in the stratum above the thermocline. For more
than 50 percent of the items used in the subsequent correlations, the
difference in temperature between these tvro depths v;as less than 0.1° C,
and for only' one -sixth of the items vras the difference greater than 0.5° C.

159

up to select' the records of stage XI to 'be used in the analysis: (1)
At least four-fifths of the pilchard eggs of a given day's spa;vning must
have reached the stage, (2) there must be at least tv;o specimens of the
stage in each included record, and (3) there must be a reasonable cer-
tainty about the length of time of development, based on evidence of as-
sociated stages of development .li/

''iThon the length of time of development of each record of stage XI
meeting the criteria v/as plotted as the dependent variable against the
temperature of the \vater at the station of collection, a scatter of points
was obtained that apparently could be fitted best by a curved line. By
using the logaritlim of the time of development for plotting against the
temperature of the water a straight line distribution resulted (uppermost
curve in figure 1), which was subsequently fitted by the method of least
squares .

A similar correlation of temperature v/ith length of time of develop-
ment was also made on three-earlier stages of pilchard eggs. III, VI,
VIII-IX (combined to give more items )£/ The fitted lines for these stages,
also shovm in figure 1, have about the same slope as the fitted line for
stage XI. In fact, Vi'hon tested statistically, the four lines were not
significantly different from each other (P>.05). The slopes are defined
by the b values in the equation Y = a+bX; these values are listed in
table 3 J together with other pertinent data concerning the correlations.
There is no evidence th-.t the relation between temperature and rate of
development changtid during incubation. Furthermore, the results obtained
for the four stages are so consistent that they justify confidence in
the reliability of the method of analysis. The time (in hours after spawn-
ing) to reach each of the four stages at teraperatiires between 13.5 and
17.0° C. is given by 0.5 degree intervals in table U. Within the tem-
perature range investigated, the time to reach a stage of development
is increased by about 7.15 percent for each 0.5° C. decrease in tempera-
ture. The total time of development to hatching, could be approximated
by adding two to three hours time to the values listed for stage XI at
the higher temperatures, and proportionately more as the temperature de-
creases.

Although consistent results are obtained for the several stages,
there is a considerable spread of points about each line. The factors
causing this variability may be associated with either the tine of de-
velopment or v^ith the temperature. Variable elements associated with
time, undoubtedly account for most of the spread about the fovir fitted

-' The records of stage XI meeting the above criteria are indicated in
table 6 by an asterisk.

2' All records of these stages that could meet the three criteria pre-
viously described for stage XI v/ere used in the correlations; these
are indicated in table 6 by an asterisk.

140

FIGURE

V)

<r

O

I

o

1.40 -

1.30

in

•

•

1.2 0

•
•

•

•

I.IO

1 1

1

1

1

- 7

3i

liO

lao

14.0

i&o

16.0

17.0

TEMPERATURE (°C)

FIGURE 1. - HSCafflSSIOK DIAORAM SiOWIWO WE VmXSWCS. OP TBIPai/lWRS ("C) ON TOS TUB OP DSVSLOHOII
(LOO. HCURSJ OF FIUHMID EDQE TO lOU) SSICTSD STAISS OF DEVELORtBII.

141

o

(1)

s

•H
4^

o

rH
II

w

Pk

B

nJ

0)

-tj

II

ii

>H

!><

u

3)

•H

h '

o

CO

tJ

o

•H

m

CO

«
Ul

0)

EH

>»

•H

r~

to to

cr>

a> iJ

60

C! Td O

ITn

•H 1) CD

O

+J CD

•

o

fw -H -P

1

?1 «!-( rH

o 3

(x, g

v^

»]

o

cu

lr^

o

o

(-4

iH

^

OJ

>JD

CO

X!

VD

CO

60

CO

r-*

o

<-H

OJ

o

Lr>

H^

0)

o

fH

CO

r—

o

o

SU)

(H

«

cti

lr^

f-<

oj

-p

pH

1

CO

ixi

t-4

1

o

r-l

•?>

in

o

H

o

to

CO

o

t-i

r—

O

r^

o

r—

1— t

VD

r^

OJ

i-i

VD

OJ

>

J-

r^

h-

UD

O

o

0)

^

.-I

oJ

o

W3

l-H

1

«i

+J

en

l-H

o

r—

1 —

KO

o

>

o

r^

r^

KD

r^

^

CTi

^

CTi

t^-

cr.

(U

CO

O

f^

CTi

ur\

f^

y)

O

ir\

ro

o

o

03

•

•

•

+J

\r\

rH

OJ

o

CO

(H

I

IM

o

CO

rH

r^

o

M

o

r^

O

in

OJ

l-H

cri

OJ

iH

r^

f~-

VX)

CM

rH

OJ

^S>

1^

<u

r-i

CTi

OJ

OJ

O

o

W)

f

■

(ti

J-"

rH

OJ

o

■<->

H

1

w

to

:=51

S

•

0)

■p

■p

w

•H

IX

|!«

cd

,o

0)

.

{>'

o

CD

s

"^

I

CO

S

•H

■P

to

(D

(h
O

1^

0)

Id

w

a ^51

142

Tabic i|. — Tine (in hours and minutes after spi-vvning) required to reach
a staf;e of devclopncnt at different temperatures

Tecperaturo

stage

III

Stage
VI

Stage
Vlil - IX

Stage
XI

17.0°C

.-

26:17

-

51:31

16.5°C

-

28:05

U1:0U

55:07

16.0°C

13:U9

30:01

hhi03

58:58

i5.5°c

1U:56

32:OU

U7:1U

63:05

1^.0°C

16:09

3U;16

50:U0

67:29

iU.5°c

17:27

36:37

5U:21

72:12

iU.o°c

13:52

39:08

58:18

77:1U

i3.5°c

20:23

hi: 50

62:31

82:38

148

lines in fi(^re 1. First in importance is the time duration of a stage
of development which occupies several hours, whereas a sample may be col-
lected during any part of the stage. Consequently, the estimate of age
of any given specimen would err, on the average, by a quarter of the time
occupied by the stage in the egg's development - a matter of three hours
at most. Second, there is the period of time over v;-hich spawning is spread
(chiefly 8:00 p.m. to midnight), which is represonted in our calculations
by the midpoint of this time (10 {00 p.m.). This might result in an error
of as much as tviro hours in a few instances. Third, not all egfjs develop
at the same rate even under identical conditions, although the variability
resulting from this cause is probably slight. Fourth, there is a possi-
bility of an error in the day of spavming to which a stage is referred;
however, an error involving tvrenty-four hours time should result in a more
narked discrepancy than any that have been observed. The temperature
data used would also contribute to the variability fn so far as it dif-
fxired from the actual temperature under \/hich the majority of eggs wore
developing at a station. This could result (1) From the concentration
of eggs occurring outside the 10- to 20-mcter zone in waters of different
temperatures than those occurring in this stratum, and (2) from the tem-
perature data not reflecting the mean temperature for the period of de-
velopment, as v^ould result if the temperature of the water changed ma-
terially during the period from such causes as the stirring action of
strong' winds^/ or upvrelling.

However, the variability was not sufficient to destroy cither the
correlation^ or the consistency in slope of the regression for the four
stages.

Since the Arrenhius equation Has been used by previous investigators
studying development of fish eggs under constant temperature conditions,
it is- of some interest to analyze the present data using this formula
for the sake of comparison. In doing this the logarithm of rate of de-
velopment (log l/t) is plotted agaihst the reciprocal of the absolute
temperature (1/T° abs.). ^ii"hon this is done for each recoird of each of
the foui^ stages of development previously analyzed, and a line fitted
to the s<:atter of points for each stage by the method of least squares,
as is shovm in figure 2, the four fitted lines are, of course, still
nearly narallcl, and are closely comparable to the lines derived in fig-
ure 1,1/

2.' The temperature would be Icn/ered if deeper, colder v/ater was mixed
with the upper I^^ers through wind action; contrariwise, the tempera-
ture could be raised somev;hat by a mixing of warmer^ surface water
with the waters of the 10-20 m. stratum by 'vj'ind action that left
depths belov/ this undisturbed.

7/

Both methods are Included in the paper since, for our purposes, the
correlation shorm in figure 1 is more usable, v^hile the Arrenhius
formulation is needed for comparison of the pilchard with other fishes.
The Arrenhius method uses the reciprocals of the values employed in
the first method with the exception that the tomporaturo is expressed
&& absolute temperature.'

144

To compare the effect of teraporature on pilchard development vd.th
that of development of other fishes, the temperature characteristic (ivfu),
also knovm as the thermal increment, has been calculated for each of
the four stages of eggs and given in table 5» Since the temperature char-
acteristic depends primarily on the slope of the lines, and since the
four fitted lines, although nearly parallel, are not exactly so, each
stage has a somev;hat different Mu value. These vary from 25,800 (the
value furthest out of line) to 22,100. Since the individual slopes, are
not significantly different from the common slope (four lines fitted
simultaneously) the latter may be taken for computing the average tempera-
ture characteristic for the v^rhole period of development, v^hich is 22,^00.
This is of similar magnitude to the fev; Ivlii values that have been deter-
mined for the development af eggs of other teleosts, which range between
16,700 and 214., 900. Hence the pilchard is not atypical, and the observa-
tional evidence here used gives substantially the same conclusions as
the more commonly used experimental evidence.

SIMIARY

1. The ratu of development of pilchard eggs as influenced by the
temperature of the water v/as studied from samples and data collected
in nature during the 19U0 and 19i4.1 surveys of pilchard spavming off
southern California.

2. Evidence based on the time of occurrence of newly-spavmod pre-
cleavage eggs in the samples indicates that pilchard spavming is largely
limited- to the four-hour period between 8:00 p.ia. and midnight.

3. Experiments conducted on the vessel at sea shovrcd that the pil-
chard egg ordinarily required about three days to develop to hitching.

k- It is usually possible to separate the eggs according to the
several days of spawning represented in the samples , and to determine
the probable length of time each stage had been developing.

5. The temperature range over v/hich pilchard, eggs v/ere found de-
veloping in nature was from 12.5°C. to 17.6°C. Hovrcvcr, the majority

of pilchard eggs vrcre taken at temperatures bet'v.'een 13.5° C. and l6,0 C.

6. The Correlation of temperature with the rate of development

of pilchard eggs was analyzed for four stages of eggs, Sir.iilar results,
were obtained for all,

7. There is no evidence that the relation between temperature and
rate of development changes during incubation.

8. The time required to reach a stage of development increases
about 7.15 percent for each 0.5° C. decrease in temperature.

9. The temperature characteristic (Mu) for the pilchard egg during
development is about 22,500. This is of similar magnitude to the few

lAi values determined for the development of other fishes,

146

.00344 .00345 .00346 .00347

.00348

.00349 .003 50

Uj 'ABS

FIOURE 2. - IffiBUBSSICN DUORAM 310UINO THK DUUIiNCS OF TBWPSRAmRl (lA^rtt. ) ON WI :«TB OT
DEVnOFHSilT (LOO.l/tlOURS) OP FIUHARD BIOS TO FOW SSJL'IU STASSS OF t^MtUSKXt.

146

en
o

a)

e

•H

-IJ

hi.

o

r-\

r-i

II

t«

X)

§

P

^

E

X

0)

x>

+j

en

,Q

CO

ro

II

i-t

>*

II

X

0)

•H
<H

Pi
O
<M

CD
-l->

Ki
Tj

Pi

o

•H
(0
03

UJ

t

EH

>

<)

rH

tn tn

Q) p)

C -d o

8

•H i; cu

1-4 -u p!

+5 CO

b •'-^ ^

fVl

^ tM iH

C\J

O ^

fe s

•H

CO

MD

o

f-H

60

r-t

r**'

o

O

1
60

(T\

ir>

tu

o

^

ir>

in

r^

o

O

f^P

1— 1

f^

t>-

1 —

60

cr,

O

CO

O

r^

CVJ

60

C\J

LTn

jj

O

i-i

iH

J-

O

W5

60

ITv
i-l

1

d

KJ

1

-rf-

o

l-H

J-

I-I

H

ir\

1

o

I-t

«^

1 —

rH

rH

>

-d-

J-

OJ

f—

C\J

rH

O

'X>

CPi

r~

f^

r^

(1)

o

1 —

VO

o

C\J

O

M

o

Ol

1^

m

o

•.

CO

•

•

•

1

r^

-tJ

CO

irv

o

C\J

CO

I-)

ir\

o

1— 1

rH

iH

k^

C\J

1

o

VX)

r^

-d--

r^

•

(I;

60

^

>^

O

^

o

o

tiO

;:C^

li^

r-{

CTi

CJ^

o

CO

o

vu

r-i

t^-

1^

rH

-P

o

^

iH

J-

O

••

in

•

•

60

in

I

CD

H-t

(T>

o

H

O

iH

1-4

>X)

cli

m

r~

0)

r-l

J-

C\J

r^

ir\

O

o

W

N^

f—

VJ3

o

60

o

CO

o

60

m

^.o

1^

60

■»^

o

r—

OJ

ir\

o

*

CO

•

60

•

60
rH

I

o

ir\

^.

m

S

-p

<u

w

♦»

0)

t4

pj

IX!

\>i

CO

,o

>

£

O

cu

l^'

-p

-p

CO

-p

o
I-I

O

u
u

CO
CO

-p

CO

147

APPENDIX

Tho development of the pilchard egf^ was separated into eleven stages
during the investigation of pilchard spawning, a brief description of
each of which follows e

Stage I

In stage I v;ere included all pilchard eggs that were not undergoing
development, either because they vj-ere so recently fertilized that cleavage
had not yet begun or because they vrere unfertilized. Pre-cleavage eggs
are usually smaller in size than are pilchard eggs undergoing devclopmentj
the difference is not in the yolk mass, Virhich is fully as large in the
pre-cleavage eggs, but in the width of the perivitelline space, which
nay be nearly wanting. Ilany eggs in this stage have an accumulation
of protoplasm at the animal pole to form a blastodisc.

Stage II

The period of development of the blastodermal cap is covered in
stage II. Although this stage includes the early cleavage divisions,
no pilchard eggs have been observed in our samples v>rith fevirer than about
128 cells. The lack of material shoviring early cleavage divisions is hard
to explain, ^specially since stage I eggs were taken previous to initiat-
ing cleavage. Such divisions must take place very rapidly, indeed.
While the individual cells arc fairly large and quite evident, the blas-
todermal cap has a rugged, berry-like appearance. As the cells become
minute and indistinct through repeated divisions, the cap assumes the
shape of a smooth, roimdcd, lenticular dome. The pilchard egg floats
with the blastodermal cap dovmwards. At the opposite end of the yolk
mass is the single, large oil globule. This stage persists until seg-
mentation cavity is formed.

Stage III

Stage III covers the period of development from the first appearance
of the segmentation cavity to the definite establishment of the embryonic
shield. The segmentation cavity is formed v;hen the blastodermal cap sepa-
rates from the underlying periblast in the middle portion of the cap;
the' cavity forms soraev/hat eccentrically, leaving the blastoderm on one
side of the cavity somev/hat thicker than elsewhere. The thickened area
is the; region of the developing' embryonic sheild. However, the outline
of the embryo is not definitely defined in this area until after this •
stage. The germ ring begins the envelopment of the yolk v/ith a cellular
sheathj by the end of the stage the ring has enclosed a third of the
yolk mass,

Stago iV

This stago begins whon the general outline of tho embryo can be dis-
cerned along the median line of the embryonic shield. Tho germ ring at
the beginning of the stage encloses a third of the yolk mass; the end-point

148

of the staf^o is arbitrarily fixed at the period of development when the
germ ring encloses four-fifths of the yolk. By the end of the stage
the embryo extends about two-fifths of the vray around the yolk, v:ith the
cephalic region both thicker and wider than the remainder of the embryo.

Stage V

The period of development covered by this stage occupies the final
enveloping of the yolk by the germ ring and the closure of the blastopore.
The germ ring continues its centripetal gi-oirth until the yolk is completely
encompassed by a cellular sheath; the final closure occurs just back of
the posterior pole of the embryo, 'wliile the germ ring is completing
this closure, there is a thickening all along the ombrj'-o, although the
resulting dorsal ridge is still most pronounced in the cephalic region
and tapers off posteriorward, so that even by the end of this stage
the tail is hardly differentiated from the surrounding tissues, and the
embryo itself is a simple, undifferentiated' ridge in which somites are
not yet defined. Eggs are considered as belonging to this stage until
the optic vesicles arc distinguishable in the head of the developing embryo.

Stage VI

This stage begins after the closure of the blastopore when the out-
line of the optic vesicles can be seen in the cephalic region of the
embryo; it ends when the tail starts to separate from the yolk. The
optic vesiclos occupy fully half of the head during this stage; the somites
take shape, being evident all along the embryo except in the imr.iediate
vicinity of the head and tail; the embryo extends about two thirds of
the v;ay around the yolk; the single oil globule in the yolk is a little
beyond the posterior end of the embryo and retains this position in the
yolk mass subsequently. This stage is of longer duration than any of
the preceding three. By the end of the stage, the pupils can be, discerned
in the eyes, the embryo is about of equal thickness along its entire -
length, v/hile the tail is somcvirhat swollen at its tip.

Stage VII

This stage commences v/hon the tail begins to separate from the yolk
mass; it covers the early growth of the tail until the free portion is
about of equal length to the head of the embryo. Although the demarca-
tion of the beginning of this stage is rather definite, its termination
is somewhat arbitrarjr. All the remaining stages arc separated primarily
on the degree of development of the tail, as it offers the best landmarks
on which to base separation into stages. During Stage VII the tail re-
mains of uniform thickness along its length; the posterior end is rounded
and even appears somei/hat swollen in some individuals; the fin fold is
barely discernible, extending aroxond the free portion. of the tail and
along the dorsal side of the body to near the region of the head; there
is either no differentiation of somites in tht; free portion of the tail
or at best a fevj- somites are formed at the base of the tail by the end
of the stage; along the rest of the body the, somites stand out sharply
when viewed from above. By the end of this stage in most specimens the
free tail extends up to, but not beyond, the position of the oil globule,

149

stage VIII

At the beginning of this str.ge tho developing tail is free fron the
yolk mass for a length greater than tho length of the hcadj by the end
of the stage about one-third of the body at the tail end is separated
fro"i the yolk. The fin-fold is bccnning conspicuous, although it is b-i-'oly
half as Vifide as the enbryo by the end of, the stage. The intestine can
be seen as a line along the ventral side of the developing tail, and tho
position of the vent is indicated in the fin-fold. Somites are formed
along the tail, at least as far as the end of the intestine. - The tail
tapers posteriorly, but the tip is still rounded.

Stage IX

Approximately one-third of the body is free from the yolk at the
beginning of the stage and the tail is often bent- to one side^ by the
end of the stage the tail is always curved, either to the right or left
of the yolk, and comprises about f/ro-fifths of the total length of the
embryo. The fin-fold is about hair as v/ido as the body at the beginning
of this stage, it is somevThat broader along the free ventral portion of
the body than along the dorsal side; by the end of the stage the ventral
fin-fold is nearly as vjlde as thvi body proper, During stage IX pigment
spots (melanophores ) may appear on the dorsum in two parallel rov/s, one
on either side of the notochord, with the spots more niuiierous ii-.mediately
behind the headj on some specimens pigmentation does not appear until
after this stage. The embryo conforms less to the shape of the yolk
than previously and is beginning to straighten in its middle portion;
at tho end of- this stage the tail is curved to one side of the yolk at
about a U5'^ angle to the plane of the embryo.

Stage X

The free tail comprises at least tvro fifths of the length of the
body at the beginning of the stage and is as long as the remainder of
the body by the end of the stage. During stage X a fundamental change
in the orientation of the embryo is completed, a change bugun in the
previous stage. This is accomplished by the straightening of the body
of the embryo and the upward flexing of tho tail. The embryo as a whole
is no longer curved aro\ind the yolk; rather the plane of orientation has
been rotated a con^leto 90°, Only the head end remains curved around
the yolk, the remainder of the body being straightened out into one plane,
although it is U-shaped due to the flexing of the end of the t^il? During
this stage the tail increases, in length, until its tip is nearly as far
forward as the base of the head; the fin-fold is about as wide as the body
along the dorsal side and somev/hat i/ider along the ventral margin.

Stage XI

This is the last stage before hatching. At the beginning cf the stage
the tail stretches as far fonrard as the base of the head, and the pos-
terior portion of the body free from the yolk sac is longer than the por-
tion carrying the yolk sac, and this disparity increases during tho stage.

150

The yolk mass is novr oval-shaped j the fin-fold is wide, bein,^ 1 1/2
to 2 times as wide as the body; the tail is divided into somites
to its very tip, which is now somewhat pointed. The eyes are still
colorless, pigmentation taking place after hatchin;j. This stage
ends with the hatching of the egg.

'ig.

1

II

2

II

3

II

h

It

5

It

6

II

7

II

8

II

9

KEY to Plates

Plate 1

Stage I

Stage II, fairly early

Stage II, late

Stage III, intermediate, dorsal view

Stage IV, beginning

Stage V, beginning

Stage V, imiite iiately be'fore closing of

blastopore, ventral view
Stage VI, early
Stage VI, late

Plate 2

Fig. 10

-

Stage VII, early

" 11

-

Stage VII, late

II 12

-

Stage VIII, late

" 13

-

Stable IX, late

•• lU

-

Stage X, early, dorsal view

" 15

-

Stage X, early

" 16

-

Stage XI, early

II 17

-

Stage XI, intermediate, dorsal view

" 18

-

Stage XI, late

151

PLATE I

152

PLAT E 2

153

V£)

CD
EH

IH

P.
O

O

Eh
CO

X

t«!

X!

CD

EH

I

0)
P. Q'

a> h

S o)

•H tn

Eh

a

O
•H
-P

CO

to

J-

.-1

CM BO

C\J

f^

rH ,-1

»

rH

O C\J

♦

.H

rH

*

OJ #

»

*

OJ
CV)

J- ir> OJ

CO

to

OJ

OJ

c\j
»

t^\* r-i * * # c\j

# »

VD r— m to r^ o i^

rH * C\J r-l J- #

* * *

OJ

jd- CM in

00 c\j c\j

OJ I^

in

CM

O

O

CM

Lr^^D
IjC^

rH J-

O
O

»

O CM

CM

LC^

J- rH

CM i-H

•CTi

LC^ r-H

CT>

CM

hH

in

CM

rH CO
CM

CM
f-l

*

ON

r-

o

in

CM
»

r-\ -t

CM

OJ

t^

CM

ir>

^o

1^

CM

CM

r^

*

»

*

r-l

k-O

r-i

VX)

:5

f<^

IC> CM

m I

r^

60

CM

CT\ CM

f^ CM -^J- O

I — r— iH

o #

I — VD

*

60

in

VD

CM

VX)

in

CM

inoCM CMJ- OC3>r^60OC0 inviD CJN CM UD CM a^ t — VI3 in

cr\^ vD cr>CMJ- invnuD i-t cri(~nrncou3vx) r^c— ct>J" i —

rH CM CM 60 m in -If

rH in J- ^ 1^ CM o

o ^- J' cr\

rn^ r<^ m,:f m J" in ^^^J:^ r^ urx in in j- in in in j- in j-

oinooininooinootnooooinoooo

CM -::t rH r^ J- rH hnOint-Oro>rHN^OO O in J- CM in j-

in in m in in^ U5

r^ i-\ r-\ r^ r-^ <-\ ,-i

o o o

I o in o
' J- inid-

in in inun

in o in o

K-\ CM rH o

^ Cr^ J- 60CM^X>CM60'rHCM60rHt — r^LnrHO'^TltOcnLn
rHrHCM t-\ C\i r-^ i-\ t-i r-{ r-i '^i r-\ r-{ r-\

r-^ OJ t^^ in 60 cr> o rH CM N^ J- in r>- 00 cr> o rH CM ro-d-

OOOOOOOrHr^r^^^r^r^^^rHr^CMCMCMCMCM

rH >JD CM vx> t-\ m r — <

rH CM !

00 60 CM

invx) r~- 60 cr» o rH

OJ CM CM CM CM t-n r<^

CM t^^ m
m t-n mrn

u

u

p<

cvi

p,

o

01

0

u
o

to

CO

o

O +J

>j o

o

c

•H

o

•H 'D
tj <U

m
o >

•H CO

cd

rH +3
0)
^1
(h
O

o

o

o

rH
>

t5

^

u)

C
•H

CO

a;-

CD

u>

CO o

P) rH

rH

CD O

(h Ch

'i-

CO

03 TJ

.H CI)

^1 +3

OJ CD

4J y

U3 .H

CO 'd

CO

03

CO
>>

CO

cO O

a +^

(3

03 -rt

'J

«) d

+3 -rl
CO 4^

♦

154

B

x) o
cj 6oo!

TO
9-1

X

r-H rH

r —
*

rH

r^^.J-

OJ nH
■*

OJMD

^r^ cm

LfN fH
.H

VD CM

CM
.-1

CM rH

CM

«

»

rH CM rO Cr\ rH r-H

OJ

i-H *

1 —

O
i-H

to
*

^<-^ rH to

»

.H

60 in
■-1 »
♦

Pj

s

a:

CH
t

P< <D
>

0)

a

o

si

-t-J
CO

CO

Lr\

CM

W

CM

r^

*

♦

t-i

I —

LOi
CM
»

r<^

CM r-^ I^ CM

ITS
*

CM

*

CM J-

OJ
»

U5

r-\ CTl

VX>

CO

60

o

*

u^

rH

»

rH

r^

OJ

'^

60

^<-> OJ LTv

CM ir^

o

J-

CTi

»

*

O rH

CM

*

CM
CO.

I — CM

^

CO r— cr.

O J- UD

Lf> cTi J" oOcr»ocoi — co&oovx)WrHeovDOOrHcrieo

vnvnOCnOOCOCMLPiCOrH r^ J- .J- r<^ r^ t^vX> J- LPi OJ '^

Ln ir\LO

LP> O LPi O LC^
;d- LPi K^v r^ tr\

r^^ J- ,^ LTMCM^ LTMn r^^ Lr^Lr^l^^ lp\ j- ^ J- irMt^ lo^

rHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrH

l^^OOOOOLr^Olr^oooOOoOLfMr^oooo

CMrHrHinojLPiCMOJJ-^ rHr^OJ-rHj-J" CMr^rH CM^<-^

r^ CM 1^ CO CM
CM rH

^r> >- CO a^ o
r^ r^ ix-N r^ J-

r— C3^r— CM r<M — J- oj^ r-^vD OJ rH vr> CM I — rH r^ k^^^o rH j-

rH r^OJ rH 'AlrHOJOJ r-\ l-i r-\ C^ <-A r-t i-i

r-i CM- rr-\^' LTNVr) CO CTv O

OOOOOOOO

OJCMCMCMCMCM""'

CM r— cr. O rH CM ^0-=^ U^ CO C3^ O
— ^....rHrHrHOJOJOJOJOJOJCVJCMj;^
OJCMOJCMCMCMOJCMOJCMOJCMOJOJCMCM

u; o
li) -H

^ "cti

05
O
-(-^ CD

cd

o

-U 03
CO >-.

ni

pi

rH O

tJ O

o
o

o

O

o

I
>5

d
0;

■H fn

CO CM
■U rH

CO v_o

CM

1) •
C5 CO

o o

+^ a

•r) 0)
(U

c! •

CD a;

P, fi

CO .H
rH

to U

'\> OJ

CO C
CO

OJ O

rH rt
•rl

J- OJ

^ !o

(0

o

■4J

0)

o

CO O
O

o - ■
? o

a;

■ H
rH

'd

UJ O
fn -P
CO

O

O

O

CJ
<0
CD

O
O

CM
tM

to

!>s

CD

o

to O

•H O

JJ >>

f^fl CD

CO td

0) W)

tJ .rl

155

X

m
S

o

r-H

(1)
>
0)
Q

"^ Cm

|o
(5 dj

fiO

I

U3

rH

CD
E-i

I

0.

p^

a,
S

•H
EH

a
o

•H
■U

ID

+3

CO

O

60

♦

CT^co

r<^c\J

rH OJ-:f

o

rH

in

O i-t rH
OJ rH rH

rH

cu o J^

■to OJ

c\j m LO

OJ

ro

1^

»

O^i^D

J-
♦

r<^ rH rH

*

ViD J- a>
»

»

to

KD

!?^

60

H

in

'60

in|,:±
♦

CM

fn

C\J r^ O^ CTv

COS

fee

rH

OJ

in

in

Ln

H» o

' OJ

r^ CTN Oi

*

OJ ^X)

in

in

rH
rH
♦ •

OJ

GO

60 MD ITN

60 ^ rH

rn

VD

o r— o o

OJ rH 60

rH r<^ en rH

in

60

CVJ

^ r-\ mvD vo OJ v£) OJ OJ rH J- invjo OOJ- t^6o or<^60vr)o^o>vO o>vi) ino^vri ojJ-
h-T m ^-vr» cnojojrH 6or--ina^ojLn ctimd to ctvvd jt J- r-- J- o md vr> r-t ojkd r--^ 6o rH

J- in^ in J- ininininror^oj i<^ m J- -d- r-^-, r-~\ j- j- j-* ^ in in^' in in KMn in m in J-

oinooooLnooinoinooininoinooinoooinininLninininoin
OJ J- in in J r^ rv^ o o rH rH r^3- ^jd-inrHOr-oino^OrHin in J- oj in rH in j- r^

c\JJ"OOinrHinorHj- CO 1^ J- o^ m r^^ invx) ct\ en J" km^ r^ en J- rn ;-< vd to ro r^

OJ r-i rHrHOJ rHOJ r-i r-A r-t r-i OJOJOU iHOlrHrHrHrH

rH OJ ^ invD ^- 60 o^ o rH OJ r^ J- invD r- 60 en o rH J- inviD r^ en o rH r^vxi t— to en o
r^rnr--\r^hr>,r<-\K>irn^ O O O O O O O O O rH r-{ r-\ r-i rH.rH.rHOJ.OJ OJOJ OJ (M OJ ro
OJ OJ ojoj OJ OJOJ•OJ■c\»■K^K^m t^'t^ r<^mmi^rnrnrnrnmr^i^fnKNr<^m'^ ro rn m

156

•4->

s

Ph
O

I-)

<u >
5P

•H
-t-> Vi

rt o
o
o ®

I t«

• 'n
Kn ^

0)

>-<

EH

X

ir\ t<\c\i <j\

X J- i-i ou J- 1^

><

P*

s

<u

Eh

I

(U
Jh

Ph

<u

B

o

•H
+3
Cfl
■P
CO

CM

60 ro

CVJ

LP\

CM

to

J-

♦

»

CM t<->
♦ ♦

CM

CM iH

♦ \<-^

I*

CMVjCJ

r-4 IJD

WD

»

ISO

ITS

»

Hs

CM

irNv^

cv*

UTv

r^ cv

CO

CM

CM

CM

CM

CO

r— CVJ

J-

N^

CO

#

rvj v£)

1-1

CMrH COCMkD.I-^ i-l O

-CTi

C\J rH CM CO cr> CO CM J- CU

* U^ r-H iH ^ »

t~~ CO

r^ O
CM

»

o ■

r i
1

^

t

»

«

CM ^

CM

60

J- 60

f<^

BO (MUD 60 CJ^O 60 CO O 60 60 CO I — (T\ CM ^ O ^S} OOJ"j:t-CJ-. BO. Lricr.l~^'-M CMJ" 60 O I — CT>
rO J- J- I-OJ- Lf^ LTv f^'60 vr>. r-H VT) O^ LTN rH 60 60 r^ CM h— 60- CM VX> O%60.J-. <3M^ O 60 VD. CM Lf\ Cn

^* LC^ LfA J-* Lo Lr\^' ^ tr\ ir~jt cm t-f^ r^ ro ro i^ r<^>-o Id- t^^ J- r^ J- u^ ^<-^ invx" lc^ r^ ^-^. ^

r-liHi-Hi-HrHrHrHi-HrHi-HrHiHrHrHr-HrHrHf-HrHrHrHr-lrHrHrHi-HrHi-H/-HrH(HiHrHrH

lr)L^^OLf^lr^ljr^oooCOOOOLf^L^^l^^L^^lr^oOOOLr^oooQ.op lo ^O ^C

CM r^ CM rH r-l rH ir< O "rHT O CM' ,1^ O J" i-H' rH rH- CM- f<^^ r^ (^>d^- -CM- rH. r<-\ rH^ O CM J.j:t.r^r^

CTN t^ C7^lr^ r^ J- crsr--f-^l^r— r^LOOJ-60f^roi6060 CMr— CO.d-^^^^COOJCMr^^-CMCMCM
rH r-^CM rH rHCM t-t r-t ,-t r-t ,-t r-i CMrH CM r-i r-i r-i CM r-i CMrH

rH CM K>i^ lOMD t— 60 O M CM r^ J" U^UD r- 60 C3^ O M CM r^^ ir^U) T— CO a^ O rH CM N-\ J" U^

tA t^ hO hr>j M~\ ^r^ r^ ^r>,^ OQOCDQOOQOrHrHr^rHrHrHrHrHr^rHO^O^CUCMrj

157

Kl

X

X

+3
c!
<u
B
P.
o

>

o

0 CO

o
o

U3

0
r-i

EH

s

EH

I

<1>

tl '
ft

s

Eh

f3
O

.H

iH U)

r~- c\J

OJ

1 ^JD

o

r-t >J3 (^

rH C\J

LO OJ

O

»

O^ iH CJN

•

CM ir^

* rH

«

1 —

*

CM rH

^'^

vr> CM

rH r^

*

♦

CM

hO r~-vx>

CM

VD

CM

O

O

r—

i-^»

CM

♦

(M

r—

»

*

*

»

CU rH

cr>r~-

CM

BO

J-

r-*

i>^r^

CM

Ji-

*

♦

*

r-i

*

o

lr^

<7^

*

crs

LPl

r^

J-

^

vo

IVJ

»

»

r<>

r—

LTn

r-{

H

60

ON

r<^

rH

CM

rO O

CM »

♦

t^

m Lp»vx>

ir\

CO

t<^

*

ir>

CM

r-t t<S

fwinvD CM

CM

t<^VD CM rH ^ 1^ r— ^ h<MsoiriCMOrHCjO^ r^vx> CMHOr^OOJ- Or-tcr>0 CM C^^v£) t— ro CTi W

O O LTvrr^rH cTiO tovo^ p<^vx) cor — incT^. oo CM^ rMCOa>iajM-\r~-or^ ir\^ \jr\ir\cr\ O ^

<D KD ir^vx> ^ ITNVO U^^ ir\ ir\ ir*^ J- lfMr> f^ r<^^ cm i^ ix*\ rr\ J* trs rr> iri-^ f'^-::!' ^ ^ in IfMSO

: r-i r-t r-* r-^ r-i r-i r-\ r^ r-\ r^ f-^ r-i rA r^ r-^ r^ rA r-i r^ r-\ r-i r-i rA r-{ r-f r-i r-t r-i r-t i-< r-< r-{ r-( t-t

OOOQOOlTMnOOi

rH J- r<^ J rH K^ r-t CM rH f<A I

oirMr\OLr%oomoootr\om(

O KVd- OOCMCMCMrHJ^r-(CM^rH<

imoOLTiOLrimoo
1^ rH OJ- r^r^cM (^ir\

■ CM r-CM i<^crir-l r— CO k^^ cr»f-M*-0 mo cm r^cMVO •-{ umticm rH

r^ rHrH CM i-i r-i CM rHrHCMrHrHrHr-t

CM O CM r^ CM^
CM rH

mJ-

03 1^ 60 CTN O rH CM rr\^ mvi3 ^- 60 cr> i
CMCM CM CMr<^t^K>K^r^rOi^r^t^ r^-

CM r^ J ITWO r— 60 CTN O rH CM rv^ J- mvT) r- 60 CTN O
^ r IQOOOOOrHiHrHrHrHrHrHrHrHrHCM

O o o

.158

C

a>
e

P\
o

i-H
0)

(D
«

Cm
O

Ti a

P! TO

•H r/)
■(J

a
o
o

X

ft

El

<u

EH

I

0) tJ
f-i (U
ft t»

0) O)

o

CO

OJ
♦

C\i

H

U5

t^^

<vD CO

LO

N^

CO

CO

r^

c\j

CM

OJ

OJ

CM

C\J
OJ

♦

I —

»

OJ

OJ

^ o^

OJ

OJ OJ

ir\

OJ

CVI

Ji-

ll^

CO
OJ

CO

Lr^

|-^^^^a^K^cr^coa^oJBOL^^r^oco^^^<-^f^l — DOJ-r— coojcTNOJOOJO-=l'f^Lr\J''^'^'-^f^
Or— oo oioj^ r^i — o^o^ojt^ r^ J- OJ r-t ioiu3 cr\>X) ojLOOJ-oja^crNLr>ojr— r-iojvO'^rH

vr) r^ LPivX) ^JD vjD vn Lf>vD ^ LOV.O V£) J" LPiVD LO U^^ J- LPivX) j:t i^ r^^ lr^ J- ro LO ro^ (v-> ijOi LTv 11^

rHiHrMrHrHrHrHrHrHiHr-HrMr-lrHrHrH<Hr-lrHi-Jr-lrHrHi-lr-lr-li-Hi-li-l(HrHr-l|-HfHrH|-l

oooooooirMr\ooooooirMi^ooooLPvLrimooLriif^ooL0Ou^gi£>Lr»

rHOJrHLPiOOrHLO. LP. J- ^ rH r-l LPi J" OJ r^ iS^st i-HLOOJ^ I^Or-l Ojj" OJJ" OO r^-3" -J f]^

o^ K^ K>, r~- ro J- crvK>covr)OJOir\OojJ- o^ J^ O k^ '-i f-~ i^^ o oj r~- oj r— cr<^ loud oj r-- ro

OJiHrHOJ OJrHrH OJrSrHOJ rHOJ i-H OJrHOJ i-lrH c-t rHr-H

rH OJ r^^ Lf^vO 1^ CO Cn O 1-4 OJ t^-:^' LPvVX) h-COCTNOr-HOJi-IOJ hoJ- LOUD I — CO O <-H OJ f^ J- LPi
OJ ajl.\J OJ OJCVI OJOJ Ojr^rOt^rOt^r^r-o^N^,-^ l-r^^ ^ ^ OOOOOOOOr-lr-lrHr-HrHr-H

LOlr^Lr^lr^lr^L^^L^^lr^^^^L^\Lr^Lr^LPlLr^L^\L^. LPvLOir^LPvU^ ltnvd vduimD'^'^o>X)vX)'vD1vDvx)Ududmd

159

0)

a

■ rH
Pi

o
o

VD
0)

r-t
EH

(V

s
p
o

>

Q
o

X

X

X

OJ

EH

1

U

-rt

ft

0)

0)

t

P.

(V

•H

in

t-)

Pj

o

^

■IJ

CO

CO

<\)

C\J

'\o

CM

.-1 r— >^
r-t c\j l^^

*

r~-

J- rH r-l

J-
•

o

LPi

*

OJ

J-
*

LT^

^
^

OJ tv^

ir\

LO

Vi3 OJ
r^60

cr> fH

f<^

CVJ

J- CM r>-

CM &0
CM

CM CO CO C7> OJ CTNVI3 O KD CMrHCOOeOJ" COCOCMOCO CMQOLf^CM '.-O KO O <£><Si OJOCM
O i-l M3 00 CO O CO (^vr> ir\ i-\ LTiVD LO J- 0^Lr»O>~.0B0r— rOCMJ- CMCOr^r— CMLfNror^

OLPvOLPiQOLOiOOOOOOLrvOLC^O
rO CM iH rH J^ LTS J- K>jrHir>OljriiHCMmOr-t

oomooomoLPiO o
CMf-HJ K^LPlLc^cMr<^r^ cmj^

u^ o

OjK^Lrv(^corv^O>OUDOviDr<^coO

CM rH rHrHO iHrHCMiHCM

CM r^ CM
rH OJ

CM
CM

r^ Cri-::t O

rH CM

CM

CM
CM

r^U3 rH ITA rH
rH rH OJ

^ I-- cr> O rH CM r<-\^ lovo r— 60 g^ O rH CM ^<~\^ lr^vx) r-- co cri O cm rH cm ro^ lo>^ r~-

rH<HrHCMCMCMOJCMCMCMrMCMCMiv^r<^h.-^rr>jr^t^rr>)r/-^K-\ rr^ J- ^rHrHrHrHrHrHrH

160

x;

X

S

9y

0) '
>

0)

G

0)
EH

I

P^ 0)

>

S 0)
•H m

o

+3

CO

O >^ CO .
iH 60 LTN

* * ::t

S'

cu

^

O »

♦ <-*

J- J* rH r«- m o iTNvx) m^^ r>- in
r-i60ccioj(<>c\jr-Oi-iM cr\#

* 4

CM t-H
rH ♦

CTi CM

C\J

r-H

V£>

r<^ rH

»

OS

VX)

s^

r —

OJ

J- 1^

60

OJ

i-l

ro ro

C\J

iH

OJ

OJ

r^ OJ

OJ

CM >r<,

hO r~.

ITMTMn

rH rH

r^

OJ

00 OJ iH VD iH

60 J-

OJ

CTi
CM

OJ

o

\p

OJ

*

CTi r-t

m rH

o

ITN

CO UD cr.
r<^ OJ

150
OJ

\0 r-i

^^ o

o

in^oj
in

OJ

»

r<^

m

OJ

(^ rH

GO

O rH rH

cr.

CTi

OJ

OJ

Jit r-^rH

r<^

rH to OJ

i-i O r->

COVX)

f<^

f^

OJ OJ

i^in

>X5

J-

cri

rH ^ 60

rH^

r~- J* OJOLTirHOCTNOJrH LTN J" OJ OJ 60 OJ VX) OJ irNv£) OJ 'X) r^VD \XlVX>Cr»OJOJOJ\OOOJOJ

&C.J- OJ m t«^^ o cr>ojLfNrH Lr\r~-rH o oj \s\ (y\ zo ^^ <j:> r-— vd 60ooo606or~r>0'-Hr— r^ui
ir~\yo LO LPv LTi LP> m jd" VX) uD r— U3 ixMrMn lo t-^vD md vx) vX) ltnlti mvD t~-»x) in^x) r — r^vx) vx> md

r-tr-tr-ir^r-ir->r-i<-^i-tr-\r-ir-lrHr-it-ir-irHr-{i-^f-ir-ir-tr-tn^r-ii-*r-ir-tt-<r-lt-ir-^t-f

mLTiLPvLrvooooomo

o Lom. LPiLPiOLPio o LP^mo

OJrHj-OjJ-^OOjOOJ<Hr-H

OOLPiO LrvLTiirxoOLno

^ O VX) rH rH
r-\ OJ

I I^CjOI OJroOMDrH

rH rH rH OJ rH

t-l I — I — U) OJ bO

CW r-i r-\ •-* ,-{

rH I — I

?^

OJ h- OJ

r— J- 60 K>0

r-f rH OJ

rHOJt-r^J-ir\vX)r~-COOt-HOJt-oJ- rnvX) l — O r*J OJ r<->^ LOvX) r— rH OJr^J-mOrHOJ t<\jt
OJOJOJOJOJOJOJl\Jr^K>,h<^r^r^r^r^ i^-:J- ^J-^3-:fJ-J- LfMrMTvLTi mvx> vu VX> VD VD /

; — r — r— I — t — r— I — i — r— t^— r--f — r^-r^r — r — i~~ r— » r^-i — r-— r^-r^-r^-i — r— r~— i — r-— r — r— f^r — r>- /

^-i^-^'-^r-^l-^^-^l-ir-it-^r-\r-^r-^r-^r~^r-^r^^r-il-^r-^r-^r-^r-^r^^r-^f-^i-^•-t•->l-ll-^r-^r-^r-^

161

i-<

E
o

>

o

C! CO

•H

o

o

rH
Eh

Eh

I

0)

B

o

•H
•4-5

CD

4J

CM ai

LO 60 r—

rH

♦

#

oj CM f^

r-- * rH

CM OJ

lOCTl rH 0>
I — hO I —

»

H

00 r<^

OJ rH

* *

^5

LPi r^ ir\
OJ J-
OJ

CM

r--\

to

CM to CM ir\

r-+*

ir\j-

CTN

cr>

CM

»

OJ u^

o

ir>

5S^

rH f^

»

CM c^^

mi — ) —

C\J-d- CM

*

ir\ CM U3

rH CTi Cr\ CO "^ ^ CTn 00

# J- rH rH

CT\ OJ CTi CTv r^<Ci
rH r^ r^ m

I ^ ITi rH ir\

f^

o

00

00

CM

r^m

CM

»

00

CM

00

CM

*

in

g^rH

rH ^

r-\ ^ r^

1^ r-i

^n

l^<0

CO

fo

CT^

ITv

CM i-f r-UD

LTN f^ rH CTx rH

CM

ir\

00

f^

00rr>,OC3^rHLnjd-J-^l-^0000rHOr--Lr\OI — OVDOJ-CUm^rHOCr\Vj3O00vr)J-00O

O 00 cj^rH I — vr> r— oovx> cmj- oj- r~-oo r^oo'J- r~<^ rjcMLTi-d- oj>jDvDJ-vr>OJ-oocor— oo
j' hA rA^" -d- ^ J- J-' Lr» in^' J-' J- J- J-' LP* rO'-JD LTi ITN LPi J" ^ ir> ir\\.o KD \s> \rs ir\^ j- md >Xi U3

r-^r-ir-ir^f--tr-tr-^r-ir-\r^i-Ar-tr-^r-ir-\r-{r-tr-(r-ir-ii--lr-{r-<r-tr-ii--{r^i--ir-tt~t(-*r-\r-tr-l

o mLrvO loo o O o o LrMr>irMoo lpvlpilpiq o Lomo mLnLr»o ijnLr>Lr\0 ltno lpilpi

t>^rH LOJ- mo LOimLTvLTiO t^OJr^r^^ a r^-? OO-^i-^mCM "■

OJrHOJC\J-=J-^mrHj-CM

ir\ rH OJ 00 CM ru

rH CM r-t r-\

rH r^ J- Cr» I — OJ rH V.D rH rH

OJ

OJ

O ro 00 CM 00 O >X> CTvVD CM 00 J" O
OJ rHr-HO OJ r-i r-l r-*

r— O 1^ C7^
O CM rH

•-I OJ (^ J- Lr\vX) t-^ O rH C\J h<~»J- ITN'^O I — 00 O rH OJ f^ J" LOUD ^- O rH OJ r^ J" lOiVD r-- rH CM f^
r^rHr^rHr^r^rHCMCM0JCM0J0J0JCM0Jr<-^^<^^O^'^r^r-r^K^ r^ J- ^J-J-J-^J^J LTNLOLPv

cnCT^cJ^cr^CT^cnc^^cr^CT^o^cr^cT^C^^cT^O"^C^^c^^c^^cnc^^CJ^C^^a^c^^c^^CT^c^^c^lCr^(T^c^^c>^a^o^a^

rHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrH

162

+3

c

0)

s

o

rH
0)
>

«
O

^^

•H

d
o
o

(U
i-H

EH

X!

X

>.

>vD

>

>

P.

B

0)
£-•

I

0)

p<

S
•f^

Eh

a

o

•H
+J

CO
+J

w

CO CM OJ CM

Ln

rH >.D .H
LO J- J-

CM

♦

CM O

r^

CM

«

cr. cr\ ro

rH r-l CM

*

»

*

bO

a> CM

r^

^ vrsin,

CM

CM

J-
o
t-l

CM r^ a>

J-

J- J-

cr»

uO

OJ

CM CM rO'

I cr>

!50

Lf>

^

J

60

CM

O rH

O

CM

»

CM r^

C7>

o

CO

r^ CM
O

CM

lf^

O CM l-OMD J-UDO CMtOrOLTNCMCMCMCOOvnOLnr^CM L0>,0 I — Lr> M cr> o o J^ 5^*^

BO r^co^ri cMvjD covo LTv r^^ r— I — CM to r-i r^vo t-~i — cjnO o LOLf>L£^cM f^'^cr>cr>'^

rH rH rH rH iH rH r^ rH .-H 1^ iH .rH iH H "H "-H rH r-H tH 4H 7-1 rH rH rH r-l rH rH r-H .-H rH i-l i-H

CUnH^ CMCMt'-^Or-lr^Lr^ rO_-t J r^OL^^^^^O.-t^^OOJr^O-J•^^L^^Li^fMf^<-''^'-'

TO CTi LTN O LPv O CVJ
i-l (H rH CM

r— CM

F^ t^ Lf^ O
rH CM

CM
H

iH H CM

r— J- r— J-

O CT^ U^ O LC^ CTi h^ I

CM r-^ rH O r-i ,

rH CM "-D

^ 11^ O rH CM r^J- rH CM M-, J- LT^UD r- O rH CM r^^ LTv-.O f^ ^O O rH OJ r^ J- LOv.0 t^ O
u-> LOv£> ^M^vD^r^r^r^rHr^r^^HCMaJCMrvJ'MC^JCMtMC^JK>,^<^^^K^^^^2^^;~^ f;^-^

(J> CJ> CJ^ Cr^ C^^ CJ^ CJ^ rH rH rH rH rH rH r^ rH rH rH rH rH rH rH rH rH 1^ rH f^^
rHrHrHrHrHrHrHCMCMCMCMCMCyCMCM-vJCMCMCMCMCMCMCMCMCMrvJCMCMCMCMCMCM

163

X

X

X

+5

c
w

a

p.

c

f-1

01

>

0)
!=1

o

0)
W)
CD
+^

EH

I

a 01

o

•rt

CD
■P

4

CM

»

r^

*

CM

VX>

CM

ir>

VD r^ ao,

f^

O CM

60

GO

r<^

1^

m

v^ (r>

IC\

CM r^

o

CM

<T\ r—

ro iH

ON

Kb

t— o

O r^ O J- ITNUS >X>rOCM«3Ov£>J-6060Vi),'^LriC\JOLnOJ- h— "^ -ij- i^VX) O CO CM O O

CT>o^r^o>^ LPiCM boj- o^t — o ru Lr\o f^i~^i — o uMvC co o lo i — vx) lti r<->i\^ ,^ o i-i i —

vr> ir\ LO LC"\<^o I — u:) ltivjd cm >.o r^vo vo lo^ f^^ i^ vjd vri vx) lpi ir\ r<^ j- ^ ud vjd ltwo J- ltn

rH(-HrHrHHfHHrHlHrHrHrHrHiHHHlHrHiHrHrHr-IHl-HrHrHiHrHrHrHrHr-(l-l

LTiO
H O

LTNO O O O O iTNQ O LTNO O Lr\ iCMr\ ir\ LCwjT^ ci LTxO LPvO L^^Lf^LC^lr^o l^^LC^lr^Lr^o
lr^cM^^lf^^JrurH.?r^J■o^<^L^^^r^^r^,r^^^r<^cMOr-^CMJ■-:}•0!^lc\Jlr^Lr^r^cMCM

CM

60 j-

r^ O ^ H tr>,jt Cr>^ O CTi^ B0CM^^O^^^^CMr^lr^CM^B0^Or— CMHU3COi-l*^
CMfMiHrHCMr-t OrH rHcHCM i-iCM r-^ r-\ r-i r-i r-i CMrH

(H CM M~\ J" ir>VD t^ iH CM r^ J* iTv O r-l CM t^,Tf r-" CM r^-d" LTi^-O r»- iH CM l^^ ITNVi) ^- 60 O rH CM 1^
^-=^-^ J-^^^LTMrMrMTN m<vO l^v^<.,0'vOl-^^^r^r-I^H,-^^-^CMCMC^JCMCMCMCMCMr^^O^O^^
cr! d 7^ ■"• ■"• H '-' I"" '-' H ""• f-' •-* "-I r-i iH iH r'^r<^r<^r'\r-^rv^r^r^tArOr^r-OK^rOi~^r^r<^r^i^
CMCVJC\JCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMroCMC\J.MCMCMCMCMC^JC\jCMCMCMCMCMCMCMCM

164

o
o

EH

e

O)

>

o

CO

M

X

!x!

Eh

I
0)

M tJ
ft 0)

a> u
B 0)

•H w

Eh

a
o

•H

cd
■p

CO

UN

CM

*

ro 1^

OJ

trv

o

OJ

♦

t^

fVi

60 C\J

CM

CM

»

cu

in ,

CM

ir\

CM

O r~- CO 60 >^0 V£) J- K> J- O -=J- <r>U3 V£>6060i-ICMO<MEO-d-r^ir\«og
J- («_ 1-^ If^ ro, rn,3- O J- CVJ rH i-H m r— VX) rO J" V^VDCTMHCMOf-l^^

LTv t<A J-* vo u> r^VsD r— LTv utn J- l^Mr^ in ir\KD udj-J-vd\^ i — -^ miTMr*

rHrHi-lrHrHrHf-liHHtHHrHrHr-irHrHiHi-lrHrHrHrHHrHHH

oir\iOLrNOLnLr>oiJt^inLniirvoinoooooooL0OinoLr»
m in J* cM-d-J- inr<^minc\j rH Oi-to^ o f^f^f^f^° 7^-^ "If^^

rHCM rH CVJrHH CVJrHCM rHiHr-l CUrHHiH

jt invD r»- o r-i CM M-\^ invT) t»- h cm r^^ in o i-i c\j t«-\jt -=»; ^f^ ^ |£^
r^ rojTi^ t^S ^ ^ -d-\d- ^ J* .=>■ usin in in ir\"-o v>d »;£^v£ h h cm k\

CM O? C\? CM^ CM CM CM C\J CM W CM CM CM W CM CV fU C^ CM fV) CM f^ CM C\J CM CM

165

0)

O

CP.

I

si
o

u
ai
S

o

to

CM

ci,

CM

O
(4

cd

o

J-

4^

o

J-

CTN

ro

O

CTN

O

=f

cr.

S

CM

rH

>-3

CM

1

1

o

r~

iH

r^

CM

•H

p.

^

^

■5?

a

S

o

^^

I

r-l
CM

cd

(M

I

00

C7N

CM

rH

I

CM

J-

O

VD
rH

EH

CO

•H

(A

O
•H

o

(1)

o
o

c;
o

-•rl
CD
C«

-d-

J-

-*

J-

in

o

O

o

CM

CM

"Xi

<.o

^£)

^X)

r^

J-

CM

J-
I

1

•

1 —

rH
t

cr.

rH
1

CM
1

CM
1

CM

1

rH

1

r-H

1

rH

1

1

1

r-\

1

1

o

O

O

O

o

r-\

r-\

rH

rH

H

CM

N^

^

m

U5

r^

CT\

rH

r^

ir>

iH

rH

OJ

CM

CM

o

CO
«

0)

■d

0)

r-\ OJ

r-^

LTv '03

cr.

CNJ

1^

CM

CM

166

3nLI0GR>J'Hy

3udd, P. L.

1917. Development of the eggs and earl^'' larvae of six California
fishes, California Division of Fish and Game. Fish
Bulletin No. 56, JO pp.

Clark, F. N.

193h- Maturity of the California sardine (Sardina caerulea),
detennined by ova diameter measurements. California
Division of Fish and Game. Fish Bulletin No. U2, U9 pp.

Crozier, IT. J.

1926. On curves of grov.iih, especially in relation to temperature.
Journal of General Physiology, Vol. 10, pp. 53-73.

Cunningham, J. T.

1889. Studies on the reproduction and development of teleostean
fishes occixrring in the neighborhood of' Plymouth. Jour-
nal of the Llarine Biological Association. N. S. Vol. I,
pp. 10-5U.

1892. The reproduction and gra\rth of the pilchard. Journal of
the Llarine Biological Association. N. S. Vol. 11^
pp. 151-157.

Fish, C.J.

1928. Production and distribution of cod eggs in Massachusetts

Bay in 192U and 1925- Bulletin o^ the United States Bureau
of Fisheries. Vol. XLIII, Part II, pp. 253-296.

Scofield, E. C.

I93U. Early life history of the California sardine (Sardina
caerulea), \Tith special reference to distribution of
eggs and larvae. California Division of Fish and Game.
Fish Bulletin No. Ul, UQ pp.

Scofield, E. C. and Lindner, U. J.

1930. Preliminary report of the early life history of the California
sardine. California Division of Fish and Game, Vol. 16,
pp. 120-12U.
Silliman, R. P.

I9U3. Thermal and diurnal changes in the vertical distribution of

eggs and larvae of the pilchard (Sardinops caerulea). Journal
of Marine Research. In press.

Tibby, R. B.

1937. The relation between surface water temperature and the distri-
bution of spawn of the California sardine, (Sardinops caerulea).

California Fish and Gaiae, Vol. 23, pp-. 132-137.
Wilson, H. V.

1891. The embryology of the Sea bass (Serranus atrarius ) . Bulletin of
the United States Fish Commission (for 1889) Vol. IX, pp. 209-2 77.
Worley, L. G.

1933. Development of the egg of the mackerel at different constant ten-
perat\ires. Journal of General Physiology, Vol.l6,pp.8Ul-857.

2566b 167

1

5o A Method of Computing Mortalities and Replacements

By
Ralph P» Silliman l/

CONTENTS

Preface Page

Introduot iono ooooooo»o»»oooooo«ooo X69

Basic assumptions oooooooosoooooooooo 169

Known quantities and relations hi pso oooooooooo 170

Determining total mortality oooooooooooooo 171

Separating fishing mortality from natural mortality o o 172

Determining rate of recruitment eoooooaooaoo 175

Effect of errors in original data «o>ooo<>oaoo 177

Summaryo »o«oooooo«oooeeooooooooo 1/9

Literature cited aooeo«oooooooooaaooo 179

l/ Now Chief, Section of Anadromous Fisheries, Branch of Fishery Biology,
"" Fish and Wildlife Service, Washington^ D. Cu

168

INTRODUCTION

The place of estiinates of mortality and replacement in a management
program, and their position in the investigation of the pilchard resource
of the Pacific coast, are pointed out by Sette (19Ji.3). Derivation of
such statistics for fish populations has been investigated by JBaranov
(1918), Thompson and Bell (193[i.), and Ricker (19i40), whose theories under-
lie the computations described herein. The last named student has de-
veloped concepts not included in the v;ork of the earlier authors and
gives mathematical formulations in a form more readily adaptable to the
data available for the pilchard population than do the others. This
report describes the application of Ricker 's formulations to the pilchard
data, and also develops nev; equations v/here necessary. To avoid adding
to the symbols in the already confused set employed in fisheries statis-
tics, those used by Ricker are adapted for this report.

This analysis was conceived and carried out under the inspiration
and guidance of Oscar Elton Sette, in charge. South Pacific Investigations
of the United States Fish and V\,''ildlife Service. For critical reading
of the manuscript and suggestions for improvement, the author is grateful
to Drs. Vit. j5. Ricker, M. 3. Schaefer, J. L. Hart, and Frances N, Clark.

BASIC ASSUI.1PTI0NS

In treating fisheries statistics by the method used herein it is
necessary to make certain simplifying assumptions. These have been set
forth by Ricker (p. UU) and may be condensed as follows:

1, The amount of effort expended toward catching fish 'is distrib-
uted uniformly over the geographical range of the species during
the fishing season.

2. There is no competition between iinits of fishing gear for fish
to be caught during one and the same instant.

Of the two types of fisheries postulated by Ricker, his "Type II"
most nearly conforms to that of the pilchard fishery. It differs from
his "Type I" in that natural mortality and recruitment take place during
the fishing season. The listed assumptions are given by Ricker as ap-
plying to his "Type I" fishery, but are later implied to be inherent in
"Type II" also (Ricker, I9U0, p. 59).

They do not hold strictly true for the pilchard fishery, especially
with regard to the uniform distribution of fishing effort. This does not
preclude the use of the involved formulations, however, if it is kept in
mind that they apply only to the extent to which the basic assumptions
hold true.

169

WOmi QUANTITIES hlW RELATIONSHIPS

Use of the series of formulations given in this report is based
on the assuKiption that the fisheiy has been stalbilized at two different
levels of exploitation during its development, Although the pilchard
fishery has never reached complete stability, there were two periods dur-
ing its history 'Within which there was little shift in the total catch,
and apparently little trend in the total fishiiig effort expended. The
first period occurred when the expansion of the industry was temporarily
halted by the economic depression of the early thirties and the second
included the recent years of the fishery, durjjig which the total catch,
after having reached ;j maximiiin in the 1936-37 season, has fluctuated
around an annual take of about half a million tons per year.

Uniform recruitment, as v.-ell as uniform exploitation, are requisites
of a stabilized fishery. During the early years of the pilchard fishery,
most of the recruitment was furnished 'by unusually successful age groups
or broods, which occurred at intervals of about four years. Four "super-
abundant groups" have been recognized by Clark (1936) during the 1^ sea-
sons, 1919-20 to 1932-33. To average out those fluctuations in recruit-
ment, the first period used in the study was chosen to include the passage
through the fishery of two of the successful broods (designated groups
"C" and "D" by Clark, 1936) and included the fishing seasons 192^^-26
through 1932-33} thus representing a time somewhat prior to the depres-
sion. During the more recent period (1937-38 through 19Ul-l|2) the dom-
inance of unusually successful groups has largely disappeared, (Clark,
1939 p. 21) indicating more imiform, recruitment.

For each of the two levels of exploitation under consideration, it
is necessary to have information on: (1) Age composition of the stock a-
vailable to the commercial fishery, (2) total catch per season, and (3)
relative amounts af effort (product of gear' and tine) expended per season.
For the more recent of the two periods, age composition was determined
from scales. During the earlier period scale readings were not taken'
and it was necessary to deduce age composition from length composition
by a method involvirig growth rate and other characteristics of year classes.
The total catch records were obtained from the weigh-in reports of the
industry and converted to numbers by means of length distributions in
samples of the catch in conjunction v/ith previously determined length-
weight relationships. Relative amounts of effort were calculated ^y,
dividing the total catch by the computed catchrper-unit-of -effort.-'

Numerical values obtained for the three items of information are
given in the accompanying Table 1. The data on age composition are from
samples taken in the commorcial fishery throughout each fishing season,
exclusive of the summer fishery carried on in the Pacific Northwest.

-' This computation is in process of revision (19U3).

170

Samples for the eight Si.,'asons of the :.arly period v/ore corabint^d to dis-
count the effect of fluctuating recruitment. P'or the latter period, v/hen
recruitment was more uniform, ago readings for the last season (19U1-U2)
only were used._/

Table 1. — Numerical values detennined for age compos ition, relative
fishery effort, and catch.

Period

• Hundreds of fish in samples of age (to last
birthday):

iRela-
tive
fffort

I

1 Catch,

'billions

l': 2 ': 3 i U

. 5

' 6

7

8

J

9 :10

11:12

13
13

!of fish

1925-33

U7.'l06:'558:*338

202

120

72'

U3

25 :13

loi 5 .

3

1

1.66

193 7 -U2

•58:527:310: 56

■ • • 4

lU-

2

1

• h

• U.98

The numorical values given in Table 1 represent the results of a
canvass of the data available at present. Many shortcuts and approxima-
tions have entered into their derivation, but they have been used in
the computations to obtain a preliminary estimate of the changes which
have taken place in the pilchard stock, and to shovi the way in vifhich
more precise data may be used when available. The analyses necessary
to the derivation of such precise data are nov^r being carried on by the
United States Fish and VlCildlife Service and the California State Fish-
eries Laboratory.

DETERI.IINING TOTAL MORTALITY

The rate of disappearance, or total mortality, of the year classes
may be determined from the relative age composition of the stock. As
pointed out by Baranov (1913), in a stabilized population, the numbers
of fish of each age, starting with the youngest commercial -size, are the
same as the numbers of a given year-class surviving each year since its
entry into the fishery. Therefore, the effective rate of disappearance
may be determined from the age composition of the stock at any time* after
stabilization at that rate. The simplest method is to plot the logarithms
of the numbers of fish of each age, so that a constant percentage decrease

w

A recent study of the method of taking scale samples for use in the ,
age analysis indicated that there v/as a tendency to select the larger
fish of each sample (scales were taken from only 10 fish out of each
50 measured) for the removal of scales. This may have caused an in-
dicated mortality rate lower than the actual. Computations to correct
for this selection are now being madet

171

will be represented by a straight line. Figure 1 shows in this fashion
the numerical values used in this paper. It will be seen at once that
in both periods the numbers of fish over age two decline at a constant
rate^/, while those of ages one and two years do not. It was assumed
that the one- and two-year-olds are not fully available to the fishery,
and the line was fitted to the other ages by inspection. The lines shovm
indicate an annual decline in logarithm of .22!^ for the first period and
.700 for the second. These may be translated into mortality rates (a)
by the following formulations ;

Let 3]_ and So be the relative stocks of any tvro successive ages.

Then S2 = S-l - aS]_ = S]_(l-a)

and 1 - a = — z
Si

taking logs log (1-a) = log $2 - log 3^

or log (1-a) = -(log 32-log 3]_) = -(annual decline in log)

Substituting numerical values given above, and letting subscripts refer
to the two periods, we have:

log (1-a-, ) = -.225 1-a-, = .60 a-[_ = I-.60 = .UO
log (l-a2) = -.700 l-a2 » .20 32 = 1-.20 = .80

In other words, the total mortality, or rate of disappearance, as indicated
by the age compositions of the stocks, was I4.O percent for the first, and
80 percent for the second, period.

SEPARATING FIS'JING MORTALITY FROi! NATURAL MORTALITY

The total mortalities determined above may be broken down into fishing
mortality and natural mortality if the latter is assumed to be the same
for the two periods. In the absence of any evidence of a change in the
rate of natural mortality, this assumption was made, and the method noted
below was adopted.

This depends upon the fact that, v.-ith the same natural mortality (n)
for two periods, a given pair of total mortalities (a-,, a ) combined with
a given ratio between the two fishing mortalities (m-, , 0.2; is associated
with a certain ratio betvreen the amounts of effort (I-^, I2) expended per
year during the two periods. The procedure is to determine what ratio

(R * rlf ) between the two fishing mortalities gives the known ratio between

__

-The regular decrease in numbers of the fish in each age class for the

earlier period is a result of the method of age analysis employed,

rather than a natural phenomenon.

178

- tt

10

u a

o <o

o

'O ■ — ^ ^ cd

^

•H D.

a n

o ^
o

O Q) -H

•H CO

•a -p

O tH •

£ -H O-
■P Ql

*< ^ ^'

§ , "J

o > Ju

0)

lO

CM

x^

I -

d

90

o

9

o

o

o

o

tr

ce

lU

UJ

Q.

a.

1-

z

Csl

J.

«0

^

J.

OJ

CO (O

^

X

0>

<
o

K

m

<
o

^^

w
tr.

<

UJ
•0<w

UJ

n

CM

CD

4>

(U

o

•rH

o

B

<u

13

CO
o

o
n
o

a

o

5

m

o
o
+>

03

+»

o

m
Pi
O

o

>

a)

s

•H

M

Q ® «? ^^ ^ O CO
loeiftjojojai-- — — — —

HSU JO StlBSNON 3AI1V-I3d dO SNHimVdOl

173

fp

the two amounts of effort (^y^ «> U)* Natural mortality (n) is pu.t in terms

of R, a-[_ and ap as follows : 1

a, = m-, + n - m-,n

(Ricker, p. 60, 1st paragraph)
a2 = mp + n - mpn

But, by definition: R = !iir and mp ?= Pto,

Substituting in the second equation above v/e have: ^2 ~ ^i + n - Rra,n
Multiplying the first equation by R: Ra-j^ => Rm]_ + Rn - Bm-^n
Subtracting from the last equation the one above it;

Ra-, - ap = Rn - n = n(R - 1)

Rai - a? , >
and n = ^^_ ^ < (I)

fp
Now -^ must be gotten in terms of R, n, a]_ and a2 as follows :

Starting again with: a]_ = ra-|_ + n - m^^n

and transposing: a-j^ - n = m^^ - inj_n =_ra-,(l - n)

a - n

Dividing by (1 - n) : m, =^-i (II)

-■- 1 - n

Also, by definition: mp = Rm-j^ (III)

f„ log (1 - mp) , ^

Finally: T^ ■ log (1 - m^j (I^)

This last formula was developed by Ricker for fisheries of Type I,
However, as shown by Schaefer (19l;3), it may be used equally well for fish-
eries of Type II.

Starting with a value of R of 2.U and substituting our previously
determined values of a^^ (.UO) and ap (.80), 2 j^ay be computed from the
above equations as follows ; 1

rr^ 2.U(.U0) - .80 _ ,
(I) n = 2.1; - 1 '^^^^^

(II) ra^ :, 1 _ .111^3 •^'^^^
(III) ra2 = 2.1;(.3226) = .77U2

rr.r^ £2 log (1 - ■77U2)

(^^) f •- log (1 - .3226) ' ^'^^

174

Table 2. — Substituting successive values of R at intervals of .1, we have

the foiioiving:

R

n

\

m^

^2

2.li

.11U3

.3226

.77U2

3.82

2.5

.1333

.3077

.7692

3.99

2.6

.1500

.29U1

.76U7

I4.15

Obviously, the value of R most closely approaching the true value is
2.5, since the value of |a of 3.99 is closest to the empirical value of
U.O. Considering the variability of our data, any interpolation of values
of R would be superfluous. Therefore, referring to the center line of
the table, we now have natural mortality and fishing mortality for the
two periods.

DET'jIUvilNINa RITE OF RECRUITIENT

To obtain the number of recruits entering the fishery each year,
we must first obtain the size of the available stocks for the tvro periods.
These may be computed from the total catch statistics, if it is known
what fraction the catch is of the available stock in each period. Ricker
calls this fraction, "rate of exploitation" (/u). Its value may be de-
termined from our previously computed values of total mortality (a), fish-
ing mortality (m), and natural mortality (n) as follows:

m
AX = ffl + n (m + n - mn) (Ricker, p.60, par. 2)

But also: a = m + n - ran (Ricker,.. p. 60, par. 1)

am
Substituting: /u = ra + n (v)

Using the numerical values of a, ra, and n given above, we have:

_.kO(.3077)_

(V) /u^ = rjO?7 + .1333 ^ .2791

. 80 r. 7692)
(V)/U2 = .7692 + .1333 = .6819

Since yu represents the fraction of the stock caught by the .fishery,
the available stock (3j^) may be computed frora the catch (c) and rate of
exploitation (/u) by the siraple relationships:

c = /u3^ or S^ =^ (VI)

175

Substituting the values of /u and c previously determined:

1.56
(VI) *S^j_ = .279 = 5.9? billions

U.98
(VI) S.5 = ■."SW » 7.30 billions

These stocks may now be broken dovm into their component ages in
accordance viith the previously determined relative age compositions.

Table 3. — Numbers of fish of each age obtained

1 Billions of Fish of Age (to last birthday):

Period

1

> 2

• 3

• U

•?

6

i 1

8

• 9

• 10

: 11 .

' 12.

: 15"

1
2

-.18,

' .68
3.98

!2.0U
i2.3U

1.23
.h3.

'.7k
.10

I .02 '

'.26

! .16

' .09 .

.05

!.0U i

.02

.02

Since the fish of age one are not fully available to the fishery,
it is necessary, in order that the recruitment to age one may be deter-
mined, to work-backward from the first age class that is fully available,
age three In our example. Let s^ and s^ be the available stocks of any
two successive age classes, and 5]_ and S2 be the total stocks of these
two age classes, respectively. If the stock of the second age class is
considered fully available, Sg = S2. s-j^ is subject to the full mortality
rate a, but the balance of the age class (S]_ - s-[_) is subject only to
natural mortality n. Formulating these statements we have:

Sg = S2 = S]_ - as]_ - n(S-, - s-, )

Expanding: S2 =^ S-|_ - as-, - nS-, + ns.

Transposing; S2 + as-|_ - ns-j_ = S, - nS,

Factoring:.

So -♦■

s^(a - n) = S-i_(l - n)

so + S-. (a - n)
Dividing by (1 - n): S-,_ = -^ — T~-li

(VII)

Substituting the empirical values previously determined and letting
subscripts refer to ages, we find for the first period:

ions

r^TTT\ Q 2.0k -f .68('.liO - .13)

(VII) S2 = 1 - .ij = 2.55 billi

(VII) Si = 2,55 \^j;Q|£^Q - -^^^ = 2.99 billions

176

And similarly for the second period:

(VII) S_ = 2.3U ^ 3.98(.0O - .13) = 5.76 billions

(VII) Si - ^-76 ^^'hkUdO - .13) ^ 6.9? billions

In other words, the estimated annual recruitrtient to age one during
the second period was about two and a third times what it was in the first
period.

EFFECT OF ERRORS IN ORIGINAL DATA

As stated earlier, the numerical values used in the foregoing compu-
tations have been obtained by approximations, and may be considerably
in error. It is, therefore, desirable to knovj- what effect errors in the
various original data have on the final results.

The curves in figure 2 indicate the values of natural and fishing
mortality which v/ill be obtaijied virith values of effort ratio and total
mortality centering- about those used in the foregoing computations. It
v;ill be noted that the effect of a given deviation is always greatest
on natural mortality, and least on fishing mortality, in the second period.
This is favorable to the practical application of the computations, since
the rate of natural mortality, in itself, is of relatively minor importance
in settling questions relating to the coraraercial fishery. On the other
hand, rate of fishing mortality under current conditions (second period)
of exploitation is of the greatest interest, since it leads to the esti-
mation of the size of the stock of pilchards from which the commercial
fishery must take its catch.

Considering the relative seriousness of various types of errors,
it is apparent from figure 2 that errors .in computation of total mortality
are more serious than those in the effort ratio. For instance, the value
of the effort ratio used in our computations was U.O. PYom figure. 2
it may be seen that, using total mortalities of ,U0 the first period and
.80 the second period, a change in effort ratio from k-0 to 3.?, a decrease
of 12.5 percent, would cause an increase in the computed fishing mortality
for the second period of only 1.9 percent (from .77 to .785). However,
retaining the effort ratio of l|.0 and changing the total mortality for
the first period from .UO to .35, again a decrease of 12.5 percent^ would
result in an increase in fishing mortality for the second period of 2.6
percent (from .77 to .79). In other words, a given percentage error in
total mortality wbuld produce about l.U tiraeis the effect ou confuted fish-
ing mortality as the same percentage error in effort ratio (for the ex-
ample given).

In general it may be said that fairly large errors in effort ratio
would not seriously affect the results of the computations, provided that

177

3.6

3.7

3.8

3.9 4.0 41 4.2 43 44 4.5

E

■

40-85

o

U

B- TO

.3S-.80

N.76
i.72

§70

40-.30

.48-80

«0 ea

■

-.68

£6

-^•38

.40-.75

-.^..^

o
234

S32

^

_

^

__^

.40-.S9

l-*.28

a:

§.26
^04

' ■

.^^"^

.35-.80

,

, ^

.4o-.ao

.45-.80

X

2.22
b.

.20
.28

.40-.r6

4S-.80

.26

■" "

.22
.20

>-

t.16

-1

^.lA
a:

.40-76

^^

^

40-.80

"^

^

-^^

.40t85

^

---^

.35-.80

3. OB

1-

$.06
04
.02

'

Stv'iit

s on the

Pacific P

Llchard or

Sardine

^^

^. ^

ffipurdinops caerulea^
V - M method of Cooputing

^

"^^

ty Kalph P. Silliman

«

L5 3

.6 3

7 3

8 3

RATIO

9 4

OF EF

.0 4
FORTS (

f./f.)

2 4

3 4

4 A

Figure 2, Natural and fishing raortallt
talitlas. Total mortalitlea for the fi
hand end cf each curre.

lea for various combinations of effort ratio and total mor-
rst and second period respectively are /?:lven at the right-

178

the estimates of total mortality were accurate. On the other hand, a
relatively small error in oomputud total mortality might cause errors
great enough seriously to affect the final interpretation of the results.
Thus it will be desirable, in the more precise determinations to be made
in the future, to place special emphasis on accuracy in computing total
mortality.

sui^caRY

1. If it is assumed for a given population that fishing effort is
uniformly distributed, and that instantaneous coB^ctition between units
of gear is absent, it is possible to estimate natural mortality, fishing
mortality, and rate of recruitment.

2. The source data required are: (1) Age composition of the com-
mercial stock, (2) total catch per season, and (3) relative amount of
effort expended per season. Each of these must be known for at least
two periods during which the fishery was relatively stabilized.

3. Total mortality may be estimated directly from the age composi-
tion of the stock at any time after stabilization.

U. Natural mortality may be separated from fishing mortality by
assuming various ratios between the fishing mortalities for the tviro periods,
and making successive trials until a combination giving the knovm ratios
between the amounts of effort for the two periods is found.

5. Rate of recruitment may be estimated by extrapolating the natural
and fishing mortality rates from the first fully available age class

to the entering age class,

6. Errors in the source data \/ill affect the estimate of natural
mortality the most seriously, and the estimate of fishing mortality during
the second period the least seriously.

LITERATURE CITED

Baranov, F. I.

1918. On the question of the biological basis of fisheries. Nauchnyi
issledovatelskii iktiologisheskii institut, Izvostia I (1),
pp. 81-128, 12 figs,

Clark, Frances N.

1936. Interseasonal and intraseasonal changes in size of the California
sardine (Sardinops caerulea). California Div. of Fish and
Game, Bui. No. U7, 215 pp. U tables, 10 figs,

1939. Measures of abundance of the sardine (Sardinops caerulea)
in California voters. California' Div. Fish and Game, Bui.
No, 53, h$ pp., 19 figs., 7 tables.

179

i

Ricker, ¥. E.

19U0. Relation of "Catch per unit effort" to abundance and rate of
exploitation. Journ. Fish. Res. Bd. of Canada, Vol. 5, No. 1,
pp. I4.3-7O, h tables, 3 figs.

Schaefer, M. 3.

I9U3. The theoretical relationship betv/cen fishing effort and mor-
tality. Copcia, 19U3, No. 2.

Sette, 0. E.

I9U3. Studies on the pacific pilchard or sardine (Sardinops caerulea)
I - Structure of a research program to determine how fishing
affects the resource. Special Scientific Report No. 19.
U. S. Fish and Wildlife Service.

Thompson, Vf. F. and F. H. Bell

1931;. Biological statistics of the Pacifip halibut fishery. Int.
Fish. Comm., Report No. 8, h9 pp., l5 tables, 18 figs.

180

6o Thermal and Diurnal Changes in the Vertical Distribution of Eggs and Larvael/

By
Ralph Po Silliman

CONTFINTS
Preface Page

Intr OQUO tjXOn© oooooooooooo«oooooeooooooooo XOt«
MeTJilOdS oooeoooooooooooooooooooooeooooo

Periods and area coveredo oooooooooooooooooooooo
Clas s it i ca ti on' oi inaterxai.o oooooooooooooooo»oooo

Standardization of hauls o ooaoooeoaoooaoooaoeoco

hiiiect Ox 'Ceinpera uure oeoooooooooooooooooeoooo

Diurnal changes oooooooooo

Summary and conclusions aooo«o«oo>

Xji teraijUre citjecia ooooooooooooooeooooooooooo

Appendix A - Tables of detailed towing data

182
182

ooooooooooooeeooo

oooeoeooooooo

OOOOOOOOOODO

185
188

200
189

Appendix B - Lists showing time and location of hauls,

depths fishedf and numbers of eggs and larvae taken, o « o e o o

198

l/ This reports without appendix A, was originally published in Sears
Foundations Journal of Marine Researchj, Volo Vj, NOo. 2, under the title,
"Thermal and Diurnal IShangeB in the Vertical Distribution of Eggs and Larva©
of the Pilchard (Sardinops caerulea)." It is reproduced by permission of
the editors of that publication«

The original data used were collected as part of a joint program of
oceanographic research carried on by the United States Fish and Wildlife
Service and the Soripps Institute of Oceanographyo Mro 0<> Eo Sette, then
in charge of the South Pacific Investigation of the Fish and Wildlife Service
suggested the investigation of the particular problems discussed in this
report^ and gave much helpful advioe© Sorting of the plankton oolleotions
was accomplished with the assistance of Works Progress Administration
projects Noso 10917 and 65-2-08-286o

181

DJTRODUCTION

Knowledge of the vertical distribution of pilchard eggs and larvae
is of interest from two standpoints: first, in adding to the general knowl-
edge of the early life of this species in the seaj and second, in defin-
ing the vertical range of its early stages so that regular surveys of the
spawning grounds may be designed to sample all strata in v;hich eggs and
larvae are to be found. In reference to the second point it v;as desired,
if possible, to relate vertical distribution to physical characteristics
of the sea, so that the depth to which nets ;must be sent might be predicted.
It was felt that the concentration of organisms at a given depth would
be controlled chiefly by temperature, density, and light intensity. The
analyses described in this paper have been designed to discover and de-
fine the relationships between these three variables and the vertical dis-
tribution of pilchard eggs and larvae,

METHODS

The field technique employed throughout the experiment was to make at
each station successive net hauls, each sampling a layer of water below that
of the preceding haul. In 1939, hauls were approximately horizontal. In
19U1, however, oblique hauls were adopted to preclude the possibility af
missing concentr,ations of lg,rvae in the gaps between hauls. Each oblique
haul extended diagonally from the top of the layer to be sampled by the next
haul, to the bottom of the layer sampled by the preceding haul. The "sur-
face" haul was made with an open net, while for the others the net was low-
ered closed to the desired level, hauled for from 15 to 20 minutes, closed,
and brought to the surface.

Closing nets used vfere identical, except for dimensions and materials,
with the one described by Leavitt (1935), and the tripping device was the
one figured by him in a later paper (Leavitt, 1938, fig. 2). Sizes of nets
and materials used are given in the list of stations at the end of this paper.

Temperatures were taken in 1939 v\rith reversing thermometers, and in 19U1
with a bourdon-type bathythermograph modified from the one described by
Spilhaus (19U0).

Y7ater flov/ was measured in the 1939 hauls by a meter consisting of
an impeller and counter. '*3tray angle" (departure of the tovdng wire from
the vertical) was measured v;ith a pendulum protractor.

PERIOD AND AREA COVERED

Stations were located within a radius of 120 miles of San Diego, Cali-
fornia, and were in an area somewhat to the south and east of the areas of
heaviest pilchard spawning as indicated by other surveys, Four series of
hauls were made in April, 1939, and SSven in April, May, and June, 19l;l.
Exact locations, dates, and depths fished are given in the list at the end
of the paper.

182

CLaSSIFICATION OF MA.TERIAL

SgRS and larvae viere of course treated separately. Also, to take ac-
count of the fact that motility increases with size of larvae, these v.'ere
divided into tivo categories: "large" and "small". A dividing line of 8 imn,
had previously been used in other vrork, vfith '^vhich it was desired to compare
the results of this analysis. Since this line did not split a mode in the
length frequency distribution of larvae in the vertical serial hauls, it
was retained for the present study.

Some larvae of the "small" group v;ere lost through the meshos of the
one-meter nets used. This, however, does not necessarily invalidate com-
parisons between numbers caught at different levels with the same net, since
the loss should be a constant percentage of the total number caught.

ST.'U'JDAHDIZATION OF HAULS

Because of unavoidable variations in the speed bl the ship and time

of hauling, the amount of water strained by the net varied from haul to

haul, even within a single series. These variations were discounted by

putting all the hauls in each series on the basis of a standard volime of

water strained. For the 1939 hauls, flow of -vatar through the nets i:as

measured directly oy a current meter, but for the 19U1 hauls, the relative

, Wi tan/1 i Wi
amount vras computed fro.n the formula ■»»- = L . virhere --- = the ratxo oe-

"'2 tan/2 ^

tween the amounts of v;ator strained for any tv/o hauls,, and /i and /2 = their
respective stray angles. Actual volumes of ■vfater strained averaged around
UOO cubic meters for the one-meter nets, and 100 cubic meters for the half-
meter nets. The numbers of organisms in the hauls were multiplied by fac-
tors (Vs^ v/here Vg = standard volume of water and Vj^ = volume strained in

Vh
given haul) v;hich made the numbers the same as if the standard volume of
water had been strained, and had contained the saine-concentration of organ-
isms as the water actually strained.

Because of the great range i:i nuribers of organisms (from 6 to 6,000
in a single series of hauls) they could not be represented graphically
on the same scale. Since the change in concentration from one series to
another was not of interest, but only the changes froia haul to haul within
a series, the graphing difficulty v:as overcome by representing pach vertical
series by a polygon of equal area. To do this, the product of hunibers of
organisms by thickness of layer sampled was obtained for each haul. These
products were sujniaed f-or each station, and the original counts of organisms
v/are multiplied by factors which made the sum of such products a constant
for all stations. Besides overcoming the graphing difficulty, this compu-
tation placed the, numbers at each level for all of the stations on a com-
parable basis, permitting direct comparison of catches at different levels
even though they vfere not taken at the same station.

To avoid excessive randoa variability froa small numbers., series con-
taining less than 100 eggs, 50 small larvae, or 5 large larvae, were omitted
from the analysis.

183

In order to construct the polygons mentioned above, it v.-as nGcossary
to compute the average depth for each haul. I-'or tho 19.39 hauls, when the
stray angle ivas measured tv/o or throe times durllis a haul, the angles were
simply averaged and the depth computed from the average angle and the length
of the towing wire. In 19Ul, when angles yfera measured- each minute during
a haul, it v;as possible to construct a plot of the course of each haul,
(fig, 1). Average depths were computed from the plots by measuring the
area bounded by each one and its baseline, ;OTd dividing by the Length of
th3 baseline.

EFFECT OF TEIffERATUIU!

Inspection of the vertical profiles for relative numbers of eggs and
larvae, togetjier with the cori*esponding temperature profiles (fig. 2) in-
dicates a possible relationship betvrcen temperature and vertical distribu-
tion. Such a relationship could arise as the result of the seeking of an
optimvmi temperature by the larvae, and by the adult fish which lay the eggs.
Since pelagic fish eggs tend to remain in v/ater of the same specific gravity
as that in which they were fertilized (V/alford, 1939), the vertical distri-
butioft of eggs shoiild tend to reflect that of the parent fish at; th^ time
of spavrtiing. This relationship is complicated by vertical turbulence above
the thermocline, v;hich tends to disperse passive bodies like fish eggs.

T'^atever the nature of the relationship betv/een temperature and the
distribution of pilchard eggs, the larvae may be expected, to respond directly
to changes in temperature, since they are capable of locomotion., Again
the relationship- is probably complicated by other factors, such as the amount
of fodder organisms present.

To measure the correlation between temperature and concentrations of
eggs and larvae, the relative numbers v/ore plotted as regressions on degrees
centigrade (fig. 3)» Hauls above a depth of 10 meters were omitted because
of the reversals in egg and larvae profiles which often occurred in that
layer. Also excluded ;vere zero hauls below the first. Inclusion of either
of these two categories of hauls v.'ould obscure the decrease from the maximum
concentrations dovm to zero, v;hich is the point of chief interest in the
regressions in so far as thoy arc to be used in determining the depth of
net hauls for r^jgular surveys. Correlation coefficients v;-erc .590 for eggs,
.55? for small larvae, and .5U8 for large larvae, corresponding to proba-
bilities of .010, .OUl, and .02U of chance occurrence. Combining these
probabilities by the method of Fisher (1936, p. 105). indicates, that the
result, as a whole, is highly significent (P = ,0009).

Of special interest in the foregoing correlations are the intersections
of the computed regression lines v;ith their baselines, since those should
indicate the temperature belovf T;hich wc vrould not expect to find eggs or
larvae. The indicated lainimura temperatures are 10.5 for eggs, lii.l;° for
small larvae, and 12.0° for large larvae. Since thxa regressions for large
and small larvae v/ere found by Fisher's (1936 p. 1U6) "t" test not to differ
significantly in slope, they were combined, giving an indicated minimum of
12.0° for both size categories.

184

diurnal c^uifCiza

To determine the extent and nature of diurnal changes in the vertical
distribution of pilchard larvae, two special pairs of serial hauls were
run in June, 19i;l, the first of each pair being occupied around midnight,
and the other around noon. Due to unavoidable novements of the vessel the
two members of each pair v/ere not located at exactly the same point geograph-
ically, but approximately the same position v;as maintained and they were
probably in water of similar characteristics.

One of the chief differences betvreen the night and day hauls was that
the former contained' many more large larvae than the latter (fig. k)' This
could result from any one of three causes; (1) location of the day hauls
in an area of lesser concentration thaji the night hauls, as a result of move-
ment .of the ship through the ^vater between the two series; (2) migration
of most of the larvae below the levels fished, during the daytime; (3) dodg-
ing of the net by the larvae, in the daytime.

The first explanation is a highly improbable one, since the lai-ger
catches at night have been observed many tines previously, for other plank-
ton organisms and larvae of other species of fish. Also, our regular sur-
veys of the spavming grounds show the night catches of larvae to be con-
sistently larger than the day.

The second explanation has been most \.ldely entc-rtained by others in-
vestigating vertical migrations. For instance, it was advanced by E. S.
Russel .(1926) to explain the larger catches of clupeoid and gobioid larvae
at night, but he v;as later forced to abandon it when subsequent daylight
hauls near the bottom (Russel, 1930) did not average larger catches than
the ones in the upper layers of irater. For our ovm data, the taking of at
least tv;o blank hauls (fig. k) below those which contained any larvae, argues
against the probability of a concentration of larvae belov: the levels sampled.

Elimination of the first tir/o explanations loaves only the third -
dodging of the not. in the daytime. FJussel (1926) and Johansen (192.t) admit
this possibility, but doubt that fish larvae could swim fast enough for
successful dodging. On this point it may be well to consider hovrfast it
would be necessary for larvae to svam in order to get out of the way of
the net. All nets tov;ed in the conventional manner give warning of their
approach by means of the piece of line i.'hich attaches theiii to the towing
v/ire. In our gear this line v;as about $ meters long, v;hilo our average speed
of towing v;as around 50 cm. per second. The larvae v/ould thus have about
10 seconds to sv;im out of the iiray. Since the maximum distance they vrould
have to svfin v;ould be the radius of the net, or $0 cm., a sv;imming rate of
5 cm. per second would be sufficient if the direction of svdraraing vrere favor-
able. This does not seem, an incrcdibl-; rate for pilchard larvae of the size
in question (over 8 mm. in length). Likewise it does not seem unreasonable
that .they should sv;im away from tho center of disturbance, and therefore
in a direction favorable to escape from the oncoming not.

186

J

RELATIVE DISTANCE TOWEO „

ngun 1. Course of serial hauls for a t/plcal siatlcn (No. 204S). "Balatlv*
dlstaae* lu cooputea froo the product of tine and avars^ tangent of aagl* of
■tray for each one-alnute intarual of the hauls. Depths we coaputed froa ooalna
of angle of etray.

RELATIVE NUMBER OF ESSS

\

rifure e. Uelbtive nuni&ers or aggs tM Iwwe UKen at aacn depth, BroKw lln* ihoia tenpanttur*,
ror Uis Urge Urv&a t«o eerlas taktn uvuid noon h«v« bMn oalttaa bdoause ol Uta affaet of ll{9tt,
Thay ara afioan ui rigura 4.

186

suaiaw Ni Hidao

o m o in
cj — —

in

in

m o K> o

M - - cvi eJ - -

SNSINV9U0 JO suaennN baiivou

187

i

In the day hauls the large larvai^ vfcre not only less numerous than
in the night hauls, but also appuared to bo distributed in soraev/hat deeper
levels . A greatc-r relative number of larvae at the deeper levels in the
daytime than at night might r^jsult from more successful dodging of the net
in the t^etter illuminated upper layers of i;ater. A greater absolute number,
however, could result only from dovmward migrauion in the daytime, providing
the overall concentration •v/ere not greater. Such a dov/n;;ard migration
might virell occur as a result of negative phototropism. Comparison of the
night and day hauls for stations 2Up2 and ZW^h (fig. h) reveals that there
actually were more larvae at the deeper levels in the day hauls. Four lar-
vae were taken in the day series at S? meters, and one at 73 meters, Virhilc
none virere taken at either of these levels in the night series. The signi-
ficance of these differences vias tested by means of Bayes' theorem, (using
the formulation of Pearson, 1930, page Ixx) considering a success, the pres-
ence of a larva at the indicated depth, ahd assuming the sfuae overall con-
centration for day and night (in other v/ords assuraing that the lesser num-
bers in the upper layers in the daytime were due to dodging of the net).
The computation gave values of P of .06 at S? meters and .$0 at 73 meters.
The latter, of course, does not indicate a significant difference betiTeen
day and night hauls, but the former, closely approaching the conventional
significance level of .05, gives some indication of an actually deeper dis-
tribution in the daytime.

In applying this result to the prediction of vertical distribution,
a knov.'lodge of the penetration of light at various tines of day is neces-
sary. Unfortunately, no empirical data for the area of our survey are
available, and we are forced to argue by analogy with data collected else-
v;here. G. L. Clarke (193U) has sho^m in his graphs that submarine irradi-
ation in the Atlantic Ocean reaches nearly its maximum value shortly after
sunrise, and is maintained there until shortly before sunset-. Accordingly,
extension of the regular hauls below 73 meters between sunrise and sunset
would seem to bo necessary in order to make svir^ of sampling all strata in
which larvae were' to be found.

SmMiklU Aid CONCLUSIONS

Analysis of catches of pilchard eggs nnd larvae taken in serial- hauls
in 1939 and I9UI indicates the following relationships and effects:

1. A positive correlation of c one enti-at ions of eggs and larvae
With temperature, at least \.-ithin the range of 10° to 17° C.

2. An apparent dodging of the nets by the larger larvae, in the
layers of ii/ater ^vhich are illuminated in the daytime.

3. A negative phototropism for the larger larvae^

Such relationships with physical conditions as have been indicated
by the present study may, of course, hold true only for the particular set
of occanographic conditions prevailing at the times and places where the
data were gathered. For instance, the relationship between the horizont-al
distribution of pilchard eggs and temperature as indicated by our regular
spavming surveys for 19U1 was quite different from that found in 19U0.

N 188

Hov.'evor, the data for tho present study were gathered in t^.'o different years
(1939 and 19Ul) and over a considerable ar.ca, so that they may permit of
sone generalization.

It may have been noted that the concentrations of cgs^s and larvae are
correlated \.dth depth as well as temperature, since the latter two are them-
selves highly correlated. However, there are some instances where the deeper
extension of v;-arm water corresponds v/ith a deeper distribution of eggs or
larvae (cf. fig. 2). Also, there is some logical basis for a correlation
with temperature, v^rhich might act either as the cause of direct reaction
of the larvae, or as an indicator of density differences affecting the dis-
tribution of eggs.

It is not intended to argue that tenporatuTG itself necessarily controls
the vertical distribution of pilchard eggs and larvae. Their location may
be determined by complex reactions of the spawners to a combination of physi-
cal conditions, including salinity, food, density, etc. To the extent that
these would induce spasming in a homologous portion of the layer of water
above the thermocline, temperature m;iy be merely an indicator and not a cause.
This, of course, doiis not detract from the usefulness of temperature as a
guide to vertical distribution.

As a basis for predicting the proper depths to v.-hich to send nets on
regular surveys, the regressions of egg and larvae concentrations on tem-
perature indicate that hauls should be made deep enough to go below temper-
atures of 10.5° C. for eggs -md 12.0° for larvae. Distribution of the large •
larvae during the daytime show^s the need for fishing to a depth greater
than 73 meters between sunrise and sunset.

APPENDIX A

TABLI5S OF DETAILED TOV/ING DuIa

Explanation of tables : The hauls at each station and the observations
during each haul are arranged chronologically and numbered consecutively,
beginning with one. Time is ship's time, ■v'ith hours numbered from 1 for
Ija.m. to 2U for midnight. Meters of wire out (m. out) refers to the length
of towing viire from the surface of the sea to the point of attachment of
the not. The stray angle {°/) is the departure of the tow'ing wife from
the vertical, expressed in degrees. A dash (-) represents a missing ob-
servation.

189

(4

(1)

to

.f:'.

I ^ Jt Jt

•4J

O Q O O
rH iH rH rH

E

•H
EH

CM OJ t^ M^
0^ CTi CTi o>

^

1^

O^

00 O CO
1 -:ru3^ 1

o

+3

O O O O O

cr> cr> CTN o^ CTN

d

CD
r-H

e

>x> r^ rH r— c\j

rH rH OJ CVJ rO

• • 9» •• •• t*

«>D

o^

O OJ t^

1 J--d- J- 1

u

c

43

O O O o O

0/

6

•H
EH

o c\J cr> oj irS

•• •« •• •%' «■
*«D vo >>D vo ^J^

ITi

o-^

'5^s<,

to

B o

555S5

0)

e

•H
EM

rH CO r— rH to
LfMfMTV O O

*• w^ •• •• ••

ir\ ir> LTWvD >-o

• • •*

-4

o

O UrMTv

1 jT fr\iC\ 1

a o

O O O O O
r^ ro r^ ^<-^ 1^

en

r-i

e

•iM
EH

IrH rH rH CVJ ?\J

cr.

BO

r<^

■f:*.

1. i^r^ifv 1

4>

O O O O O

cu rvj CM OJ rvj

e

•H

EH

vo ''~0 vo rH
«« •• »• •• ••

(\J

o-^

r>-if»c\j o
1 jit f<MfMr\ 1

O

*>

e g

o o o o o o

rH rH rH rH rH r-4

+>

to

0)

a

EH

mvD o u?>o>>o

rH rH C\l CM CVJ t-<^

r-(

O^

1 1

a 0

o

'• ■• T*

1^ ro

fi

•H

o

EM

1

1

rH
0)

•s

Em

a

EH

ir\rH

o
ai

CQ

,o

O

rH OJ r^^ ir\VD

o

■H

CO

u

o

VD

OJ

to

cr<

•o

ir4

15

CO

IH

o

cd
•d

a

.rt

o

EH
I
I

CM

CD
Eh

VD

LTv

-4

9 o

BO O O Q
O O O O

CM Oi CM CU CM

a

.rl

EH

-4

irwO Q ITi CM
ir> inO O rH

I ^ J-

aS

a

Eh

--4

t^

CM

+5

a o

CD
EH

--)

a o

I

O O O O

o o o o

' r-t r-i r-t

-O to r^AJ
H rH CM r'^

rH^
1 J- 1-^ I

O O O O
t^-f— J — I —
■ • •• •• ••

is vDvo rl.
7J 7? C^ ri

?5^

§?^^^

£H

^^"3..

o

rM o

o o o

CM CM CM

B^
"^

-< H '^

I I

a n

o

a

•rl

EH

C
Oi

<0
X>
O

CLi

CM CM

r\ in

r-t CM i^-rt in

^-v

c:

r-

■H

+5

S

cO

iH

---1

r--

a.

0

1 in 1

u

+J

000

o

• !:!

000

«H

VD

a 0

ru CM cu

VD

(iy

60 rH 0

e

CM r<Mn

<i/

.H

*• f* •*

•f

C-i

r — r— f^

--I

VD J-

0^

1 tnin t

QJ

■tJ

0 Q 0 0'
0000

cjy

• 0

in

a c

i-t r-{ r-i r-^

a

^SS"^

•H

•• •• f* ••

U

EH

in in in in

o-f

VD 0
1 r<Mn

-g

+3

000

H

■4-

a.o

r> t^h-

•

rt

•♦ •*

»• «» 49 ••

(r\

li/

^<-^VD to

f<^

rH

a

inino

m

r»

•r4

•• ♦• f»

rH

tS

EH

^ ^ ir\

O

oH

inr--
1 -=»■ rn 1

+>

rH

• ?»

§§§s

p.

K>

a 0

(p

>o in in rH

-<!

E

0 0 r-i CJ
•• •• •• «•

^

EH

;d-^^-d-

-4

.mm

N

0-

( -:* in 1

*f «f •• «-*

c!

' 4J

o

• ?»

0000

-H

CM

a 0

CM CM CM CM

**

«* ••

<a

0)

rH fv^cT\ cr>

+»

•S

inintno

OT

• • •• *9 *•

EH

CM CM CM m

h

«•

»* ••

*f •* •♦ «t

O

--I

Vi

r

1 1

cd

• HJ

■p

a 0

in in

CO
-d

y

t-i

5

0)

a

CM ^»^

•rt

EHj

CM CM

>

»• »♦ ff •f

O

EH

1
1

c

•rl

•

■»3 »1

l<\

CO 0)

m

^1

r-i at K^^

r-t

n^ 3

^

ca C

CO

,0

Eh

0

1

;

190

X

u
o

Cm

o

u

.^1

o

w

r

<M

in
-u

fC

•H

o

J-

CO
EH

.. . .

<. ,.

..

■--1

r-\ CM

o-

I Lr^J■ 1

+-^

O O O o

• p)

o ci o o

Jd-

e o

ru OJ CM <M

LOiVD r^ CM

c.

J- J- u^^o

S

• •• •• ••

•f^

CM nj CM r^

a-

E-<

rH H rH tH

•■^1

CJ> ITN

,g

1,

1 J- ir> 1

H

+J

o o o o

h

• ^

o o o o

1— 1

-1

r^

S c-

i-f r-f r-t rH

CM r^ CM CJN

ca

Oy

Lr> LTl o o

K

e

• • 4^ ^t *•

•H

O O iH r-4

E^

l-l rH iH /H

..

• V ••

• *' .V •• ••

--4

CO

c

1 J- I- ••

■ij

• ;»

o- o o

CVJ

a c

CM CM CM

O'

Nnr~-» 00

B

cH'r-l CM

•H

• • «• ••

■ EV

cr^ fcr> G^

•-4

o

.i ■! .. ..

.. .i

..^'

• ^

l-l

a o

CM fM

n>

^ hr

fl

3 IfN

•A-

^ '••

E-t

WD PO

• • ,\- A- •«■

.%• .»•• .*• A<

fl

C

•H

^ h

CO O

^1

iH CM t^J*

fj P

o c

rO

o

O

VX>

o
en

cr<

o

60
t-4

Eg

o

cci

•r»

o

.CO

U3

--1

o

sncocMOOtoototacM^
r^ r<~v i-<^ 1-^-? CM i~^ CM r<^ 1^ r^N 1

o CO r~vx) J- CM cH o covxTirs'tn

O C^^C^^CT^C^^C^^O•^CT^CO bo CO'tO

rH ■

e

ft

erir-H hOLPil — CTNrH N-MXM — CJ~iO
ir»00000<-HrHr-f r-f -i-r CM

iHr-lf-HrHrHrHr-tr-lr-lrHi-HrH

in

c^.

to J- r— rO CM 1^^ p-H rH VJD
CM t^^ r>^ 1^ ^<^ 1^ ir% p^ K^ 1 r<^ I

c

lo^ ho, (M iH o cj> CO r-ivo LO-m
t~— 1 — ,r^ 1 — 1 — f*-vj3 vn vxi V.O vx) AXi

^COO CVi^VDCOCDCM r^vo r—
OJ CM ro 1^ K> (^ f'-\^ .d^ J- J- J-

rHrH-HfMrHpH-HrHrHr-tiHrH

;d^

^^-

r^ iH O rH r^- ltn M-\vr> o m j-
CMr<^r^M-NCMt-^r<^f<^r<^f<^KN I

O CTv CO l--V£) LTv J- r<-> CM rH O O

kT) irMr>LfMr\irMr\u^urvmLr. ir\

fH

1

EH

^LTiLrMTMnLnOOOOOiH

^

1

r^ H-> r^ f^ (^ i^^rj- ^ ^ j± ^ ^
iHrHr-tcHiHiHpHfHrHrHiHlH

CM

O

kD cb .zf U3 viD cTiuD J cn r- M-

^<^CMp<~^CMK^CMCM^r^f<^CM^^ 1

1

\r\Jt f^ CM rH o CTi CO r— vo m ltn

0)

• a'

■H

1^

rrs IXS r — CTi rH t<-\ ir\ t~- O^ rH t<^-d"

r^rHr^r^cM(^J^vJCMCMK^r^^<^

hrtv t-^x f^ 1^ f^ f^ i^ f~\ f<~\ 1^ r<^ 1^

rHrH. HrHrHrHrMrHiHrHrHrH
• • •# »■ •• •• >• •> "• •• "^ •* •^

IT\f>-P^rH O Cr>0C3 CM l-^jt rH
.If ru W li^T^t^rH J- f^'r^ K\ 1

a-?

o a^ CO r**-vo Lrv,3" r<-\ cm rH o o

K>CM(rMrMCMCM-VJCMCMCvJCMfM

•• -3-

a

COOCM-rt-vOCOOrMj:*VOCOrH

OiriiCucMCMqvJtMCMCMC^CMrA
rHr-IrHrHrHnHrHrHr^r-«rHrH

c

c

•H

-(J

t

o
m

O

'-4

cr cc o t«- r>-vD (M o 1^ CO ••

N^^ .3^ CM ;=f CM r^^ f^ rO OJ 1 1'

a t

r-i rH -H rH

a

EH

u

QrHrHrHrHrHCMCMCMCMCMf^

• •

rHrHiHrHr^rHr-lrHrHrHrHiH
rH^rH r-\ rH rH 7J,H rH H rHrH rH

r^ (\i rA.^ muD r^ 60 CTl O rH OJ

nH rH t—i

191

J

^

cH

oj iv^ tr\ J- r-- 1— 60 o ?o r-- itnv^ bo co ^
J- r^ J- u-^^ J J- lo J^ J- ^ ^ J- ^ J- 1

4J

o o o o o o iTMo Lr> i^ Lp> o o o o o

rHnHrHrHnHr-IOOOOOOOQOO
i-lr-HrHrHr-lf-lrHrHrHrHi-tt-lrHpHr-lr-l

EH

rvj 1^^ LO'^ j^ w cr> O i-i oj kn^ lPivd qo
r<-^ r^ f-AK^ r<^ (^ r^ r^i J- J- -=f ^ ^ .:* ^ J-

oooooooooooooopo

c\jc\jcocvjfuc\iajc\jojruajiMc\it\jtvjcM

-4
o

r^iH O t^ior— BOJ- O ii^v£> CO ^ O r-- o>

^ ITMI-v^ ^ ^ ^ 3- ITN J- ^ ^ LfMrN.=J- J- ■

o

•H

1

tH

!

irMTMOLfMrMOvO O O O O LPltOLfMOLOl

bococo60totowD60«ocoto r^.r— r^ 1 — r—

E

C-t

t)OOOOOr-liHrHr-trH,-ltH .-H r-l rH

ooooopoooooooooo

(Mojc\jc\jc\jaj(rvjc\j(\/c\j(\iOJajc\jc\joj

tr

-~4

60 ir\ OJ cr> 00 vx) r— J- co oJ- ai h-mrnj-
[^ ^ vD LO J- J- Lr\ J- J- Lf\3- u^JJ- J- ir^ j- i

K o

ooooooooooir^LTitrMTN ir\ m ur\

o
w

p.

1)

ir^vD r— CO o^ O rH cv r^ J- IjTvvX) f~- co g^ O <-♦

a^ 0^ CTv cr\ CTi Cr\.<3^ CT^O^a^C3^C7^C^^O^CT^C^>0^
rHrHrHrHiHrHrHrHrHrHrHiHrHr-Hi-HrHl-H

^

^ to irHvO J- KN CM Oi Q CO J- VjD VO ViJ CTl

CO

4^

oooooQOOOirMfNuiLfMrNLr\Lriur\

s

ro r^^ J-^J-J-J-^^J--3- ir\Lf>. iTMrMTy

J-

OOCOCOCOCOCOCOBOCOCOCOKlCOCOOOlOtp
rHrHrHrHrHrHrHrHcHrHrHrHrHi-HrHi-Hr^

rH

'

■^

rHCM rHirM-^KMr>OJ- >-J± CO VX) I^ LTv

^ Lr> r ir\fy u^ J- iTN Lf% J- jd" J- ^ ^ ir-, J- •>.

o

Kl

a g

ir^ooooooooirMrMrMTMrMn in trv

CMr^r^r<>,r^r^r^t<^r^OJfVjroc\4(MCV)C\lC\i

•H

^4

a

•H
EH

CO C3^ O rH C\J f^^ LTv^X) 1 — >0 OA O rH OJ h^rf

'<

COCO6Qft0COCOCOC0rO60W)CO60COeOCO60
rHrHrHrHrHrHrHr-trHrHrHrHrHrHrHrHrH

o

C\J

4*

r-H

r— r^ o h-'X) ooojQJ- o^r-tcMcvjin

a 3

irMrMriir^irMTMriLTiLrYOOO oooo Q

r^r-{r-i,-<r-{r-ir-tr-<rHr-{r-i<-\r-t, r-i,-{i-\l-i

+3

•
a

.H

i~r>, ,^3- j-^jj-j-j-^j-j- Lr^^Mr^LP^lf^

t^r — r^r— r~-r~r— 1 — r^r-i — r~-t — i — i — i — f^

r-*r-tr-tr-tr^r-ir-{r-\r-\r-<r-^r-tr^r-ir-\r-<r-{

^4
o
(n

■p

K
•H

o

1

(U
i-H

EM

C

OJ 60 MD O CTl 60 q^-^ J- (M So ^--d- V£) KM —

;i- ^ LTvvx) ^ ^ .3- ir\ 1^ J: ^ ^ ^ ^ J- ^ 1

■♦J

O^J-J--=l-^^-^-*J' -=^^ J- ^ -:*• ^ O

a

•H
EH

O rH OJ f^^ IfNV^ t^ 60 CTl O rH Oi l*>^ ITy^
r^r-tt-ir-{i-ir-{r-tr-{r-tr-tr-*r^r-ir-1iHr-tr-t

o

•H

>■
u

o

1

r-\i-ir^r^r-^<-ir-*l-t

192

o

«

u

c

ni

S)
t/3

CTi

5>

o
o

■H

-p

TO
CO

o
c

E-l

en

^4 K>, r-4 U^vD CO r-i QJ ro tr (5> 0^ O ru v^ q)

^ ooooooQOOQOOogog

a Or-Hi-li-lrHr-tf-tcHr-lt-lrHr-H

kr>

-4

a§

--J

oi oj cv clj c\J t>J CM ru c\i cvi cvj c\J CM CU ru CU

t— H o ^ g> J; \g '^o g^ r;:~\^ f^-s i^r^ ^

oooooOLr\ic>'^LPvLr\00
S"";H^!=ioQQqqqq

o o
o o

C\J «<,^ LOV£) r— to CTiO rH CM K%^ £*^'^

LrMriLPiuTNirMrsLfMnoooooooO

CMCMCMCMCMCMruCMl^r<^l^ff^fr»ff^':f>fr»
CMCMCMCMCVJCMCVJCMCMCMOJCVJCMCMCMCVJ

aoiHtor— cMOr^LnaMO?9Ci>'^^i£* ■

r^ r^ r-r> J- ir^^ f^-:* ^ ;d- ^ -3" t^^ J 1

S o

ITN

■~-J

lOirMTMi^LfMrNO O O O O ITMCMrMrvLTN
COCOCO'SObO'SOBOCObOCiOBOr^l — I — I — r^

CM r^ J- U^VJD t^ T» CTv O rH OJ 1^ J- Lf><^ r--
CM CMCM OJ OJ CM CM CM ^'-^^<^^<-^^f^l<^K^^O^O

CMCMCMCMCMCMCMCMCWCMCMCMCMCMOJCM
CMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM

00 lr^ J- r^cMvjoV;£^o^vDrr-wr--c\jCM

CM f^ j:!- r^ r<^ r^^ mvn ^CMJ-J-CM^li>l

l~)r-lOOO0000l^^lr^LPlL^^Lf^l^L^^lr^

rH CM r<-i^ lr^vr> i— bo cr. o r^ oj t^^ l£)V;D co
UA Lr> lr^Tr^tr^ UMTMfMOOOOOOOOO

r^

„^^^^^rHrHrH.-lCMCUCMCMCMrvJCMOJ
ScMOJCMOJCMCMCMCMCMCMCMCMCMCMCMCMCM

o

f3 o

r^ OJ OJ CM CMCM c^i^i-'^h^r^'^r'^f^l^ r^ J

^4

a o

tM

o c^^ O lr^ 1^ 1-1 vx) cr. CO J; vxi iTv 1^^ or; CM

CMCMCM?0CMCMCMCMWfUWCMCMWCM0JOJ

oOa cnco ,^ M lorG o r-^to ^ q! q J;^
r^^ 1^ r-^^ J- ^ J- J i-O r^-=r -^ -=1^ -=t -fC ,',

°°^^°^^^r^C\JCMCMCMCMCMaiCM

(0

e

EH

M en O
LTMO O

rH CM ro J- ltn^S ?r ^ 9^ 9 d 2J ^"3

-~l

+3

a o

-4
o

+5

a?

^

O rH

CM CM

CMCMCMC\JC\J?UCMCMCMCMCMCMCMrM

r-^ r^ 1^ J; ro «> !i>-3: (^^ ?i) -?, V ,

irMrMrMrviTMrMTMrMr^ggoooooo

r^.r-i rHr-lrHHHr-(HHr-<r4Hi-jH<-^'-^

O r-t CM

•• •• •♦

009.

CM CM CM

iv>^ invD r-io en g H CM k>^ in>5

'T? -r^ ':?'::•::'::* "r^i^ i^ f. -^ 1 -5 5

K^ fi-N^^ CM J- J- 1^ CM r<^ r^ 1^ r^ r^-^ J.

I

CO

o

o^^^^^^^^^^^^^ <=>

60 CTiO

•* •• ••

CM ix^ J- i^*-o Cr 52 SD 9 ^ ?^ ^

O 00 OOOOOHr4HH

H CM ,<^^ invi) h-M cng H w r-.^ inv;^ i;^

195

1

60

o>^

c\) c\j ^ iTi i7\ ir\ iTv J- LT-,^ J- ^ J- -d- tr> J- cm

4J

• • ••

<D
E

•H

OOOOOOOOQOOQQQQQQ
HrHrHrHrHrHr-lrHr-li-trHiHiHrH,HrHrH

ij- LfNvjj r— to <T^ 0 r^ cu 1^^ ir\vI5" r^ cb cT. rH

OOOOOOiHrHrHiHrHrHrHrHrHrHrU

r^ ^'^ r<~\ N^ f<^ t'^ t^ r>'S r'A r^ K^ lA t<^ t*^ 1^ ^'^ ^'^

I--

OOOOoOLriirMfMr. looooopo

rHrHrHtHrHr-H'OOOO 0"0 O O O O O

0/
El

• H

It ur^Kn r— 60 o^ d rH (\j t^^ ^r^•sD h- 00 (!a 1^ ^

rHrHrHrHrH,HajnjC\jajC\JCUOJOJf\JC\Jt^

o

•r-l

cMruc\jrvjajc\jCMroc\jc\j(rvjojc\jajajc\jaj

^

--4

c

0 tocur— irMXMr\r^u^ r^t— r— vD 0 iTNirS

r-t

■D

u

4J

LTVLOLPvLOLClLPvO O O O O Lr^Lr\Lf^^r^Lr^L^>

10 f<i ; 0 to to '0 to to to 60 bo ^- r- r— r- r~- r^

1y
S
•H
EH

^ r— n CTi 0 iH c\j r-^^ lr^vr> r^ to ct\ 0 '-^ c\j
r<^ h^ K>, r^jd^ J- J- ^ ^ ^ ^ --"*, J' -^ ir> O-MO

iHrHrHrHrHrHrHrHrHrHrHcHrHrHrHrHrH

VX)

p.

ITl

0

0 r-(T\&o 1 — hoLPvtovo rH^vo 1 — a^^-r<-^Lf^-H

4J

OOOOOOOOO^^^l^^lr". u.tOLT. u \ LT, m

•H
EH

0
J5

■HrHrHiHrHrHrHiHrHrHrHrHrHrHrHiHrHiH

^

--4

0 0 to to 0 cr\^s? ir> i-r->, rH r— ir> to u) \r\^ c\j to

J-

2?

4J

LTNLT. lr^lr^L^^tr^lr^lr^lr^QOOQQQQQQ
r^f^t^i-^r^r^i^l^ f^.-J J" ^ ^ ^ ^ ^ r3- r*

nj

w cr. 0 rH CM t^^ itnvjD r— t-o ctn 0 rH aj r^^ lpi
cu ru to t^ ro r-^ t^ r'-\ KM^o ^<-^ ro-5 J- ^ jd" J* ^

1

•H

6h

r\j(Majc\jajaiojt\j(MOjc\jc\jc\jrvjc\jc\jc\jCM

.-I

•-J

r->

0

^ 3- J- ,5 3^ K-i ro^ J-^^j-^J-^ifMO J ,H

E 0

000000000 LT. lOLfMTMrMriLOLPvirN
t^r^rOt^K>ror<^r<^r^0JC\Jt\ifVJCVJC\JCVJC\i(\l

'^j

^ Lf>l^ r-- to 0> 0 rH nj 1^^ LPv^X) r- to CTi rH l\J

CM

c.

c\jc\jcu'Ajc\jrjoj(MCMtMOjc\jCMc\j(\jajc\jru

OJ

0

r— to vo r<-> 0 0 Lf^ ic\^ vx)J-LrMr>Lr>c\iLr\cv 1

+5

E g

LPvLP\LOLOL£Mr\ir\irMrio 0 0 0 0 0 0 0 0

rHiHrHrHrHrHrHrHiHrHrHrHrHrHiHrtrlirHtH

0

0
G

•H
L-H

0 rH OJ t^^ LPikn r— to cTi 0 rH nj f'^^ ir\<..o r«-

iHiHrHrHrHrHrHrHrHrHC\JC\JOJIMOJ(\IC\J<M

OJCMCMtXjrvJCXJCMCMCMOJCMOJIMOJOJfMCMCM

0!

«

.a

:*

c

EH
1

rH

0

-u

o^^-^st^^^^^^.=i-^^^^ 0

•H

r— bo a> 0 iH c\j t>^J* lOivx) r- to o^ p rH oj -d

c\jCMOjajcMrac\jcMCMC\jCMruc\jc\jcM(\ico
fMOjc\jc\jc\jr\jtvjc\jojojaiojc\joacMc\jc\]

to

rH

1

to

t»

u
to

>
0

0

C
•H

13

rH c\j t^^ Lr\»x> 1^ «3 <r\ 0 rH nj r<^ J- LOVD r~- to

• rHrHrH'r^rHrHrHrHrH

194

d

o

a

r-H

U
O

tw

VJD

o
uo
cd
P-

<o

to

to

0)

CM
O

CO

w

J-.
o

o
E^

CTv

0)

r-1

EH

cr>

r^ ^ ,-:^ J- J .=r ^ Lo li^^ ^ J- ^ ^ ^ -^- .^^ i;;^

-p
a o

oooooooooOqoooooo

t^ Jt u^<ij r^ CO o-v Q r4 CM K> J- Lr\<^ f^ ?& 2

a o

-^

^ ^ ^ ^ ^ ^ r-i r-{ r^, rj\ r-\r-* rH rH r-i r-\^ r-{

^ A' r^* r^ ^l>^ !^-3 ^ ^ ^ "^'5 i" ^>'5 ^
^ ^ lo Ln,-i- 3^ ^ ^. ^ ^. ■^- '-^ ^ -^ -^ ^. '^^.

-» •• •• >• »• •• •• •• •• •• •• •••••••;• • ^^

QOOOOOOOOOOOQOQQQ

■^^r^r^^^r-^r^r^n^^-^ .-li-H i-t rH i-( r-l rH

cj r^ J lr^MD r— to CT-, o rH OJ r<^^ i^^ "^ '^

O OOOOO p. Or-lrHl-HlTlH'-^r^ ■-''-'

^* ^ rH i-H I-t rH r-H rH fH rr( rrl T* T* Ci "^ H TJ .;,

iK<-o >i) ir» to o^ f-^ J- to o q> i£>J^- 1£> i£>^ SS hS.

a* •• •• •• •• •• •♦ •• *• •• *• •• •• "• U ** 11 11

J- tr\V£) 1^ to CTi O rH CM K%-^ LTvVO I — «J CTv O CM
3 r-l iH I-rSlnKjCMCMCMOJCMOJCMrjOdl^l^

^^^^,-|rHrHrHi-4nHrHrH-rHrHrHrHrHrH

kXJ

lo

-p
FJ o

irMrMr\irM.rMfMr>0 ooot^^lr^lr^L^^Lf^l£^

.^^55'i)'5^^^:55^^^'°^^.^'i?^

o

eg

oor)OOOOOOLr>"^UMfMrMr^LrMr^u_\

--J

rH r-( rH rH rH rH

lrM^^^vr) ir\r— >jd ir»^ ^ lir !£^ Cr ^ !i> !i>-3;

L^Mr^Lr^Lf^L^^l^^t^\lr^lr^

iTNVT) r- CO cr< O r-( CM r^^ l£^^-0 [:: ^ 2? ^ Cl JfJ*

OOOOOrHrHrHrHrHrHrHrHrHrHCMCMCM

r-\ r-\ rA r-\ r^ •-< r-\
r-\ rA r-* rA r-\ l-\ <-*

r^ rA <-\ rA rH,.H
r-\ rA r-\ r-* r-\ rA

LTi o o r-^ LTv CM r^ o J- r^ ^-^ "^ 2^ $ "^

f^ 1

doooooOOOooopirMrMTMrMoiri

ro

ir\>aD h— CO cr> Q ^ cv

--1

EH

--J

^r^^- ti^vp r- CO cr> O H f^

ocDoooooogggooooooo

yrjZlJ^—l^rHrHrHiHrHrHrHrHr-lrHrHrH

[?,^ J- Lf\ ir>j:J- ^ ^ ir\^ J- J LP. ir\^ r<^^ t

°°S,^^^^^^!5i^c^JCMOJCMCMOJCMOJal

OOOOOrHrHrHrHrHHrHrHrHrHOJOJOJ

ooooooooooggggoooo

(^rH LT^J- O rrvO'rH CM^Ci! [Ar^Q H^ J;

a o

tH

o

,LfMrMrMrMP>ir\Lr>ooooggggg

^^^^rHrHrHrHrHrHrHrHrHrHrHrH

0^a^

I — nn rr. O rH CM hr> J" LTvvn f~^0 OsO t-^C^
Cr^CT^a^CT^CT^Cna^C^^(J^C^^C^^C7^0^C^.C^\0^

I K^

ro O to LT, - O g tr> OJ CM ^p^^ "^ .

o^ ^ J- J-

J- ^ J- J- ^- -=»• ^ -=1- -^ -^ ^: <^

EH

^ o

"^

O rH CM 1^^ "^--O tr 1° P^ O |-4 CM Kjv J-
OOOOOOOOOOrHrHrHrHrH

<io CO cT^ en CTv CJ^ a> cr> cr> CTi cr> cr\ cTi cri cr^ en cr>

^CM ^<^J u-Nv£)r-to a^Or^ CMr-.^Lr.^;:;-w

195

o
o

ON
rH

J,

ci
c

1/3
l-<

c

«H

Cfl
4J

CD

Ml
•H

C
EH

••

•-4

o

B ?*

• • ••

• • ••

.•1

W 60 OJ -=»■ 'O U> Cp !.0 rH M3 50 OJ ^ O VT) -:J- VX)
h<-MrMr> J- J- ,-f J- -d^ h<~> t^ J- 3- ITvif J- rH

OOOOOOOOQ li'.'^'to O'O o oo
-1 rH rH rH rH rH ,0 .O .O.Q .O .'^TS C^ CTi Ox CT. '0"^

»^ ** •• •« •» •• •• >• •. •• ,^» - ^* «• «• >« »• ••' ••

C^ Q rH CV f<^^ LTN^JTi, Ir— lO. crv O rH C\J'r*^-:*V5

i-l iH rH pH rH rH rH .H.V-t .rH *H (H: rH rH H H rH

•* •• •• •• •• ■• •• •• *•«'*••"*••*'•«**»•' "V***** •-••*/••. .»*»

rH LTMv-^ (\J Lf^t^t-rLTvJ^: LTv iS rH O r— :3 C^^
O O CJ o o o o,o' Q O DO Q^(r> Q Q Q

T-\ <-\ r4 r-\ r-\ r-\ r^ r-\ t-^ r^'r-<t-\' 'r^' "rH 'r-^ r-V i-<

t>-

e

c

O^ O rH OJ r^^ LTv"^ h-r IC' O^ O rH nj ^r^ J- VO
ITl O O O O O. O. .O. 0,.O..O..rH rH rH rH rH rH
• • •• •• •• •• •• «« •* •• •• •• '•• '••"••* «r •«-••

Ar-Kr-\^^r-\r^r^T-^^-^r^T-^r^r^ r^ r-\ r-\ r^

r-K r^ J- J- J- J- J- ^. ^ 3^. i^r* ^ -d- ^ -d- OJ

• 4J

G o

ooooooirMr\ir\LfMPvoOQOC>o

rH rH rH rH rH rH O.O.O OOOOOOOO
r-\ r-\ r^ r-K r-K r^ r^ r-{ f-\ r-^ rA T-A'r^ T-* r^ r-\-r^

EH

O rH rVJ cr^ J- Lr\<.i3 >- -^0 g^ O rH OJ (<r^ J- irNvD
C\J CMOJOJOJ OJOJCVI.OJCXJ M-\ t^ r^ f^ ri-M-f^ ro

ajcxjojojcucuoicvojcvi, <\i. ojojojojc\jc\j

VD

C

c\j LPi^ r— -0 CO OJ t^- o ^-u■^ J" oj o f^ o c^

^ ^ J r^ r^ r<^^ ^ 103 J" J" J" 3" J" ^ f^

LrMr>irMr\uMrvo o c o o icwxwcwrws \ \r\
!•' 0 "o i.j ro ij "<j u;i '"o »c 'c r— r— r — r~-p— r^

(D

J- Lr^^.^ t-- tp o^ o rH c\j ^r^ j- lpooj r~- tc" cr, o
j-j-j-^ifjd- LTiioLOirMrMriLrMrMr-. Lr.o

EH

oJaJOJoJOJOJ'^JC^c^^oJoJOJaJOJi^J^uoJ

t

1

pi

in

-4

c

K^ ro J- LP> J^ ly^ H^ rr\Si J- 1^ r<-> t^-^ ^ J- J- J-

• -9

e 8

oooooooooLTiirMou^LfNLPitfMrMrv

-fl wo VX) 'J5 VID VJ^ VJD VJ3 VX> LTMXMr^ LPi LTl LT. LP, LTi LPi

•H
EH

OJ K\^ ir\u> r^ lo cr\ o rH OJ r<-\jd- Lr\>-o r— '.-o o

rHrHrHrHrHrHrHrH0jrV0JOJOJ0JOJOJt\jN-\

C\JC\JC\JOJOjaiOJOJOjOJCUOJOJCMOJOjOJOJ

^

B

EH

0> O rH OJ r^^ ITN^O 1-^ CO CTi O rH t\) r^^'J LTlVD

r^ J- J- J- ^ .* ^ ^ ^J- .^ ^ irMrMrMr\ LTsLO u-\

ojW(^rui^rGc\jruoJoj'rooJojojojoJ';Jc\i
ojojojojojrvjojojcviajojojc\jojoj(\joioj

r^

O

o o eg '0 CO LTivx) 1 — OJ 'o o Lr\ o o OJ o Id- i-o

•■9

B g

c~ C'ooo.ooooLr^u^LrvLrvLriioLr>uMr>

i^r'^r<^r'>t<^i~r\r<^f<^r^OJ OJ AJ otAJ CU t\J CO oj

c,

C

•H

EH

"-0 0-\ O rH OJ t<>, J- LTiU!! r— CO en O rH OJ 1^^ lO
OOrHrHrHrHrHrHrHrHrHrHOJOJOJOJ.OJOJ.

OJ OJ OJ OJ CM.OJ.OJ.OJ OJOJOICMOJOJOJCUOJOJ

OJ OJ OJ OJ OJ OJ OJ oj'oj'f\j"c\j ■oj'oj'cu oj cuoj 'aj.

(M

o
B ?

(D
E

EH

r— iTN o ir\ OJ 'r^lrf- ir>cv 'i — ooJLr^ooJO^o

i-^ K>i 1^ J- ^ J- ix^ r^3^ r^J* J- -* r^^ J- J- ^

lr^Lf^lrMr^l^^l^^t^^'lP\Lf^O O O O o OO.O o
r^r-^t-\r-{t-\r-\<-\r^r-\r-\t-\T-\r^r^r-{<-\r^r^

1^- !-0 cr> O rH OJ rr\':d- Lrvvr)"^- r_g cr\ O rH OJ. t■r^^
l«^ r^ r^^ ^ J- J-, J- .-J- 3^ J- 3^ ^ LTMrMrMPv Ln

rH rH rH rH rH rH'rH" "rH rH rH pH rH rH rH rH rH rH rH

OJ OJ OJ 'AJ OJ OJ.CVJ, C\J (\J OJ OJ OJ OJ OJ OJ OJ OJ OJ

rH

C

O IPi i-r\ J- LP\ LfA O O rH rr^ "' r— O J- (-"^^
lO rH OJ K>f^ K\^ IP) J- J:J- J- l-r>,^ ^ ^ 1

B^

O ^ J- J- -If 3- J- ^.^ J- J* ^ ^ ^ ^ O

01

n

0 O^ O rH OJ r^ J- LnvO ^- '.: crs O rH OJ K^

O O rH rH rH rH rH ri.rH r^■rH rH OJ OJ OJ OJ

OJ OJ OJ OJ OJ OJ OJ Oj'OJ'OJ'OJ'OJ OJ OJ OJ OJ

1

u

o

1

r^ OJ r^j- ir>^ r— .0 o> o rH OJ iA J- iP>k.o r— w

19C

CTi

J-J

rH OJ ^<-^ t<^ J- ^ ^ J ir. J- m lo J- ^ JF ^ i^ ^^

OOOOOOOOOOOOOOOOO
rHrHrHiHr-(rHOOOOOCriCria>Cr. o>cri
ajC\JC\JC\JC\JC\JOJ^ru<\jajHrHr-(rHr-l,-l

S

•H
EH

to en o rH OJ rojd- ifNvx> ^--^oc^^Q rH cvi r<^m

•• •• «• •• •• •• *• «• v« •« «• •• •• »« •• •• ••

ojc\Jc\Jojajc\jc\jc\ioj(\jojcvjcMCuajojaj

r-tr-ir-lr^r-ir^t-i'-ir-it-ir^r-iiHf-ir-lr-ir-t

to

c

*• >«

«)

a

•H
EH

r^ J- LfNM3 ir>r— ^^P^^OQC\J^r^lr^^^J■rHl^^

OOOOOOOOCOOOOOQQQ

rHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrH

zi- lTvvjd r— ijc on O »-» oj ro^ m^D r— ic cr\ OJ
;d-j-j-^^j- irMr\LjrMrMrMrMrMrMrMr>o
•• *« •• •• •• •• •• •« «* •• •• •* •• »• •• ••• **

'-{r-l'-*r~\r-ti-{r-tt-ii-tr-ir-\r-1'-^r-if-ir-{0J
r-ir-\r^r-*r-lr-ir^i->r-tr^r-*r^r-tr^.i-ii-ir-t

a
o

r—

--*

o

CO CT\KD Lr\ LTv J- VO rH W I^ lAu:! iH ijo, lA Q VJi?

<D

B

•H
EH

OOOOOOirMfMrMTMnOOOOOO

rHrH>-HrHrHrHOOCQOOOOOOO
rHr-lrHrHrHrHrHrHrHrHrHrHf-lrHrHrHrH

cn O rH OJ r^^ ir\^fi r— JO CT^ o r^ cm t<^ J- u)

OrHrHrHrHrHrHiHrHrHi-HCVICNJOJOJOJCU

^r-ir-tr-ir-ir-tr-ir-^r-ir-ir-trHi-ir-ir-ir^t-i

<o

O

s^^^^^^i^^s^^.^^^?^:^

a 3

LrMTMrMTMnmo o o o o o trMf\ifMfMri
CO to to tc to to so tc tc to to to 1 — r- r— r- 1 —

S

1 — to CTv O rH OJ I^^ mvO r— to C^ O rH OJ hr>,
1^ rn, r^^ jt^^JStJ^Jt^^^ LTMrMntrN

OOOOOOOcJOOOOOOOOO

rHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrH

03
0)
CO

in

o

5^J-^^-*J- i-oi^a-^J-^ r<Ar<-\(^^^ r^

-"1

QOOooooocLriLfMrMriLfMiMrMrvLn

(0

B

O rH OJ K>,^ LPiVO r~ iO lS\ O t-t OJ l>r% J- LTvVD r—
rHrHrHrHrHrHrHrHrHrHOJOJOJOJOJOJOJOJ

oobooooooooooooooo

rHrHi-HrHrHrHrHrHrHrHrHrHrHrHrHrHrHrH

J-

o

CM r^ m J- 1^3^ J- ^ ^ ^r^^ J- ^ J- J- IV-, r^ J

• -a
Si

a

•H
EH

oooOOOoooLTv'J^i^LrMrMjrMrMrMn

CM

O
•r-t

O^CT^c^^a^c^o^cT^'T^c^>cT^a"^o^c^^c^\CT^a^cr^Q

f^

o

njJ-ooj-*ojQifSirMr\Lr\ i — ^ rr\ o oj f\^
f^ J- m J- r^ r^^ J- J- r«-v J- ^ ^ J- J- ro r^ r<-v

a 1

l~^'~^r^t'^f^r<^l^f^l^CM OJ CM OJ CM CM OJ CVJ CM

a

• H
EH

vx) (^ to o^ O rH OJ ti-\^ mvD r-- to o^ o r-t cm ^

rHrHrHrHCMCMCMCVtCMOJOJCMCMOJr^K^'^'^

+:>
W

O

OJ

--4
o

O CM to J- o J- m.Q CM ^ ir> r-- in OJ irw^ r-- r--

si

0)
E

•rH
EH

LrMrMrMrMrMrMmrMrvooooooooo

CO

O rH CM 1^^ invXI 1^ to CTnO rH CM t^-^ L^^ ET"

iTMrMrMfMrMpi m'm rnnni OOOOOOOO

pctotocotoaotot;,totocricr\cr\crvCTMT\<rMTv

^

rH

O

OrOOCMJ-VD ojJ-J- CMin ^-\^ l~~ i£\ .

o

e g

Ojt^^J*^^^^^^^^^^ O

1

a

•H
EH

^ r— to C3^ O rH CM rO^ rnv^O h- to iT. Q rH

OJ CM OJ CM r«-M^fni<^r^r^t^tOf<^m^.=J"

(U
i-H

Eh

tototototototatoootoEototototcao

i

(1)

0)

O

O
•H

>

(h

,-i CM rnJ- LfM^ r^fto ctnO rH cm r^J- io>^ r— «)

rHrHrHrHrHrHrHrHrH

197

APPENDIXB

LISTS S^roi.aWG TILS- AND LOCiiTIOH OF HAULS, D2PTH5 FISHED,
AND NUISERS OF E^XrS AND LARVAE 'PAKEN

Data are given in thj foliov/ing order: Station number; date; time
interval; latitude; longitude; type of net; depth interval fpr each haul;
numbers of eggs; length and number of larvae for larvae of each .5 mm.,
interval of length. Tine of day is given to the nearest hour vath the
hours numbered consecutively from 1 for 1 a.m. to 2U for midnight. Depths
are in meters and lengths of larvae in millimeters*

Excmplo: (Station) 1832;- (Month) h/ (Day) 10/ (Year) Ul; (Stoj-t)
llh- (End) l5hj (North latitude) 33:13; (Vfest longitude) 1L8:26;. (Type
of net) B; (Upper limit of stratum fished, depth) 0- (Lower limit of
stratum fished, depth) 13; (Number of eggs taken) E 161; (Larvae) L:
(Length) 2.5, (Number) 3; (Length) 3.0, (Number) 3?; (length) 3o, (Num-
ber) h; (Lir.iits) 17-26r (Number of eggs) E 103: (I,arvae) L: (Length)
2.5, (Number) 22; (Length) 3.0, (Number) 366; etc.

Net types- are as follows: Type A is a closing net of one meter di-
ameter mouth; made of No, 2l|XXX grit gauze, 'vvith the last half meter of
No. 56XX>[ grit gauze; and with a detachable "cod end" of No. 56XXX grit
gauze. Tj-pe B is a closing net of one-half meter diameter mouth; made
of No. liOXXX grit gauze, v/ith the last 65cm. of No« 56XXX grit gauze.
Typo G is a closing net of one meter diameter of mouth; made of cotton
scrim with openin,gs approximately 1 mm» square, with the last half meter
of No. 56XXX grit gauze; and with detachable "cod end" of No. 56XXX grit
gauze. Approximate sizes of openings of XXX grit gauzes are: No. 2I4.,
1.10 mm.; No. I4.O, .65 mra. ; No. 56, .U? mn.

F8Aj U/19/39; 3h-9h; 32jU8, 117:!42; A; 0-3: ^ :-377; L: U.O, 1; U.5,1;
5.0,Uj 5.5,2; ;?. 0,2; 8.0,1; 9.5,lr6-8: E 2879; L: li.0,1; U.5,3; 5.0,7;
6.0,1; 6.5,2; 7.0,1; 7-5,1; 8.0,2; 9.0,2; 10^0,1; 10.5,1; 7-l6: E 3878;
L: 3.0,2; 3.5,1; U.0,3; U.5,8; 5.0,9j 5.5,2; 6.0,2; 17;25: E 1265;
L: 5.0,1; 8.0,1; 30-3U; E 56; L: 0; hh'h6: E 5; L: 0; U5-60i EO;
L: i|.5,2; 108-115: E 6; L: 0.

F8B; h/19/39; l5h-19h; .32:U8, 117:UU; A; 0-2: E 2892; L: 5.^,lj 6.5^1;
7.0,1; 15-17: E 1359; L: 3,5,1; U.5,1; 29-35: E 88; L: 0; 53-^8? E: 8;
L: 0; 68-72: -E 0; L: 0; I6ii-l83 : E U; Lt-O.

F9A; U/20/39; 2h-8h; 32:30, 117,32; A; 0-5: E 52j-L: 3.^,$; U.O,ij .ii.5,2;
5.0,U; 5.5,6; 6.0,1;..6.5,3; 7.0,2; 12-lU: E I9I4J L: 6.0,1; 28-32: E 5U;
L: 0; U5-57: S 1; Li 6.0,1; 56-59: E 1; Lj 0; 109-^:-: E 7; Lt 0.

F9B; U/20/39; 9h-13h; 32:29, 117:39; A; 0^2;, S 9lOj L: 5.5,2; 6.0,ii;
6,5,5; 7vO,r,«.8»0,3j 8,5,;L; ?.0,1; 9. 5, 2j 10.0,1; 11,0,1,- Ih-J^; E iil37;
L: 3.5-,l; U.0,6; 5.0,1; 6.0,1; 57-66: .E-6; L: 0; 126-m9: E Uj.L: 0.

liHjnly one stray angle reading.

198

1832; U/lO/Ul? Ilh-l5h; 33:13, 118:26; B; 0-13: E 181; L: 2.5,3; 3.0,35;
3.5,U; 17-26: E 103; L: 2.5,22; 3.0,366; 3-5,187; U.0,2; li.5,6; 5.5,1;
28-39: E 11; L: 2.5,1; 3.0,U7; 3.5,31; U.0,2; U.5,6; 5.0,2; Ul-55: E 0;
L: 0; 53-66: E 0; L; 3.0,2; 3.5,2; 69-83: E 0; L: 2.5,1; 3.5,1; 5.0,1.

20U6; U/30AI; 17h-21h; 32:29, 119:26; B; 0-3: E 0; L: 3.0, 3; 3.5,3;
ii.0,1; U.5,2; 5.0,33; 5.5,36; 6.0,9; 6.5,1; 7.0,1; 9.5,1; 6-12: E 0; L:
3.5»2; U.0,2; U.5,2; 5.0,31; 5.5,51; 6.0,55; 6.5,20; 7.0,3; 7.5,U; 8.0,2;
8.5,2; 9.0,1; 9.5,1; 10.0,2; m.0,1; 15-25: E 0; L: 2.5,1; 3.0,1; 3.5,2;
U.0,1; h.S,3; 5.0,27; 5.5,37; 6.0,7U; 6.5,23; 7.0,9; 7.5,5; 8.0,5; 8.5,2;
9.0,2; 9.5,2; 10.5,1; lU-29: E 0; Lt 3.0,2; 3.5,11; U.0,11; U.5,12; 5.0,38;
5.5,29; 6.0,15; 6.5,3; 23-U3: E 0; L: 2.5,3; 3.0,19; 3.5,30; )4.0,l6;
14.5,11; 5.0,25; 5.5,3; 6.0,6; 6.5,3; 7.5,1; UU-62: e 3; L: 2.5,1; 3.0,13;
3.5,11; U.O,U; U.5,1; 5.0,1; 65-92: E 18; L: 6.0,1.

20h3; 5/2AI; 20h-2Uh; 32:50, 118:13; C; O-h: E 90; L: 3.0,1; 5.0,20;
5.5,30; 6.0,32; 6.5,20; 7.0,38; 7.5, U35 8.0,53; 8.5,U8; 9.0,33; 9.5,17;
10.0,19; 10.5,6; 11.0,lUj 11.5,13; 12.0,7; 12. 5, U; 13.0,5; 13.5,3; lU.0,1;
IU.5,1; 15.0,2; 7-13: E 16; L: U.0,1; U.5,h; 5.0,21; 5.5,19; 6.0,12;
6.5,6; 7.0,10; 7.5,9; 8.0,lU; 8.5,15; 9.0,23; 9.5,13; lo.o,26j 10.5,12;
11.0,7; 11.5,11; 12.0,1^; 12.5,9; 13.0,7; 13.5,6; li;.o,3; ll;.5,3; 15.0,1;
16.0,1; 17-26: E 1; L: U.5,1; 5.0,2; 5.5,1; 6.0,1; 7.0,1; 9.5,1; 11.5,1;
12.5,1; 13.5,1; 25-33: £ 1; li U.5,1; 11.5,lj 12.0,1; lU.0,2; 1U.5,1;
15.5,1; 30-53: E 0; L: 11,5,1} 52-73: E 0; L: 0; 68-98; E 0; L: 0;
97-1U2: E Oj L; 0.

•2U52N; 6/I7-I8/UI; 22h-3h; 32:23, 117:52; C; 0-3; E 1;.L: 11.0,1; 12.0,5;
12. 5, U; 13.0,3; 13.5,3; lU.0,2; IU.5,1; 15.5,1; 18.5,1; 7-12: E 0; L:
8.5,1; 13.5,1; lU.0,1; 15-2U: E 0; L: 9.0,2; 11.0,2; 11.5,1; 12.0,1;
12.5,2; 13.0,1; 13.5,1; 15.0,2; 22-35: E 0; L: 13.0,1; 36-52: E 0; L:
10.5,1; 11.5,1; 1U.5,1; U8-8U: S O; L: 0; 57-101: S 0; L: 0; 90-1U6:
E 0; L: 0.

2U52D; 6/18/Ul; 9h-lUh; 32:17, 117:52; C; 0-3: E 0; L: 0; 6-12: E 0;
L: 0; lU-21: E Of L: 0;. 25-36: E 0; L: 11.5,1; 38-U3: E 0; L: 11.5,1;
12.5,1; 52-67: E 0; L: 17.5,1; 18.0,1; 68-81: E 0; L: 18.0,1; 96-152:
E 0; L: 0; 125-156: E 0; L: 0.

2U5UN; 6/I8-I9/UI; 21h-2h; 32:12, 118:38; C; 0-U: E 8; L: U.5,1; 5.0,2;
6.0,1; 6.5,2; 7,0,1; 9.0,1; 9.5,1; 10.0,1; 10,5,3; 12.5,1; 7-13: E 15;
It U.5,1; 5.0,2; 6.0,1; 6.5,2; 8.5,1; 9.0,2; 10.5,1; 11.0,1; 12.0,1;
12.5,1; 18-25: E 1;,L; 5.0,1; 7.5,1; 8.0,1; 9.0,3; 9.5, U; 10.0,1; 11.0,2;
12.0,2; 12.5,1} 13.0,1; 13.5,1; lU.0,2; lU.5,2; 15.0,1; 2U-32; E 0; L:
U.5,5; 5.0,2; 5.5,1; 7.0,2; 8.5,1; 9.0,2; 9.5,3; 11.0,5} U.5,35 l2.o,3j
12.5,3; 13.0,U; 17.5,1; 38-52: E 0; L; 10.0,1; 11.0,2; 12.5,1; 52-69:
E Oj Lx 0; 66-1081 E 0; L: 0; 90-1U9: E 0; L: 0; 122-208: E-0; L: 0.

2U5UDJ

12.

10.

E o'i it 0} '120-198 i'e Oj ii~6»

199

LITERATURE CITED

Clarke, G- L.

193U. Further observations on the diurnal" migrations of copepods
in th(3 Gulf of Maine. Biol. Bull., Vol. LXVII, No. 3, PP.
U32-UU8, 6 figs.

Fisher, R. A.

1936» Statistical methods for research vorkcrs. London.

Johansen, A. C.

1925. On the diurnal vertical moven3nts of young of some fishes in
Danish waters. Modd. fra komm. Havundersgelser, Sorie;
Fiskcri. Bind VIII, Nr. 2, pp. 1-23. 5 tables, 11 figs.

Leavitt, Benjamin B.

193^. A quantitative study of the vertical distribution of the

large zooplankton in deep water. Biol. "3ull., Vol. LXVIII,

No. 1, pp. 115-130, 5 tables, 3 figs.
1938. The quantitative vertical distribution of nacro-zooplankton

in the Atlantic Ocean basjji. Biol. Bull., Vol. LXXIV, No. 3,

pp. 376-39Uj 3 tables, 5 figs.

Pearson, Karl

1930. Tables for statisticians and bionetrlcians. Part 1. Third
Edition. Cartridge University Press.

Russel, E. 3,

1926. The vortical distribution of the marine macro-plankton,
III. Diurnal observations on the pelagic youn'? of teleos-
tcan fishes in the Plyrnouth area. Jour. Marine Biol. iiss.
N. S. Vol. lU, pp. 3B7-U11;, 13 tables, 8 figs..

1928. The vortical distribution of th.-; marine macro-plankton.

VIII. lAirther observations on' the diurnal behavior of the
pelagic young of tolcostean fishes in the Plymouth area.
Ibid. Vol. 15, pp. 829-850. 10 tables, 5 figs.

1930. The vertical distribution of th^^ r:iarinG raacroplankton.

IX. The distribution of the pelagic young of the teleostcan
fishes in the daytime in the Plymouth area. Ibid. Vol. 16,.
pp. 639-676, 6 tables, 7 figs.

Spilhaus, Athelstan F.

I9U0. A detailed study of the surface layers of the ocean in the

neighborhood of the Gulf Stream with thu aid of rapid mciasur-
ing hydrographic instruments. Sears Found. Jour. Ueir, Res.
Vol. Ill, No. 1, pp. 51-75, 18 figs.

Walford, L. A.

1938. Effect of currents on the survival of the eggs and larvae
of the haddock (Melanogrojimus aoglef inus ) on Georges Bank.
Bull. U. S. Bur. .sis h. Vol. ATjX, No. ^9, pp. 1-73, 5 tables,
50 figs.

200 77171