+; 1
f ' Hah)
+ Wea)
Heel detray
vt Fahd
hat
Hatt
\ i t
i '
i aie at
i he elt ay Was m

' etl iee Manor iia heeds

’ hein

1 ow ibtabebe
Ait | yy ihe Mattei
ei ey Sew ibe elay)
j fp cheba
‘ i tabarade
‘ yiera Wekins
Crheadta gs,
“abe aiieets ?.
Ie

veh

" vy
CT Aaa D rate alals
pa esata sft!

‘aMelal Hie eae?
Hales Lee

‘
Sf
‘
t
, 1
¥
‘
4 ,
~“ ue
1
E ‘ ‘
;
1 Jala
1 sieht gt
j i! F
f H
Hut i
‘ a)

"fA

N
‘ DEPARTMENT OF COMMERCE,

BULLE TIN

OF THE

ONT ED STATES
BUREAU OF FISHERIES

Or. xl
1925

HENRY O'MALLEY
COMMISSIONER

ses) We

WASHINGTON
GOVERNMENT PRINTING OFFICE
1926

SD S1MMOD “10. THAN

aay mes pe seit nt eat

INVES mms VT ALIWa

ay FO

rats Cans 9

YAIIAM'O “4MSH
AAMOAZIMMOD
AOTOVIEAW
; _ SORO OMIT THEM AIVOD

age!

CONTENTS
&

VARIATION IN THE MAXIMUM DEPTH AT WHICH FISH CAN LIVE DURING SUMMER IN A MOD-
ERATELY DEEP LAKE WITH A THERMOCLINE. By Frank Smith. (Document No.
G7 ONissued September 15," LOZ4 1 ee Tn a ee te
BLACK TUMOR OF THE CATFISH. By Raymond C. Osburn. (Document No. 972, issued
VI c ae AAO Di) emi meas VRE ees aon SQ acbls ae ea Bip uo wi a ee
GROWTH AND DEGREE OF MATURITY OF CHINOOK SALMON IN THE OCEAN. By Willis H.
Rich. (Document No. 974, issued March 24, 1925.)__________._______________-
SEASONAL DISTRIBUTION OF THE PLANKTON OF THE Woops Hotz Reaion. By Charles J.
Fish. (Document No. 975, issued April 2, 1925.)_________.-_---_-_-----__---_-
DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES. ANINVESTIGATION OF ENZYMES
IN FISHES, WITH COMPARATIVE STUDIES ON THOSE OF AMPHIBIANS, REPTILES, AND
MAMMALS. By Walter A. Kenyon. (Document No. 977, issued April 2, 1925.) ___
GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM, SILIQUA PATULA (Dixon).
By F. W. Weymouth, H. C. McMillin, and H. B. Holmes. (Document No. 984,
ISAC ANU? PABA 5 IANS) ae ae Sa ep gp a cS Se TC
FISHES OF THE REPUBLIC OF Ex SAaLvapor, CENTRAL AmmrRIcA. By Samuel F. Hilde-
brand. (Document No. 985, issued August 28, 1925.)__.______________________-
FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS, 1897 To 1922. By Leonhard Stejne-
ger. (Document No. 986, issued October 2, 1925.)_____________- Sei oto oa Sane
GENERAL INDEX

Page
1-7

9-13

15=90

91-179

181-200

201-236

237-287

289-332
333-347

ERRATA

Page 33, fourth line of legend inside Figure 6: inside should read outside.

Page 72, line 11: 52.5 per cent should read 47.5 per cent.

Page 123, Figure 25: Milercerttwm campanula should read Melicertum campanula.

Page 130, line 7 under the heading ‘‘ Annulata and Vermes”’: Tomopteros should read Tomopterus.

Page 155, Figure 55: The symbol forPalemonetes vulgaris should be — — —.---— instead of __.e__..
IV

VARIATION IN THE MAXIMUM DEPTH AT WHICH FISH CAN
LIVE DURING SUMMER IN A MODERATELY DEEP LAKE

WITH A THERMOCLINE
&

By FRANK SMITH
Professor of Zoology, University of Illinois

&

Contribution from the University of Michigan Biological Station and the Zoological Laboratory
of the University of Illinois

a

During seyeral summer sessions the writer served on the staff of the Univer-
sity of Michigan Biological Station, which is on the shore of Douglas Lake, about
17 miles south of the Straits of Mackinac. During the last few years of service he
was responsible for instruction work, which included among other things a study
of the habits and distribution of fishes in the lake. This led to a desire for a more
precise knowledge concerning the vertical distribution of fish in places where the
depth is sufficient to permit the establishment of a definite thermocline. During
the latter part of the sessions of 1920 and 1921 attempts were made to determine
the maximum depth at which fish could remain alive for any considerable length of
time in such places and to ascertain what relation such maximum depth might have
to the position of the thermocline and to the correlated oxygen and hydrogen-ion
conditions. These attempts led to results sufficiently definite and interesting to
make publication seem worth while. It became evident that one could predict,
within 2 or 3 feet, what such maximum depth would be if furnished accurate data
on the temperature, amount of dissolved oxygen, and the acidity conditions exist-
ing at levels having 2 or 3 foot intervals in the upper part of the thermocline. From
the temperature data alone one could predict nearly as closely. Variations in the
depth at which the thermocline is established are accompanied by similar varia-
tions in the maximum depth at which fish can remain alive.

Douglas Lake is approximately 4 miles in length, with its long axis extending
in a northwest-southeast direction, and the average width is somewhat less than
half as much. With the exception of about a half dozen small isolated areas, in
which the water is 60 to 90 feet in depth, the water of the lake is too shallow to
have a thermocline established. In the deep places or holes, as they are termed
in station parlance, a thermocline is ordinarily present during July and August,
when the station is in operation. The depth at which the thermocline is established
varies considerably in the different holes and in any given one of them is subject
to some fluctuation. The depth is apt to be greater in the latter part of the season,
and long-continued, strong winds tend to depress the level of the upper part of the
thermocline, if not of the entire stratum. For the observations described in this

103591°—24t (1)

9) BULLETIN OF THE BUREAU OF FISHERIES

paper the two deep places nearest to the biological station were selected. One of them,
which will be designated as Station A, is but a quarter of a mile away, and the other
one, Station B, is about a mile from headquarters. Station A differs from the other
stations in being partially sheltered by a high wooded bluff from the full force of
the frequent, strong, northwest winds, while Station B is in a position to get the
maximum effect of such winds. The thermocline at Station B was ordinarily
several feet deeper than that of Station A.

The general plan of operations included the submergence of a series of wire
cages attached at different levels to a rope which was suspended from a wooden
buoy or float. The float was anchored in the desired situation, and the rope to
which the cages were attached had a weight fastened to its lower end to keep the
rope taut and maintain a constant depth for the cages. The cages were made
of galvanized wire netting of about 14-inch mesh and were cubical in form, with
each surface 1 foot square. Fish of several kinds were placed in each of the cages,
and they were let down to the desired depths, where they remained until the next
visit. Temperature readings and tests for the oxygen content and the hydrogen-ion
condition at approximately the same levels as those at which the cages were placed
made a part of the records. Subsequent visits at varying intervals of time were
made for the purpose of examination of the condition of the caged fish, the removal
of dead ones, and the addition of others. Not infrequently additional chemical and
temperature examinations were made for comparison with the former ones.

For hearty cooperation in the provision of equipment and in the securing of nec-
essary assistance thanks are due Dr. G. R. La Rue, the director of the biological station.
Dr. P. S. Welch generously made the few chemical tests that were necessary the
first season. In 1921 Dr. Minna Jewell, who was a member of the station corps
of assistants, rendered most valuable service in making all of the chemical determi-
nations and records. For this work she was particularly well fitted because of
much previous experience in such chemical studies on this lake and elsewhere.

During the first season the observations were of a preliminary sort for the
purpose of finding owt whether the limiting depth is very definite, and whether
the fish that were obtained in shallow water were suitable for the tests at greater
depths. The results seemed to show that the lower limit of the body of water,
which had the conditions necessary for sustaining the life of fish, was very sharply
defined, and also made it apparent that the fish from shallow water were suitable
for the tests.

The first test was made July 26, 1920, at Station A. Data of Doctor Welch,
taken three days earlier, showed a thermocline between 45 and 53 feet, with plenty
of oxygen at 45 feet. Two specimens of Lepomis gibbosus were placed in each of
three minnow pails at depths of 40, 55, and 65 feet at 4.30 p.m. At 6.45 p. m. of
the same day in each of the pails at the two lower levels one fish was dead and the
other one nearly dead. Those in the pail at 40 feet were in good condition and were
still living 24 hours later.

On July 28 at 10.40 a. m. yearling bullheads and suckers were placed in the
pails at the same depths as before, and at 7 p. m. those in the two lower pails were
dead, while those in the upper one were in good condition, and most of the latter
were still living 24 hours later.

VARIATION IN DEPTH AT WHICH FISH CAN LIVE i

At Station B, August 3 at 10.35 a. m., four wire cages were placed at 40, 49,
55, and 65 foot levels. Each contained some minnows, perch, and a young small-
mouth black bass. The cages were examined at 1.45 p.m., and it was found that
all of the fish in the two upper cages were living while those in the two lower cages
were all dead. Late in the afternoon of the same day, with the aid of Doctor
Welch, records of temperatures were made and water samples taken from the 49
and 55 foot levels. The temperature at 49 feet was 624°; at 55 feet, 51.6°; and at
65 feet, 50°. The oxygen content at 49 feet was 5.28 c. c. per liter and at 55 feet
2.6 c.c. At 49 feet pH was 7.6, and at 55 feet 7.1. These data indicated that
the lowest limit at which fish could live was somewhere in the thermocline.

Tests made August 6 at Station A, in connection with class work, indicated
that fish could not live at as great a depth as at Station B, and also that the thermo-
cline was not as far down. In the afternoon of the following day, with the aid of
Doctor Welch, who made the chemical examination, the following data were secured:

August 7, 1920, Station A

Depth

45 feet 474 feet | 50 feet | 523 to 70 feet
Berm WCrauurery ty een ek oes aos ncca seg ectans ages sons sens tanesepsacnscaesceas | 65.8 to 70.4__ _ - 64 58 51.6 to 46.5
Muy PONY CC PONLGl e226 ie ook cs ga vecenabe doesn sagt eaewsvedencreccndwncs | 5.52 to 7.29____ 4. 33 2. 28 | 0.028 to 0
Weert oe ae Sees oe Ve tS cha ue cad idesacncnebeciasessawencnseonnnaens | WeSitO See enn ee iva ao 7.0 to 6.9

At 4.25 p. m. fish were placed in baskets at 374, 424, 474, and 534-foot levels.
At 8.30 p. m. the fish in the two upper baskets were in good condition; some of
those at 474 feet were all right and some were ailing; all at 534 feet were dead.
During the two following days (August 8 and 9) the tests were continued, with some
shifting of depths of the cages, with the general result that it was shown that fish
lived very well at 45 feet; at 474 feet some would live for a day or more but not
in good condition; and at 50 feet, or below, all soon died. A young sucker and a
young bullhead were left at the 45-foot level until August 17, when they were still
in good condition and were released.

During August, 1921, tests were made at Station A on five different dates and
at Station B on three dates. Two of these latter tests were unsatisfactory because
of the continued high wind and rough water which subjected the buoys to which
the cages were attached to constant motion. This presumably accounted for the
death of most of the fishes in the upper cages, which, under normal conditions,
should have lived. In the tests made in 1921 the cages were placed at 24-foot
intervals. Where slight discrepancies occur between the depths of the cages and
the levels at which temperature and chemical records were made, they are due to
the fact that on checking up the distances on the rope to which the baskets were
attached and by taking for a standard the rope used in taking water samples for
analysis such discrepancies were found to exist. Different ropes of varying texture
and previous treatment often differ in the amount of change in length which they
undergo when subjected to tension under water.

4 BULLETIN OF THE BUREAU OF FISHERIES

The chemical data in the following records were all obtained by Dr. Minna
Jewell. The first of the tests in the summer of 1921 was at Station A on August 5,
when the following data were secured:

Depth

37} feet 40 feet | 424 feet | 45 feet | 47} feet | 50 feet

Temperature, S yeyegs abs S50 sei ee ss OT eet see 70. 6 68 63 59. 5 55. 6 53.6
Oxygen: Cpemliters sont) stele ae ae ee eee eee 4. 24 2735 0.95 0. 09 0 0
pHi Sst RA ie we b> art ee a ae eee 7.8 Ar) 7.3 752 it 7

Fish in cages placed at 38 and 403 feet remained in good condition until the
end of the test, several hours later. All of the fish of each of two different sets,
including perch, minnows, and mud minnows, died within two hours when put in
cages placed at 43 and 454 feet. One week later the thermocline was somewhat
deeper, as shown by the following data:

August 12, 1921, Station A

Depth
40 feet | 42} feet 45 feet | 47} feet 50 feet | 52} feet
‘Temperatures oi. ok a ec ee ge ee ee es > Se 68. 5 68 65. 4 58. 3 54.9 52.6
OxVeenscscnper liter). anes ae arte SDE Ser eee ass 5. 47 5 1:7 0. 96 0 0
DH isha a a ee eS 8.4 8 7.3 7 7 6.9

At 3.08 p. m. about 15 perch (young of the year) were put in each of the cages
and placed at 42, 444, 47, and 494 feet. At 4.23 p. m. all of those in the two
lower cages were dead and all of those in the two upper cages were in good condition
except one that had died in the uppermost cage. Those left in the two upper cages
were in good condition when examined again at 7 p.m. At 7.45 a.m. of the follow-
ing day 9 fish were in good condition at 42 feet and 8 at 444 feet. At 1.45 p. m.
only 4 were living at 42 feet and 1 at 444 feet. A test made August 16 and 17
gave data similar to others and need not be described.

Another depression of the thermocline is shown by the following data:

August 19, 1921, Station A

Depth

45 feet | 474 feet | 50 feet | 52% feet

Tem Perature; Wiese oe eotee = ak eee ee eee en et Sn yen apie (eer eee ee eee ee 66. 2 63 54.9 53. 8
Oxygen? ¢!esper litents = 2 LBs. SE a a ee 3.6 1. 84 . 59 .42
10) © Bake oA ale apiece ele Mtoe eet pe dota a pig cannes ee oy eS PR ae 8.1 7.4 7.0 6.9

The above records were made at 6.30 a. m., and at 8.15 a. m. fish were placed
in three cages at the following depths: In a cage at 45 feet were four perch, a log
perch, and a young smallmouth black bass. These had already been in the cage
at that depth for two days and still seemed in good condition. A cage containing

VARIATION IN DEPTH AT WHICH FISH CAN LIVE 5

a young bullhead, a young smallmouth black bass, and three 2-year-old perch was
placed at 474 feet. A cage with similar contents was placed at 50 feet. At 11.20
a. m. the fish at 45 feet were in good condition. At 474 feet the bass and one
perch were dead but the others seemed to be in good condition. At 50 feet all the
fish were dead. At 3.30 p. m. all of the fish at 474 feet were dead and those at 45
feet were in good condition. At 6.40 p. m. those at 45 feet were still all right, and
in the cage at 474 feet were placed young perch, young Lepomis gibbosus, and
young rock bass. At 7.30 a. m. of the following day all of the fish at 474 feet were
dead, and in the cage at 45 feet a young bass and two perch were still living, but the
others were dead. The fish of that cage had been confined at that depth for three
days. The lowest limit at which fish could live was very definitely between 45 and
474 feet, and in the upper portion of the thermocline.

The last test of the season at Station A showed a still further depression of the
thermocline and a corresponding change in the depth at which fish could live.

August 22, 1921, Station A

Depth

45 feet | 47} feet | 50 feet 55 feet 70 feet

TOI DOLALIITe sy heres Oe emer nin, eee ee ean eee eee ese ee eee 64. 4 64 61 51.4 48.8
On PerCrC MDE MUCracncndcn’ cu fonate fanbecnsccneccabcctscuwsucsenessetesse 4. 84 3.70 DF a ee eee See 0
Dees see re eno coon ae ence ees sas besn sates cecal cecceeseses cle. 8.4 8.0 ES eh eee 6.8

At 1.05 p. m. fish were placed at 45, 474, and 50 feet, as follows: A cage placed
at 45 feet contained 7 young perch, 1 yearling sucker, 2 yearling rock bass, and 2
young bullheads. <A cage placed at 474 feet contained 5 young perch, a half-grown
sucker, a young Lepomis gibbosus, 2 minnows, and 3 young bullheads. A cage
placed at 50 feet contained 2 young perch and 3 young bullheads. The perch used
were mostly 2-year-olds and the remainder were yearlings. At 3.15 p. m. the fish
at 45 feet were all in good condition, and those at 474 feet were also, except the two
minnows, which were ailing. In the cage at 50 feet the perch were dead and the
bullheads nearly dead. At 7.05 p. m. the fish at 45 feet were in good condition and
were transferred to Station B for a test to be described later. At 474 feet the two
minnows were dead and the L. gibbosus was ailing. These were discarded and the
others, still in good condition, were transferred to Station B. The fish at 50 feet
were all dead. The lowest limit at which fish could live at Station A was about
7 feet deeper than on August 5, two and one-half weeks previous.

As already stated, high winds and rough water made two of the three tests at
Station B in 1921 quite unsatisfactory. The fish were necessarily confined in a
relatively small “live box”’ in a rowboat for an hour or more, while journeying to
the station and while temperature and chemical records were being made. During
all of this time they were being jostled about because of the motion of the boat, due
to the rough water. After they were submerged, the cages containing them were
being continually jerked about because of the action of the waves on the float from
which the cages were suspended. Although unsatisfactory, the records are not
without significance.

6 BULLETIN OF THE BUREAU OF FISHERIES

In connection with the first test of the season at Station B we have the follow-
ing data:
August 13, 1921, Station B

Depth

47} feet | 50feet | 524 feet | 55 feet 60 feet

Temperature, CH 2_ jh. Jb 5esh. welch be Loy es ee ae sae SU 66.3 ))4_.-- 2 56. 5 53.9
Oxygen;cVcsiper liter oie as ee a, ee a een oe 4. 69 2. 95 0.77 0.3 0. 06
DHE 255i. eter teers. dei eee et Pay! eee yh vaedr sy, 8.2 7.5 12 7 7

At 3.15 p. m. four or five perch (1 or 2 years old) were put in each of the four
cages, which were placed at the 474, 50, 524, and 55 foot levels. About one hour
later the cages were raised and examined as well as the rough water would permit
without undue rough treatment of the fish. Of those in each of the two upper’
cages about one-half seemed to be in good condition and the others were ailing.
Those of the two lower cages were all dead or dying. The cages were next raised on
the morning of the 15th, when the fish were all found to be dead and more or less
bruised.

The last test at Station B and the last of the entire series was in some ways quite
significant. Fish were placed at 7.30 p. m. on August 22, but the records were
obtained on the following day.

August 23, 1921, Station B

Depth

52h feet | 55 feet | 573 feet | 60 feet | 62} feet | 65 feet

Temperaturengh oS 5 hee EE Re eee eerie ease seve | eat se es 65 64.5 64 6 54.6
Oxyvenscicuperiliter S22 ee ee ers eae eee 4. 87 4. 87 4. 87 4. 65 1, 93 0. 35
DEY De eee FE Fe Re POS I Ren) ee 8.3 8.3 8.2 7.7 .3 7.1

Five bullheads, 1 smallmouth black bass, and 8 young perch were put in a cage,
which was placed at the 524-foot level. The fish that had been in a cage at the 45-
foot level at Station A were placed at 55 feet. The fish and cage that had been at
474 feet at Station A were placed at 574 feet. Several young bullheads of the
season were also included in this cage. At10a.m., after the data in the table above
had been secured, the cages were raised, and it was found that the fish that had been
at 524 feet were in good condition, with the exception of one perch that had died and
was removed. Those that had been at 55 feet were in good condition, except a
perch and a yearling sucker that had died and were removed. Those that had been
at 574 feet were in good condition, except a perch that had died and one that was
ailing. These were removed and several fresh specimens were added to this cage,
including a few perch, a smallmouth black bass, and two mud minnows. Each of
the cages was now placed at a level 5 feet below that which it had previously occu-
pied. The one that had been at 524 feet was now placed at 574 feet and the others
at 60 and 624 feet, respectively. At 2.10 p. m. the cages were again raised, and it
was found that in the one that had been at 57% feet all of the fish were in good con-

VARIATION IN DEPTH AT WHICH FISH CAN LIVE i

dition except two perch. Of those that had been at 60 feet all were in good condition
except one perch that had died and another perch and a rock bass that were ailing.
Of those that had been at 624 feet all were dead except a few bullheads which were
nearly dead.

The fish in the 60-foot cage had been confined for over 24 hours, having been at
the 45-foot level at Station A for 6 hours on the previous afternoon and then trans-
ferred to Station B, where they were at the 55-foot level for over 14 hours and
then at the 60-foot level about 4 hours; and yet the majority were in good condi-
tion, although they had originally been obtained from shallow water. Some of
the fish in the 624-foot cage had had an experience similar to that just described,
and others were fresh specimens put into the cage only about 4 hours before; but
the difference of 24 feet meant the difference between life and death. The data
in the last table furnish the explanation. The fish in the 624-foot cage were in
the thermocline with reduced oxygen and unfavorable hydrogen-ion conditions.

A similar change in conditions 20 feet nearer to the surface was found at Sta-
tion A on August 5, with a corresponding effect on the fish. Fish could live at
404 feet but died at 43 feet, just as in the last test they could live at 60 feet but
died at 624 feet.

The results of these observations very definitely answered one question that
was in mind when the tests were planned. The lowest limit of the body of water
that supplies the conditions necessary for fish to remain alive and in good condi-
tion is in the upper portion of the thermocline. In this upper part of the thermo-
cline there is a marked decline in the amount of available oxygen and a rapid
adverse change in the hydrogen-ion conditions, indicated by the reduction in the
numerical value of the pH index. The expression ‘‘upper portion of the thermo-
cline” is somewhat indefinite, and further tests may show that a statement placing
the lowest limit of the habitable region in the upper quarter or upper third of the
thermocline layer is warranted. No attempt was made to determine which of
the various factors involved may be most potent in fixing the lower limit of the

habitable part of the lake. ;
SUMMARY

The results of a considerable number of tests made in Douglas Lake, Mich.,
show a very definite and close correlation between the lowest limit of the body of
water in which fish could remain alive and the upper portion of the thermocline.
During August, 1921, the position of the upper portion of the thermocline at one
station shifted from a depth of about 40 feet to one of 47 feet, and there was a
corresponding variation in the depth at which fish could live. At another station,
where the depth of the upper portion of the thermocline shifted from about 50
to 60 feet, there was a corresponding change in the depth at which fish could remain
alive. They remained in good condition in cages placed at the depths mentioned,
while those in cages placed at a level 24 feet lower down soon died. In each instance
the fish in the lower cage were in water having decidedly less oxygen and_ less
favorable acidity conditions, and these factors, rather than that of mere depth
alone, determined the lowest limit at which the fish could live. No attempt was
made to discover which of the unfavorable factors was most potent.

5d

Va ie TMs on te

Pan 1. ahd

modi fiyos pM a vgceht otater Hh deat is 12 iad Vib dal) Mees. ; he
pee sa a oe yand Nope # bie i fq3i Mody ‘at bog bot had
sour coittwiehesiiliped wat a Senay heck sow Ussesl (cota

5

eats ‘ rey fyyass fen at wef low pa a | Ay

Ee ROT shy OH veig iff ay earl @ Toh iiiate

L
biaere ish foal pont Be onld\ tae avn voy arsine: (oh
lo dite erort hy hroehic fo ah
* hey 4 ‘i y JA a = ae] ‘ r r *
lis Ags CUTE OL ais» gOS horgpatehe roe er erst rob iia. Han vodt

LOCH ine eon di Der 2) tren p Se

Hi ih soc toet or ST fi BO vay SUNG at eth atin
Rie RE ead ba wl! Ase rSiedl ETB 6 tas ar shite ads cs avi teat it
Le Hep eeoatan: FGhARR ol ov fet geD Ub asthe» ra sha eine 2
VO Terohh bites, sobre bop et iTanys teers bee, (Fore an fey Wb vert sve:
Si ha Toot san oa hak bl) oY Qaaeee 194k OS Stabibauy “ph re
ty eit Dion: tek’ lath otra a = np fs ae

WE 53 EP ek PER ON ne eel ange ROR oti eae
creel! OT 1b See te SO Se Oren Boxy

iu es ' i 1 Ps
yah ey PSUR Pea 1 ALCL fhe oe oe rahe y's aids ji2 4y tel oy

s 7% 7 S ri rt
SG OME ty eee Oe RET TAPE Sel ress ed Ce Cre seh |

$a

{ ion et = -s ‘ ys .
MALT to re TES CRORE UCT SPREE ONT hha deter: Tee oe ied a: SE
: : aes We

's & “pe ‘a ‘ f » Pee A Fate

YAAMMUS

eed. MEAP Sorte eee bee $33

eat VR) + gets asd Band iit Ferd “e abt qeb'a uw ber Tt
TAY seit Diiths det Sip hae (reals pak Saye KOU eK ARSE
Loathis Br GRY pel te sete TAG ip eu old) fea, sepa,
Hk ROL Lay TERT be ae aT Op iad wareanse NHN. ws nee vth ye,
) dt #4 jy yatiicd sts $i4 ee esteatts 3 Frown, ah fh pia ven eed. pe
bai aeoxt aca IN eCaL jf AR des “rl, peta ) Bion: fey Bein
a), CGR Gath irelohaad bee Ay OT wR ala
ie vents tee ddd ia) to luce ee aes eta “eres h epee
avy ey aA arid f gid: f fell ait aby. 48 told Bena
meine) chine Wear ban Higtony | Sielaenee ati, mild. hi:

neo

BLACK TUMOR OF THE CATFISH '
ad

By RAYMOND C. OSBURN
Professor of Zoology, Ohio State University

&

A conspicuous and very unsightly disease of the common bullhead, or horn-
pout (Ameiurus nebulosus LeSueur), appeared a few years ago in a pond at Wau-
quoit, near Falmouth, Mass. The tumorous excrescences produced by the disease
are so striking that it seems almost impossible that no mention of it has been made
in the scientific literature. Ichthyologists and fish-culturists consulted have pro-
fessed their ignorance of the existence of the disease, and examination of the litera-
ture of fish pathology proved futile.

The keen-eyed Thoreau, however, did observe the bullheads of the Concord
River to be affected with this trouble more than 70 years ago. I should have over-
looked this reference to it but for the kindness of Henry W. Henshaw of the United
States Biological Survey, who called my attention to certain statements in Thoreau’s
“Journal.”’ Under date of July 10, 1852, that writer states: ‘One of these large
pouts had a large velvet-black spot, which included the right pectoral fin; a kind
of disease which I have often observed on them.”’ The location was on the Concord
River. Later (July, 1858) Thoreau again wrote in his “Journal”: ‘TI see a pout
this afternoon in the Assobet”’ (a tributary of the Concord) ‘lying on the bottom
near the shore, evidently diseased. * * *. Nearly half the head, from the snout
backward diagonally, is covered with an inky-black kind of leprosy, like a crusta-
ceous lichen.”” There can be no doubt that Thoreau observed the disease which
forms the subject of the present paper, but it seems strange that so long a period
should elapse before such a striking malady should again come to the attention of
a naturalist.

In the summer of 1917, while I was spending a short time at the Marine Bio-
logical Laboratory at Woods Hole, Mass., the late Vinal Edwards, collector for the
Bureau of Fisheries, brought me a specimen of the diseased fish and asked my
opinion of the nature of the growth. I took the opportunity to visit the pond where
the disease occurred and collected some more material for preservation for sec-
tioning and made some photographs. The best I could do at the time was to pro-
nounce the disease some form of a tumor, and this opinion was concurred in, with-
out any further information, by several pathologists to whom material was sub-
mitted.

In the summer of 1919 the Bureau of Fisheries made possible my further study
of this disease, and I spent six weeks at the Woods Hole station. My results,
though not complete in every detail, have been satisfactory.

1A very brief account of this disease has already been published in Bureau of Fisheries Doc. 896, Progress in Biological

Inquiries, 1920, p. 17.
21734°—25f 9

10 BULLETIN OF THE BUREAU OF FISHERIES

So far as known, the disease at present occurs in only one pond, a small natural
lake of some 10 acres, situated on the property of Dr. L. C. Jones, of Falmouth,
Mass. Doctor Jones first observed the tumors and submitted specimens to Mr.
Edwards. I owe him many thanks for assistance in obtaining material and for
various observations.

GROSS DESCRIPTION OF THE TUMOR

In the younger stages the tumors appear as intensely black areas of variable
size and form in the skin of the fish (figs. 1, 2, 83, and 4). With later development
these areas become thick and warty looking, with a very irregular surface, which
sometimes, especially on the lips, may be fimbriated (fig. 4). Though at first the
skin is only slightly thickened (figs. 1 and 7), the tumors may eventually rise 6
or 8 millimeters, or even more, above the level of the surrounding skin (fig. 8).
They are soft to the touch and, in older tumors, the slightest pressure causes the
extrusion of any inky black fluid. This color, as will be seen later, is due to the
presence of enormous numbers of a minute black coccoid bacterium.

Any part of the skin may be affected, but the tumors are more commonly found
on the fins and lips, parts which are especially subject to abrasion and are thus
exposed to direct infection. In the majority of cases the skin on the outside of
the body only is affected, but not infrequently tumors on the lips spread to the
inside of the mouth (fig. 4, tumor on upper jaw). Vinal Edwards told me that he
had seen one specimen in which the tumor had extended across the roof of the
mouth and down the esophagus into the stomach. I observed one case in which
the tumor, originating on the side of the body near the pectoral fin, had spread for-
ward through the branchial chamber, involving part of the gills, and extended across
the floor of the mouth to the lower jaw. In one case a tumor originated on the roof
of the mouth.

Development proceeds so slowly that the rate of growth has not been deter-
mined. The tumors have been observed in all stages of development, from small
spots less than 3 millimeters in diameter to areas involving more than one-third of
the skin of the fish, and from cases in which the skin was scarcely thickened to
those in which the tumor projected 8 or 10 millimeters above the natural surface.
As a rule the larger tumors are more mature and much thickened; but this does not
necessarily follow, for I have seen cases in which a tumor covering several square
inches was scarcely raised above the surface (fig. 1) and others in which the tumor
was as high as its diameter.

In spite of the tendency to spread laterally, there is no evidence of metastasis,
and the deeper tissues beneath the tumor do not seem to be invaded; at least, in
none of the material sectioned is there any evidence of such an invasion below the
lower layer of the dermis, except that in older tumors the connective tissues between
the muscles may be involved to some extent. Even in these cases the disease does
not appear to penetrate to any considerable depth. On the fins the tumor frequently
affects the entire substance, and a tumor originating on one side will penetrate to
the opposite side. Perhaps this is to be expected, since the fins consist almost
entirely of connective tissues.

Buti. U. 8. B. F., 1925. (Doc. 972)

)

stage, scarcely raised above the

Fic. 1.—Extensive infection of a 9-inch fish. The tumors are still in the ‘young’
surface. The areas of dense black seem to indicate four centers of infection.
Fic. 2—‘‘ Old” tumor involving all of the adipose fin and extending downward on both sides. <A slight margin of
younger growth is seen about the base of the older, dense black portion.

Fic. 3.—An “old” tumor involving the caudal peduncle and the base of the caudal fin, extending on both sides. The
highest portion of this tumor is nearly half an inch above the natural dorsal line.

Fic. 4.—Tumors on the lips. That on the lower jaw is ‘‘older’’ and somewhat fimbriated on the surface. That on
the upper jaw involves a portion of the oral surface, The left maxillary barbel shows infection in a very young
stage.

Buuu. U. 8. B. F., 1925. (Doc. 972)

Fic. 5—Microphotograph of a section through an older but as yet unbroken tumor. The epidermis is
still continuous over the surface. The black area indicates the thickening of the dermis—5 millimeters
in thickness at the thickest portion shown. The portion marked X is shown enlarged in Figure 6.

Fic. 6.—Microphotograph under higher power of the portion marked X in Figure 5. The extent of the
dermis is evident, and comparison with the right end of Figure 5 will indicate the amount of thicken-
ing above the normal. The extrusion of a lobe of the tumor into the epidermis is shown a little at the
right of the center. The rounded black object in the epidermis at the left of the center is a section of
another such lobe, and a third portion shows near the left border.

Fic. iia Mero pactoEraDy of section of adipose fin, fully infected, but in an only slightly advanced
condition.

Fic. 8.—A similar section of an adipose fin at a more advanced stage of the disease. The connective tissue
of the fin is entirely involved. The rounded cavities are large cystlike spaces filled with the bacteria,
most of which were lost out in making the slide.

Fic. 9.—A ‘‘young”’ tumor at the base of the caudal fin, experimentally produced by inoculation, as

indicated in the text. Two months after inoculation.

BLACK TUMOR OF THE CATFISH 11

MICROSCOPIC STUDY

Under the lower powers of the microscope the sectioned material of the tumor
appears as a mass of small black bodies, usually more or less rounded, so thickly
distributed among the connective tissue fibers of the skin that they almost com-
pletely obscure everything else. These bodies appear to be cystlike masses of the
cocci, and scattered individual cocci may also be observed everywhere in great
numbers.

In smears made from fresh material the same cystlike bodies are present, along
with scattered cocci, and under the oil immersion the cocci within the cyst may
often be seen in active Brownian movement. It was at first supposed that this
movement might be due to some activity on the part of the organism, but in smears
made from material macerated in strong potassium hydroxide the same movement
was evident. In fresh smears the cysts may be seen to break under slight pressure
of the cover glass, liberating large numbers of the cocci. The cocci are densely
black with an endogenous pigment, and they are so abundant that it is difficult
to examine sectioned material with satisfaction, even when cut to only two or three
microns.

From my observations it appears that the tumor begins its development in
the outer layer of the dermis and later involves all of this layer, expanding it to
many times its original thickness. The dermal blood vessels are enlarged, but I
have seen no evidence of bleeding, even in old tumors. As the disease progresses
small fingerlike projections of the tumor invade the epidermal layer and make their
way to the surface (figs. 6 and 8). Eventually the epidermis sloughs off over the
older portion of the tumor, leaving a ragged surface formed by the dermal connect-
ive tissue fibers and the substance of the tumor. A slight pressure of the tumor
at this stage causes the extrusion of the inky-black fluid containing the cocci,
as indicated above.

STUDIES AND EXPERIMENTS

As some form of fungus had been suggested as the cause of the disease, my
first studies were directed toward this point. Ordinary fish fungus (Saprolegnia
sp.) is not black, but some species of Aspergillum and some other fungi do produce
black pigment. Some fishes were found with fungus on the tumors, but a little
study revealed the fact that this fungus was not black and that it sometimes occurred
elsewhere on the same fish where the skin had been abraded and where there was
no evidence of the black tumor. Moreover, when growing in the tumor, the fungus
was not found in the deeper portion. Careful search was made for any fungus in
sectioned and fresh material and by macerating with strong potassium hydroxide,
which breaks down the animal tissue but leaves fungi intact. Fungus as a cause
of the tumors was thus thrown out of consideration.

The question of an animal organism as the causative factor was also con-
sidered, especially as the Myxosporidia often produce tumorlike growths, but no
evidence of any such could be observed.

The hypertrophy of the pigment layer of the skin, due to stimulation by an
organism or otherwise, at first seemed a possibility. The extension of the black
tumors into unpigmented areas of the mouth, gill chamber, pharynx, and even

12 BULLETIN OF THE BUREAU OF FISHERIES

down into the stomach (according to Edwards), and the origin of the tumor in the
mouth, as observed in one case, did not favor this view. The presence of the black
granules throughout the tumor and the growth of black colonies in culture rendered
the idea untenable.

That the minute black granules observed under the higher powers of the
microscope could be bacteria did not at first seem likely. Bacteria which secrete
any form of pigment intracellularly are rare, and the granules are so minute (aver-
aging only one-third of a micron) that little could be made of them, even under the
oil immersion, though diploid individuals were common enough. The conviction
was finally forced upon me that we have to deal with a bacterial tumor of slow
growth and of a rather benignant type, which, however, bids fair to destroy the
catfish in this pond and which, if permitted to spread, may become a menace over
a larger area.

Cultures and inoculations were made to determine whether the organism could
be grown or transmitted. In this work I was fortunate to have the assistance and
advice of Dr. W. W. Browne, of the College of the City of New York, and Miss
Helen Mitchell, of Yale University, bacteriologists on the station staff for the
summer of 1919.

It was found very difficult to grow the organism in, cultures, but a measure of
success was achieved by combining extract of catfish tissues with agar, according
to a commonly approved procedure of bacteriological study. The growth was so
slow that at the end of two weeks the largest colonies were only 4 or 5 millimeters
in diameter. Contamination of the cultures was thus almost unavoidable, though
a few of the many cultures made were successful to the extent above indicated.
Incubation, at higher temperature did not seem to hasten the growth, and cultures
at air temperature grew about as well. The colonies in culture have the same
intense black color as the tumors, and, on microscopic examination, the cocci
showed the same size and appearance.

Inoculations of uninfected catfish of the same species from another pond were
made by the following methods: (1) By hypodermic injection of the black fluid
from a tumor; (2) by scraping the skin and rubbing in portions of a fresh tumor;
and (3) by grafting portions of tumor under the skin. The catfish so inoculated
were left at Woods Hole until the end of October, two months after inoculation,
when they were preserved in formalin and forwarded tome. Some of these showed
undoubted tumors of small size, indicating that the disease is directly transmissible
(fig. 9). Unfortunately some of the fishes were allowed to become mixed up after I
left the laboratory, so that it is impossible for me to state whether one method was
more successful than another, though there are specimens of all three types that
produced tumors.

The organism causing the tumor appears to be a very minute Coccus, and
seems to have been hitherto undescribed. It is perfectly round; it averages about
one-third of a micron in diameter, ranging from one-fourth to one-half a micron;
and it secretes endogenously a black pigment. Unstained smears show the organ-
ism, because of the pigment, as well as stained preparations. In smears the indi-
vidual cocci are not black when viewed under the oil immersion lens, but are dis-
tinctly dark. In mass they appear absolutely black. Diploid forms, indicating cell

BLACK TUMOR OF THE CATFISH 13

division, may be observed commonly both in living and preserved material. Not
being a bacteriologist, the writer will not attempt to describe or name the species.

As to the fatality of the disease, there is no definite information. Doctor
Jones informs me that he has seen no dead fish, but the nature of the pond is such
that they would not readily be observed, for the shores are brushy to the water’s
edge and there are no beaches. Moreover, there is much vegetation in the pond,
' especially near the shores, which would help to make the discovery of floating dead
specimens difficult. Certainly the disease is not rapidly fatal, for fish with well-
developed tumors have been kept under observation, in aquaria in the laboratory
for six weeks without showing any changes except for some slight growth in the
area and thickness of the tumors. It would thus appear that there is no active
systemic poison or toxin liberated, at least at the time of the year during which
these studies were made. Fish with large tumors were taken with hook and line,
even those with large tumors on, the jaws accepting the bait readily. However, it
was noted that old tumors, from which the epidermis had been lost, were frequently
infected secondarily with Saprolegnia, and there seems little doubt that this fungus,
which only enters abraded surfaces, would soon cause the death of the fish.

Doctor Jones believes that the disease has spread very rapidly since the sum-
mer of 1916, when its presence was first observed. In 1917 about half of the fish
taken had tumors of various sizes and degrees of development. In the summer of
1919 it was a difficult matter to find any uninfected catfish in this pond. No doubt
the habits of the bullhead in schooling together make the transmission of the dis-
ease an, easy matter. It was not noted on other species of fish in this pond during
my work there nor on the same species of catfish in neighboring ponds. That this
disease is capable of attacking the bullheads in other ponds is evident from my
inoculation experiments. Under date of February 2, 1921, Doctor Jones wrote me
as follows:

A man who has been fishing through the ice on this pond tells me that he caught a pickerel
the other day which was covered with the same dark colored spots observed on the hornpout.
Unfortunately he had disposed of the specimen, but so far as I know this is the only instance
where the disease has been observed on any fish except the hornpouts.

_ As the evidence in this case is not complete, it may be well to suspend judg-
ment as to whether the disease is capable of attacking other species than the bullhead.

In the course of my work no measures satisfactory for the control of the dis-
ease have been suggested. Salt and potassium permanganate solutions, both of
which are useful in the control of Saprolegnia, were tried on individual specimens,
but without any noticeable beneficial effect.’ Of course, it should hardly require
mentioning that no species of fish from this pond, whether showing such infection
or not, should be planted elsewhere.

ad

A
What

‘ a il hie ee

yee

i eae
. i ainugynn oe

GROWTH AND DEGREE OF MATURITY OF CHINOOK
SALMON IN THE OCEAN

&

By WILLIS H. RICH, Ph. D.

Assistant in charge of scientific inquiry, United States Bureau of Fisheries

Sad
CONTENTS
Pago
Ht ROGUIC LIONS ayes kes tee. ett ee ede es ol is Se ee ee ee eee 15
Wie th OCs tarp eat se Syncs 2 peewee. ergo Foe So ee ER ee 18
Determination acess = es See ee Se eee ee ee ee ee ee 18
Wetermination on relative wmaturityere- == ]22 eee - oe oe eee eee eee eee 20
Correlation between size of eggs and size of fish_______--------------------------- 27
Fish taken in the Columbia River and in the ocean off the mouth of the river____-_-_---- 28
Relative maturity of chinook salmon taken in the ocean____-_--------------------- 29
Ac ezorimmature tight. oe Sin Ne oo os ee Loe a eee eee oa 39
Percentages of immature fish at different times___=___...-_---------------------- 40
Percentages of fish more than one year from maturity__-----__------------------- 42
ADUNGdANCeL ON Une VALIOUS AGC PTOUPS == S52 ass en eae = ee ee ee eee 43
(CTR ON ih jac ae cas tae an rest a Ps SO Pie Nok ONO, nea A em rs an ee 48
Hishetakenune Monterey Bay. 42.2214 22 2k? So de tlt Se oes eee ese 63
S17 Caneel Uren Sy PON hip mere WOURe Re yh Ieee ees Ae Pee Lee RS 63
Percentages of the, various age proups...-.-..-.--.-.......--.--.---4----#4---=-- 64
Relativemaaturit yee 2 = eee ae sey oe Se ee ee ee eae 66
HishyrromWrakesibay: anGilorb Brages 2 so85 252 222 ee ee ee 67
Sua a aya ame COM CIUBION Sa este ee ee ee ee ee ee 69
Supplementanvatables@sc he ot) 3 bet 2 eh 8 a ek See ee oe 75
IBS TOLL O Saray layer ee ener aye et ate Ma SNe ey pace ee ee ee Meee eee ee 89
INTRODUCTION

The rapid increase in the amount of trolling and purse seining for salmon which
has taken place along the Pacific coast during the past few years has directed atten-
tion toward this phase of the salmon industry. The general consensus of opinion of
those interested in conservation is to the effect that these methods of taking salmon
are, under certain circumstances at least, destructive and undesirable. The fishing
on Swiftsure Bank, off the coast of Washington, was reported on a number of years
ago by Dr. C. H. Gilbert (1915), who found that the silver salmon (cohoes) taken
by purse seines on the bank, “especially in the first part of the season, are far from
having attained their full growth, although maturity is but a few months distant.”
Of the spring salmon (the chinook) he makes the following statement:

The spring salmon is taken in large numbers and furnishes a somewhat inferior product, with
soft flesh, little oil, and poor color. Several thousand young of this species are captured during
the season, 2-year-olds about a foot long with white, soft flesh—a total waste. The numbers of

these are relatively small, as the great majority of the salmon on the bank are in their last season,

but the loss is nevertheless serious and deplorable.
1G}

16 BULLETIN OF THE BUREAU OF FISHERIES

In 1920 Smith and Kincaid reported on the immature salmon taken off the mouth
of the Columbia River, Grays Harbor, and Neah Bay during 1918. Later in the
same year (1920) a more complete report by Smith gave the results of his investiga-
tions during 1918 and 1919. In both of these reports Smith gives the results of
chemical analyses, which show that the product resulting from the canning of the
immature fish is decidedly inferior to the standard grades in fat and protein content.
He also presents figures that show something of the extent of the operations of both
the purse seine and the trolling fleets, and concludes that “the taking of immature
salmon in the Puget Sound and on the banks along the coasts of Oregon, Washington,
and Vancouver Island is responsible for a great loss in one of the important food
products of the region.”

The “outside fishing,” as it is commonly termed, has been increased by the
addition of both purse seines and trolling boats, but the exact amount of this increase .
in the different regions where such fishing is carried on is difficult to determine.
The purse seiners frequently move from one location to another, fishing wherever
the greatest returns can be obtained. The trollers not infrequently (at least in
Oregon and Washington) fail to take out licenses. Much of the trolling is carried on
outside the 3-mile limit, and some of the trollers argue that it is not necessary for
them to take out licenses to fish except in territorial waters. Furthermore, one
fisherman often takes out licenses for more than one kind of gear—as for trolling
and gill netting, or trolling and trapping—and trolls only a small part of the season.
These factors make it practically impossible to determine with any accuracy the
amount of increase in this outside fishing, but there can be no doubt that it has de-
veloped within the past few years into a very important factor and represents an
increased drain on the resources of the fishery which may easily result in serious
depletion. The number of trollers operating off the mouth of the Columbia River
during the three seasons, 1919 to 1921, inclusive, was estimated at between two and
three thousand, and the Columbia River District Troller’s Union alone contained
some 1,500 members. At the same time there were probably not less than 40
purse seines operating in the same region. This is in strong contrast to the con-
dition in 1914, when there were only a few dozen trollers and possibly three or four
purse seines operating off the mouth of the river. ;

One of the chief objections that has been raised to this ‘outside fishing” has
been that it was destructive of immature salmon. The most casual observation
shows that a considerable percentage of the fish taken outside are relatively imma-
ture. They average decidedly smaller than the fish taken inside, are feeding
heavily, and observation of the gonads discloses the fact that these organs are
relatively undeveloped in many of the individuals. The taking of.fish one or
more years before they will become mature reduces the population that will form
the basis of the fishery one or more years later, at which time fishing operations of
dangerous intensity will be brought to bear upon the reduced numbers. Such an
increase in the intensity of fishing that will affect the abundance of salmon previous
to maturity, when they are comparatively small and of poor quality, and acting
on a resource already showing evidence of depletion, is unquestionably a dangerous
development.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 17

In a preliminary report dealing with the subject of this paper (Rich, 1921a)
the writer made the following statement:

The determination of the age by means of scale studies will not alone give a sufficient index
to the degree of immaturity since there is such a wide range in the age at which these fish reach
the spawning stage—from 2 to 6 years. If the percentage of individuals of different ages among
the mature fish were constant, it would be possible, from a determination of the percentage of
fish of different ages taken by troll and purse seine in the ocean, to estimate the percentage of fish
of different degrees of maturity. This, however, is not the case. ‘The percentage of fish of the
various age groups varies greatly at different times among the mature fish and also among those
taken in the ocean. Presumably these variations, at least among the mature fish, are due quite
largely to racial differences, but our present knowledge of the various races of chinook salmon is
far too limited to aid in segregating the races from mixed lots. Even if our knowledge of the races
were complete it might well be that they could not be identified and segregated accurately and
fully enough to serve this purpose. It is apparent that some other means than the determination
of the age is necessary in order to learn the percentages of mature and immature fish taken in
the ocean and their relative maturity.

It has been found, as will be subsequently shown in detail, that the size of the
eggs gives a fairly accurate means for determining how soon the females will mature.
During a field trip to Monterey, in 1915, it was noted that there was considerable
variation in the size of the ova, and a series of egg samples was collected. It was
thought that from a study of the eggs something might be learned of the percentage
of mature and immature fish taken by troll in Monterey Bay. These eggs were
collected from the cleaning tables and no attempt was made to collect data and
scales from the fish from which the eggs were taken. Nothing was done with this
collection until the summer of 1918, when the matter was again brought up for
consideration. On examining the samples, the fact became apparent that several
reasonably distinct sizes of eggs were to be found, which presumably represented
different degrees of maturity. (See Table 23, p. 86.) In order to secure more
complete data, another collection was made at Monterey in the summer of 1918,
and in this case measurements and scales were secured along with the egg samples.
The study of this collection confirmed the earlier observations as to the presence
of fairly distinct size groups of eggs, and the examination of the scales showed
a high degree of correlation between age and the size of the eggs. (Table 25, p. 87.)
Extensive collections have since been made from the fish taken by troll off the mouth
of the Columbia River, and it has been found that, while the determination of
maturity is not as simple a matter as it had first appeared, the general features
will hold good.

It is obvious that this method is applicable only to the females and no method
has yet been devised for accurately determining the relative maturity of the males.
It was thought at first that the percentage of males of various degrees of maturity
could be calculated with reasonable accuracy from the percentages of males and
females found among the mature fish in the stream if once the percentage of mature
and immature females taken outside was known. If it were not for disturbing
factors this could easily be determined, for each separate age group, by means of the
following proportion: Percentage of males taken inside : percentage of females
taken inside :: percentage of males taken outside which are mature : percentage of
females taken outside which are mature. It has been found, however, that the

18 BULLETIN OF THE BUREAU OF FISHERIES

proportions of males and females found among the mature fish taken inside the
river at different times during the season is so variable as to make a calculation of
this nature very unsatisfactory on the basis of the data at present available. No
attempt has been made, therefore, to do this. The males undoubtedly show a de-
cided tendency to mature younger than the females, and for this reason the total
percentage of immature fish taken outside is somewhat less than the percentage of
immature females. Just how much allowance should be made for this factor it is
impossible to say, but in any event it would not alter materially the general con-
clusions arrived at from a study of the females alone. It may be mentioned that,
from the standpoint of conservation, a knowledge of the effect of a fishery upon
the supply of females is of much more importance than a similar knowledge of
the effect upon the supply of males.

In the preliminary report mentioned above an outline of the method used in
determining the relative maturity was given, together with a brief summary of the
more important results that had been obtained at that time. The following report
embodies the results of more recent studies and gives in detail the data on which
the conclusions are based.

The writer wishes to acknowledge his indebtedness to Dr. C. H. Gilbert, of
Stanford University, whose advice has been constantly available throughout the
preparation of this report. Many helpful suggestions have been obtained from Dr.
F. W. Weymouth, of the physiology department at Stanford University. Some
of the earlier work was done at the Hopkins Marine Station at Pacific Grove,
Calif., but most of it was done in the laboratories of the zoology department of
Stanford University.

METHODS

DETERMINATION OF AGE

Although a detailed study of the life history and scale growth of the chinook
salmon still remains to be made, the main features are sufficiently known to per-
mit of a reasonably accurate determination of age by the usual examination of the
scales.

The determination of the age of fish by means of a microscopic examination
of their scales has been used so extensively during the past two decades that a
detailed description of the method seems unnecessary. It depends upon the fact
that the rate of growth of the fish varies materially at different times of the year.
During the spring and summer, in general, growth is more rapid, and during the
fall and winter it is slower. The scales of many fish, including the salmon and
trout, bear series of concentrically arranged ridges on their outer surface. On
account of their concentric arrangement these are known as rings or circuli. The
scales increase in size with the growth of the fish, and circuli are added at the mar-
gins. The scales are never normally shed, but increase in diameter by these accre-
tions to their margins, and in thickness by additions to the inner surfaces. The
markings formed by the circuli are therefore persistent throughout the life of the
fish. The arrangement of these circuli is characteristically modified by variations
in the rate of growth of the fish. During the more rapid growth of the spring
and summer the rings are spaced relatively widely, but during the period of slow

GROWTH AND MATURITY OF SALMON IN THE OCEAN 19

or of perhaps no growth, which obtains during the fall and winter, the rings are
crowded together closely and are frequently more or less broken and imperfect.
The complete year’s growth, therefore, consists in a ‘‘summer”’ band of relatively
wide rings followed by a “winter” band of narrow rings. By counting the num-
ber of “‘summer”’ or ‘‘ winter” bands the age can readily. be determined.

Weymouth (1923) has given a brief though admirable discussion of the various
methods of determining age. In answering certain criticisms that have been made
of the method of age determination of fishes by means of scales, he makes the
following pertinent observations:

These objections are not valid, however, in the case of a number of species of fish, where the
annual nature of the marks rests on no assumption of any kind, but on direct observation. In
some a study of the scales throughout the year has clearly shown that the ring is formed during
the winter and only once each year. In others it has been shown that the number of rings agrees
with the known age of fish kept in captivity or of marked fish recaptured after known periods.
The soundness of these conclusions is not affected by the fact that in certain species the rings
are less distinct and hence in these cases may be an unreliable guide to age, nor that incompetent
or hasty workers may have reached incorrect conclusions in any species.

_ Although this method of determining the age of fishes appears to be simple
and of easy application, in practice serious difficulties are frequently encountered.
The relative approximation of the rings is merely a reflection of the rate of growth
prevailing at the time the rings were formed. Not infrequently factors cause
variations in the rate of growth which are unassociated with the seasonal varia-
tions responsible for the formation of ‘‘summer” and “winter” bands. As a
rule, the cause of these incidental variations is unknown, although the present
writer has suggested causes for some of the minor variations observed in the scale
growth of chinook salmon previous to or during their seaward migration (Rich,
1920). Snyder (1922 and 1923) has also discussed the causes responsible for
unusual checks on the scales of chinook salmon in the Klamath River. In the
case of the chinook salmon the determination of age is further complicated by the
fact that the young fish, at least in the Columbia River, may migrate seaward at
any time after they emerge from the gravel of the spawning beds up to an age of
18 months or more. There are, therefore, many difficult and puzzling characters
in the scale growth which must be worked out before a completely satisfactory
analysis can be made, but a full consideration of the problems connected with age
determination in the chinook salmon lies outside the scope of the present paper.
The age determinations will, therefore, be presented without further discussion of
the details. It may be remarked that the author, in the course of an intensive study
of the scales of several thousands of chinook salmon from the Columbia River,
has found nothing that conflicts with the main conclusions reached by Gilbert
in his initial paper dealing with scale studies (Gilbert, 1913). In all cases where
the interpretation of the scales was in doubt the individuals have been omitted
from the tables. This procedure may have eliminated some few small categories,
but these would not materially affect the main conclusions that have been reached.

The lack of data bearing on the types of scale growth presented by the fish
of the various coastal streams and of the various tributaries of the Columbia has
made the study of the scales of the fish taken in the ocean an extremely difficult

20 BULLETIN OF THE BUREAU OF FISHERIES

matter. Numerous more or less distinct types of scale growth could be described,
but the description alone would have little bearing on the subject matter of this
report, in view of the fact that it is at present impossible to assign the various
types to particular streams or tributaries. It has been necessary, therefore, to
omit a detailed segregation of individuals on the basis of racial differences and to
segregate only into the main age groups. With the exception of age, the only
other feature of the life history which has been used as a basis of segregation has
been the general type of the early growth. Gilbert has shown in the report above
cited (1913) that the chinooks of the Columbia River exhibit two distinct types of
“nuclear growth” (the growth of the first year), forming what he terms the “ocean
type” and the ‘‘stream type” of nuclei. The ocean type of nucleus is large, with
rings that are relatively widely separated (figs. 27 to 33), and indicates that the fish
migrated to the ocean early in its first year. The stream type of nucleus, on the
other hand, is small, of closely crowded and more delicate rings (figs. 34 to 40),
and characterizes the scales of those fish that have spent the entire first year in
fresh water. An age group, in the sense in which the term is employed in this paper,
comprises all of the individuals in a collection that are of the same age and that
have the same general type of nuclear growth. Thus there are fish with ocean
nuclei from 2 to 6 years old, and also fish with stream nuclei from 2 to 6 years old—
a total of 10 age groups in all. Illustrations of typical scales of each group will be
found in Figures 27 to 42.

DETERMINATION OF RELATIVE MATURITY

Mention has already been made of the fact that the determination of the rela-
tive maturity has been based upon variation in the size of the eggs.

The egg samples, as collected in the field, consisted in a small portion of the
ovary. When the eggs were relatively small and the ovary but half an inch or so in
diameter, a section of the ovary about 1 inch long was taken for a sample. When
the eggs were larger, as in the nearly mature fish, a section of the ovary approxi-
mating 1 cubic inch in volume was taken. These egg samples were tagged with
serially numbered tin tags and the number of the egg sample added to the other
data recorded in the book in which the scales were preserved. The samples were
preserved in 10 per cent formalin.

The size of the eggs has been determined by measuring 10 of each sample
and taking the average. The larger eggs—those over 1 mm. in diameter—were
measured in a simple device, which consists essentially of a small trough, V-shaped
in cross section and with closed ends, which is graduated in millimeters. In use
this is partially filled with water, the eggs are placed in a row in the bottom of
the trough, and then are carefully pushed up to the zero end of the scale by means
of a small piece that fits the bottom of the trough and on which is graduated a
vernier, enabling one to read accurately to tenths of a millimeter. The reading
is made when the first egg is in contact with the zero end of the trough and the eggs
are all just in contact with one another. If too great pressure is applied, one or
more of the eggs will be pushed above the others, so that the error in procedure
is readily detected. When this happens it is necessary to push back the vernier,

GROWTH AND MATURITY OF SALMON IN THE OCEAN 91

readjust the eggs, and repeat the operation. The measurement of 10 eggs by this
scale gives, by simply moving the decimal point one place to the left, the average
size of the eggs to hundredths of a millimeter, a degree of accuracy which is un-
necessary in the great majority of cases. In preparing eggs for this measurement
it is necessary to free them very carefully from the ovarian membranes, so as not
to break the delicate egg membrane and yet to clear them of all shreds of tissue
which might tend to affect the measurement. The smaller eggs—those less than
1 mm. in diameter—were measured by means of a microscope fitted with an.eye-
piece micrometer, carefully calibrated against a stage micrometer. In using this
method it was necessary, of course, to measure the 10 eggs separately, and then the
average of these measurements was found.

If one examines the eggs of chinook salmon taken in fresh water during their
spawning migration, it is found that the eggs of different individuals vary only
slightly in size. A similar examination of the eggs of females taken in the ocean
discloses that there are wide variations in size. A group with eggs approximately
the same size as those of the fish taken at the same time of year in the river can
readily be selected. In addition to this group, however, many of the females
have eggs distinctly smaller than any found among the spawning run in the river,
and it is possible many times to separate these smaller eggs into two or more groups,
even without careful measurement. When this observation was first made it seemed
probable that each size group of eggs indicated a different degree of maturity,
so that an analysis of the relative maturity of the fish in a given catch could be
accomplished merely from an examination of a series of the ovaries. The assump-
tion was that the fish with the largest eggs would mature and spawn during the
year in which they were captured, that those with eggs falling within the next
smaller group would mature during the next year, and so on. It has been found,
on closer analysis, that while the size of the eggs alone forms a very satisfactory
diagnostic character for distinguishing between fish that will mature during the year
in which they are captured and those that will not mature for at least one more
year, it can not be depended upon to distinguish between those that will mature
in one year from those that will mature in two or more years. It has been found
that the eggs grow in proportion to the rest of the fish until, approximately, the
beginning of the growing season, which is destined to end in the maturing and
spawning of the individual. With the onset of this last growing period, however,
the rate of growth of the eggs is relatively accelerated and a differential growth
sets in, so that the eggs gradually increase in size relative to the size of the fish.
As a result, the eggs of the maturing fish are relatively larger than those of the
immature specimens. This fact makes it possible to distinguish between those
individuals that are maturing and those that will not mature for at least one year.

The observed differences in the size of the eggs of the immature fish taken in
the ocean are due only to corresponding differences in the size of the fish. As will
be shown later, there is a high degree of correlation between the size of the eggs
and the size of the fish. The observation that the eggs may be grouped on the basis
of size is dependent upon the fact that the differences in size between the younger
age groups of chinook salmon, as in many other animals, is often so marked that

29 BULLETIN OF THE BUREAU OF FISHERIES

the fish themselves fall into fairly distinct size groups corresponding to different
ages, and the size groups of eggs is merely a reflection of this condition.

The acceleration of the growth of the eggs during the last growing season is
shown clearly in Figure 1. This graph shows the proportional changes in the size
of the eggs of fish with ocean nuclei during their third year—first, for those mature
fish taken inside the river; second, for the mature fish taken outside; and third,
for the immature fish taken outside. The points on this graph indicate the position
of the weighted arithmetic means of the logarithms of the egg sizes.

—S LEGEND —
Ciawe — Jake insiae,

is t ---0---Skiftre - sehen atwae
3 % ——-~.—- Inmate - taker olsige.
F. % 6
oN Y
ot}
bes XY,
NaN
Wk
aN)
0% s
NX
yi.
SSS.
Ss
SiN:

Qg vo <0 Jo 40 JO 60 70 oo 9 700 Ho 720, 630 fQ KO
D4r of SEASON.
STAY SPINE W774 Avcus7 SEPTEVPTEL.

Fia. 1.—Changes in the sizes of the eggs of females in their third year, ocean nuclei, during
the course of the fishing season. Columbia River and the ocean off the mouth of the
Columbia

In addition to the fact that the eggs of the immature fish are distinctly smaller,
it is apparent from Figure 1 that the slope of the curve for the mature fish is much
steeper than that for the immature fish, indicating unmistakably that the eggs of
the maturing fish are increasing in size, not only actually, but relatively, at the
more rapid rate. If the two curves were projected back into the months preceding
the opening of the fishing season, they would evidently meet some time in March
or early in April, which would indicate that the differential growth probably begins
at about that time.

This differential growth of the eggs of the maturing fish is shown in another
way in the following table, which gives the proportional size of the eggs in relation
to the length of the fish in a few selected groups chosen largely because they were
better represented than the others.

1 See pages 24 and 25 for an explanation of the use of logarithms in this connection.

GROWTH AND MATURITY OF SALMON IN THE OCEAN DH}

TasBLe 1.—Relation between diameter of eggs and size of fish (Columbia River chinook salmon)

Geometric
Average Average mean of

Number = Fi A
~ | lengthin | logarithm | diameters DX100
Age group Date taken peels centimeters} ofegg | of eggs (D) L
(ZL) diameter in milli-
meters !

18 48.11 T. 9588 0. 909 1. 890
11 50. 40 1. 9480 . 889 1. 763
In third year, stream nuclei, immature_ 5 Be 4 . Hb . ne 1, 909
5 f 2.017
4 66. 00 1150 1. 303 1. 975
100 61. 00 1043 1. 272 2. 085
32 61. 61 0981 1. 254 2. 035
In third year, ocean nuclei, immature_- 2 an 3 en i bi 1. 930
. . 390 1.977
ue 70. 72 1500 _ 1.412 1. 998
18 79. 56 3174 2.075 2. 610
10 86. 40 5100 3. 230 3. 740
In fourth year, ocean nuclei, mature__- ia be oe ae 3.415 4. 055
4. 370 4. 890
63 93. 50 7243 5. 300 5. 670
13 96.15 4054 2. 542 2. 645
In fifth year, ocean nuclei, mature_-_-_- a = ae th ; is $ Hie
19 100. 16 7553 5. 695 5. 680

1 The geometric mean of a series of measures is the number corresponding to the arithmetic mean of the logarithms of the
original measures. Since the data on egg sizes has been handled in the logarithmic form (see pp. 24 and 25), it is convenient to
use here the geometric rather than the arithmetic mean.

Dx 100
L
centage of the length of the fish represented by the diameter of the eggs. The
actual percentages are so small that it is more convenient to handle the values in

this way.

The table shows that among the immature fish there is little variation in the
relative size of the eggs as compared with the size of the fish. The ratio is prac-
tically the same in immature fish of the different age groups, and within a single
age group there are no significant changes in the ratio during the season. This
signifies that the size of the eggs and their growth is closely proportional to the
size and growth of the fish. In the case of the mature groups, however, a constant and
marked increase in the ratio evo ioe is clearly shown as the season advances—
an increase which is indicative of a distinct differential growth, resulting in the
rapid increase in the relative size of the eggs as maturity approaches.

In May there is comparatively little difference in the relative size of the eggs
of the mature fish and those of the immature fish. The value of the ratio for
immature fish is very close to 2.0, and in May the ratio for mature fish is only
about 2.6. During the season, however, the value of the ratio in the case of the
maturing fish steadily increases until it is nearly three times as large as that for the
immature fish. Evidently the differential growth of the eggs, which takes place
during the last year, has not progressed far by May. Scale examinations show
that the new growth of the year, if apparent at all, is at this time just beginning
to show as a distinct band of wider rings at the margins of the scales. (See figs. 38,

The ratio, , given in the last column of this table, is 10 times the per-

94 BULLETIN OF THE BUREAU OF FISHERIES

39, and 40.) It seems altogether probable that the differential growth of the eggs is
begun simultaneously with the onset of the growth period of the last year.

Another disturbing factor, which undoubtedly makes the accurate determina-
tion of the relative maturity from the size of the eggs more difficult, is the fact
that the various collections contain a mixture of races, which, as has already been
mentioned, have not been segregated, and it is known that there is considerable
variation among different races in the size of the fully mature eggs.

In determining the relative maturity it has been found necessary to compare
the size of the eggs of the fish taken in the ocean with the size of the eggs of the
undoubtedly mature fish taken in the river at about the same time. Moreover, it
has been found necessary to consider the entire distribution of egg sizes as ex-
hibited in each collection in determining whether any particular group of fish was
mature or immature. The tables have been carefully scrutinized and the size of
the eggs in the particular group under consideration compared with the size of
the eggs of the other fish taken in the same collection. If necessary, comparison
has also been made with the size of the eggs of the undoubtedly mature fish taken
in the river. In making these comparisons, not only the mean, but the range of
the distribution as well, has been kept in mind. If a frequency distribution of
egg sizes was found to be unimodal, the position of the mode and the extent of the
dispersion were considered in relation to the position of the mode and the disper-
sion of other groups—especially those known to be mature. If the given distri-
bution were bimodal or multimodal, the position of each of the modes and the
distribution of individuals about these modes has been considered separately and
compared with the mode and dispersion of other groups. It has proved impracti-
cable to publish all of the detailed tables, but representative examples will be
found on pages 81 to 88. The interpretations of the degrees of maturity have all
been made through a study of such detailed tables.

In this study the logarithms of the observed egg diameters instead of the
actual measurements have been tabulated and plotted. This has been done as a
simple mechanical means of reducing all measurements to a strictly proportional
basis. As has already been stated, it was thought during the early stages of the in-
vestigation that the egg size would indicate whether the immature fish were des-
tined to mature during the year following their capture or not for two or more
years. With this in mind it seemed essential to emphasize not the actual but the
proportional variations, since a difference that would be relatively insignificant in
the case of the larger eggs might be decidedly significant in the case of smaller
eggs. For example, a difference of 0.5 mm. in eggs averaging 7 mm. in diameter
would be of no great significance, while with eggs averaging only 0.7 mm. the
same difference might be of decided importance. Therefore the logarithms of all
measurements were found and the tables prepared from them. Although the final
results of the study do not entirely justify the use of this method, the labor in-
volved in retabulating the data, recalculating the constants, and redrawing the
graphs is so great that it has not been attempted, particularly in view of the fact
that such modification would in no way affect the conclusions that have been
drawn. If it be remembered that this use of logarithms is only a means for show-

GROWTH AND MATURITY OF SALMON IN THE OCEAN 25

ing percentage variation rather than actual variation in the size of the eggs, any
confusion will be avoided.?

The tables and text figures have been arranged with class intervals of 0.02 in
the logarithm of the diameter of the eggs. This is the same as saying that the
mid-value of each class has been made 4.713 per cent greater than the mid-value
of the next preceding class.

If the variations in the sizes of the eggs are such that the relative maturity of
the fish can be determined from their study, several consequences will naturally
follow, which may be used as criteria in determining the validity of the method.
First, it must be possible to separate the fish taken in the ocean into at least two

7ihers wt wheel of Werrendile. Depa
AY E192

Taken by Fell a? the revit of tha
Cotrlis Rrrer

Y4ely LE, HUB

3s

NMUTBER or = SNEIVILALS,

7p eae 2, CO Ae 2S 2, “a? ao 02
Locnerwre ee DIAMETER of EGGS.

Acruaa DIAPIETER of £665 mm.
Ka 425 65 20 25 Jo IS 4.0 x7) 62 7O AQ

Fic. 2.—Distribution of egg sizes in two typical collections, one containing only mature fish
and the other both mature and immature specimens

groups on the basis of the size of the eggs—one group corresponding approximately
to the undoubtedly mature fish taken in the stream, and the other group char-
acterized by distinctly smaller eggs and composed of fish that would not mature
during the year in which they were taken. A comparison of almost any of the
tables on pages 81 to 88, which show the variations in the size of the eggs of fish
taken in the ocean, with a similar table of fish taken within the river at about the
same time, will show that this is the case. One such example is shown in Figure
2, which shows the distribution of egg sizes in a collection made just outside the
Columbia River, July 28, 1919, and also the similar distribution in a collection
made inside, at Warrendale, Oreg., on July 16, 1919.

Figure 2 shows clearly that the fish taken outside the mouth of the river are
sharply separated into two main groups—one with larger eggs, which closely agree

‘For the benefit of those unfamiliar with the use of logarithms it may be stated that the logarithm of 1 is 0, that the loga-
rithms of all values lying between 1 and 10 are less than 1 and greater than 0, and that the logarithms of all valueslying between
1 and 0.1 lie between 0 and —1. ‘The logarithms of values intermediate between 1 and 10 appear, therefore, as simple decimal
fractions. Those of values intermediate between 0.1 and 1 are customarily shown as a decimal fraction preceded by the figure
1 over which is placed the minus sign, thus: 1. The logarithm of 0.9 is therefore written 1.9542 and the logarithm of 1.1 is written
0.0414. It is also customary to abbreviate the phrase ‘‘logarithm of” to ‘‘log’’; thus, in this paper ‘‘log D’’ has frequently been
used to indicate ‘‘the logarithm of the diameter.”’

26 BULLETIN OF THE BUREAU OF FISHERIES

in size with the fish taken inside the river at Warrendale and which are undoubt-
edly mature. There can be no question that the fish with the smaller eggs would
not have matured during the year in which they were caught. The details of the
distribution of egg sizes among the various age groups in these two collections
can be seen by referring to Tables 17 and 18, on page 81.

Secondly, if it be true that the variations in egg size form a valid criterion of
the degree of maturity, it should be found that among the fish taken in the ocean
some of the age groups will contain both individuals with large eggs and others
with small eggs. Those with large eggs are destined to mature during the year in
which they were caught and those with small eggs would not have matured for at
least one more year. This condition will inevitably result from the fact that the
fish do not all mature at the same age. Considering the fish of a single age group,

Taken af Porierey, Cate
Sune (9-21 79/8.

NevaB eve 08 LYUVIAALS.
8

O. : BES FO Q40 iO.
LOCARITHIT of LIAPIETER of ZEG6S.

ACTUAL LHAPILETER of L£6GS ri.
40 1.25 x ZS AO. Si 20 %F 40

Fic. 3.—Distribution of egg sizes among females in their third year, ocean nuclei, taken at
Monterey, June 19 to 21, 1918

say three years with ocean nuclei, which may be found together in the ocean, some
of them will mature during the year as 3-year fish while others will not mature
for another year, as 4-year fish. If the size of the eggs forms such a criterion as
is claimed, it should be possible to segregate these two categories on the basis of
egg sizes. It has been found possible to do this more or less clearly in a number
of the collections that have been studied. For example, in Table 18 (p. 81) it
may be seen that individuals with large eggs (maturing) and those with smaller
eggs (immature) are to be found among the fish in their third year with scales having
nuclei of the ocean type. The same thing is true of the fish in their fourth year
with scales having nuclei of the stream type, given in the same tabie. Similarly,
in the collection at Monterey (June 19 to 21, 1918) the 3-year fish with ocean
nuclei are well separated into two groups on the basis of egg size (Table 25, p. 87).
The frequency distribution in this last-mentioned collection is shown in Figure 3.

GROWTH AND MATURITY OF SALMON IN THE OCEAN PAE

CORRELATION BETWEEN SIZE OF EGGS AND SIZE OF FISH

A careful examination of the data has been made in order to determine to what
extent the size of the eggs is correlated with the size of the fish, and, further, the
extent to which this factor may tend to produce the observed differences in egg
sizes.

A very definite positive correlation is found to exist between the size of the eggs
and the size of the fish when fish of the same age group and the same degree of
maturity are considered. ‘The coefficient of correlation (7) has been computed for
two typical distributions—one of mature and the other of immature fish. Table 2
shows the correlation between the size of fish and the size of eggs in 66 females in
their fourth year, ocean nuclei, which were taken by troll off the mouth of the Colum-
bia River, August 13, 1919. There are obviously two distinct groups represented
in this tabulation, which are readily distinguishable on the basis of egg size. One
group of 63 individuals, in which the logarithm of the diameter of the eggs is greater
than 0.55, and another containing only three individuals, in which the logarithm
is less then 0.35.’ The group with the larger eggs is mature and the coefficient of
correlation between size of eggs and length of fish was found to be 0.6210+0.0531,
a high positive correlation.

TaBLe 2.—Chinook salmon taken by troll off the mouth of the Columbia River, August 13, 1919

Females in their fourth year and which migrated as fry ole wan ocean nuclei) tabulated to show correlation between length
and size of eggs

Centimeter length (mid-value of class)
Logarithms of diameter of eggs Total
(mid-value of class)
77 79 | 81 83 85 | 87 89 | 91 93 95 | 97 99 | 101 | 103 | 105 | 107

2

l

2

2

1

3

if

6

2

2

8

4

(f

IDL

6

2

66

Note.—In this and ali other tables of this report an asterisk (*) has been used to mark a break in the continuity of the
table. It indicates that one or more classes have been omitted from the natural sequence just preceding the class marked.
A double asterisk (**) is used to mark such breaks as are of particular significance.

The coefficient of correlation between size of eggs and size of fish has likewise
been determined for a group of 36 immature females in their third year, ocean
nuclei, taken in Monterey Bay, June 19 to 21, 1918. The data showing the dis-
tribution according to size of eggs and length of fish of all the females of this age group
contained in the collections are given in Table 3. Two distinct groups, which are
well separated on the basis of egg size but overlap almost completely in length,
are also shown in this tabulation. The coefficient of correlation for the immature
group is 0.7680+ 0.0461.

28 BULLETIN OF THE BUREAU OF FISHERIES

TABLE 3.—Chinook salmon taken by troll in Monterey Bay, Calif., June 19 to 21, 1918

Females in third year, and which migrated as fry (scales wan ocean nuclei), tabulated to show correlation between size of fish
and size of eggs

Centimeter length (mid-value of class)
Logarithm of diameter of Total
eggs (mid-value of class) Oe
57 | 59 | 61 | 63 | 65 | 67 | 69 | 71 | 73 | 75 | 77 | 79 | 81 | 83 | 85 | 87 | 89
Ry) Reese es ese) Pe) Nad a mth Pee ergo ae 1
249 | Se) EE Pl RO | op es |e es Naea eS 5
Seely sek eel ee Se QU oe soe Se SR 2 [Le 3
Le PRS a COPE ees eg | ee | ae ee
ieee! oule ae Len. i Group 1 (386 indi-
5 a eet Peeve biota Mewes oon Ni cial eeato a 5 viduals).
pe Sele ol: YAO Wea see, [ac aoe 3
a oe eae 1 1
eee ele 1
E_o|2 gee|LES [Sooo wee |ooe aloe al pA ae ee 3
pl See he pee ee eee See CREE Ly SSS 2
pee |e wosl|bsee[eeeele ply ane Fa aps Sean Oe A 1
Eee PEEP ELT EN Ad ee iy eae 4|Group 2 (15 indi-
mas a ee See ee pl | eel aged fl 1f viduals).
lee 24/8 3S | aed eae ee by SESE bee 2
Bee Zeu|Ssealo. cen ee [ond ee te 1
1 Ye 2 259) 4 5 2 8 6 [3 5 3 |. 51
Gromp lose = oi ccee eee ue ee 7 i sa nV 0 a0 VU ae 3 Bal i Be Lae ps 36
GIOUD pees n= sees eee Beja eell cee Peeeeey eee ge A 1 2al2 COUP Sa} Ne 15

These examples indicate clearly that within a single category (that is, among
females of a similar degree of maturity and of the same age group) the size of the
eggs is dependent to a considerable extent upon the size of the fish. The regression
coefficients have been calculated for the two groups for which the coefficients of
correlation were determined. From these it was found that in the first group
(mature fish taken off the Columbia River in August) the logarithm of the diameter
of the eggs was, on the average, increased by 0.005850 for each increase of 1 cm.
in the length of the fish. In the second group (immature fish taken at Monterey
in June) an increase of 1 cm. in the length of the fish was accompanied, on the
average, by an increase of 0.005535 in the logarithm of the diameter of the eggs.

It may readily be seen, however, from an examination of Tables 2 and 3, that
the groups distinguished as mature and immature are sharply and widely separated
on. the basis of egg sizes but overlap almost completely in so far as length is con-
cerned. Within each group there is the distinct tendency for the larger fish to have
larger eggs, but this is quite inadequate to account for the greater difference observed
between the mature and immature groups. These can only be explained as the
result of the differential growth of the eggs during the last year.

FISH TAKEN IN THE COLUMBIA RIVER AND IN THE OCEAN OFF THE
MOUTH OF THE RIVER

In discussing the various features of the life history, which will be treated in
this paper, the data on the fish from the Columbia River, from near Fort Bragg
and Point Reyes on the northern coast of California, and from Monterey Bay will
be handled separately. Each age group will also be given separate consideration.
The oldest fish will be treated first, since the results of their study will aid in inter-
preting the data from the younger age groups.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 29

RELATIVE MATURITY OF CHINOOK SALMON TAKEN IN THE OCEAN

It has been shown above that the determination of the degree of maturity is
accomplished primarily by comparing the distribution of the egg sizes of a group
of fish with the similar distribution of fish known to be mature. It has been pointed
out that the size of the eggs changes during the season; that there are differences
in the size of eggs of fish of different ages, due to the difference in the size of the fish;
and also there are racial differences in the size of the eggs. It is obvious that these
factors will make it impossible to draw a sharp line of distinction, which will hold
good throughout the season, between the size of the eggs of mature and immature
fish, and that the determination of the relative maturity of the fish taken in the
ocean must depend upon a comparison of the distribution of egg sizes in the various
age groups with the distribution of egg sizes of unquestionably mature fish taken
in the river at about the same time of the year. This has been done in the deter-
mination of relative maturity given below.

FISH IN THEIR SIXTH YEAR, OCEAN NUCLEI

Only two females of this age group were taken in the collection studied. One
was taken on July 28, 1919, and the other September 18 or 19, 1919. Both were
undoubtedly mature. The logarithm of the diameter of the eggs (log D) was 0.69
in the first specimen and 0.77 in the second. No fish of this age group appear in
the collections taken inside the river near these dates, but four specimens are found
in a collection made June 24 and 25, 1919, and three in one made July 3. The
average log D in these collections was, respectively, 0.55 and 0.657. The maximum
age attained by the chinooks of the Columbia River is 6 years, and comparatively
few fish of this age are found. It is, therefore, quite to be expected that any fish
of this age group which might be taken in the ocean would be maturing.

FISH IN THEIR FIFTH YEAR, OCEAN NUCLEI

As has just been indicated, very few fish with ocean nuclei older than 5 years
are found in the Columbia Riverrun. It was, therefore, to be expected that nearly,
if not quite all, of the fish of this age group which are taken outside would be matur-
ing, and such was found to be the case. No immature fish of this age group were
found. Figure 4 shows the average log D for the collections of fish taken outside
and also for those taken inside throughout the season. The figure shows clearly
that there are only negligible differences in the sizes of the eggs of the fish taken in
the two localities.

FISH IN THEIR FOURTH YEAR, OCEAN NUCLEI

The determination of the relative maturity of the fish of this age group has
proved more difficult and the results are less satisfactory than with any other cate-
gory. The number of fish belonging to this age group, which were taken in the
ocean and which were with certainty immature, is small. There must, however,
be a considerable proportion of the 4-year fish with ocean nuclei that do not mature
during their fourth year, since there is a rather high percentage of fish in their

3210°—25¢

2

80 BULLETIN OF THE BUREAU OF FISHERIES

fifth year, with ocean nuclei, found both in the ocean and within the stream. No
satisfactory explanation of this can be given, although the possibility of some sort
of a selective migration naturally suggests itself. Figure 5 shows the distribution
of egg sizes for this age group during the season, and it may readily be seen that,
with the exception of the groups marked by a double circle, the range of egg sizes
shown by the fish taken outside is quite comparable with the range shown by those
taken inside the river. The groups indicated by a double circle are those that

AcTruaL DIAPIETER of £GG6S -7-70m.
£06 0 LVALTIETER ow LE6G6S.

°© Soker js.
© Jekerr ovukyde.

(] Gagle soecimer.

o Zk SOn = 7S 5 700 TET 750
DAY of SGEASOV
PTAY SOME Vad AYLGOST _,\ SEPTEAOEE

Fic. 4.—Sizes of eggs of females in their fifth year, ocean nuclei, from the Columbia River

and the ocean off the mouth of the river. In this and other similar illustrations the average

size of the eggs is indicated by the circles or dots; the vertical lines extending from these

points show the limits of twice the standard deviation (+ 2c). Ninety-five per cent of the

cases will fall within the upper and lower limits indicated by these lines. The standard

deviation has, with a few exceptions, been calculated only for those cases in which at least

20 individuals of the particular age group in question were contained in the collection
may be considered as immature, although the interpretation, especially in the case
of those collections made during May, is somewhat doubtful.

A correlation table has been prepared (Table 4), giving the sizes of the eggs
found in the fish of this age group contained in a collection of troll fish taken off
the mouth of the Columbia River, May 17 and 18, 1920. Reference to this table
will show that the fish are quite distinctly separated into two groups on the basis
of egg sizes. The log D of one group is greater than 0.30, averaging 0.4254, and of
the other group is less, averaging 0.2486. The modes of the two distributions are
distinct, and it seems probable that two categories are represented—one of mature
and the other of immature fish. This interpretation is further indicated by the

GROWTH AND MATURITY OF SALMON IN THE OCEAN 81

fact, apparent from Figure 5, that the size of the smaller eggs agrees with the size
of eggs in other later collections, in which there is no question as to the correct inter-
pretation. Reference to Figure 9 shows also that these small eggs are approximately
the same size as those of the immature fish taken at the same time, which were in
their fourth year, stream nuclei—a group fairly comparable with the one under

ee eI
© SYitiure ~ Foken preside.
0 Stalure —Takera ouside
© Lamaire Takes oGrde.
C] Sxl GEecie/?.

a en

|

i}

|

!

;

F H

N i

6 |

Ns !

Ny | |

Sy

QF
aed
oN
SS |
ay
2
XY 4

DAY of GEASON.

PTAY SUNE SLY AUGUST. __\SEPTEPIBER

Fic. 5.—Sizes of eggs of females in their fourth year, ocean nuclei. Columbia River and the
ocean off the mouth of the Columbia

discussion. The distribution of egg sizes in the case of these 4-year fish with stream

nuclei is similar to that of the 4-year fish with ocean nuclei in that it does not seem

sharply to separate the two groups of mature and immature fish. The factor

Dx 100
L

, described above and illustrated in Table 1, has been calculated for that

82 BULLETIN OF THE BUREAU OF FISHERIES

group of 4-year fish with ocean nuclei that had the smaller eggs, and was found to
be 2.24.3 This is somewhat larger than is characteristic of immature fish (about
2.0), but it is nearer this figure than that which is characteristic of mature fish taken
in May (2.6).

Taste 4.—Chinook salmon taken by troll off the mouth of the Columbia River, May 17 and 18, 1920

[Females in their fourth year, ocean nuclei, tabulated to show correlation between length and size of eggs]

Centimeter length (mid-value of class)

Logarithms‘of'diameter of |. = ee eee
eggs (mid-value of class) Total

1

4

5

3

1

.33 1
30 2
3 5
2 2
; 1
WAG Bee es ee otc Rs Ne | ee opr Is ces | | 2 ee | ee eel er 1 Lewes |e ee ee eee 2
46245-22225 Boe cones eeee | SSS Eee Eee eS Be L DA es es Dea Dea Oe Eee & Doe ee eS 3
ATi oc wees oe SERS eS ee See tone ee eon eee Jess oe | sone oem | econ | eee 2 lecceul |e AS ak all ee 2
AOE So es SB Ss Sc See ce ee ee ee eae eee roee 2G) Seren beeen pees Page SE bo ee 1
Dom ee tne ie. Se ee cal: Meee | eeeenet | O eeemoee 1 yi ea Pips ane | eae Pagmipamseen| Vales fa ey eae) bet 1 1 Fae ae 3
Potala es foo Lk eee i 2 (}y| Peep 9 4 4 4 5 i eras FS 4 2 Toseee 36

The relatively slight differences noted between the size of the eggs of the mature
and those of the immature fish in this collection are probably associated with the
fact that the collection was made in 1920, while most of the collections that have
been made the basis for this study were made in 1919. It is not improbable that
the differential growth of the eggs, which takes place during the last year (and
which has been shown to be mainly responsible for the observed difference in the
sizes of the eggs of mature and immature fish), may begin at different times in
different years. It should also be noted that the fish taken in the ocean during
1920, which were available for study, were selected, very few being under 8 pounds
in weight. Laws passed by the Washington and Oregon Legislatures prevented
the sale of immature salmon under this weight, so that few of the smaller fish found
their way to the canneries. This factor alone, by tending toward a selection of
gear that would take the larger fish, might well be responsible for the difficulty en-
countered in determining the relative maturity of the fish found in this collection.

In most of the other collections no particular difficulty has been encountered
in determining the relative maturity, but the percentage of immature fish is, as
has already been mentioned, much lower than would naturally be expected. Figure
30 shows a scale from one of the immature fish of this age group taken on August 13,
1919, and Figure 31 a scale from one of the mature fish taken at the same time.

FISH IN THEIR THIRD YEAR, OCEAN NUCLEI

The distribution of egg sizes during the season for the fish that are in their
third year, with ocean nuclei, is shown in Figure 6. The difficulties encountered in

3 Average log D=0.2485. Geometrical mean of the diameters=1.772 mm. Average length=79 cm. 17.22.24.
d

GROWTH AND MATURITY OF SALMON IN THE OCEAN DO

the interpretation of the data in the case of the 4-year fish with ocean nuclei are
entirely lacking here. Two distinct distributions are clearly shown. In one the
eges are small, the average log D being in every case less than 0.20, and the increase
in size observed as the season advances is very slight. It has already been shown

LEGEND —
®° Sigitre- Joker mk.
© Kendiure-Sehen zs.
0 SGftre - Fike ovf5r2e.
© Lermciire- kerr vrsce.

ee ye eee)

°

Cl Uae ower.

ACTUAL LVALTETEX of L66S - rr.
£06 of DISPUTE of ZGGS.

A2S

------+4@-------
1}

Om Tea ie ns, 0 25 50
LAY of SEASON

CZAY. ANE en e AYGUST ,\FEPZEPIEEL

Fic. 6.—Sizes of eggs of females in their third year, ocean nuclei. Columbia River and the
ocean off the mouth of the Columbia

in Table 1 that the factor ee for fish of this age group contained in several

of the larger and typical collections, is close to 2, which is characteristic of other
groups of undoubtedly immature fish.

84 BULLETIN OF THE BUREAU OF FISHERIES

In the other distribution the eggs are distinctly larger, the log D ranging from
about 0.34 at the beginning of the season to about 0.80 by the end of the season.
This increase in size during the season is conspicuous and is in marked contrast to
the condition observed in the group of immature fish. So far as the available data
show, the segregation of the fish of this age group into mature and immature groups
may be accomplished with very satisfactory accuracy by means of a study of the
egg sizes. The percentage of mature and immature fish taken at different times
during the season is treated on page 40 ff. Figure 28 shows a scale from one of the
immature fish of this age group, which was taken on July 2, and Figure 29 one of
the mature fish taken on the same date.

It will be noticed in the collection made inside the river on July 7, 1919, that
one immature fish of this age group was taken. It will presently be shown that there
were other immature fish taken at the same time which belonged to other age
groups—2-year fish with ocean nuclei and 3-year fish with stream nuclei. The
explanation of this occurrence and other associated facts is interesting in its possible
bearing on the habits of the chinook salmon.

These immature fish observed inside the river were taken by seine on Sand
Island, a low, level, sandy island several miles in length and about 1 mile in width,
which is situated in the estuary of the Columbia River, only a mile or so inside the
mouth of the river. The seining grounds are toward the upper end of the island,
approximately 5 miles from the river’s mouth. On the Washington side of Sand
Island is Baker Bay, a broad, shallow stretch of water in which are situated most of
the salmon traps in use on the Columbia River—about 200 in number. Under
ordinary circumstances the water of this lower part of the Columbia estuary is fresh
or only slightly brackish, depending upon the tides and the flow of theriver. At the
time of the “spring” tides, however, especially when the river is at a low stage,
the lower part of the estuary, up for a distance of several miles, becomes very brack-
ish, approximating the salinity of pure sea water. On such occasions it frequently
happens that small, immature fish are taken, both in the seines operating on Sand
Island and in the traps in Baker Bay. It is generally believed among the fishermen
of the lower Columbia River, and some statistical evidence is available which would
indicate that the belief is well founded, that the salmon enter the river in greater
numbers on the spring than on the neap tides. Apparently the immature fish
to some extent join those that are maturing in a general movement into the river
on these high tides, but do not pass beyond the upper limit of the approximately
pure sea water. It is at these times that immature fish are taken on Sand Island
and in Baker Bay. Presumably the immature fish return to the ocean again as the
water freshens with the receding of the tide, while the mature individuals remain,
in large measure, to continue their spawning migration.

In this connection it is interesting to note that the stomachs of the immature
fish thus taken inside the river are frequently filled with food, usually in an advanced
stage of digestion. This is in contrast to the condition in the mature fish, the stom-
achs of which have never been observed to contain any food.‘

4 The writer has described (Rich, 1921) an instance, apparently unique, in which the stomachs of mature salmon taken in
fresh water contained food. In the case described chinook salmon were found feeding on eulachon ( Thaleichthys pacificus) in the
Cowlitz River, a tributary of the Columbia, some 70 miles above the ocean.

GROWTH AND MATURITY OF SALMON IN THE OCEAN , 35

FISH IN THEIR SECOND YEAR, OCEAN NUCLEI

All of the females of this age group taken in the ocean have been found to be
immature. This was to be expected, since it is known that no mature females of
this age are found among the spawning runs within the river. Figure 7 shows the
distribution of egg sizes during the season. Comparison with Figure 6, which shows
the size of eggs of the 3-year fish with ocean nuclei, will show that the eggs are dis-
tinctly smaller in the case of the 2-year-old fish. This would be expected from the
fact, previously demonstrated, that the size of the eggs is correlated to a high degree
with the size of the fish.

The question arises as to the relative maturity of these 2-year-old fish as compared
to those that are 3 years old. It has been shown above that the method used for

.
S
10 wy e%
w !
Ni
§ H ©
y = i © ©
fom’) a Peat
i) K i]
NaN
R i |
sel y fe
S Neal 1
x
MANS]! TLLECENO—
4 4 1
x 8 1 | 1 © Lardlire - Jhere toasike
as & Nip ' - 0 Lormraiure- Jokere aise.
Fi i
7 “hai ae a aL "75 S”:C«OO n/n SS,

LAY of GEASON
TAY SOVE Say AvCUusS7T POT LP IESE

Fic. 7.—Sizes of eggs of females in their second year, ocean nuclei. Columbia River and the
ocean off the mouth of the Columbia

determining relative maturity is inadequate to differentiate between degrees of
immaturity, although it is, in most cases, an entirely adequate and reliable means
for distinguishing between fish that will mature during the year in which they are
taken and those that will not mature for at least one more year. Although it is
impossible to establish the fact with certainty, the probabilities are all in favor of
considering the 2-year fish as, on the whole, one year further from maturity than
the 3-year fish taken at the same time. Certain obvious possibilities for error are
inherent in such a treatment, especially when comparatively few data are available,
as in the present case, but it is felt that the picture presented by this means will be
much nearer the true state of affairs than can be arrived at by any other method.
This allowance will, therefore, be made in the later discussion of the percentages
of mature and immature fish taken at different times of the year.

86 BULLETIN OF THE BUREAU OF FISHERIES

FISH IN THEIR SIXTH YEAR, STREAM NUCLEI

Only three specimens of this age group were found among the troll fish included
in this study. Two of these were taken May 8 to 10, 1919, and one August 13 to
17,1918. In the discussion of the 6-year-old fish with ocean nuclei, it was mentioned
that 6 years is the maximum age of chinook salmon in the Columbia River, and it
follows that fish of this age could only be mature. The size of the eggs indicated

1
co)
i)
1
!
1
i)
1

“LEGEND ~

~

© Mature - taken msde °

© Mature - raken ovtside

© Immature - taken inside

© /mmature - taken outside
‘L1 Sigle specimen

q

g

ACTUAL LVAITETER of LG6G6S --mrm7.

N

206 of DIAPIETER of LEGS.

AS

: ST SIEET a HsOln 75 70 ayEEe 2)
DAY of SGEASOW.
VLAN SINE SULy AUCGOST. _;SEPTEPIBELE

Fic. 8.—Sizes of eggs of females in their fifth year, stream nuclei. Columbia River and the
ocean off the mouth of the Columbia}
beyond question that such was the case. The eggs of the two specimens taken in
May had an average log D of 0.58, and the log D of the specimen taken in August
was 0.75. The log D of the fish of this age group taken inside the river, of which
there were 7 specimens, ranged from 0.51 to 0.75.

FISH IN THEIR FIFTH YEAR, STREAM NUCLEI

Figure 8 shows the distribution of egg sizes for fish of this age group during the
season. Only three immature fish were found. Two of these were taken in the

GROWTH AND MATURITY OF SALMON IN THE OCEAN 3837

lower part of the Columbia River estuary, May 17 and 18, 1920, one in a trap in
Baker Bay, and another by gill net. The other immature fish was taken by troll,
June 4, 1919, The figure shows how distinctly these immature specimens are
separated from the maturing fish on the basis of egg size. There can be no doubt
of their immaturity. Figure 38 shows a scale of one of the immature fish taken
inside the river in May, and Figure 39 one from a mature fish of the same age
group taken at the same time.

A

LECEND—
© SYa/ure - taker? wmside .
0 Mature - faker ovksite.
© Larmaiure - there on5;Ze.

q

ACTUAL LYAPETER of L665 - rm.
£06 0o« DYAVIETER of L6GS.

[] We seecimen.

a

“oO Fae HRS) ee) aVOOR ET 125 750

LAY of OLEASOW.
PTAV SUNE SLY AUGUST PTEPIBEL.

Fic. 9.—Sizes of eggs of females in their fourth year, stream nuclei. Columbia River and the
: ocean off the mouth of the Columbia

FISH IN THEIR FOURTH YEAR, STREAM NUCLEI

The distribution of egg sizes found among the females of this age group are
shown in Figure 9. Two distinct groups are apparent, in one of which the average
log D is less than 0.30. In the other group, the log D ranges from 0.40 or 0.50 in
the early part of the season and to about 0.80 at the end of the season. The first
group is composed of immature individuals and the second of those that are matur-

88 BULLETIN OF THE BUREAU OF FISHERIES

ing. Figures 36 and 37 show scales of immature and mature fish, respectively,
taken by troll on July 28, 1919.

FISH IN THEIR THIRD YEAR, STREAM NUCLEI

The fish of this age group resemble the 2-year-old fish with ocean nuclei in
that mature females have never been reported. None of these fish taken outside
was mature, as is apparent from Figure 10. The question of the percentages of
these fish that will mature during the following year as 4-year fish, and during the
second subsequent year as 5-year fish, arises here as in the case of the 2-year fish
with ocean nuclei. As in the case previously considered, it seems probable that
these 3-year fish with stream nuclei are, on the whole, one year further from maturity

LS

Rae

bs
12s Qe.

NS a

Sy

ny

NIN

AN | “Le6£ND

SN

S 87. © Lamettre -tohen cetide
wR 8

x N [] Grgle seecimen,

DAVKS of SEASON.

STAY SINE SLY Acsusr EP TIERE

Fic. 10.—Sizes of eggs of females in their third year, stream nuclei. Columbia River and the
ocean off the mouth of the Columbia

than those in their fourth year. Two-thirds of the 4-year fish with stream nuclei
were mature, so that it appears probable that approximately two-thirds of these
3-year fish with stream nuclei will mature during the following year and one-third
during the second year following. A very few will not mature until they are in
their sixth year.

FISH IN THEIR SECOND YEAR, STREAM NUCLEI

No females of this age group were found in any of the collections, although a very
few males were discovered. One was taken inside and three outside. Among these
three, however, two were contained in a special collection, made in 1914, of fish
selected for their small size—the “grilse.”” Mature males of this age group are
extremely rare in the Columbia River and apparently are seldom taken by troll.
In so far as their maturity is concerned, it seems fair to assume that they will mature
one year later than the 3-year fish with stream nuclei and two years later than
the 4-year fish.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 39

In the preceding discussion of the various age groups, no attempt has been
made to show the percentage of mature and immature fish in each, nor the percent-
age which the immature fish formed of the total number in the different collections.
These data have been compiled in separate tables for ease in making comparisons.
Table 5 shows, for each collection of fish taken inside the river, the percentages of
the total number of females which are composed of the mature and immature
fish of the different age groups. Table 6 gives similar data for the collections of
fish caught in the ocean. A discussion of the composition of the entire collections
as regards age groups will be given later, but attention is first directed to a discus-
sion of the age groups most commonly represented among the immature fish.

AGE OF IMMATURE FISH

Reference to Table 5 shows, as would be expected, that very few immature
fish are taken inside the river. It has already been explained (page 34) that a few
immature fish are occasionally taken in the lower part of the Columbia River
estuary. Thecollections of May 17 and 18, July 3, and July 7 were all made in this
part of the river and are the only ones to contain immature fish. Such immature
fish as were taken belong to the 2 and 3 year groups with ocean nuclei and to 3 and
5 year fish with stream nuclei.

Taste 5.—Percentages of females of different age groups and different degrees of maturity taken in
the river at different times during the season !

Ocean nuclei in— Stream nuclei in—
Second . z Fourth| Fifth | Sixth | Third {Fourth | ‘ 7 Sixth
Date year Third year year | year | year | year | year Fifth year year
I I M M M M. I M I M M
|
8. 3
| 2. 5. O
CLUE) T1CG} F250 iby fe Si et) ee emo 2.1 ZOS Sal ee eae, ee ee Bhid ie eee eae rh Oy Gt Bee ees
MUNC MG ONG 200 2. oo coc eeccanc|nacacan-|eteaacas|ace-caas 40.8 yb Ea W485 |2-escces 180. | fess see 1.2 3.7
PURE hs 7 ele Ap a en am ee ie eel (oe 2.0 47.0 33. 3 5.9 2.0 (toy eee papa 20K esaeeee
ulygpeeee eee oo 2 22 ot chee 12.2 2.4 2.4 48.8 CG) eee 9.8 254) || See PA hy Paes
UV ALG ree en tee ee eid see 2) ee 4.3 41.3 PS Dae Pape al | Sie |zoeeeose 4.3 2.2
J liltys 20 beeen une oe ea Bak wee (Ee ee esos 34.8 G5.2t| Samana | Bee eee |e = | oe ees | ae eee
Pa Sena ee et case eta |seaswese)asceceudlecomeass 80. 0 2OAOU Se Nice Serer eee sea reer eel eee eee
PAG Gn, We ane eo cree alee c dows fone nccaclesceanas 55. 3 O4e2 ||Sescesce|beeescas o0! | saaeeces 2iOnleaoneen
NI oe ects ea es ee hs ea 19.5 61.1 1 RO | eee ene cee VERY seme 2.4 2.4
SOD tel 2oeea ere ee ae are ee eee oe tte eee 16.7 57.2 LS Hots eee [eee yh et eee eee (fe! 2.4

1 The initials I. and M. stand for immature and mature, respectively.

Among the fish taken in the ocean (Table 6), it is seen that by far the greater
part of the immature fish are in their third year, with ocean nuclei. Fish in their
second year, ocean nuclei, are the next most important component. Fish in their
third year, stream nuclei, come next in importance. In addition to these age
groups, which include the greater part of the immature fish taken, a few immature

40 BULLETIN OF THE BUREAU OF FISHERIES

fish in their fourth year with ocean nuclei, fourth year with stream nuclei, and fifth
year with stream nuclei, are taken.

Taste 6.—Percentages of females of different age groups and different degrees of maturity taken in
the ocean at different times during the season !

Ocean nuclei in— Stream nuclei in—
Second Bes 7 aie | Fifth | Sixth | Third |. : Sixth
Date year Third year | Fourth year year | year | year |# ourth year] Fifth year year
I I M I M M M. I I M I M M

Mayes ito Ol 22a: 22. seen ses 11.'5' | 60:65); 0. 2)|tes se 10.9 026) sees L059 i 25 eee! 1:2 | eee 1.8 1.2
MiaWal8 pl G20 be tae ees we olla se 2070) °7.. O13 20.5 L252 i) Cena arte e Lider Patera see 5; Gi Soe eee
ia ods eit Se ee Ae 25.0 | 43.7 GS Sag] cas el ee See 6230. soos San Ree ee
SUNG Ae 2 Se os AE a eee 26:47) 47..0i|2522 22 WO |ooneee 1 Olieeossse= 1628, 2 02 0u|e sees Lb Aer Sse eee
June W022 Cy PFS oo ae 33: 8 ||) 83:3) |sasene|- see ~ TR | pes Be 2 Sere 7 Bap We eee cn | eee | Heer OF || ad Sh
VUNG Dba e oo eae cece ae eee Lae 19:35 M6; 51) S625) Seen * 32.3 Fe Bi A Feet eat 12 On eee LOS) Lea 3.2 122. ee
Junei2552 23 hese eek ee Co eerie De | ar a 28464) cele ee ee ee 14) 3 5|ee ceo LASS) RS See oe eee ee
July; 2) 222 FoF oe ae ot ba ee 14511) 26:34) £5315|)) 1.104) 2272 LIGIG esses ee COO) RU [Frye ly pss BECO) ieee eee
Uuly 282 ee oe ee Sake Asiz2: | eroleleen sos 50. 0 18. 4 10 Tan 105) 4220. je ater 120) eae
7-8 0-23] GO ae RR Pb 6.3] 9.5] 3.2] 66.4 Ge eae aune LO") 1.70%" 162 3: [Ses aa ei at eee
Aur Sto lytie 2 Se fo cee 4.9} 3.9 | 20.6 1.0 | 18.6 ISG.) oot ee 2 Pa ee 2. 94a ce see 26. 5 1.6:
Sept. Sand 92). eee eee 16::7,|685an| ete. |S too ee eee 8.3 16.7, || --- .23|4---.2|-So2e5 eee eee

1 The initials I. and M. stand for immature and mature, respectively.

PERCENTAGES OF IMMATURE FISH AT DIFFERENT TIMES

The composition of the various collections as regards mature and immature
fish is best shown in Tables 7 and 8, which give the percentages of mature
and immature fish found in each collection taken inside and outside, respectively.
In Table 8 is also given the estimated percentage of fish that would not mature for
two or more years. This has been derived by assuming that in any one collection
the 2-year fish with either ocean or stream nuclei would, the next year, show a
similar percentage of mature and immature fish as did the 3-year fish with the same
type of nucleus at the time the collection was made. As has been mentioned above,
this assumption is not necessarily justified, but it is believed that the general results
of the application of such an assumption will approximate more closely the true
state of affairs than could otherwise be reached.

TaBLE 7.—Percentage of mature and immature females taken inside the river at different times during

the season
Percentage Percentage
Date Date
lee Mature erred Mature
May lOtetees ohooh ot ee ee ms Aer ee LOO! |e SualiysB eee al ee eee Oe ee ee 2 98
Milas 3 Siemt soot hss on os eae Benes we eee ee LOO} odie ee es eee ee eal ae Se 25 75
May 16 Sree eee A la ee ee eee al LOOP iakur yet 6 tes Pie RUNS ee Ney 2nd | ee 100
May Tijandlt822__¢ 2.2 _ es! eaten 5 OSH daly Zell! See Se A ie a eee 100
DY) Dea are We Uae oe See eee ae anes | canes T28 100
May20 and! 3tir2t 2 ee eee 100°)|Aag; (622 0S20 Seb tee 2) ee eee Se eee ee eee 100
MAUGHES so UN Te eo ae ee eee 100
Juries 0: 22h Pee Sa A DSN EL SY AI EA MOON | MAUS 22 2S Ce ee ee Sane ee es 100
JUNG 1620 eek OOS ae 2 eee een a eee 100
JUNE 16 andl oe SO AI IR TOO W|4Sept.el2i2.. fv e dee ee ee aha a 100
pune: 24 an de25etee ASR e ee eeua ee 100

GROWTH AND MATURITY OF SALMON IN THE OCEAN 4]

TasBLe 8.—Percentage of mature and immature females taken outside the river at different times during
the season

Percentage due to mature Percentage due to mature
during— Total during— Total
percent- percent-
Date s age of Date 2 age of
Tear imma- rear Tact Can imma-
taken ex one ture fish taken Neds Poeoue ture fish
(mature)| ¥ (mature)! °° vee
May 8 to 10________- 16.9 pls 11.3 SS zai tly 2 oe sade 48.5 37.6 13.9 51.5
Mia yi24 02k 6.3 1.9 21.8 OST Oy 28 once ect eecee 75. 5 19.6 4.9 24. 4
NUNC fee fe-ssolos. Se ilps 55.9 42. 6 OSLO LAUR. 18.2.7 evosscens 86. 4 12.7 .9 1396
ame TO £2 fet. li 1 TBE 55. 5 33.3 88.9 |; Aug. 13 to 17___..._. 88. 2 10.9 8 1alers
RUNG 2 eA Ses 61.3 29. 1 9.6 38. 7
Jumeiz5: 20.2 2k 42.9 Ce ty) Pear eee 57.1 || Sept. 18 and 19_____- 8.3 75.0 16:7 91.7

Reference to these tables shows that, with the exception again of the three
collections made in the lower part of the Columbia estuary, the fish found inside
the river are all mature. This is so obviously in accord with the familiar facts
of the life history of the chinook salmon that it would be quite unnecessary to
present the data given in Table 7 were it not for the unusual presence of immature
fish in those collections made in the estuary and the desirability of presenting a
table that may be compared with the similar table for the fish taken in the ocean.
In the case of the fish taken outside the river, it is apparent that during May and
the first half of June the percentage of immature fish is high. This percentage
gradually falls during the latter part of June, July, and August, until by the middle
of August only about 10 per cent of the fish taken outside are immature. The
one collection made in September shows a high percentage of immature fish again.
The number of fish included in this September collection is too small to be reliable
(12), but it would not be surprising to find that at this time of year most of the
fish taken outside are immature, since the height of the run of chinooks in the
Columbia comes during August, and the result of the entrance into the river of
most of the maturing fish would be to leave mainly immature fish outside. At
the time these collections were made there was little outside fishing done during
the fall months, so that it was impractical to gather adequate data.

The percentages of mature fish taken outside are shown in Figure 11. It is
evident that, with the exception of the last record (that made in September),
there is a fairly steady increase in the percentage of mature fish in the collections.
The trend of this increase has been calculated, omitting the record for September
and that for May 18, 1920, which has previously been shown to be unreliable for
this purpose, and assuming that a straight line will approximate the most probable
change in the percentage of mature fish taken. The trend has been calculated
by the method of averages, * weighting the various points by the number of indi-
viduals contained in the collections that established the points. The trend, shown
on the graph by the broken line, fits the later part of the data remarkably well,
but it fails to fit accurately the data for the early part of the season. It is probable,
however, that this trend represents the true tendency as accurately as could be
done by any such generalization of the observed data. It is quite conclusively

5 For a description of this method see Lipka, Graphical and Mechanical Computation, Wiley, 1918.

42 BULLETIN OF THE BUREAU OF FISHERIES

shown that during May less than 20 per cent of the fish taken in the ocean are
mature; during July, between 50 and 75 per cent; and during August, between
75 and 90 per cent. It has already been mentioned that few data are available
for September, but those that are available indicate a return to the low percentage
of mature fish found during the early part of the season.

PERCENTAGES OF FISH MORE THAN ONE YEAR FROM MATURITY

Table 8 shows that the percentage of fish taken in the ocean which would not
mature for two or more years is considerably less than the percentage that would
mature in one more year. The data are not reliable enough to serve as a basis
for definite conclusions, but in general it appears that fully two-thirds of the imma-

8

PLR CENTAGE.
8

“o : | fale
-
7
oan

WLE6EWO ~
—_O—_ Zewre Ferd/es

Jags Ckvlated Tend

0 10 2£o Jo FO FO 60 70 GO 90 400 40 720 430 “FO £30 60

DAY of SLASON
YONE SULY GUST. SEPTEMBER.

Fic, 11—Percentage of mature fish taken off the mouth of the Columbia River (exclusive of
the data from the collection made in 1920)

ture fish found in the commercial catch would reach maturity during the year fol-
lowing that in which they were taken. These percentages of immature and mature
fish obviously do not represent the exact percentages of mature and immature fish
that go to make up the schools in the ocean, but only the percentages as taken by
the trollers. Undoubtedly there are much higher percentages of the younger age
groups actually present outside the mouth of the river than would appear from this
study of the catch. The taking of the small fish is relatively unprofitable to the
fishermen if larger ones can be caught, and this would inevitably lead to such aselec-
tion of gear or of the locality in which the fishing is carried on as would make it
possible to take more of the larger fish.

Tt should be noted that, in designating as ‘‘mature”’ fish taken in the ocean,
it is not implied that such fish are ready to spawn or even to enter the river. The
term is used merely to indicate that the individuals would mature and spawn during
the year in which they were taken. There are many individuals especially during
the early part of the season that are not ready to begin the spawning migration,

GROWTH AND MATURITY OF SALMON IN THE OCEAN 43

even though they would have matured during the year in which they were taken.
For instance, many of the fish with ocean nuclei captured during May and June,
and which have been considered as mature, would, in all probability, not be ready
to enter the river until August or September, and would be feeding and growing at
_ arapid rate during the interval. Gilbert has shown, in his paper on the salmon of
Swiftsure Bank, that maturing cohoes taken at that point increased nearly 100 per
cent in weight during July and August, just prior to their entry into fresh water
for the purpose of spawning. On account of the range in the age at maturity of the
chinook salmon, it is impossible to get as reliable figures for this species. It will
presently be shown, however, that there is a distinct increase in length of the chinooks
taken outside the mouth of the Columbia during the season. In the case of the
4-year fish with ocean nuclei—the age group that contains the greatest number of
maturing individuals—the length increases from about 80 cm. at the opening of
the season to over 90 cm. by the close of the summer season on August 25. Since
the weights vary as the cubes of the linear dimensions, this increase in length indi-
cated an increase of nearly 50 per cent in weight during the summer. At this rate
a fish weighing 16 pounds at the opening of the season on May 1 would weigh 24
pounds by the end of August. The 3-year fish with ocean nuclei increase in length
from about 60 cm. to about 75 cm. during the season. This indicates an increase
of approximately 95 per cent in weight, a figure quite similar to that given by Gilbert
for the cohoes, which are also 3-year fish.

In résumé, this inquiry into the relative maturity of the chinook salmon taken
in and near the Columbia River has established the following facts: (1) The fish
found in the river are, with very few exceptions, mature; that is to say, they have
definitely left the ocean and entered upon their spawning migration. A very few
fish taken in the extreme lower end of the Columbia River estuary, and only under
exceptional tidal conditions, are immature; that is, they will not mature and spawn
during the year in which they are taken. ,(2) In contrast with this condition it
has been shown that the fish taken in the ocean near the mouth of the Columbia
River contain varying proportions of fish that will not mature for at least one more
year. (3) Theseimmature fish taken in the ocean belong mainly to the following age
groups, arranged in the order of relative abundance: Third year, ocean nuclei;
second year, ocean nuclei; third year, stream nuclei; fourth year, ocean nuclei;
fourth year, stream nuclei; and fifth year, stream nuclei. (4) The percentages of
mature and immature fish taken at different times during the fishing season have
been determined, and it has been shown that a very high percentage (approxi-
mately 90 per cent) of the fish taken in the ocean during the early part of the season
are immature. The proportion of immature fish gradually lessens as the season
advances, until in August nearly 90 per cent of the fish taken are maturing. There
is some indication that in September there is a return to the conditions found in
May—a high percentage of immature individuals.

ABUNDANCE OF THE VARIOUS AGE GROUPS

The variations in the percentages of the various age groups taken inside the
river at different times during the season are striking, especially when compared
with similar data for the fish taken in the ocean. Table 9 gives for each collection

44 BULLETIN OF THE BUREAU OF FISHERIES

of fish taken inside the percentage of fish belonging to each age group, and in the
last two columns the percentage of fish with ocean and stream nuclei. The data
on the percentage of fish with ocean nuclei and with stream nuclei are also shown
in Figure 12. In the figure the trend, as shown by the broken line, was determined
by twice smoothing the original data by threes,° and then further smoothing “by
eye.” A high percentage of fish with stream nuclei is found during the first month
and a half of the season, but during the latter part of June the percentage of fish
with stream nuclei decreases rapidly and fish with ocean nuclei constitute from 80
to 90 per cent of the take after the first of July. The percentages of fish with
ocean nuclei are not shown in the figure, as they are merely complementary to those
for the fish with stream nuclei.

TaBLE 9.—Composition of collections of fish taken inside the river as regards age groups

Percentage with— Total percent-
age with—
Date Ocean nuclei in— Stream nuclei in—
: a Ocean Stream
Second| Third | Fourth] Fifth | Sixth |Second| Third | Fourth] Fifth | sixth | 2UCle! | nuclei

year year year year year year year year year year
IMigyi 10: 2220s 22) 2 Pe See eee ee DAW Se y= all S= SN Sa eee ole LE 80. 5 LSD | ook es 2.4 97.6
IV Syl oe eeen Sone eee 4.3 1.4 USSU iba ey A) (a a a 25.9 49.7 13.0 0.7 10.7 89.3
Mayui6 ou not 8 a ee a ee te ae eS ae 82.5 Wsbi} 22S Sa| sleeees 100. 0
May ty and 18) 000200. leeeeetecloee ss oe 5.6 3 ar gy eels al ea a al atte a 1b ba | 792:63| 22 eases 9.3 90. 7
Mayi27- 2 ah 22 2s Ps) eee Be py aeeeres PB yal Names Soe a 2:3 34.1 20. 4 4.5 38. 7 61.3
May 30 and 31________.__- 4.8 yl sae pe ae 9 al a 2] Pe 1373 12.3 W.2 1,2 6.0 94.0
MUNG AO toe etc ot see cet caltec aero eae 25.0 LOROG| Paes aoe |e tee 25.0 35.0 5.0 35. 0 65.0
JUNO 6: wie oe oe ot ee Bae ee ele ae. oe ee P48) eee ST 10. 2 48.0 28. 6 4.1 9.1 90. 1
June 16'and 172250 2 = ae es its} 8 aU ey fa) ieee | He 8.7 5257 20. 5 .8 17.3 82.7
June 24:8nd 25.8 ss sas ee 3.7 14.8 25.9 9.3 9.34 |/2 25.5 2t2 5.6 24. 1 5.6 1.8 63. 0 37.0
July Be Paes soy So eee 2.0 47.1 33; 3 5895 eat arse 2.0 7.8 2.10) |5o ke eee 88. 0 12.0
SUNY iS. ee ee 30. 5.9 25.9 PAIR) DAG | betray ee 10. 6 20 12 i|caeeeee 85.9 14.1
July AG. SSE A aS 19.7 170 24.0 15.3 b IPR 03) | es eee 15.3 5.5 bea Ars) 77.6 22.4
UL yA 28 eee ee eed | eee 4.0 42.0 50. 0 ZAOY ea ene S| See 2. 00 oles ae 98. 0 2.0
AUS ND scoso. Beane meee 20434 tees Beye 16. 7 4.2 4,2 1255 4,2 402, \Se eee 72.0 28.0
AUS. 622-2255. eee ee 10.5 6.3 46.3 291.5) | ote EE ES 312 Be Ve Ouleect sece 92.6 7.4
TAVIG 22 ee eas eee oe 10 22. 4 66. 0 Pe? A eat Se kd rl of Moe ly gp} 1.0 1.0 92.7 7.3
Septsde2 2 oot eo tee |e 220! 59. 0 G5 3i |e eee eee RS 3.2 4.2 8) 87.3 iP e

Although it appears from these data that stream nuclei predominate during
about half of the season, it must not be inferred that approximately half of the total
number of fish taken during the season are of this type. The fish taken during May
and June form a comparatively small part of the total pack, although an important
part, owing to their fine quality. The great bulk of the run occurs during July
and August. Reliable statistics are not available, but it is probable that, in most
years, not more than 25 per cent of the total pack on the Columbia River is taken
during May and June. On this account the percentage of the total pack formed
by fish with stream nuclei will be much less than the percentage formed by fish

6 This smoothing was accomplished by smoothing independently the series of time values (10, 13, 16, 18, 27, etc.), and the series

of percentage values (97.6, 89.3, 100.0, 90.7, etc.). The resulting smoothed time values were then paired with the corresponding
percentage values and the pairs used as coordinates for the determination of the ‘‘smoothed’’ points.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 45

with ocean nuclei, even though the two types are predominant for approximately
equal times.

Among the fish with stream nuclei, those in theirfourth and fifth years are most
common. A few in their sixth year are taken, but constitute a relatively small
part of the total catch. Three-year-old fish are quite common, but this age group
contains only precociously mature males. Two-year-old fish with stream nuclei
are very rarely found.

The fish with ocean nuclei are likewise predominantly 4 and 5 yearsold. Fish
in their sixth year are comparatively rare, but those in their third and second years
are fairly common, especially during the last half of the season. It has been shown

—LEGEND —
© Observed valves.
----- end

SFERCENTAGCE.

DAY of SEASON.

STAY. SUNE CSZ LOK.

Fic. 12.—Percentage of fish with stream nuclei taken inside

above that some of the 3-year fish with ocean nuclei are mature females, but most
of the fish of this age group are males, and all of those in their second year are the
precociously mature males locally known as “‘grilse”’ or ‘‘jack salmon.”’

The variations In the percentages of fish of the different age groups found among
the fish taken in the ocean is strikingly different from those observed among the
fish taken inside the river. The data are presented in Table 10 and Figure 13.
The trends shown on the graph represent the position of the averages given at the
bottom of the last two columns in Table 10. In view of the fact that there were
no consistent changes in the percentages of fish with ocean and with stream nuclei,
it appeared that the averages would represent the trends as accurately as could
be done by any other means.

8210°—25t 3

46

BULLETIN OF THE BUREAU OF FISHERIES

TaBLE 10.—Composition of collections of fish taken outside the river as regards age groups

Date

PERCENTAGE.

Total percent-

Percentage with— age with—

Ocean nuclei in— Stream nuclei in—

Ocean | Stream

Second} Third | Fourth) Fifth | Sixth | Second] Third | Fourth] Fifth | Sixth | DUclei | nuclei
year year year year year year year year year year
13,2 A[PpBGN2 + | (9 s72804| ce Ola fa| denen 2 /MUb uae TOP) 5\ | ah 1h MO.zaa lene 22. 4
ee 42,4. | 26:8) ||'913¢Gi |Pamn ee sl an wanes RE he Tae 54g) | Vemeeee MSDN 17.3
23, 3i| 4343" ls, 20/0 ru] eeuadllll eames | CE rainoen 1303 0)[ oe cassie 0 oe a 86.7 1B.3
24.2 | 47.5 | 3.3 By esky i Gey We aby, Weg 75.9 1
1: Fal WDB: sol Big oul ward | RE cera | uur 253 Orr] seas erty cranes aalcabe le 75.0 25.0
14.9 22.4 29.8 Se ONn We 5 en he See 17.9 10. 4 P56 |. Bases 70. 2 29.8
27. 30M cord 1) 22753 wn mane | 2s Tei et) ek tes lpa| eroneee e e 63.8 36.2
14: 10h |fegiea™ ||) 9303) al ieatTite | 2 ee | Olea TAQ? |mstoak ORG) jee ee 79.9 20.1
3: Oll|f6r4't | »B0s0minass6. | 1s ON aon 710 | e350 0) eet sO is |e 88.9 1.1
Dilee|, bsSmileneoiemiiaare |. es elie TO Pier Pe ae 91.6 8.4
ANU OAE 5! || LONGI ALSTON) | kaa mes nan 19 | 29 | 965 | 1.0 | 67.7 32.3
ewe ya [acts pea| ame le ni iT | eee 1010" ||! 100) 03430 eee 76.7 23.3
16.00| 31.80] 23.80] 5.85) .64| .07| 11.55| 652] 3.66] .14] 78.0 22.0

——o— — eon rweki
——-—_ Srream pened
~---—----- Trends

mane J

oF iG: AS oy
MAY AUGUST

40 60

DAY

STAY. SUNE SEPTEIIBER

Fic. 13.—Percentages of fish with stream and ocean nuclei taken outside

One of the most notable differences between the collections of fish taken in
the ocean and those of fish taken in the river is the fact just mentioned, that in
the former there is no such decided and consistent change in the percentage of fish
with ocean and with stream nuclei as has been described for the collections of fish

taken inside.

The percentage of fish with ocean nuclei taken outside varies from

63.8 to 91.6, averaging 78 per cent, and of fish with stream nuclei from 8.4 to 36.2,

averaging 22 per

cent. These figures are somewhat different from those given by

Fraser (1920) for the chinook salmon taken during 1915 and 1916 in the Straits of

Georgia, near Vancouver Island, British Columbia.

Fraser found 34.6 per cent of

GROWTH AND MATURITY OF SALMON IN THE OCEAN 47

fish with the stream type of nuclear growth and 65.4 per cent with the ocean type.
In 1917, however, the same author (Fraser, 1921) found exactly the same per-
centage of fish with the ocean type of nucleus which we have recorded here—78
per cent.

Contrasted with this condition of relative stability in the percentages of fish
with ocean and with stream nuclei, which is found among the fish taken in the
ocean, are the marked variations found among the fish taken in the river and which
are illustrated in Figure 12. There would seem to be but one possible explanation
of these facts, namely, that the fish taken outside the mouth of the river repre-
sent a fairly accurate sample of the entire population of Columbia River chinooks
contained in the ocean, and that these average percentages of fish with ocean and
with stream nuclei represent, approximately, the percentage of fish with stream
and with ocean nuclei that would go to make up the total of the run in the Colum-
bia River. It is, of course, possible that races other than those belonging to the
Columbia River Basin are taken off the mouth of the Columbia River, but since
the Columbia is by far the most important chinook salmon stream on the coast it
can hardly be doubted that the great majority of fish found just outside are native
to this river. The fish found outside therefore represent the supply of fish from
which the runs of mature fish found in the river are drawn. It would be expected
that the removal from this supply of considerable numbers of fish with stream
nuclei during the early part of the run, and of greater numbers of fish with ocean
nuclei during the later part of the run, would tend to upset these ratios in a sys-
tematic manner, but this effect is not noticeable. The high percentages of imma-
ture fish taken outside would tend to obscure this effect, since the migration of
mature fish would affect only the percentages of fish with ocean and with stream
nuclei among the maturing fish. It seems probable, however, that with many more
data than are available, some systematic fluctuations in the proportions of fish
with ocean and stream nuclei might be shown.

On account of the fact that the variations in the percentages of fish with
ocean and with stream nuclei among the fish taken in the ocean are practically
negligible, the average percentages of fish of the different age groups taken through-
out the season is significant. (This, as mentioned above, was not true of the data
from the fish taken inside the river, on account of the great variations in the per-
centages of fish with ocean and with stream nuclei and the differences in the rela-
tive importance of the different parts of the run. See page 44.) These data are
given in the bottom line of Table 10, and from them it is evident that among the
fish with ocean nuclei the 3-year group is most abundantly represented, forming
31.8 per cent of the total take. The 4-year group comes next and forms 23.8 per
cent of the total. Fish in their second year with ocean nuclei form the next most
important group, with an average of 16 per cent. A few 5-year fish and a very
few 6-year fish appear in the collections. In the case of the fish with stream nuclei
it is seen that 3-year-old fish are most numerous, followed by 4, 5, 6, and 2 year
fish, in the order of importance. Fraser (1920, p. 172) gives the percentages of
fish of different ages of ‘‘sea type;’’ that is, those whose scales show the ocean type
of nuclear growth, as observed in the Straits of Georgia. The figures are very

®

48 BULLETIN OF THE BUREAU OF FISHERIES

similar to those obtained in this study of the fish taken off the mouth of the Colum-
bia River, as may be seen from the following table:

Tasie 11.—Percentages of fish with ocean nuclei in the various age groups

Age
Locality
Second Third Fourth Fifth Sixth
year year year year year
Offimounthiof Columbia River 2223 222-502 2s ne on ee eee eee 20. 5 40.7 30. 5 7. 50 0.8
Straits{of/Georgia'(from Fraser)i2 220 7.2222 2202) 1 ee eee reece ee 28. 1 42.9 27.8 bj 2) |

The correspondence is remarkably close, considering the fact that the data
were taken at different places, in different years, and were handled by different
observers.

It is interesting to compare these data with those obtained from a study of
the collections made of fish taken inside the river. Among the fish with ocean
nuclei taken in the river, 4 or 5 year old fish were found to be the most numerous.
Among the fish taken in the ocean, 3-year fish were most abundant. Also, in the
case of the fish with stream nuclei, it was found that 4 and 5 year fish were most
numerous in the river, while 3 and 4 year fish were more common in the catches
made outside. It has already been pointed out that the data bearing on the pro-
portions of fish of the different age groups present in the ocean can not be compared
in detail with the similar data for the fish taken in the river, but in a general way
it may be stated that the fish taken outside tend, on the whole, to be about 1 year
younger than those taken inside.

GROWTH

A study has been made of the increase in length of fish of different age groups
taken during the season of 1919, both outside and inside of the river. The study
was confined to the collections made during a single season in order that any yearly
fluctuations in growth rate might be excluded. It must be emphasized that an
observed increase in the average size of the individuals contained in a series of
collections is not necessarily due to growth, particularly in the case of fish taken
inside the river. The constant migration upstream after fish have entered fresh
water results in a constant change in the content of the run passing any given
point along the river. Gilbert (1914, 1915, 1916, 1918, 1919, 1920, 1922, and
1923), in his intensive study of the sockeye salmon of British Columbia, has
conclusively demonstrated the existence of distinct races characterized, among
other things, by very different sizes at maturity. Considerable unpublished evi-
dence is in the hands of the writer, which indicates that similar races are to be
found among the chinook salmon of the Columbia River. The result of such differ-
ences in size of the races running in the river at various times will necessarily be to
mask, more or less completely, the results of growth. The extent to which the
effect of growth will be thus masked will depend upon a number of factors, such
as the relative sizes and numbers of the various races found together within the
river. If large races were running early in the season and smaller races later, the

GROWTH AND MATURITY OF SALMON IN THE OCEAN 49

observed changes in size during the season would be less than the actual growth
occurring within a single race. It might even occur that fish of the same age group
taken late in the season would be actually smaller than others taken earlier in the
season, although they had had several additional months of life in the ocean dur-
ing which time they were feeding and growing rapidly. ‘The reverse condition
might also be found, in which smaller races were running early in the season and
larger ones later. The effect of this would be to augment the true effect of growth.

In the case of fish taken outside, however, it seems probable that the observed
changes in length quite truly represent the effect of growth, since fish of the same
race presumably remain within the fishing area over the entire fishing season.
That this in general is true is indicated by the fact, demonstrated above, that there
is comparatively little change in the composition of the outside schools as regards
age groups, and particularly the main types of nuclear growth, while at the same
time very conspicuous changes are taking place in the composition of the runs
in the river. The details of the growth process are necessarily lost unless they
can be worked out for single races. A mixture of several races, such as we have
in the collections available, will show only certain of the general features of growth
that are common to at least the majority of the races represented in the collections.
The results of the study of such mixed material give a composite picture, which
may not exactly represent the true condition in any one of the races represented.
Fraser (1920) has made an intensive study of the growth of all five species of salmon
as found in the Straits of Georgia, where, as in the present study, it was impossible
to segregate accurately the various races. ‘The results of his study are interesting
as showing the general trend of the growth, but it would seem that such material
can not be relied upon to show the finer details of the growth process.

In order to reduce to the simplest terms the observed changes in length, the
trend of these changes during the fishing season, from the Ist of May to the end of
September, have been calculated. The data on fish taken in the ocean have been
kept separate from those relating to fish taken in the river, and separate trends have
been calculated. An examination of the data showed that the rate of change in
length was, in most age groups (and by inference in others), practically constant
during the season, and it was therefore assumed that the trend would be best shown
as a straight line. The trends were therefore calculated on the basis of the formula
y=a+ ba, in which y is the observed length (the average of all individuals of a single
age group found in any one collection), x is the day of the fishing season on which
the collection was made (beginning with the opening of the fishing season on May
1), and a and 6 are constants. a@ represents the most probable average length
of the age group in question on the Ist day of May, and } the most probable daily
increment in length. The calculations were made by the method of least squares,’
and the ‘‘observation equations” were weighted according to the number of indi-
viduals found in each age group in the various collections. This method of direct
weighting is recommended in some of the texts on the method of least squares.
The mathematical work of calculating these trends was accomplished in most
cases by the use of logarithms, and the results were checked by means of a 20-inch

? See any of the numerous texts on the method of least squares.

50 BULLETIN OF THE BUREAU OF FISHERIES

slide rule. In the case of a few of the age groups in which the numbers involved
were small, the calculations as far as the determination of the constants from the
“normal equations” were done entirely on the slide rule, which gave exact figures
to a sufficient number of places. The calculation of the constants from the ‘‘normal
equations”’ was in all cases done by the use of logarithms. The following table
(Table 12) gives the values of the constants a and 6 and the values of y when z is
100 (the most probable average length on the one-hundredth day of the season—
August 8), for each age group, both for fish taken in the ocean and in the river.

Figures 14 to 21 show, for each age group, the average length observed in each
collection and the calculated trend of the growth during the season, calculated
separately for the collections made in the ocean and in the river. The upper and
lower limits of twice the standard deviation is also indicated in the same manner
as in the previous figures showing the variations in the size of the eggs. For the
purpose of this study, both males and females have been included. The various
ate groups will be discussed separately.

Tasie 12.—Data for the trends in length of chinook salmon taken during 1919

Number
Age group of speci- a b
mens
FISH TAKEN IN THE OCEAN

Ocean nuclei
Second ivearsi. 2-222. Le eee eee be ae Sa cee net acaeesaen eecseenes 121 42, 97 0. 0888
Thirdtyear:: 2). 595.5 Ee ee Cee ek SER a Se ee ee 322 60. 20 - 1193
MOUEtD YORE i202 25.2rscs co eee ee ee ee ee eee 194 77. 65 . 1410
Fifth yearsccs-522. 52055 eae SR ee es eee 37 92. 02 . 0610

Stream nuclei
Mhirdi year. 28 2. eh es 2 Sy I eal LY ie ee SE 2 ee ee eee Wy Rae 117 48. 57 . 0979
WGUTUH WOO co oso a hence e oe ae ran oe ee ea ete eee ee See ee oe en 40 72.10 . 1169
Wifth syoar 2525. 255 3 ae re ee ee ee ee ees ee eas il 83. 50 . 1064

FISH TAKEN IN THE RIVER

Ocean nuclei
Second year. ssice ook es Aes PE eS eet ea eee ee ee 90 44.54] —. 0252
Mhirdiyeate. = 2223s TS Fee 8 eS EE Seas SEER ee PRT Pe 99 48. 98 . 1989
Hourth'year-- 2205 2202 See Or as er a ae a eg Ua ae sean en een 319 79.10 .0771
HMifth!-veari cee Nye A ee AES Ate AY CE AONE Re Oe ee errs 166 111. 10 —. 1051
Bixthiyears 2 2 Salt Wee ae a eee Sere ena anes eee oetn an prensa Sane 19 104. 70 . 1546

Stream nuclei
GAN) og Mast) ghee eA i as Pe el See Fak A ae all SiN a al ea a a 115 47. 30 . 0956
Mourth year. <8 sss oe ee oe ae ah ae ed rec eee 489 75. 00 . 0671
Mlith Veer ee ee eee eee ee eee e ee eee ee 147 92. 30 . 0832
Sixthyearc- o: 2.4.2 Fon ees ae ee ee ee ee eee 17 112.05 | —. 0600

1 The fish of this age group taken on June 16 and June 16 and 17, 1919, were unusually large. Both collections were taken well
up the river, one at Seufert and the other at Warrendale, and it is difficult to escape the conclusion that these fish are representative
ofalargerace. Omitting these two collections the constants become: a=100.5, b-=0.0015, and the value of y when z is 100 is 100.65.

FISH IN THEIR SECOND YEAR, OCEAN NUCLEI

Figure 14 shows the variations in the length of fish of this age group. The
differences between the fish taken outside and those taken inside are conspicuous.
The trend of those taken outside is upward, from nearly 43 cm. on the Ist of May to
nearly 52 cm. on August 8, 100 days after the opening of the season. The change in
length during the first 100 days of the season has been used simply as a matter of
convenience in calculation and discussion. In the text figures the trend is extended

GROWTH AND MATURITY OF SALMON IN THE OCEAN 51

beyond this point. The points that show the average size of the fish of this age
group taken in the various collections are distributed quite closely about the trend.
The fish that were taken in the river, however, tend to be smaller toward the end of
the season than at the beginning, as is shown by the downward slope of the line of

:

LEGEND —
0 Jeken oviside —— Trend
e Jeker sweside,
(] Gwgle specimen.

LENGE TA - cr77,

DAY of SEASON

SATAY SONNE Sev AVEUST | SEP:

Fic. 14.—Length of fish in their second year, ocean nuclei. As in the illustrations showing
sizes of eggs, the upper and lower limits of twice the standard deviation (+ 2c) are shown
by the vertical lines extending from the points that indicate the position of the mean. The
same conventions have been used in the other illustrations of length

trend, and the deviation of the points from the trend is distinctly greater than in the
case of the fish taken outside.

FISH IN THEIR THIRD YEAR, OCEAN NUCLEI

Figure 15 shows the variations in the length of the fish in their third year, ocean
nuclei. The trends for the fish of the two groups—those taken outside and those

52 BULLETIN OF THE BUREAU OF FISHERIES

taken inside—differ materially, although in quite another manner than was noted in
the fish with ocean nuclei in their second year. In the 3-year-old fish, the trend for
the individuals taken inside is much steeper than that for those taken outside, indi-
cating a more rapid increase in size during the season. The trend for the fish taken
outside starts at about 60 cm. on the 1st of May and is increased to a little over 72
em. by the 8th of August. As in the case of the 2-year-old fish, the variation about

M5

|
—LEGEND—

° Taken ovlide, — — Trend

° Joker inside, Trend

O)Sigle specimen.

i f6fwo—
0 Taker ovkive, —— Trend
°* Token t131he, ——— Trend,
(Sigle yoecimen.

4) ca 77) 79 100 125 150
DAY af VDEASON. LAY of SEASON.
¥ Lone veer AQLCOTT | SLOT LIAY. SINE OTA a AUGUST SEPT.
Fic. 15.—Length of fish in third year, ocean nuclei Fic. 16.—Length of fish in fourth year, ocean nuclei

the trend of the points indicating the average size found in the various collections is
less in the case of the fish taken in the ocean than among the fish taken in the river.

FISH IN THEIR FOURTH YEAR, OCEAN NUCLEI

The data and the trends for the fish of this age group are shown in Figure 16.
The trend for the fish taken outside starts at 77.7 cm. and increases to 91.7 cm. by the
8th of August. The corresponding trend for the fish taken inside is much less steep,
starting at 79.1 cm. and reaching a value of only 86.8 by the same date. As in the
two age groups previously discussed, the deviation of the points from the trends is
greater in the case of the fish taken inside.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 53

The study of the fish of this age group taken in the river is complicated by the
fact that two quite distinct races are found running simultaneously during the latter
part of August andSeptember. A discussion of the racial differences observed among
the chinook salmon has not been attempted in this paper, but in this particular case
the differences are so striking that a failure to segregate the two races would be mis-
leading. Tables 19 and 20 on pages 82 and 83 give the length-frequency distribu-
tions in detail for two large collections made, respectively, on August 22 and Sep-~
tember 12,1919. Anexamination of these tables will show that the length-frequency
distributions of the 4-year fish with ocean nuclei are distinctly bimodal. The
bimodality is apparent, not only in both collections but also in the two sexes of each
collection, as well.as in the distribution in which males and females are combined.
The cumulative evidence of these facts points strongly to the fact that two distinct
races are present. Additional evidence of a very satisfactory nature, which sup-
ports this interpretation, is available from the scales. Two quite different types of
scales are found among the fish of this age group: One, typical of the smaller fish, is
characterized by a comparatively small ocean nucleus, which is usually not sharply
defined from the succeeding growth of the second year, and the winter bands of the
second and third years are relatively close together. The type that is associated
with the larger fish has a distinctly larger nucleus, which is usually well set off from
the rest of the scale. The second and third winter bands are more widely separated
and are frequently less sharply marked than those of the other type. Figure 41
illustrates the scales of the first type—associated with the smaller sized fish—and
Figure 42 illustrates the scales of the second type. The evidence given by the length-
frequency distributions and by a careful study of the scales indicated that there was
very little overlapping in the sizes of these two races, and consequently the line of
separation has, somewhat arbitrarily, been set at 90 cm. All of the fish above
this size have been considered as belonging to the larger race, and all below it as
belonging to the smaller race. The overlapping of the two races is quite obviously
so slight that there will be little error in making this assumption. In the text figure
the two races have been shown separately in the last two collections. Both races
were included in the calculation of the trend.

FISH IN THEIR FIFTH YEAR, OCEAN NUCLEI

It has been shown above that there were comparatively few fish of this age group
taken outside the river, although it is well represented among the fish taken inside.
It is significant, therefore, that the trend, as determined from these relatively inade-
quate data from outside, is quite similar to the trends as determined for the other
age groups found among the fish taken outside, which were much more adequately
represented. On the other hand, the trend for the fish taken inside is reversed, the
larger fish occurring during the early part of the season, although the group is well
represented both as regards number of individuals and number of collections. The
deviations from the line of the trend are again greater in the case of the fish taken
inside the river. These features are shown in Figure 17.

54

LENGTH ~ crm.

: 0 Jokers ovisthe———= Trend

BULLETIN OF THE BUREAU OF FISHERIES

—LLEGEND—

Taker snside, Pend
[] Simgle yeecimen.

DAY of SEASON.
AVEUST EP.

STAY. SONNE veLy
Fic. 17.—Length of fish in their fifth year, ocean nuclei

GROWTH AND MATURITY OF SALMON IN THE OCEAN 55
FISH IN THEIR SIXTH YEAR, OCEAN NUCLEI

This age group is so poorly represented among the fish taken outside that it is
impossible to learn anything of the general trend of the growth. The group is by
no means adequately represented among the fish taken inside, but for the sake of
completeness the trend has been calculated and is shown in Figure 18. It can not,
however, be considered as reliable.

ne7

420

LENGTH ~ cm.

—LEGEND —~
0 Takers owlside.
Joker: waste, —— Trend.

[lagle jececmer,

oO 2S 60 yom 100 725° 430
DAY or SEASON.
PTA. SUNE Vezy UGUST | SLEPT:

Fiac. 18.—Length of fish in their sixth year, ocean nuclei
FISH IN THEIR THIRD YEAR, STREAM NUCLEI

The data for this age group are shown in Figure 19. The trends for the fish
taken inside and those taken outside run close together and almost parallel—the only
example among all of the age groups in which this is the case, although it might well
be expected that this would happen in the majority of instances, were it not for the
influence of factors other than growth, such as the presence of a succession of various
races. Asshown by these trends, the fish of this age group averaged about 48 cm. in

56 BULLETIN OF THE BUREAU OF FISHERIES

length at the beginning of the season in 1919, and increased approximately 10 cm.
between that time and August 8. The greater variability of the fish taken inside is
not as clearly shown in this group of fish as in many of the others, but a considera-
tion of the actual deviations of the observed average lengths from the calculated
lengths (the residuals) shows that even in this instance the fish taken inside are dis-
tinctly more variable. The average deviation from the trend among the collections
taken inside is 1.8 em., and among the fish taken outside is only 0.9 em.

—LEGEND—
0 Jake ovlside —— Fend
*Sokern srrsde Trend

LENGTH ~ cre.

(] O7gle woecimers.

DAY ef SEATON

STAY SLSMVE Sey AYGUS7 | SEPF

Fic. 19.—Length of fish in their third year, stream nuclei
FISH IN THEIR FOURTH YEAR, STREAM NUCLEI

Figure 20 shows the data for the fish of this age group. As in the last group
considered, the trends of the fish taken inside and outside are in fair agreement,
although the agreement is by no means as close as in the 3-year fish. Only 40 indi-
viduals, scattered through 8 collections, represent this age group among the fish taken
outside, while there were 489 individuals in 16 collections taken inside. Yet the
average deviation of the points from the trends is distinctly less in the case of the fish
taken outside.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 57
FISH IN THEIR FIFTH YEAR, STREAM NUCLEI

Only 11 fish of this age group appear in the collections of fish taken in the
ocean during 1919. These were scattered through six collections and yet give a
trend similar to the other trends shown by the fish taken outside. The trend for

PNET ~ con.

Z

-L£6END—
0 Taker otfsice———— Ferd

°*Joker sweside, rend.

llGvagle yocenmen.

DAY cof SLASOW.

STAY SOHNE SLY AYUST | SEPT

Fic. 20.—Length of fish in their fourth year, stream nuclei

the fish taken inside is quite similar, but runs considerably higher. The data are
shown in Figure 21, and an inspection of this graph shows again the greater variabil-
ity of the average sizes of the fish taken inside.

58 BULLETIN OF THE BUREAU OF FISHERIES
FISH IN THEIR SIXTH YEAR, STREAM NUCLEI

No fish of this age group were taken outside of the river, and only 17 inside.
A text figure, therefore, has not been prepared to show these data, although the
values of the constants determining the trend have been given in Table 12. That
the trend would be reversed is indicated by the negative value for 6 as given in

—LECEND —~
ofekers ovtsrde, —-— Trend
Joker s78s1Jde, —— Frernd.

(] Sage species.

LENGTH - com.

LAY oF SEASON.
STAY SINE SLY AUGUST | S5SEPZ

Fic. 21.—Length of fish in their fifth year, stream nuclei

the table. Although, with so few individuals, such a trend can not be considered |
very reliable, it is worthy of some consideration in view of the fact that, among
the fish taken inside the river, there were two other age groups, both of which
were well represented but which showed a similar negative trend.

Several important facts are shown by these data on the length of the fish.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 59

An examination of the tables and text figures given above shows clearly that
in all of the age groups represented among the fish taken in the ocean a marked
increase in length occurs during the fishing season, from May to September, in-
clusive. Similar increases in length are shown in five of the age groups taken in-
side the river, but in three of the age groups—the 2-year and the 5-year fish with
ocean nuclei and the 6-year fish with stream nuclei—the trend is reversed and the
fish belonging to these age groups are smaller toward the end of the season than
earlier. It is also noticeable that, even excluding these age groups in which the
trend is reversed, the trends for the fish taken inside are more variable than for
the fish taken outside. The average daily increment (6 in Table 12) for the fish
taken outside varies from 0.0888 to 0.1410 cm., while that for the groups taken
inside, omitting those age groups in which the trend is downward, is much greater,
ranging from 0.0671 to 0.1989. This is over two and one-half times that shown
by the fish taken outside. Attention has also been called repeatedly to the fact
that in each age group the deviations of the average lengths for each collection
(shown as the points in the figures) from the trend (as determined from the series
of collections taken over the whole season) is greater among the fish taken in the
river than among those taken in the ocean. All of the evidence available indicates
a decidedly greater variability among the fish taken inside. A few instances of this
sort might be attributed to chance, but the uniformity of the evidence demands
some other explanation.

It seems very probable, if not, indeed, certain, that this greater variability in
the rate of increase in length among the fish taken in the river is due to the fact
(which has been mentioned above) that there are more profound changes in the
racial constitution of the fish found migrating up the river than in those found
schooled outside, and that these different races vary sufficiently in their average
size so that the effect of growth may be masked more or less completely by these
racial changes. Evidence has been presented which shows that both of these
factors are present in the migrating fish within the river. It seems hardly neces-
sary to argue that there is a succession of races during the season among the mi-
grating fish. Granting the existence of such races it follows that there must be a
succession of them in the river during the season, so that at one time the run con-
tains a conspicuously large percentage of fish from one or at most a few races,
while at other times other races will predominate. This would be particularly
true in the case of such a large river system as the Columbia with its numerous
tributaries, which probably have, as Gilbert has shown for the sockeye salmon,
each its individual race. It has been shown in one specific instance—that of the
4-year fish with ocean nuclei taken during August and September, 1919—that
distinct races are to be found among the chinook salmon of the Columbia River,
and, moreover, that these races may vary sufficiently in size so that the predominance
of one of them at one time and another at another time would undoubtedly greatly
modify the true effect of growth.

The reversed, downward trend of the fish belonging to some of the age groups—
the fish in their second and fifth years with ocean nuclei and those in their sixth
year with stream nuclei—may be explained, then, by supposing that the races that
predominate in these age groups early in the season are large and those running

60 BULLETIN OF THE BUREAU OF FISHERIES

late are small. Such characters might, however, fluctuate from year to year,
so that samples collected in other seasons might not show exactly the same effect.
It is not necessarily to be expected that the trends of the fish of different age groups
taken inside will run parallel with each other, even though they do not run parallel
with the trends for the fish taken outside. It may well happen that at any given
time one race of fish will predominate among the fish of one age group, and that an
entirely different race will be predominant in another age group. The evidence
is not available to prove this for the Columbia River chinook salmon, but Gilbert
has shown, in his extensive studies of the sockeye salmon, that different races
may show quite different distributions among the different age groups. It may
be inferred that similar conditions would be found among the chinook salmon if
sufficient evidence were available. The possible effect of such a succession of
different races should be carefully considered in any detailed study of growth in
which this factor may have an opportunity to operate.

It has been shown that there are probably no radical changes in the racial
constitution of the fish found in the ocean outside the mouth of the river. It has
also been shown that the fish taken in the ocean are, in varying measure, immature,
and that many of them are not destined to enter the river during the year in which
they are taken. In consideration of these facts, it seems reasonable to suppose
that the increase in size of the fish taken in the ocean reflects, with reasonable
accuracy, the true effect of growth. If this be true, it is evident that most of the
growth of the year takes place during the fishing season—that is, during the late
spring and summer. This is shown by the fact that the fish of any one of the
age groups are, at the end of the season, nearly equal in size to the fish of the next
older age group at the beginning of the season. This is quite to be expected since
it has been shown by numerous studies on various fish and other marine and fresh-
water organisms that the main growth occurs during this part of the year. Fraser
(1917) has shown that the chinook salmon found in the Straits of Georgia make
their most rapid growth during this season—80 per cent of the total growth of the
year taking place during the months from April to September, inclusive. These
figures must, however, be considered as very general, since in computing them
Fraser made use of the yearly increment of growth in fish of all age groups. In
his investigation of the growth of the Pismo clam, Weymouth (1923) found that the
period of most rapid growth usually came during May to September, inclusive.
The writer (Rich, 1920) has shown that the period of most active growth in young
chinook salmon previous to or during their seaward migration occurs during the
spring and summer but varies somewhat with the locality. In the Columbia
River, the most rapid growth occurs between May and September, inclusive,
and in the Sacramento River from March to July, inclusive. Weymouth points
out that the rate of growth is profoundly affected by variations in temperature,
being induced and accelerated by higher temperatures. The earlier occurrence
of the period of most rapid growth in the young salmon of the Sacramento River,
when compared with those in the Columbia. River, is undoubtedly due to the
earlier warming of the Sacramento water. Numerous other examples, showing
that the most rapid growth occurs during the spring and summer, might be cited,
and it seems certain that for most organisms the more eke growth occurs during
the warmer part of the year.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 61

Since the data on which this paper is based were taken entirely during the
months from May to September, inclusive, the trends that have been calculated
for the fish taken in the ocean represent the average slope of the growth curve
during the period of most rapid growth. In order to show the general appearance
of the total growth curve for chinook salmon, these trends have been used in pre-
paring Figures 22 and 23, showing growth curves for fish with the ocean type of
nucleus and with the stream type of nucleus, respectively. Such a use of these
data is not strictly justified, as the data were all taken during a single year (1919),
so that the individuals belonging to each separate age group came from different
broods. The fish in their second year came from eggs deposited in 1917, those in

Oe es ie
CAet ee aeaeee
aa ligne leer ed ae

90

60

70

60

LENGTH.- cm.

$0

z > —& = = & z= > Ise = = & =
3 = 7 Ss = 8 § = 3 s = a 5
SECOND YEAR THIRD YEAR FOURTH YEAR FIFTH YEAR”

Fic. 22.—Growth curve of chinook salmon that migrated seaward as fry (scales with ocean type
of nuclei). Data from fish taken in the ocean near the mouth of the Columbia River. Solid
lines show the trends of growth during the fishing season; broken lines indicate roughly the
probable course of growth during the remainder of the year; circles show the size of chinook
salmon taken in the Straits of Georgia, as given by Fraser. Fraser’s measurements did not
include the caudal rays, while our’s did

their third year from the brood of 1916, and so on. The most logical method for
preparing such a growth curve would be to have data over a series of years so that
the growth of the fish derived from a single brood year could be obtained. Lacking
this, it has seemed desirable to present in this form such data as are available.

In the illustrations the calculated trends are shown by the solid lines, and the
trends for successive years have been connected by dotted lines. No data are °
available upon which to base the form of these connecting lines. They have there-
fore been drawn in “by eye.” The resulting growth curve probably represents
fairly well the average size of these fish at different ages and at different seasons.
Some error is doubtless due to the use of a straight-line trend as showing the size
during the period of rapid growth. The growth during the first and the last parts

3210°—257 4

62 BULLETIN OF THE BUREAU OF FISHERIES

of each growth period is probably slower than indicated, while that at the height
of the growing season is somewhat more rapid. The data available, however,
do not lend themselves to a more detailed analysis.

The general form of the growth curve thus obtained is typical and consists of a
series of steps, rising rapidly during the summer season and more gradually during
the rest of the year. Similar growth curves have been presented by a number of
authors and have been obtained by the writer in an unpublished study of young
steelhead trout. For the sake of comparison, the data given by Fraser (1917)
for the size of chinook salmon taken in the Straits of Georgia have been indicated

Le
1
pe
BEE@EBe
CV

v,
=a
a

100

LENGTH - C1.

50

= S & = > LN ES = = =
= x = &

S NS 8 SS S BS S BS By S
THIRD YEAR FOURTH YEAR FIFTH YEAS

Fic. 23.—Growth curve of chinook salmon that migrated seaward after spending one year in
fresh water (scales with stream type of nuclei). For explanation see Figure 22

on the figures. Fraser gives the size of fish of the different age groups at the end of
each year as calculated from scale measurements. Since he has considered April 1
as marking the beginning of the new growth period, his data for the length of fish
at the end of each year have been placed on the ordinate corresponding to this date.
His data are not exactly comparable with ours, since his measurements do not
include the caudal rays, while ours do so. On this account his figures fall con-
sistently below those for the Columbia River fish but near enough so that it seems
probable that the differences are due almost entirely to the different measurements
used. There is very little variation in these differences, Fraser’s figures being 6
to 10 cm. lower than ours.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 63

The agreement in the data from the Straits of Georgia and from the ocean
near the mouth of the Columbia River is remarkably close, considering the different
conditions under which the data were collected. The growth is apparently nearly
identical in the two localities. It seems evident that the growth curves given in
Figures 22 and 23 represent quite accurately the average size of chinook salmon
during their life in the ocean.

The results of this study of growth may be summarized as follows:

1. Examination of the trends of the variations in size of the different age groups
during the season has shown that the variations are much more irregular among
the fish taken in the river than among those taken in the ocean. Evidence is pre-
sented to show that these irregularities are due to variations in the racial consti-
tution of the fish taken in the river. On this account the data from this source
are not suitable as a basis for a study of growth. The data obtained from fish
taken in the ocean is, however, suitable for such a study.

2. A growth curve has-been based on the trends obtained from the study of
the fish taken in the ocean. It has the form typical of organisms the growth of
which continues over a series of years and shows seasonal fluctuations in rate.
The agreement with similar data obtained by Fraser in his investigation of the
salmon of the Straits of Georgia is very close—a fact which indicates that the growth
curve presented in this paper may be relied upon to show the average size of chinook
salmon in the general region of the Columbia River and Puget Sound during the
greater part of their life in the ocean.

FISH TAKEN IN MONTEREY BAY

A few data are available from the chinook salmon taken by troll in Monterey
Bay, Calif. <A collection of scales was made during the summer of 1915, and at
the same time a collection of egg samples was made. The egg samples were merely
taken from the cleaning tables and, since the value of the size of the eggs in deter-
mining the relative maturity had not then been established, no attempt was made
to take data and scales from the fish from which the eggs came. In fact, this value
first became apparent through the study of this collection. In the summer of
1918 (June 19 to 21), a small collection of females was made and egg samples taken
and tagged so that they could be identified with the corresponding scales and data.
At the time this collection was made it was the intention to obtain many more data,
but the fishing happened to be unusually poor and only 63 females could be obtained
in the time available. The study of these few data has proved interesting, although
they form an inadequate basis for final conclusions. More extensive studies of the
fish taken in Monterey Bay are being made by the California Fish and Game
Commission under the direction of Prof. J. O. Snyder, assisted by E. A. Mc-
Gregor, to whom the writer is indebted for the privilege of examining some of
their unpublished data.

SIZE ;

The length-frequency distributions for the various age groups found in the
collections made in 1915 and 1918 are given in Tables 22 and 24 on pages 84 and
86. The number of individuals belonging to each age group and the average lengths
are given in Table 13.

64 BULLETIN OF THE BUREAU OF FISHERIES

TasLe 13.—Number of specimens and average length of salmon of each age group taken at Monterey

June 16, 1915 June 19 to 21, June 16, 1915 June 19 to 21,
1918 1918
Average Average Average Average
Num-| length, | Num-| length, Num-|} length, | Num-| length,
ber centi- ber centi- ber centi- ber centi-
meters meters meters meters
Ocean nuclei: Stream nuclei:
Second year_--------- 30 63.27 2 47. 00 Mhirdiyears---- 2. 2-5" 1 59. 00 1 57.00
‘Thirdtyear 2228252204 28 76. 07 51 72. 64 Fourth year-_--------- 6 81. 33 3 80. 30
Ourthicyear® 22-232 22 25 87. 72 6 91. 30 Bifthyearsee ree 2 93.00 =. 28 |eeeaeee
Mifth'year v2 tb % 22 14 96: 60-2. = = tess, eee

A comparison shows that these figures are, in the main, in fair agreement with
the data from the Columbia River. A slight tendency is apparent, especially
among the fish of the younger age groups, for the fish taken at Monterey to run larger
than did the Columbia River fish in 1919, but many more data than are at present
available will be necessary before this can be asserted with any certainty.

PERCENTAGES OF THE VARIOUS AGE GROUPS

The collection made in 1915 can not be considered as showing the proper per-
centages of fish of the various age groups. It was taken at a time when the fish
were so segregated at the cannery that arandom sampling wasimpossible. Individ-
uals were sorted into three lots, according to size. One contained fish less than 5
pounds in weight, which was, at the time, the legal limit below which the fish could
not be sold. Many fish of this size were taken by the fishermen and were brought
to the canneries, only to be discarded. The second group contained fish between
5 and 15 poundsin weight, and the third group those above 15 pounds. In collect-
ing these data it was necessary to handle the various sizes separately, and it was,
therefore, impossible to get a fair sample of the take as a whole. The collection
made in 1918 contains practically all of the females that were brought to one of the
large canneries and to several of the smaller fish dealers during the period of three
days in which the collection wasmade. It therefore represents a fair random sample
of the take at that time. The percentages of fish of the different age groups are
given in Table 14. The table also gives the percentages of fish of the different age
groups as found by Mr. McGregor of the California Fish and Game Commission.
His collection contained over 200 specimens taken between June 4 and June 22,
1921.

TaBLe 14.—Percentages of fish of different age groups in Monterey Bay

Age group

Date Ocean nuclei Stream nuclei

Second Third Fourth Fifth Sixth Third Fourth Fifth

year year year year year year year year
June: 19'bo21 191s ek wie oe eee ee ee 3.2 81.0 OH Bs | Lee oo alive PSS 1.6 CY (| eee en
Jtme 4: tO 22: 1921 CSE ae eerie ESS FS 139 70. 8 13. 6 6.6 0.5 1.9 3.3 14

GROWTH AND MATURITY OF SALMON IN THE OCEAN 65

There is a remarkably close agreement between the figures for these two collec-
tions, when it is considered that they represent different years and the work of two
entirely independent observers. The proportion of fish with ocean and with stream
nuclei is almost identical—6.3 per cent of the 1918 collection and 6.7 of the 1921 col-
lection having scales with the stream type of nuclei.

It was shown above that the percentage of fish with stream nuclei among the
schools off the mouth of the Columbia River was close to 22 per cent, and the fact
was mentioned that Fraser has found that 34.6 per cent of the chinook salmon taken
during 1915 and 1916 in the straits of Georgia showed the stream type of nucleus.
Although in 1917 Fraser found that the percentage of fish with the ocean type of
nuclear growth was 78 per cent, identical with our findings in the fish taken near the
mouth of the Columbia, it would seem probable that the salmon in the Straits of
Georgia tend to contain a larger percentage of individuals that have remained an
entire year in fresh water before migrating seaward. Gilbert (1922a) has found
that the chinook or king salmon of the Yukon River invariably have scales with the
stream type of nucleus.

These data indicate with considerable certainty that latitude, or more properly
the differences in climatic conditions obtaining in different latitudes, has a definite
effect upon the early history of the chinook salmon. In the more southerly latitudes
the tendency is for the young fish to migrate soon after they become free-swimming,
early in their first year. Under the influence of the more severe climatic conditions
of the north, which cause later hatching and slower growth, the tendency is for
greater and greater percentages of the fish to remain in the home stream during the
first year.

A similar condition has been found to exist among the salmon of Norway.
Dahl (1910) gives tables showing for three localities and for two years the per-
centages of these salmon that migrate seaward after spending 2, 3,4, and 5 winters in
fresh water. He concludes:

A close examination of these tables shows that the age of the smolts at migration varies
between 2 and 5 winters. In the south the smolts are generally young, but the farther north we
go the more pronounced is the tendency for the fish to remain longer in the river before migration.

In this connection it is interesting to note that in a large river basin such as
that of the Columbia River a distribution of races is found which is suggestive of
this distribution in latitude. It is certain, from unpublished data in the hands of
the writer, that as a rule the races in the Columbia which spawn in the higher
tributaries are predominantly those which have the habit of remaining in fresh
water during the first year and whose scales, therefore, show the stream type of
nucleus. The races in the lower tributaries, on the other hand, migrate, as a
rule, during the first year, and as a result the scales of the adults show the ocean
type of nuclear growth. It is evident, therefore, that the effect of altitude on the
early history of the chinook salmon is very similar to the effect of latitude—an
increase in either altitude or latitude resulting in an increase in the number of
young that remain in the stream during the entire first year. This similar influence
of altitude and latitude is well known from the numerous studies of the geographical
distribution of land animals and plants, and it is interesting to note the parallel
effect on racial habit in such an anadromous form as the chinook salmon.

66 BULLETIN OF THE BUREAU OF FISHERIES

RELATIVE MATURITY

In the matter of relative maturity the salmon of Monterey Bay do not differ
greatly from those taken off the mouth of the Columbia River. The distribution
of egg sizes, as found in the collection of eggs made in 1915, is givenin Table 23.
In this table the egg sizes are given separately for the three size groups mentioned
above—that is, for the fish less than 5 pounds in weight (Group 1), for those between
5 and 15 pounds (Group 2), and for those over 15 pounds (Group 3). Group 2 is
composed largely of fish in their third year with ocean nuclei, and the distribution
of egg sizes in this age group*has been shown separately in Figure 3. Table 25
gives the distribution of egg sizes as found in the 1918 collection. Figure 24 shows

Taken at Monterey, Calf

Yune 29, 1918.

& 3

WVUMBER OF INDIVIDUALS
: sy

Taken at Monterey, Calf
Yune 19-21, 1918.

S

480 -.90 000. 0.10 0.20 0.30 0.40 0.50 5 0,70
Locaritnm of Diameter or Eces

Acruat Diameter of Eecs mm.
4.0 L251 15 475 2.0

0.63 0.75 2,5 3.0 3.5 40 45 5.0

Fic. 24.—Distribution of egg sizes as found in two collections made at Monterey, Calif.

the distribution of egg sizes for both the 1915 and the 1918 collections. The general
appearance of the two distributions is quite similar, although many more mature
fish were taken in 1915 than in 1918, probably due to the selection involved in
making the first collection, which has been mentioned above. A distinct bimodality
is apparent in both collections. One mode appears in both at log D 0.17 and another
at log D 0.39. The two modes in the distribution of egg sizes in the fish in their
third year with ocean nuclei, as shown in Figure 3, are at the same points. Of
the two groups shown in Figure 24 the fish with the larger eggs would undoubtedly
have matured during the year in which they were taken, while the others would
not have matured for at least one more year. There is obviously a certain amount
of overlapping, which is especially evident in the 1915 collection. An examination
of Table 23 on page 86 shows that the overlapping is entirely within the group of
fish weighing from 5 to.15 pounds. The distribution of egg sizes within this group

GROWTH AND MATURITY OF SALMON IN THE OCEAN 67

is quite irregular but is obviously bimodal. It has been possible, by the appli-
cation of statistical methods, to separate the two component curves with some
degree of satisfaction, but in view of the fact that the material is selected and
therefore not representative it seems unnecessary to discuss this in detail. It is
apparent from the table that all of the fish contained in the group weighing less
than 5 pounds were immature and that all above 15 pounds in weight were destined
to mature during the year in which they were taken.

The percentage of mature and immature fish can not be determined from the
data available for 1915 on account of the selection involved in making the collection.
The 1918 collection, although inadequate, gives some indication of the relative
number of mature and immature fish present in Monterey Bay at the time the
collection was made. In Table 25 (p. 87) and Figure 24 it is shown that the lines
separating the two distinctive egg sizes comes at approximately log D 0.30. With
the exception of one individual, for which the log D is 0.29, the two groups are well
separated. If we include this doubtful individual with the group having the larger
eggs, this group will include 24 individuals and the group with smaller eggs
will contain 39 individuals. On a percentage basis, then, this collection contains
38.1 per cent mature and 61.9 per cent immature individuals. This agrees very
closely with the data from the Columbia River. In Figure 11 the trend of the
percentage of mature fish taken in the ocean near the mouth of the Columbia
River shows that on the fiftieth day of the season, June 19, the catch contained
40 per cent of mature fish. Such a close agreement as this is doubtless accidental,
but it indicates strongly that the relative maturity of the chinook salmon in Mon-
terey Bay is, during the latter part of June at least, approximately the same as
that of the salmon found in the ocean near the mouth of the Columbia River.

FISH FROM DRAKES BAY AND FORT BRAGG

A small collection of scales and eggs was made on August 15 and 16, 1918,
by representatives of the California Fish and Game Commission from fish taken
by troll in the region of Drakes Bay. This bay is located about 30 miles north of
the Golden Gate and is one of the centers from which trolling is conducted. Scales
and eggs from 12 chinook salmon taken near Fort Bragg were also taken on July 17.
Since both collections were small, and separate study has disclosed no marked
differences, they have been combined in the tables presented. Unfortunately serial
numbers were not given to both scale and egg samples, and it is impossible to
refer the scales to the corresponding eggs. The length of the fish was recorded
with the scale samples, and a tag attached to the egg samples also gives the length
of the fish from which the sample was taken, but except in a few extreme instances
this does not serve to identify the corresponding samples. Furthermore, the
number of egg samples does not agree with the number of females. On account
of this confusion of the records, a satisfactory analysis is impossible, and it has
been necessary to handle the data for size of fish and size of eggs separately.

Table 26 (p. 87) gives the length-frequency distributions for the fish with
ocean nuclei. In addition to these there were four fish with stream nuclei, all in
their third year and averaging 64.5 cm. in length. Three of these were females.

68 BULLETIN OF THE BUREAU OF FISHERIES

Comparison with the data from the Columbia River and from Monterey shows
that the average lengths of the various age groups, as found near Fort Bragg and
Drakes Bay, is substantially in agreement with the data from other sources, although,
as at Monterey, the fish are somewhat longer than Columbia River fish of the same
age groups. The data are too few, however, to give accurate results.

Table 27 (p. 88) and Figure 25 give the egg sizes found in these collections.
Sixty-four females are available, and the table shows that they are divided, on
the basis of the sizé of the eggs, into three well-marked groups—the group with
the largest eggs destined to mature during the year in which they were captured
and the others composed of immature individuals. Sixty-seven per cent were
mature. This is a somewhat smaller percentage of mature fish than was found
off the mouth of the Columbia River at the same time of year. Since most of the
individuals contained in this collection from the northern coast of California were
taken in Drakes Bay, August 15 and 16, the comparison has been made with the
Columbia River fish taken August 15. Reference to Figure 11 shows that on this

Taken at Drakes Bay
and Fort Bragg

1918-

Numser oF INndIVIOVALS

4.80 AIO 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70

Locaritm oF Diameter or Eces

‘AcruaL Diameter oF Locos mm.
0.63 0.75 4.0 1.25 45 t.7 Sime O. 2.5 320 35 40 45 50 $5

Fic, 25.—Distribution of egg sizes as found in a collection from Fort Bragg and Drakes Bay,
Calif.
date (the one hundred and seventh day of the season) about 87 per cent of the
fish taken near the mouth of the Columbia were mature. The discrepancy can
hardly be considered as significant, however, on account of the inadequate number
of data from Drakes Bay and Fort Bragg.

Since the egg samples were accompanied by data giving the length of the fish
from which they were taken, and presumably, in the main, came from the same
individuals from which scale samples were taken, an attempt has been made to
determine which age groups were represented among the immature fish. A table
has been prepared (Table 15) showing the correlation between length of fish and
size of eggs, and also the range in the size of the fish composing the three groups
distinguished by egg size. For the sake of comparison there is shown, in the same
table, the range in size of the females found in the various age groups. From this
table it is apparent that the fish with the smallest eggs were all in their second
year. They were all between 40 and 50 cm. in length, and the only individuals of
this size represented in the collection of scales were 2-year fish. The second group
of fish (those with log D between 0.00 and 0.30) agree in length with those in their
third year. There are 18 individuals in group 2, and in the scale collection 19
females were 3-year fish. The agreement between the two categories makes it
seem practically certain that group 2 is composed largely, if not entirely, of fish

GROWTH AND MATURITY OF SALMON IN THE OCEAN 69

in their third year with ocean nuclei. Group 3 (the mature individuals) is evidently
composed mainly of fish in their fourth and fifth years.

TaBLe 15.—Chinook salmon taken by troll near Drakes Bay and Fort Bragg, Calif., July and
August, 1918, tabulated to show correlation between size of eggs and length of fish

Centimeter length (mid-value of class)
Logarithm of diameter of eggs (mid-value of
8 eines) 88 Total
45 55 65 75 85 95 105

TAYE} esp PR Sr i 4 | Saeeee |e se | ey | else 3 (Group 1; 3 individuals).
fp eee ere ew at ee ek By ee cecctecece-|Lekescloesece 1 AS (cca (eee, Sel Ween 2
ALO Seen oem ae Orie Sete te ett stb oso eeSaSe|oL lsc. 1 4 2 Vi leseec- [se ocoe 8?(Group 2; 18 individuals).
Rae See enon ee ee oe re eo ree eee scenee|-scnes|>eeo-- | ee eee eee 8
ap eee eM s ha lates cha sbbeeso cbse aiendcc| veeecalawsnan|ouacec seeces 3) |eeecee | eesvet 3
amma heared rend (eee oe oe ; . . ee (Group 3; 43 individuals).
fr SIMA HONE MITTS ie fee Ee Re] 5 2| 7

Motalees vee ts 20 e225 i lt koh eet eset! 3 1 9 4 17 18 12 | 64

(Gio) 5 0) 3 | ee ae pee ee ee ee 35) eee ee | eee | ees eae | ees eee

ROU ee ee eden ca caulesenes 1 9 3 LA eerie aa (EL enemas 18

OTE OU Dsmeraeeee soe ha ease oat sore caslee cane |eoa- os |seeaee 1 12 18 12 | 43

SeconGsyealesere ent oe) Se ooo asa ct en sbosecsokece Spleen cee leanees | pee em ee ae Nie Seni (eee 3.

PE Mircevonl wn weer somite Nee les cceeeke-cleenss pee 8 3 8) sec aeate oes 19.

Hounbliny ears: SiS sfe se 2. oh. tk ket sok S/S cchesacele aches 1 2 17 10 | 30

TC CAL peer ee eee ae Pde he eee eS ah, et ee os be eee ll ese 1 et
UROL eee es eG at aoe eecalsasads ces i eee Oe 8 4 10 als 11 | 53

SUMMARY AND CONCLUSIONS

1. It has been shown that the eggs of female chinook salmon undergo a well-
marked differential growth during the growing period just preceding and terminat-
ing in the migration. As a result, those females which are destined to spawn
during the year in which they are captured may be distinguished by the size of
their eggs from those which will not mature for at least one more year.

2. Using the size of the eggs as a criterion, the commercial catch of salmon
taken in the ocean near the mouth of the Columbia River has been analyzed and
the percentage of mature and immature fish determined. It has been found that
the percentage of mature fish taken in the ocean varies greatly during the season,
being relatively small (from 10 to 20 per cent) in May but increasing gradually,
until in August nearly 90 per cent are mature. While in general, as would be ex-
pected, fish taken within the Columbia River are mature, there are times when a
few immature fish are taken by seines and traps in the extreme lower part of the
estuary. The few data from Monterey and the northern coast of California are
in substantial agreement with those from the region of the Columbia.

3. The immature fish taken in the ocean comprise, in the order of relative
abundance, the following age groups: Third year, ocean nuclei; second year, ocean
nuclei; third year, stream nuclei; fourth year, ocean nuclei; fourth year, stream
nuclei; and fifth year, stream nuclei.

70 BULLETIN OF THE BUREAU OF FISHERIES

4, The relative abundance of fish with stream and with ocean nuclei has been
considered, and it has been shown that about 22 per cent of the fish taken off the
mouth of the Columbia River have scales with the stream type of nucleus. Since
these fish are predominately from the Columbia River, this probably represents
closely the percentage of fish of this category contained in the entire population.

5. Evidence is presented which shows that the more rigorous climatic condi-
tions associated with higher latitudes and greater altitudes tend to increase the
percentage of fish with stream nuclei—that is to say, more of the young fish
remain in their home stream for at least one year after hatching.

6. Variations in size within the various age groups have been studied and a
growth curve constructed. It has been found that the variations in the size of
different races successively ‘passing up the river on the spawning migration are such
that data from this source can not be relied upon to show the growth. Similar
variations among the fish taken in the ocean are, however, consistent, and may
be depended upon to show the growth of the various age groups. Only this data
has been used in constructing the growth curve.

7. The growth is typical of nearly all organisms in that it progresses at a
maximum rate during the warmer part of the year—from May to September—
and slows materially, if it does not stop entirely, during the colder months.

The undesirable features of the fishing for chinook salmon, which is carried on
in the ocean by trolling and purse seining, are more or less obvious. A large percent-
age of immature fish are taken, which are far from having attained their maximum
size and of relatively poor quality. They are feeding heavily, and the presence of
large quantities of food in the stomach and intestines causes rapid spoiling. In
many cases these immature fish are unfit for canning. On the other hand, the fish
found in the river are, with very few exceptions, mature, and have definitely left
their ocean feeding grounds and begun the long journey to the spawning beds. They
have ceased feeding, and therefore growing, and the deterioration in the quality of
flesh known to occur in salmon during their spawning migration will soon begin.
Since the fish taken in the river have thus reached theirmaximum size andan optimum
condition for commercial use, it seems logical that the commercial catch should be
restricted to this stage in the life history.

From a business standpoint, the development of this ocean fishery would seem
to be most undesirable. The poor quality of the outside fish, when canned, can not
be questioned, and to continue to pack and market them as Columbia River chinook
salmon can not fail to react unfavorably on the reputaiion of the Columbia River
product. Many of the packers now place most of the fish taken outside in the infe-
rior grades, marketing them as chums, but there is certain to be a constant tendency
to place inferior fish with the better grades.

A much more important phase of the development of this outside fishery has
to do with its effect on the conservation of the salmon run in the Columbia River.
In order to understand this, however, it will be necessary to review briefly the recent
history of the salmon industry in this district.

With slight fluctuations, which can not be referred definitely to any cause, the
pack of chinook salmon on the Columbia River has remained fairly constant for the
past 15 or 20 years. The opinion is current, however, that the run of salmon was

GROWTH AND MATURITY OF SALMON IN THE OCEAN 71

low for the few years just preceding 1914, but in that year the run suddenly rose to
normal again and was maintained at this higher level for a number of years. This
erroneous opinion is the result of a common practice which considers only the canned
product in discussing the trend of the productivity of the Columbia River. The
figures for the canned pack alone support this contention but do not take into con-
sideration the mild-curing industry, the development of which has had a marked
effect on the production of canned salmon. Much of the mild-cured salmon was
marketed in Europe and this market was suddenly closed by the opening of the

Thousands of cases

1890 (1895 1900 (905 /H0 1915 /920

Year
Fia. 26.—Pack of chinook salmon on the Columbia River, 1890 to 1923, inclusive. The mild-
cured pack has been reduced to a basis of cases, one tierce being considered equal to 25 cases.
Data from Cobb (1921) and Pacific Fisherman Year Book, 1924. Dotted line shows the mild-
cure pack; broken line, the canned pack; and solid line, the total

World War in August, 1914. If this pack is calculated on the basis of cases of canned
salmon and is added to the canned pack, the totals do not show a marked depression
preceding 1914 nor a sudden rise in that year. There are fluctuations, of course,
but there is nothing to show that they are systematic or other than “chance”’ vari-
ations.

This is apparent from Figure 26, which shows the number of cases of chinook
salmon packed on the Columbia River from 1890 to 1923, both inclusive. The
pack of mild-cured salmon is also shown, reduced to a basis of cases, and the total
pack when these last data are added to the pack of canned salmon. The reduction
of mild-cured salmon to a basis of cases has been made by considering that one

Wo BULLETIN OF THE BUREAU OF FISHERIES

tierce is the equivalent of 25 cases of canned fish. Published data are not available
to show that this is a fair procedure, but it is believed that it gives a very close
approximation to the truth. It is stated by Cobb (1921) that the loss in weight of
mild-cured salmon ‘during the 2 or 3 weeks in which the fish lie in the first packing
may be reckoned at 30 per cent.’”’ Packers on the Columbia River usually estimate
that chinooks lose about 25 per cent of the round weight in cleaning preparatory
to canning. Although the loss is probably a little more in the case of the fish that
are mild cured, on account of the removal of the backbone, the figure is close enough
for practical purposes. If, then, the weight of the cured fish is 70 per cent of the
cleaned weight, and the cleaned weight is 75 per cent of the round weight, the total
loss in weight during the entire process is 52.5 per cent of the round weight. It
would therefore require 1,600 pounds of round fish to produce a tierce containing
830 pounds of mild-cured salmon. Canners on the Columbia River consider that
it usually requires about 65 pounds of round fish to produce a case of canned fish,
and on this basis 1,600 pounds would produce very close to 25 cases.

The Columbia River salmon fishery has for years been prosecuted with an
intensity that makes it seem remarkable that a run of commercial value still remains.
Figures, unfortunately, are not available to show how this intensity has increased
since the beginning of the industry, but there can be no doubt that there has been
a tremendous increase in the total fishing effort within the river. There has been
not only an increase in the number of men, boats, and various units of gear, but
a marked increase in the effectiveness with which the gear is employed. The motor
boat has, within the last 20 years, replaced the slower sailboat, and the length of
gill nets and sees and the size and effectiveness of fish wheels has increased. 'The
fishing inside the river was, at the time the outside fishing first began, about as
intensive as possible. Practically all good trap and wheel sites, seming grounds,
and ‘‘drifts”’ in which gill nets could be operated were occupied. The discovery
that salmon could be caught profitably outside the mouth of the river by trolling
and in purse seines offered, therefore, a new avenue of expansion in which the
fishermen so engaged did not come into direct and immediate competition with
those already established on the fishing grounds in the river. It was found that.
the area in which such fishing could be carried on was wide, trollers ranging 20 to
30 miles in all directions from the mouth of the river. Such a broad region offered
large possibilities for expansion and, since the outside fishing proved lucrative, it
is not surprising that fishermen flocked to the new fields. Since this has not been
accompanied by any appreciable reduction in the fishing effort within the river, it
has meant a sudden and great increase in the total fishing effort directed against
the salmon run of the Columbia River. Since a considerable portion of the total
pack of salmon in the river has come from fish caught in the ocean, it means, further,
that a correspondingly smaller percentage of the total pack has come from inside
the mouth of the river from gill nets, sees, traps, and wheels.

From the evidence given above it is apparent that the immediate effect of the
introduction of fishing methods that attack the immature fish found just off the
mouth of the river is to increase the intensity of fishing, not only upon those salmon
that are destined to mature and spawn during the year but also upon those that
will form the spawning run the following year, and, to a more limited extent, the

GROWTH AND MATURITY OF SALMON IN THE OCEAN 73

second year later. The full effect of the increased fishing effort during a given
year will not, therefore, fall entirely upon that year but will be distributed over
at least 3 years. Thus the outside fishing conducted in 1920 affected the run of
mature fish into the river not only in 1920 but also in 1921 and 1922. Entirely
apart from the fact that the young, immature fish produce an inferior product, this
encroachment upon the runs of future years seems an especially vicious phase of
this newly developed fishery. The full effect of the outside fishing is partially
hidden; it is not immediately apparent in a decreased run into the river. It might,
therefore, easily cause a very serious depletion before it became apparent that there
was any danger of such an outcome.

At the same time other factors that have undoubtedly tended to reduce the
supply of salmon have been at work. Many of the tributary streams that were
once used as spawning beds by thousands of salmon are now blocked by dams of
one sort or another, and other streams are made barren by the removal of quan-
tities of water during the irrigation season. Large numbers of young salmon on
their seaward migration become lost in the irrigation ditches or impounded in the
pools left in the main stream as the water is drawn off for irrigation—where they
die as the water warms and evaporates. On the whole, there is no question that
the available spawning area in the Columbia River Basin has been materially
reduced by such factors as these, and it seems probable that the encroachment on
the spawning area will continue for some years to come. We have, then, a situation
in which the continuance of the salmon is menaced on the one hand by a diminishing
spawning area and on the other by an increased intensity of fishing. The various
industrial and agricultural projects that are responsible for the erection of dams
and irrigation ditches are of such importance that it is idle to suppose that they
can long be opposed successfully by the interest of the salmon fishery. Regardless
of right or wrong, it is inevitable that sooner or later the fisheries must disappear
wherever they are directly and unavoidably opposed by the requirements of indus-
trial and agricultural expansion.

Efforts to counteract the effect of these various agencies, all of which tend
toward the destruction of the salmon, fall into three general categories: (1) Legal
restrictions. These may affect the type or amount of gear used, the area open to
fishing, and the time during which fishing may be conducted. (2) Construction of
fishways over dams and of screens to irrigation ditches. (3) Artificial propagation.

The first of these is obviously designed to reduce the intensity of fishing. It is
the oldest and still the most effective and indispensable of all means for conserving
fishery resources. Unless a sufficient number of mature fish are permitted to ascend
the rivers to the spawning areas, depletion is certain to occur regardless of any
efforts that may be made in maintaining spawning areas or in reducing the mortality
of the young fish by means of artificial propagation. |

The construction and maintenance of fishways and screens to irrigation ditches
is purely a palliative measure designed to offset in some measure the effect of en-
croaching civilization and development. They merely lessen to some extent the
effect which the building of dams and irrigation ditches has in destroying spawning
areas, and can not be expected fully to counteract the effect of this one destructive
agency.

74 BULLETIN OF THE BUREAU OF FISHERIES

In artificial propagation we have a method designed to offset the work of destruc-
tive agencies, which is at once the hope and the despair of the scientific conservation-
ist—the hope, in that it is so eminently logical to protect the young of the salmon
during the early part of their life when the rate of mortality is high, and the despair,
in that the evidence of its efficiency is inadequate and conflicting. Many instances
can be adduced showing, apparently, the beneficial results of artificial propagation,
but there are other instances in which no such results are to be observed. And if
the extravagant claims of some of the proponents of artificial propagation were true
we would long since have ceased to worry about the future of our salmon resources.
The difficulty apparently lies in the fact that, as at present conducted, the procedure
of artificial propagation is not based on scientific knowledge. With a gradual in-
crease in the efficiency of hatchery procedure, which will come with placing it more
and more on a truly scientific basis, we may hope that artificial propagation will
come to be one of the most important factors in the preservation of our fishery
resources. Noteworthy progress is being made by the Bureau of Fisheries toward
this end, and it seems certain that the future development of artificial propagation
is most promising. At the present time, however, it must be admitted that the
importance of artificial propagation as a means of conserving the supply of salmon
can not be accurately evaluated.

We are left, then, with the single alternative of maintaining the intensity of fish-
ing below the danger point if the salmon run is to be preserved. Just what this pointis
no one can tell, and for that very reason it is essential to see that the intensity of
fishing is kept down so as to provide what may be reasonably supposed to be a margin
of safety. It is the duty of all who are interested in conservation to see that this is
done—especially those officials whose duty it is so to administer the fisheries that
depletion may not occur.

Just where do these facts fit into our discussion of the effect of the development
of outside fishing on the supply of fish in the Columbia River? We have seen that
the pack on the river has remained practically stationary for a number of years,
during which time the intensity of fishing has been increased, especially by the addi-
tion of trolling and purse seining; that the spawning area is being gradually reduced;
and finally, that a restriction of the amount of total fishing effort is essential to the
maintenance of the run in the Columbia River.

It must always be an indication of danger if an appreciable increase in the in-
tensity of fishing does not provide a corresponding increase in output. It must
inevitably indicate that the productivity of a fishery is being maintained at a given
level only by drawing to some extent upon the reserve stock needed for breeding if
the race is to be maintained. Evidently a surplus of breeding adults is normally
provided in nature, and it is only from this surplus that man can draw without
immediately endangering the continuance of the supply. Any infringement upon
the necessary breeding reserve is dangerous. A slight infringement may show no
immediate effect, but if continued over a period of years the effect will be cumulative
and is certain to end in disaster. The increase in fishing effort on the Columbia
River by the development of trolling and purse seining has shown no corresponding
increase in the total pack, and we may assume, therefore, that the present intensity
of fishing is too great and is resulting in a dangerous reduction of the reserve of breed-

GROWTH AND MATURITY OF SALMON IN THE OCEAN 15

ing adults. It may well, indeed, have been too great even before the outside fishing
developed, in which case the new fishery is especially dangerous.

The intensity of fishing could, of course, be reduced by increasing restrictions on
the gear used in the river, but this is hardly reasonable, since the outside fishery was
the last to develop. It has further been shown that the outside fishing is uneconomi-
cal in that it takes the fish at a time when they are of poor quality and are much
smaller than they would be at maturity, and that it is especially and subtly dangerous
because it not only increases greatly the intensity of fishing but attacks the supply of
fish one or two years before they become mature.

It is quite possible that these reasons are insufficient ones on which to base a
legal restriction of this newly developed fishery, but if the run in the Columbia
River.is to be maintained we may be certain of one thing—if the outside fishing is not
restricted, it will be a matter of a relatively short time only before the fishing in the
river itself will have to be limited proportionally so as to supply the deficiency of
spawning fish which is certain to result from the increased intensity of fishing caused
by the development of trolling and purse seining. Efforts have been made to have
restrictive legislation passed by the State legislatures of Oregon and Washington, but
so far these efforts have met with only partial success. If it should prove impossible
to prevent outside fishing entirely, it would seem desirable to limit such fishing to
the latter part of the fishing season. This, at least, would reduce the number of
immature fish taken, would tend to improve the quality of the pack, and would
minimize to some extent the danger of seriously depleting the supply of fish before
some indication of the imminence of such depletion has become apparent.

SUPPLEMENTARY TABLES

TaBLE 16.—Constants for each frequency distribution, Columbia River collections
[F, female; M, male; S.D, standard deviation]
COLLECTIONS MADE INSIDE THE RIVER

Length in centimeters Logarithm of egg diameter
* Typeof | Year Num
Locality and date Pyicleuatenliclass Sex sen
Mean Ss. D: Mean sy 1D)
1. Ilwaco, May 10, 1919___----| Ocean____- AU Re aceen AG S3s00'n we | |p sees 0}:53 ee te eee eee ees
Stream__- 4) | Meee W7s|( (BB095', | fas ekan sc eace-|bessssccceceesce|Socsens-sccsces
eae aie 16 . 87 Be teeee setae) Us b3l eee e eee Uae
Both 330), 7182!45=-0) 49) |) ay 150! 34) Eo 2-o seek seco |- os ea enc ecole
Bal Misoez=2 2 SOOKE My as eee | Renae ee ee eeo. | See ie Poe ee
seeoces 5 | 91.80 saccedseaccee! 582 Sees es Sa
Both TA\WOQR7ON ec eeeoee Foe 23| eee aesesescsscn|oocoeese canoes
2. Astoria ,May 13, 1919_____- Ocean___-- 2)) Mieseces Great O0V 9) | Seamed oe eee eee eo ee eee oe
So) Me obee 2 5ESOOM | ig | Se eee eee eee eee Peat eee cee Se
45) Meceets Zu 83800l. © si Secen ake. | Rac eesecdaoses [scansceseececnn
Renesas Srils S700k) me 5) Sates ao eees Ut geal as cil |e ee Be ee
Both___- TE eSSR8OG) 9 catleeree seen eal Biss ee sees eee ieee ee oe
Stream ____ Bl bay Bees en 36), 4856781553) AN70S=) 87 ee oo eee eee | Pies ccteeeccen
As a shae 41 73. 73+ .61 He 801443) | Sec kee cok ac s| be 222 22 Fes
Steere 28 | 75.85 .60] 4.744 .43 498-40. 0075 | 0. 0598-0. 0054
Both OOH F450 se a6) (6640 SO once cccwemcencecl|occcnceenecnce=
SME ae be Oh S206 hy Sete ono Wee kacseseness|-oseccaueaweccs
Eee pp een 12} 87. 64 Sesesed estes 57 ee ee
Both LSt SCs 00 ws ep | Sees enone Se okt Sade woes oes oat accea ce
(il) any Ee 1610500), il eecweeeee eas |Saaeos. cease ces bec eks eee oe ose
3. Seufert, Oreg., May 16,1919 | Stream__-_- Boe Mire 4 li n05 002 4 eA BS 284225 20 ILS 2 ce [bso ae cece ek
1) See 77| 74.744 .34] 4.404 .24 . 5713+. 0039 | .0502+ . 0027
Both LSU W480 27) as 864- 19 oo oes |b 22a es ene ee
Sel Vee ee LOG OG TSO lt dese || ewe eee eee Se oe os. i] ee ee
es 93.642 .61] 4.244 .43 6718+. 0062 | .0430- . 0044
Both__- SD NO45 6252 Ou manne 4 Oto oa lcs cS sce e bon eaee

76

BULLETIN OF THE BUREAU OF FISHERIES

TaBLE 16.—Constants for each frequency distribution, Columbia River collections—Continued
COLLECTIONS MADE INSIDE THE RIVER—Continued

Type of

Locality and date nucleus

4. Ilwaco, May 17and 18, 1920
(from gill nets and traps).

Stream_.--

5. Chinook, May 27,1919 (from
traps in Baker Bay).

Stream__-.

6. Dodson, May 30 and 31,
1919 (from wheels and

seines). Stream.-__

7. Ilwaco, June 10, 1919 (from
gill nets).
Stream__-_-

8. Seufert, June 16, 1919______-

Stream.__-

9. Warrendale, June 16 and | Ocean-_____
17, 1919 (wheels and
seines).

Stream___-

1 Calculated for 27 mature specimens.

on wm OO OOD aon mo mre

POO

a

we OO

ore wo

Pow

6

Length in centimeters

Logarithm of egg diameter

Mean

96. 67 . 87

77. 29-1, 01
78. 23 . 96
77. 70+ .71

76. 52-1. 16
81. 541. 03
79. 30+ .81
103. 80
101. 56+ . 96
102. 36 .76
115. 60
109. 00
114. 00

67. 00
91. 00
114. 60
108. 80
111. 70+ .89
54, 64
72, 92+ . 72
81. 351. 03
75. 06+ . 66
102. 00
96. 00+ .76
98. 00+ . 86
123. 00

8. 2540. 73
8.484 . 62

Mean

7. 30+ .68} .5977+ .0092] . 0693: . 0065
8. 18+ . 50

7. 86-4 . 82
7. 76£ «73
8.20 . 57
“6.06 . 68
6. 00: . 54

For one other log D was 0.11; for another, 0.29.

a

heh Mn nent

GROWTH AND MATURITY OF SALMON IN THE OCEAN aT

TasBiE 16.—Constants for each frequency distribution, Columbia River collections—Continued
COLLECTIONS MADE INSIDE THE RIVER—Continued

Length in centimeters Logarithm of egg diameter
Locality and date Type at eer Sex Aue
Mean Ss. D. Mean ish 1D)
10. Astoria, June 24 and 25, | Ocean____- Dal) Miaeces 2] 50.00
1919 (seines). 3°) Meee 22. 8 | 60. 50
2) NE 3 79. 60
Wye ten 32 11 | 81.50
Both... 14} 81.10
5 2 | 112.00
3 | 96.20
5 | 102. 50
6 PMiee <225 1 | 123.00
pCa ae 4 | 106. 00
Both___- 5 | 109. 50
Stream_.-_- 3: | Weer cee 3] 49.60
4 Mo. 2.2. 8] 77.30
pea eae ae 5 | 84.20
Both___- 13; 79.90
1g oe eres BU 3 | 95.00
(ats A 1} 101,00
11. Ilwaco, July 3, 1919 (traps) .| Ocean----- Bijue fee 2. Oh eed OO} ne | One eee SO LOO TS a maa | casero eas
2) Besge.J2 24) 83. 580.54] 3.93+0.38 | .5900-+0.0046 | . 0332-0. 0032
5 i) Biot. S. 17} 99.9441.12] 6.82+.79} .6029+ .0062] .0381+ .0044
65) Bes 3 . 6570
Stream_-___ 6 (i ees 1 0100
Qo Re 2a ede 4 5950
6 ithe ees 1 6300
12. Sand Island, July 7, 1919 | Ocean__--- 21 Mise oS. 21) 43\38+: .51] 3.462: .36 [-----.-----_-...
(seines). a. 5 i. 902
Both___- 26} 44.38% .53} 4.04 .38 |------------..-.-
3 Bu STG (fase oes [ec ceseeee 22 A o8|e
i 2 One 0.11 and
another 0.51
5
An ING ee 2
5
6
Stream___- 3
4
5
13. Warrendale, July 16, 1919 | Ocean-____- 2
(wheels). 3
4
5
6
3
4
5
6
14. Ilwaco, July 28, 1919 (traps) | Ocean____- 3
4
5 LG Soares ie | Pa aie aR
Fa Pear Mame et res Seer Es a di pat ceceratres
Stream___- 4
3210°—257 5

718 BULLETIN OF THE BUREAU OF FISHERIES

TaBLe 16.—Constants for each frequency distribution, Columbia River collecttions—Continued
COLLECTIONS MADE INSIDE THE RIVER—Continued

Length in centimeters Logarithm of egg diameter

Locality and date Typeot er Sex Num-

15. Seufert, Aug. 5,
(wheels).

Stream_._-

16. Dodson, Aug. 6, 1919 | Ocean...--
(seines).

RPwDP cPWNoO

a

Stream_-.--

17. Astoria, Aug. 22, 1919 Ocean-----
(seines).

NT

S050 ty eae aet a 74
79, B+1.42] 9. 6641.00 |.......---------
ae text, Ds

cS

Odie ANG | plcse esac sees . 7316+ .0098 | .0730+ .0069
Table 19.
TS a ei (eta Se es eae es . 7100
04000 Sceecncccecs tbl olson ees ese
STS6Te ) Yeeeccetscceee . 7033
8350). *% esecccsdccuce]escecio cesses | eee ene
80F00) ) | | |2e2se-252-222 and
LOSROOF" ceca sues . 59

TOHQOEYS «a Secscecaceess

VBESOne: 4 sal see esses suse

73.1041.31} 8.94+ .93
he text, p.

Por

Stream...-

ao

w

18. Astoria, Sept. 12, 1919 | Ocean--....
(seines).

>

eR Rat alo ke) b eee ee
Table 20.

o

Stream-.-_-

oF

COLLECTIONS OF FISH TAKEN IN THE OCEAN BY TROLL UNLESS OTHERWISE STATED

ay See

19. May 8 to 10, 1919____..__-- Ocean----- 25 | Mee So V7}: 44:;532-056 1)! 08272420; 433) 2. - = os |e eee

es ee st 19 | 43.104 .50] 3.26+ .36 | 1.8100+0. 0085 | 0. 0548+0. 0060
Both_--- 36] 43.784 .40] 3 27

3 | Mies 2-- 51.) 62.40 261) 5
F..:----|' 102)|. 61,004 .38 | 5
Both_---| 153} 61.464 .31] 5

4) Mit oo BRB fo. 00 Nl PA ete ass
| eee 18 | 79.564 .71] 4
Both: -_- 21) 79.674 .63| 4

in ae eee D035 00K ieee | to oeeen ence an

Stream.-_-- ou] Vee = 34] 49.064 . 43

3.
eee 18| 48.114 .59] 3
Both__-- 52] 48.73+ .35] 3.
4) Mies Zu OOM CON Me | Ay) zeean cle One ko eee eee
Dee eae 2) 74.00 sane
Both___- 4{ 70.00 fess
5. | Fi aed. =
Cio) eee Ohl) CREO 9 dea

2 The distributions were so irregular and, especially in the case of the males, suggestive of bimodality,that it seemed useless
to calculate the standard deviation and the probable errors until a more detailed study can be made.
3 Excluding 2 individuals with log D greater than 0.30. (See p. 34.)

GROWTH AND MATURITY OF SALMON IN THE OCEAN 79

TABLE 16.—Constants for each frequency distribution, Columbia River collections—Continued
COLLECTIONS OF FISH TAKEN IN THE OCEAN BY TROLL UNLESS OTHERWISE STATED—Contd

Length in centimeters Logarithm of egg diameter
Typeof | Year Num-
Locality and date aaa (lane Sex Bee
Mean 8s. D. Mean 8. D
20) May, 18, 1920. --22-- 2st 22_ 2 Ocean._--- 3 Gey AS 16=0540) | 5 47442 OF 28) Boe ones ape ee eeeennonaae
39 | 70.744 .57] 5.254 .40] (See Table 21.) |__.__.________-
aC be A(t See $m) |e Bae ne i ee eee ee
4 QO S495 255.9 fallen os 600) | aaoeeeae acceso. |Uosoneces eae
36 | 83.00+ .63 | 5.664 .45 | (See Table 21.)
(afin Me teR hile seceaeda| lt eee te ee: 1) eek eee
5 20) 1 OO2602E 122) S12 OT lic. occ ccccesaea|acecceccnancace
13 F
33
Stream.__- 4 15
13
28
5 8
6
14
21. May 24, 1919____.....-..... Ocean___-- 2 ag eee soe 4
Both_--- 7
Ey a ee 5
Biiscose 8
Both_--- 13
AEMerSe. 2 3
Serer 3
Both_-_-- 6
Stream____ 3 || yee :
4
22. June 4, 1919... eee Ocean..--- 11
18
29
25
32
lit
3
1
4
1
Stream__-- 1
11
11
22
4 4
5 1
1
2
232 J1n0:10;:19192.2 25.8. Ocean___-_- 2 7
3
10
3 3
3
6
4 di
af
2
Stream__-- 3 4
2
6
24) JUNG21, 19192-2222. 2 ae” Ocean..--- W) 4
6
10
3 aD
4
15
4 10
10
20
fA |G ee eee 2
Stream__-- Bn fovea ee 8
Wf eecs 4
iIBOtheees 12
4 Eee 2
eae 5
Both... 7
Ra eee, il

4 Exclusive of 1 individual with large eggs; log D greater than 0.30.
5 Two individuals had small eggs, log D 0.10, and two had large eggs, log D 0.50.

80 BULLETIN OF THE BUREAU OF FISHERIES

TaBLe 16.—Constants for each frequency distribution, Columbia River collections—Continued
COLLECTIONS OF FISH TAKEN IN THE OCEAN BY TROLL UNLESS OTHERWISE STATED—Contd.

Length in centimeters Logarithm of egg diameter
. Type of | Year Num-
Locality and date amas lanes Sex Gor
Mean 8. D. Mean Sie:
25, June 25, 1919 (purse seines)-} Ocean-___. 2 3] 45.50 1. 8990
3 1 [ie SEC 0 MiSs (ps See ee ENE eI ee
4 1] 88.50 Sen cocies eee eee
2{| 83.50 0. 5430
3 B55 20M Wy d| St soo o- we alas ee eee eee
Stream ___ 3 W- S56Rb0Ue P| Se ses se 3 oe eee
1] 46.00 1. 9590
2 Bl 20 ae fo PAE oe Soho nalts oS eee eee
4 PSSA OOW tian faeces soe eo ee ee eee
1] 79.00 0. 5610
2 SLIOOME) Ths 1 See coerce. al hols oleae eae
OGrdulys2;nl O10 arene eee Ocean___- 2 14] 49.30 T. 8730
3 31 67.45 .69 | 5.7240. 49 | 60. 1085-40. 0052
4 23 | 83.78 .90| 6.37% .63|7 .53364 .0116 | .0805+ . 0082
5 11] 95.40 - 5792
Stream__- 3 7 | 55.00 4 . 0214
4 10 79. 60 8 . 4340
5 3} 88.40 . 53800
Qa Duly.28;, 1910222 -- eee Ocean___- 2) | Ee 8 ||. 55.67 T9100) ey al | tae ee eee
3H See 16 71. 01 PO. 1b8le oo 3 ee ee
A \ See 49 | 89. 33+ .64 . 6406 .0059 | .0618-- . 0042
5] Wee. 18 | 96,224 .75 . 6733 . 0094
6) | PRBS: 1 | 103. 00 0
Stream__- Sai ee | 59. 00
PO te! See Bere 3 | 82.33
6) PRS st. 1 | 87.00
28. Aug. 3-5 and 25, 1914 (se- | Ocean___- 2 35 | 51.914
lected as ‘‘grilse’’). 38} 51.374
73} 51.6384
3 2! 62.00
Stream-___ 2 2{ 40.00
3 4| 56.50
12 57. 50
Both-___- 16 | 57.25
20 AUG 13, 1010e cece onset eae Ocean___- 2) 5S E 2) 657.00
3 15 | 78.60
4 66 | 93.00
5 4 | 103.00
Stream__- 3 1] 59.00
4 7| 81.00
30. Aug. 13-17, 1918 (five taken | Ocean___- 2 5 |. 53.00 9500.) .. {|beee se eee
in September). 3 25 | 86.60+1.04 | 7. 70+ .74 [140.7580 .0035 | .0235 . 0024
4 20; 95.304 .98] 6.50 .69 15 .7426-- .0127 | .0820-- .0090
5 19 | 100.164 .99} 6.40+ .70 . 7553+ .0121 | . 0784+ . 0086
Stream __- 3 266 00M G Clas see eeee 1300s) © -% ‘ESS eee
4 Beer eOO vie Nase as esses a (ily (Uae |S a I
5 27 95. 59 .77 5.96 .55 7811+ . 0043 0332-+ . 0030
6 1 t05s00ie, easseeescs eee (D008 i eee eee
31. Sept. 18 and 19, 1919___._.. Ocean___- 2 5 | 52.20
2 56. 00
7| 63.30
3 7 | 75.28
T \ekOL 72
14 73. 00
6 1 | 103. 00
1] 99.00
2 | 101.00
3 1} 55.00
2 66. 00
3 | 62.33
4 3 | 91.00
5 1 | 103.00

6 Exclusive of 5 individuals, for which log D was 0.4340.
? Exclusive of lindividual, for which log D was 0.1900.

8 Exclusive of 3individuals, for whichlog D was 0.1766.
® Exclusive of 3 individuals, for whichlog D was 0.6167.
10 Exclusive of lindividual, for whichlog D was 0.2500.
11 Exclusive of 6individuals, for whichlog D was 0.1480.
12 Exclusive of 3individuals, for whichlog D was 0.3030.
13 Exclusive of 1 individual, for which log D was 0.2700.
14 Exclusive of 4individuals, for which log D was 0.1600.
15 Exclusive of 1 individual, for which log D was 0.2500.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 81

TasLe 17.—Chinooks taken in wheels in the Columbia River near Warrendale, Oreg., July 16, 1919

Females tabulated according to size of eggs, type of nucleus, and age

Ocean nuclei Stream nuclei
Logarithm of diameter of eggs (mid-value of class) Ate par ante aan Total
ir ; our i Six
year Fourth year Fifth year year year year

oesees 1 eee oe eae eee bee ee 2
eye nae 4 ii peowee se | Be eees Se | eee 5
1 D leeeeeeesbeudagss ji een ses pyeiee 4
1 2 Ll See no al] eierarons wall Lean 4
ie oe es 2 1) DSRS aCe) ROO Fon Ver ae ea 6
Ses oon 2 fe eee a Seca see eee 4
pee 4 2 1 1 I 9
pbelae ui 3 2 i bal (ipa a
edase ee 1 4h Pose sok | Saees | veaee 5
2 19 18 4 2 1 46

0. 60 | 0.62470. 0072 | 0. 6555-40. 0074 0. 66 0. 68 0. 67

sesoeeee 204663 . 0051 | . 04662 0052 | 3 28) ances cl ccacscs

TasLte 18.—Chinooks taken by troll off the mouth of the Columbia River, July 28, 1919, collected
at Ilwaco, Wash. Females only

Tabulated according to size of eggs, type of nucleus, and age!

Ocean nuclei Stream nuclei

Logarithm of diameter of eggs Total
(mid-value of class) Second: Third

year Third year Fourth year Fifth year year

Fourth year

Miatg 2

4

ai

9

15

11

4

6

2

4

2

2

2

alee 1

96

Mean ten. so see ee ceee ee eae 1,91 | Immature. 0. 1531 | 0. 6406-L0. 0059 | 0. 6733-40. 0094 | 0.0756 | Immature. 0. 25 |______
Miaturess.6: G161<|ps2= eae mnene ee | Benne Rees | CR es IMature.2= .68))222=25

Dtandard deviationessses-|assesces| seeeee a senne sesso! . 06182: .0042 | .0594+ . 0067 |_-_---__|_-.---------------]------

_1 Two specimens, one a 5-year fish with the stream type of nucleus and one a 6-year fish with the ocean type of nucleus, are
euuitted from the table. The logarithm of the diameter of the eggs in the case of the 5-year fish was 0.61, and for the 6-year fish
.69,

82 BULLETIN OF THE BUREAU OF FISHERIES

TaBLe 19.—Chinook salmon taken by seines on the lower Columbia River, collected at Astoria, Oreg.,
August 22, 1919

Tabulated according to length, type of nucleus, age and sex !

Ocean nuclei Stream nuclei

Centimeter length
(mid-value of
class)

Fifth
year

eae Fourth year Fifth | Total

Third year Fourth year year

Male | Male | Female Total Male | Female} Total] Female} Male | Female] Total | Female}

8 62 3 2 3
82. 50 | 79. 28-1. 42 77.00 | 87.67

Standard de-
Viatlonsssss|seeae | nee |omee ene 9: 662-1,00) |< aco oce ees ere cece ceck |e oee sel castes ee ee eee

1 In addition to the specimens tabulated above there was 1 female 103 cm. long, in its sixth year, stream nucleus.

2 The apparent bimodality of the distributions of the 4-year fish has been discussed on page 53. The various averages and
standard deviations are as follows: (1) Fish less than 90 cm. in length—males (19 specimens), mean, 77.00--0.77, standard devia-
tion, 4.98-+0.55; females (11 specimens), mean, 81.73; total (30 sepcimens), mean, 78.74-+0.59, standard deviation, 4.78--0.43.
(2) Fish greater than 90 cm. in length—males (18 specimens), mean, 101.44-+0.81; standard deviation, 5.10+0.57; females (14
specimens), mean, $7.30; total (32 specimens), mean, 99.620.57, standard deviation, 4,800.41.

GROWTH AND MATURITY OF SALMON IN THE OCEAN 83

TasLE 20.—Chinook salmon taken in beach seines on the lower Columbia River collected at Astoria,
Oreg., September 12, 1919

Tabulated according to length, type of nucleus, age, and sex

Ocean nuclei Stream nuclei

Fourth

Centimeter Third year Fourth year Tifth year year

length (mid-

Fifth year Sixth year

ValUGlOlClass)) jae SE

7 21) 35} 21 | 56 2 4 6 3 1 3 4 4 1 5
78. 80/73. 101. 31 | () | () | @) ]112. 00 |93. 00 |99. 30 87. 7 |103. 00 |92. 30 |95. 00 |105. 00 |103. 00 |104. 60 |..-

Standard de-
Viationso222)0.222|-22-. Bde OS ieee | Cese ee Se ate les ee el occ sekeealecsccal awsbouleswecee sasaecsleee

1The two types (small and large) of 4-year fish with ocean nuclei have been separated, as was done in Table 19. (See p. 53.)
The means and standard deviations are as follows: (1) Fish less than 90 em. in length—males (20 specimens), mean, 78.50--0.66,
standard deviation, 4.42+-0.47; females (10 specimens), mean, 81.40; total (30 specimens), mean, 79.53-40.56, Standard deviation,
4.55+0.41. (2) Fish greater than 90 cm. in length—males (15 specimens), mean, 102.76; females (11 specimens), mean, 97.37;
total (26 specimens), mean, 100.46+0.78, standard deviation, 5,940.55,

84 BULLETIN OF THE BUREAU OF FISHERIES

TasBLe 21.—Chinooks taken by troll near the mouth of the Columbia River, collected at Ilwaco, Wash.,
May 18, 1920

Females tabulated according to size of eggs, type of nucleus, and age

Ocean nuclei Stream nuclei
Logarithm of diameter of eggs (mid-value of class) Total
Third | Fourth Fifth Fourth Fifth
year year year year year

BRE DN NONON GONE WHR NTROWWO TOWN r

1 Group with log D less than 0.30, mean is 0.1474+ 0.0053; standard deviation is 0.0444+-0.0038. Group with log D greater
than 0.30, mean is 0.395.

2 Group with log D less than 0.30, mean is 0.2486. Group with log D greater than 0.30, mean is 0.4254+-0.0095; standard
deviation is 0.0658--0.0067.

3 Group with log D less than 0.30, mean is 0.2300. Group with log D greater than 0.30, mean is 0.4167.

TaBLE 22.—Chinook salmon taken by troll in Monterey Bay, Calif., June 16, 1915

Tabulated according to length, type of nucleus, age, and sex

Ocean nuclei

Centimeter length Second year Third year Fourth year
(mid-value of class)
Male Fe Total Male Fe Total Male Female Total
male male

1 1

2 1

7 2

2 2

3 1

5 4

2 =

1 oe

[se le eS eee en

ee ee ee

0cX = “6I6T ‘F euNnL

*IOATY VIQUIN[OD 9q} JO YINOUL 9y} YO [Joy Aq uoyR,
‘I6T @ “soT “uo Te ‘afemMeg ‘oInjeuUT ‘sneponu

uea00 ‘Ivak puodes SII UT UOTM[es YoouryO—7zZ “oI

0ZX “6161

Arne ary eIquinjog oy} Jo yNour oy} Yo To Aq wayRL “EIT'0 T “Boy “Wo 19
‘gree ‘ainyeurut ‘snafonu uraso ‘Iewak PAY} Ss} Ul UOWMTeS YoouryO—sgz “OW

'
G

6l “A “a ‘SA “TINng

‘GG

(FL6 0d)

02x “6I6T ‘Et ysnsny “ary eIquinjoD O@X ‘erl6r ‘2 Aine
04} JO WNom oy} YO [[0 Aq uoyBL, “1820 GT “SOT “UID 0g ‘e[BUlay {AY VIQUIN[OD 9Y} JO YINOUI oY} YO [for AQ UoyRT “Tero |G “BOT “WO 99
‘aIMYVUIUAT ‘snepnu Ued0 ‘IBVOA YJAMNOJ S}I UI UOWRS YOoUIYO—'0E “OMT Q[euleg ‘oINjVUI ‘snoeponu uBad0 ‘IvaA PAI SP UI UOTUTLS YyoouryDO—'6Z “OI

(46 90) =“S@6T “A “A ‘SO “TT9E

OSX “6161
OZX ‘6I6T ‘82 Afoe ‘IaaTY BIQmINIOD oy] Jo YNow yi Yo [Jory Aq UoyRL, “069'0 T ‘el ysnany “WANT viquinfoy oy. Jo YNoUT oy. Yo Tor Aq WoL “EEO T “BOT
ZO] “UD C06 ‘a[BUIOg “aIN}eUI ‘sneponuU Uva—dO ‘AvOA YY S}L UL UOW[RS YOU O—ZE “OL “wo 2) ‘opeUlag =«“aINyRUL ‘shepNu URI ‘IBaA YIANOJ S}t Ul UOWTRS YOOUIYO—TE "DI

(646 90) “S261 “A “A “SA “TIng

cexX “6I6I ‘¢ Jsn3ny “ZoIQ ‘s}1aNog wou [OOYM UT UAV, OZX “6I6T ‘82 Arne ‘oan VIquuNfoD oy Jo YJNoW oy} Yo [[o13 Aq woyRy,
“Ud 9g ‘OBJ. “oINyeur ‘sneponu urvedjs ‘1waA PUOdDS S}I Ul UOMITVS YoouIyO—'Fe ‘ONT "189'0 @ *30T “UL gOT ‘afVUIO + -aINyVUT ‘snopONT Uvad0 ‘IBAA YYXIS S}I UT UOUM[LS YooulyYO—'ee “DIT

(646 90) “S26l “A “A ‘SA “TIag

0X “6IGT “ ouNg ‘IATY vIquIN[OD
OcX “6I6t ‘82 Aine “ANZ BIqUINJOD 942 JO YINoUT 9y1 Yo [01 Aq uUoyRL “L670 C oq) JO YINoUT 9} YO To Aq udyRy “ud Og ‘aeyY “UMOUYyUN
“SOT “Ud 7g ‘a[BUIA «“OINJVUIUWT ‘SMefonU UIveTjs ‘vad GINO} s}t UL UOWU[RS YOoUTYO—9E “DIY AYINeU ‘shofonu uTveAYs ‘IVIaA PIG} S} UL UOWI[RS YOoUTYO—'SE “OTA

(p46 90d) “SZ6l “A “A ‘SQ “IIAg

OZX “OZ6T ‘ST ATT “OAT
eiquinjog ‘Avg soyeg Url sdey Ul Ueyey, “ZITO G “sO, “WO 0g ‘o[RUlOT 07x “6I6T ‘Re Ane {OANT VIQUINTOD 94} JO YANoUL Oy} Yo [fo Aq uoyry
“dN BULLI ‘snepnu ures ‘iweA YYy oy}? Ul Uomyes Yyoouryo— sg “Oly “159°0 @ “BOT “urd pg ‘TRUE “oINyeUI ‘shoapoNU UTeays ‘“IweA YAMOJ SIT UT UOUT~eS YoouryO—ze “OWT

(p46 20d) “St6t “A “A “SO “TIAg

02x

{UZ oy Jo vsoy} oie ssur peu a) WY} Os ‘pazyreys you pry J YWNoy oy} JO pueq JoUIM 9
AOU OL “GIGI ‘OL ABI OD oy Jo YNow oy. Yo | UL “OZ6T ‘ST ABI ‘“toany
“Soy CUulo ZG ‘a[eUlog = “aNyeuL ‘shepnu weatys ‘wad YIXIS ey} UL UOWM[RS YOouUlYO—'OF “4 “UD ¢g ‘opeUlog = ‘ainqeur ‘snoy:

1 os ‘unseq
) fAR

(16°20) +“S26l “A “a ‘Sa ‘TIng

OZX “Wd STO “TRIN “eT VMSA

YI porvdur0d 9q prnoys pur ‘snojonu jo dA} odie] oY} BULAvY ‘evVI OY) JO VATIL 0CX “Wd 1g ‘e[eUley “OT OINSIA IA poredu10d oq plmoys pue ‘snejonu jo edAy4
-jUsseIdel @ SI SIUT “6I6I ‘2% YSNBNY “SAGO ‘VIOjSY Iwo JOATY VIQUIN[OD 2} TleWs 94} SUIAVY ‘dvi 94} JO OATZLIUOSOIdaI B SI SINT, ‘GIGI ‘ZZ YSNBNY “Sa1O ‘VIIOISV IJVU JOATY
Ul oUles Aq UdyB], ‘“snejonu uved0 ‘IveA Y}IMOJ oy} Ul UOM[Vs YooulyO—ZF “D1 eIquinrog 94} Ul oules Aq Usaye], ‘SNajonu Ueed0 ‘IvOA YIAINOJ OY} UL UOUT[VS YoouIYO— IF “DWT

(46 90) + “S26l “A “a SN “TIAg

GROWTH AND MATURITY OF SALMON IN THE OCEAN 85

TaBLE 22.—Chinook salmon taken by troll in Monterey Bay, Calif., June 16, 1915—Gontinued.

Tabulated according to length, type of nucleus, age, and sex

Ocean nuclei

Centimeter length Second year Third year Fourth year
(mid-value of class)
Male Fe- Total Male Fe- Total | Male | Female Total
male male
1 2 eco ccect | eee ten tet celaccescee eke
1 pL ee ee 5 5
raceme Bh ee eee 5 5
Reeteese 1 J) 3 4
Sees A eee 2 2
Sewers |seooaeate ose 2 1 3
besccese|sanssasess=5- 1 1 2
Bote SOE Noe ek ed TV; |e2ctecuceezuz 1
POM? ccna |socaceccssseslsesecte 1 1
we soe docs e saayhoe feral | Secs Seok ee ee 1
a 30 17 11 28 ub 18 25
51.3 | 53.270. 46 | 76.41-0.96 | 75.54 | 76.07+0.69 | 90.48 | 86.6740. 60 } 87. 720. 67
Deeasnes 3.75 .83 | 5.864 .68 |.--.-..-| 5.444 .49 |___...__] 3.784 .43 5.00 . 48
Ocean nuclei—Continued Stream nuclei
Centimeter length (mid-value Fifth year Third Fourth year Fifth Total

of class) year

Male Female Total Male Male Female Total Female

Sh COW NOOO RE ARWHTW NER Wo PROWD

86 BULLETIN OF THE BUREAU OF FISHERIES

TasLe 23.—Eggs from chinook salmon taken by troll in Monterey Bay, Calif., June 29, 1915

Tabulated according to size of eggs and approximate size of fish

Logarithm of diameter Logarithm of diameter
of eggs (mid-value of Group, Group Group 3 3 Total || of eggs (mid-value of Gropp Group Group 3 3 Total
class) class)

peer eee Peery ee See re 1 3 5
Sees Par 1 7 17
1 1 4 6
ee |fee 1 8 14
2 2 10 10
10 10 di 8
6 6 6 7
7 7 2 4
3 3 1 1
4 4 —
44
4 1 5 0. 4264+0. 0059 |_--_-_

4 2 6
7 3 10 0, 0653+ . 0042 |---_-_

(jp See Caen 6

8 1 9

1 Fish under 5 pounds. 1 Fish between 5 and 15 pounds. 3 Fish over 15 pounds. 4 See page 66.

TaBLe 24.—Chinook salmon taken by troll in Monterey Bay, Calif., June 19 to 21, 1918

Tabulated according to length, type of nucleus, and age

Ocean nuclei Stream nuclei
Centimeter length (mid-value of class) : A Total
econ 6 Fourth Third Fourth
year Third year year year year

GROWTH AND MATURITY OF SALMON IN THE OCEAN 87

TasBLeE 25.—Chinook salmon taken by troll in eeey Bay, Calif., June 19 to 21, 1918. Females
only

Tabulated according to size of eggs, type of nucleus, and age
SS SESE ES

Ocean nuclei Stream nuclei
Logarithm of diameter of eggs (mid-value of class) ; Total
Second Third Fourth Th ird| Fourth
year year year year year

1 Group with log D less than 0.30, mean equals 0.1622-+-0.0042; standard deviation equals 0.037240.0029. Group with log D
greater than 0.30, mean equals 0.3887--0.0080; standard deviation equals 0.0458--0.0056.

TaBLe 26.—Chinook salmon taken by troll near Drakes Bay and Fort Bragg, Calif., July and
August, 1918

Fish that migrated as fry (scales with ocean nuclei) tabulated according to length, age, and sex

Second year Third year

Centimeter length (mid-value of class) Se ee
Female | Total Male Female Total

= SND PwW PHNWH WCNNw

vi 19 36
75. 2341.41} 74.89+1.37 75. 050, 98
8. 6441. 00 8. 84+ .97 8.744 .69

88 BULLETIN OF THE BUREAU OF FISHERIES

TaBLE 26.—Chinook salmon taken by troll near Drakes Bay and Fort Bragg, Calif., July and
August, 1918—Continued

Fourth year Fifth year
Centimeter‘length“Gnid-valueiof class) > |—_—— a e Total
Male Female Total Male Female Total
1
6
1
2
1
4
3
2
2
3
2
3
2
2
5
4
4
2
1 3
1 2
1 2
6 6
5 6
5 8
i dal (Nemes Wy Uh) er eaieiieeets Li Pee a aga 12
Lt 8 eens. oso | Pe eee 3
2 7
i
1
3
1
Dee ccawbesnceckcawecouecsoustanuauscees| paces secu eece Seo cook one ee eee eee 1
Total. .-2ielos- ese ees Be 20 30 50 3 1 4 105
IM Gansces Seanan aaa --| 102.70+0.98 | 97.2040.65 | 99.40+0. 61 109. 60 105. 00 108550)| aaa
Standard deviation 6.48+ .69 5. 30+ .46 6.402: .43 |..-2.52)-2 Ll - eee

TaBLe 27.—Chinook salmon taken by troll near Drakes Bay and Fort Bragg, Calif., July and
August, 1918

Logarithm of diameter Logarithm of diameter pis ae
of eggs (mid-value of Number of individuals of eggs (mid-value of Number of individuals
class) class)

}}Mean, 1.8667. 4
4
1) 1
2 2|Mean, 0.6244+0.0080.
1}y4 ean, 0.1778-£0.0095. i Standard deviation, 0.0780+0.0057.
3 Standard deviation, 0.0600+ 0.0067. 4
4 8
3 4
1 it

GROWTH AND MATURITY OF SALMON IN THE OCEAN 89

BIBLIOGRAPHY

Coss, Joun N.

1921. Pacific salmon fisheries. Appendix I, Report, U. S. Commissioner of Fisheries, for

1921 (1922). Bureau of Fisheries Doc. No. 902, pp. 1-268, figs. 1-48. Washington.
Dauu, Knot.

1910. The age and growth of salmon and trout in Norway as shown by their scales. (Trans-
lated from the Norwegian of Ian Baillee.) The Salmon and Trout Association.
London.

Fraser, C. McLran.

1917. On the scales of the spring salmon. Contributions to Canadian Biology, 1915-16
(1917), pp. 21-88, tables, figs. 1-15. Ottawa.

1920. Growth rate in the Pacific salmon. Transactions, Royal Society of Canada, Series
III, Vol. XIII, 1919 (1920), pp. 163-226. Ottawa.

1921. Further studies on the growth rate in Pacific salmon. Contributions to Canadian
Biology, 1918-20 (1921), pp. 7-27. Ottawa.

GILBERT, CHarLtes H.

1913. Age at maturity of the Pacific Coast salmon of the genus Oncorhynchus. Bulletin,
U. 8. Bureau of Fisheries, Vol. XXXII, 1912 (1914), pp. 1-22, figs. 1-29. Wash-
ington.

1913a. The salmon of Swiftsure Bank. Appendix, Report of the Commissioner of Fisheries,
Province of British Columbia, for 1912 (1913), pp. 14-18. Victoria.

1914. Contributions to the life history of the sockeye salmon (No. 1). Appendix, Report
of the Commissioner of Fisheries, Province of British Columbia, 1913 (1914), pp.
53-78, figs. 1-13. Victoria.

1915. Contributions to the life history of the sockeye salmon (No. 2). Idem, for 1914
(1915), pp. 45-75, figs. 1-8. Victoria.

1916. Contributions to the life history of the sockeye salmon (No. 3). Idem, for 1915
(1916), pp. 27-64, figs. 1-10. Victoria.

1918. Contributions to the life history of the sockeye salmon (No. 4). Idem, for 1917
(1918), pp. 33-80, figs. 1-15. Victoria.

1919. Contributions to the life history of the sockeye salmon (No. 5). Idem, for 1918
(1919), pp. 26-52, figs. 1-34. Victoria.

1920. Contributions to the life history of the sockeye salmon (No. 6). Jdem, for 1919
(1920), pp. 35-68, 6 illus. Victoria.

1922. Contributions to the life history of the sockeye salmon (No. 7). Idem, for 1921
(1922), pp. 15-64. Victoria.

1922a. The Salmon of the Yukon River. Bulletin, U.S. Bureau of Fisheries, Vol. XX XVIII,
1921-22 (1923), pp. 317-3382, figs. 276-302. Washington.

1923. Contributions to the life history of the sockeye salmon (No. 8). Appendix, Report
of the Commissioner of Fisheries, Province of British Columbia, 1922 (1923), pp.
16-49. Victoria.

Ricu, Wituis H.

1920. Early history and seaward migration of chinook salmon in the Columbia and Sacra-
mento Rivers. Bulletin, U. S. Bureau of Fisheries, Vol. XX XVII, 1919-20 (1922),
pp. 1-74, Pls. I-IV. Washington.

1921. An instance of adult, sea-run chinook salmon found feeding in fresh water. Cali-
fornia Fish and Game, Vol. 7, No. 1, January, 1921, pp. 7-8, fig. 5. Sacramento.

1921a. The relative maturity of the chinook salmon taken in the ocean along the Pacific
Coast. Idem. pp. 12-22, tables, figs. 8-12. Sacramento.

Smitu, E. Victor.

1920. The taking of immature salmon in the waters of the State of Washington. Depart-

ment of Fisheries, State of Washington, June 16, 1920, pp. 3-44, illus. Olympia.

BULLETIN OF THE BUREAU OF FISHERIES

90

Smits, E. Victor, and Trevor Krncaip.
1920. A report on the taking of immature salmon in the coastal waters of the State of Wash-

ington. Twenty-eighth and twenty-ninth annual reports of the State Fish Com-
missioner [of Washington], April 1, 1917, to March 31, 1919 (1920), pp. 39-46.

Olympia.
Snyper, J. O.
1922. The return of marked king salmon grilse.
April, 1922, pp. 102-107, figs. 40-50. Sacramento.
A second report on the return of king salmon marked in 1919, in Klamath River.

1923.
California Fish and Game, Vol. 9, No. 1, January, 1923, pp. 1-9, figs. 1-5. Sacra-

California Fish and Game, Vol. 8, No. 2,

mento.

WeyrmouTs, Frank W.
1923. The life-history and growth of the Pismo clam (Tivela stuliorwm Mawe).
letin No. 7, California Fish and Game Commission, 1923, pp. 5-120, tables, figs.

1-15, graphs 1-18. Sacramento.
am

Fish Bul-

SEASONAL DISTRIBUTION OF THE PLANKTON OF THE
WOODS HOLE REGION
&

By CHARLES J. FISH, Ph. D.,
General Assistant, U. S. F. S. Albatross

&
Contribution from the U.S. Fisheries Biological Station, Woods Hole, Mass.
&
CONTENTS
Page Page
Introduction._...._..2_....L-..------ 91 | General discussion of plankton—Contd.
Methodsaets iis puke 93 Crustacea—Continued.
ocation= 2... -2--25-2.---.-----.--2- 96 Copepoda______------------- 141
Salinity and density_________________- 98 Cirripediaes= eee eee 147
Temperature__...._..........-_-_---- 100 Arthrostraca_---------___--__ 149
General discussion of plankton________- 101 Cuma ceats 2 ee a ee 152
Diatoms and other plants_________ 104 Schizopoda and Stomatopoda__ 152
Brotozeaseia. tech: aod 2 ee 121 Vie CTU ere ee eee iss
Coelenterata_________------------ 123 rach yur 2ss= == ee 159
Annulata and Vermes_________---- 130 Pyecnogonida and Xiphosura___ 161
Mollusca: 325-5 eee 136 Chordatasees .a22 ee eg ea 162
Echinodermata_______-___-----_-- 138 1 ib ts] oe one Sirs oe ee Neen eae, 164
Wrustaceas = Ss= 525-8 ose eo eek 139 | General conclusions._____--_---------- 172
Rh yillopodasss se seen iUSI2) || Lehloylnoyse pe Volah joe Se 176
Ostracodaesee 5 2 oe see ee 140
INTRODUCTION

In the plankton section of the report of the Conseil Permanent International
pour l’Exploration de la Mer, published in September, 1922, it was pointed out that
greater attention should be paid to the seasonal variation and range of marine
plankton. As early as 1880, Prof. S. F. Baird remarked to Commander Z. L.
Tanner, after the initial cruise of the United States Fish Commission steamer
Fish Hawk, that “the profitable study of useful sea fishes can not be prosecuted
without a knowledge of their food, the food of their food, their respective friends
and foes, the habitat of the several species, and their means of passing from one
region to another in the embryonic as well as in the adult stage. The temperature,
currents, and specific gravity, also, should be studied in connection with the migra-
tions and habits of pelagic forms.” Since that time only one area of the Atlantic
coast of the United States has been investigated with the object of completely
surveying and determining the distribution of the plankton, currents, salinity,
and temperature. The interesting results of these investigations, which were

91

92 BULLETIN OF THE BUREAU OF FISHERIES

carried on by Dr. H. B. Bigelow, are published in a series of bulletins from the
Museum of Comparative Zoology at Cambridge, Mass., and a more complete
account of these investigations and explorations is now in process of publication.

It has long been known that Woods Hole occupies a unique position on the
Atlantic coast. It is the northern limit of many southern forms and the southern
limit of many northern forms. Oceanic animals, also, are often carried into this
pocket on the coast by the southerly winds and strong tides that prevail in the
summer months. Yor that reason Woods Hole was selected as an ideal location for
the study of plankton and the interrelationships of the various pelagic faunas.

Under “plankton” I have included all animals occurring in surface collections,
whether free-swimming or carried by currents. Such a broad definition includes a
great many benthonic forms carried from their natural habitat by storms or high
winds, but in a littoral region one can not always decide accurately which species
have been accidentally carried to the surface and which are free-swimming.

The present paper is the result of a continuous investigation of the plankton
in Great Harbor, Woods Hole, Mass., covering a period of two years. The purpose
was to make an exhaustive qualitative study of the plankton of this region, the
seasonal distribution of the various species, their interrelationships, and the general
factors governing their distribution.

The investigation consisted of three parts: (1) An examination of plankton
samples taken daily during the years 1899 and 1900 in Great Harbor by the late
Vinal N. Edwards, collector for the United States Fish Commission; (2) a survey of
all records of surface collections of previous years; and (3) examination of living
material taken daily in surface collections in Great Harbor, observations on tem-
perature, salinity, and other factors governing the seasonal distribution of the
plankton, and a survey of the general geography of the region as a factor affecting
plankton distribution.

The first part of the investigation occupied the entire time of the author during
the year 1921-22 and was carried on in the biological laboratory at Brown Uni-
versity. Many of the fragile animals had become disintegrated during the 22 years
in which the material had remained untouched, and the preservatives in some of
the samples had evaporated. Over 200 vials remained intact, however, and offered
ample material for study.

The second part of the work involved much time and proved to be a very
tedious task. The results, however, were very important, as they covered the
daily records of surface collections extending over a period of 15 years—1893 to
1907, inclusive. The larval fish and celenterates taken during this time had been
carefully identified by Vinal N. Edwards. Diatoms, copepods, amphipods, annelids,
and other planktonic forms were recorded as groups, the relative abundance for
each day being carefully noted. Complete records of the weather, wind, and
temperature for most of this period were available and proved indispensable in
explaining peculiarities in the seasonal distribution of many species. This part of
the work was done by Marie D. P. Fish, who aided me in the study of the larval
fish also.

The final part of the work was carried on from June 22, 1922, until December
31, 1923, at the laboratory of the United States Bureau of Fisheries at Woods Hole,

PLANKTON OF THE WOODS HOLE REGION 93

' Mass. From June 22, 1922, until May 1, 1923, observations were made daily at
the same spot where all my previous material had been taken. Fortunately a
series of collections had been made by R. A. Goffin during the spring of 1922. From
these I was able to trace the first appearance of the summer species. From May 1
to December 1, 1923, the collections were made three times a week, except during
the interval from August 22 to October 4. The records for the past summer are
therefore not as complete as those of 1922, although they serve as a basis for com-
parison.

A kind invitation from Dr. P. S. Galtsoff to assist him in his monthly surveys of
Long Island Sound from September, 1922, to August, 1923, made possible valuable
observations on the distribution of certain pelagic organisms, particularly the
diatoms, in relation to their presence at Woods Hole.

It is a pleasure to express my especial gratitude to Prof. A. D. Mead and Prof.
R. M. Field, of Brown University, who furnished me helpful assistance and guidance
throughout my work. I am especially indebted to Marie D. P. Fish for her careful
tabulation of Vinal N. Edwards’s records of surface collections and temperatures
collected over a period of 15 years. I am indebted to Dr. P. S. Galtsoff, who made
possible my observations on salinity at Woods Hole and the plankton of adjacent
regions, and I wish also to express thanks to Dr. Henry B. Bigelow, Dr. Hugh M.
Smith, Dr. Paul Bartsch, Dr. Albert Mann, and Prof. A. E. Verrill, for helpful
advice and criticism rendered at various times during the progress of my work.

METHODS

My first plans provided for daily observations on temperature of the air and
water (surface and bottom), salinity, oxygen, wind, weather, sea, transparency,
vertical hauls, and surface and bottom collections with plankton nets of No. 2
and No. 20 bolting cloth. Because of the amount of time required to identify
the many species of zooplankton and phytoplankton it was found desirable to
discontinue certain of these observations. The following schedule was finally
adopted:

1. Daily temperatures of surface water and air.

2. Salinity (at certain periods) and density.

3. Daily meteorologic observations on wind, weather, sea, etc.

4. Vertical hauls at weekly intervals with No. 20 net.

5. Daily surface hauls with No. 2 and No. 20 nets. (Later, No. 20-net hauls
were reduced to twice a week except during the diatom maxima.) Nets 3 feet by
12 inches with a brass bayonet-lock bucket on bottom were used.

The temperature was taken each day at the time of setting the plankton nets.
A series of observations later proved conclusively that at all times the bottom
temperature at my station is exactly the same as that of the surface (Table 2,p.101).
Bottom observations then were made only during periods of rapidly declining or
rising temperatures.

For a period extending over four months salinity was determined daily by
titration with nitrate of silver. When these could not be made at once, they were
preserved in the standard “citrate of magnesia’’ bottles of the sort used for that

94 BULLETIN OF THE BUREAU OF FISHERIES

purpose by the United States Bureau of Fisheries. After it was found that there
were usually no important variations observations were made only on certain
occasions to indicate the influx of Gulf Stream and other ocean water. Had it
been possible continuation of the daily tests would have been very desirable.

Observations on the condition of the weather, sea, wind, and sky were taken
daily. These factors are of great importance, particularly the winds, in determining
the distribution of planktonic animals.

Vertical hauls were made weekly, but they yielded rather disappointing results.
The water is only 11 feet deep at low tide, and for that reason a very small net of
the Birge type, with a special bucket, was adopted. The material collected was
centrifuged for two minutes at about 1,000 revolutions per minute in a graduated
glass tube, and the result measured in cubic centimeters. The figures obtained
are not included in this report because I did not have time to make individual
counts of the various species, and the total mass was meaningless, being made
up of diatoms, dinoflagellates, particles of dirt and detritus, larval copepods, larval
mollusks, and an occasional adult copepod. All the large planktonic forms had
successfully evaded the net as it was being drawn to the surface, and the resulting
mass did not give a fair estimate of the amount of plankton in the water at the time.
To get these various-sized animals, a series of nets of at least 10 different meshes
would be necessary, and even with these there would be so much overlapping that
the results would be of little value. The pump has not succeeded in overcoming
this difficulty in the case of the marine plankton. On eight occasions during the
past year I centrifuged over 100 samples taken by pump in Long Island Sound,
and invariably the deposit contained a larger proportion of smail forms and a
smaller proportion of large forms than did the vertical hauls made at the same
time. A successful method of accurately determining the real volume of marine
zooplankton as well as of phytoplankton is yet to be devised.

The most valuable results were obtained with surface nets. The waters are
so churned up in Great Harbor that there was no difference in the collections taken
at the surface and those taken at the bottom, except that the latter often contained
more sand and small detritus. For that reason the bottom hauls were discontinued.

The daily routine of plankton collecting and investigation, consisted of three
parts. First, the nets were suspended from the end of the dock by means of pulleys
attached to outlying piles in such a position that one was suspended in a northerly
direction and the other in a southerly one (fig. 1, p. 97).

When the nets were hauled the contents were emptied into a flat glass dish
entirely covered with black paint except for a small area at one corner. A tight-
fitting top completely shut out all light except in the corner over the clear glass.
A light placed at this end caused all the Crustacea, larval annelids, and, in fact,
most of the free-swimming planktonic organisms that are positively phototropic
to crowd at the lighted corner, where they could be picked out individually with
a pipette or drawn out in bunches with a long glass tube and deposited in a watch
glass or petri dish for examination. A second collection was then made from the
detritus in the bottom, consisting of dead organisms and any forms that had not
been attracted to the light. Finally, the last bit of sediment, after all the rest of
the tow had been poured into a silk bag to be strained, was placed in a dish. This

PLANKTON OF THE WOODS HOLE REGION 95

was often found to contain large numbers of small mollusks, ostracods, and Foram-
inifera.

After the living specimens had been observed they were killed with a 2 per
cent solution of formalin and reexamined. The species not readily identified were
placed in separate watch glasses and subjected later to a more careful examination
with a higher-power lens. For a general examination of zooplankton a binocular
microscope with low-power lenses (Nos. 55, 40, and 24) is very satisfactory. Smaller
forms were mounted on slides and examined with a compound microscope.

Several samples of phytoplankton were placed in watch glasses and examined
alive. This made possible a rapid survey of a large amount of material. Next
some of the material was mounted on slides, with barium mercuric iodide as a mount-
ing medium, and examined with a higher-power lens.

The common species were tabulated daily on charts, records being made of
the rarer specimens. If these began to appear frequently, they were given a place
on the chart. This method proved to be very simple and convenient. The material
was later put in 2 per cent formalin and labeled for future reference.

The direction of the currents in Great Harbor during the flood tide ©
(fig. 1, p. 97) was determined in two ways. The first method was very simple,
consisting of observations made while great masses of broken ice were floating
through the passage during the spring months. The results obtained in this way
could be checked up as often as desired. The second method was used to determine
the smaller currents near shore, and the course of the back eddy along the shore
of Nonamesset. This was accomplished by placing large quantities of shavings in
the water on a calm day and plotting the courses which they took. Theresultsmay
not be entirely accurate in minute details, but they show the general movements of
the water in the harbor during flood tide.

The combined results of my observations on material of 1899-1900 and those
of Mr. Edwards have been plotted on quadrille paper. The charts based on the work
of the past two years are on Keuffel and Esser No. 334D graph paper.

A great difficulty presented itself when I started to assemble my results. In
qualitative work the greatest amount of material possible is essential, and the
only way to obtain this is by surface towing, which obviously does not lend itself
to any accurate measurement. [ven if figures could be secured the daily variation
in the winds and tides is so great at Woods Hole that the results would be more
confusing than helpful. One can state when the first specimen of a species appears
and when its season ends, and the fact that the numbers may be increasing daily
can also be seen, but to present this information in a satisfactory manner is difficult.

The plan finally adopted consisted of the use of four categories—very scarce,
scarce, abundant, and very abundant. These served as calibrating points from
which the seasonal distribution of a species could be plotted in a fairly accurate
manner. Of course, the basis for measuring the abundance of copepods was not
the same as that for the diatoms; 500 of the former might be considered abundant,
while the same number of diatoms would be considered very scanty. Again, 50
specimens of the oceanic annelid, Tomopterus, would be considered abundant,
but 50 specimens of a common copepod would be thought scarce. The measure-
ment, therefore, is relative; that is, the symbol given to a particular animal for a

96 BULLETIN OF THE BUREAU OF FISHERIES

particular day indicates its relative abundance for that day compared with its abundance
for all the preceding days or weeks since its appearance and is not to be compared with
that of the species of any other phylum. To eliminate as far as possible the confusion
arising from daily variation, three-day averages were used in plotting the points
on the charts. There may be objections to my method of presenting the data
in graphic form where definite figures were not available. However, I feel that the
seasonal variation can best be shown in this way, and that any method which sim-
plifies the work and makes it more easily understood is justifiable. The symbols
used on the charts are as follows: V.A., very abundant; <A., abundant; S., scarce;
V. S., very scarce; and N., none.

LOCATION

All material for the present investigation, with the exception of a few observa-
tions made in Vineyard Sound, was obtained from the water at the end of the Bureau
of Fisheries dock at Woods Hole, Mass. This spot was selected, first, because it
offered such excellent possibilities for qualitative plankton investigation, and,
second, because the bottom fauna, whose larve make up a large percentage of the
summer plankton, had already been carefully surveyed.

The location is an exceptionally fortunate one for an investigation of seasonal
distribution, although impossible for a study of diurnal migration. On the flood tide
the local current rushing through the narrow passage of Woods Hole sometimes
reaches a speed of 8 miles an hour. Figure 1 shows that one of the three main
branches of this current heads directly for the Fisheries dock. Here it divides,
one half turning to the south and the other to the north. By placing nets at the
two ends of the dock one hour after the tide has turned to flood and hauling them
one hour before the ebb it is possible to have a strong current of water passing
through the nets continuously for four hours. More material can be collected in
this way than would be possible in several hours’ towing from a boat. To deter-
mine the complete pelagic fauna of a region, the largest possible number of daily
samples are needed. Even then many scarce forms probably pass through their
seasonal cycle without once being observed in surface collections.

Another advantage of the position of this particular station is the uniformity
of the plankton both during the day and during the night. Extended observations
showed that the mixture of the waters during the flood tide so churns up the plankton
that there is almost no difference between the hauls of the day and those taken in the
evening. I know of but two exceptions to this statement. These are the amphipods
and certain annelids, which remain under rocks in the daylight and emerge after
dark. Then they are picked up by the strong currents and appear in the greatest
numbers in evening collections. As these are not true pelagic animals, they do not
seriously affect the problem. Thus, the collections made at any time showed equally
well the representative plankton for that day.’

The features of the coast adjacent to Woods Hole have much to do with its
fauna. It has long been thought that the arm of Cape Cod to the east constitutes a
barrier that changes the course of the cold northern ocean current and deflects it
away from the continent. Not all oceanographers agree as to the above, but even

97

PLANKTON OF THE WOODS HOLE REGION

Fig. 1.—Currents of Great Harbor, Woods Hole, Mass., during flood tide.
, nets,

Light shaded area, shallow water; black area, land; +, rocks; @, location of plankton
Scale in statute miles

98 BULLETIN OF THE BUREAU OF FISHERIES

if this is not true, Verrill and subsequent authors, including Bigelow (1914-1922),
found that the coastal water temperatures north and east of Cape Cod were very
much lower in summer than those south of it. None of the planktonic animals
common north of the cape appear south of itinsummer. In winter, however, the
cape does not form a barrier for the neritic plankton, which often appears at Woods
Hole in great abundance.

The Gulf Stream lies about 85 nautical miles off the Massachusetts coast,
just beyond the end of the Continental Shelf. Between this warm area and the
mainland there is a broad belt extending from the north. Some consider this to be a
continuation of the Labrador current and attempt to explain faunal distribution
on that basis. Others consider it to be mainly a contrast belt between the warm
littoral zone and the Gulf Stream. According to the latter viewpoint, the Labrador
current does not extend south of Newfoundland. No matter which theory is
correct it is evident that this broad belt is affected on one side by the southerly
winds and on the other by the unusually strong tides of this region. Any forms,
then, that have blown in from the Gulf Stream will be carried farther inland by the
moving water. This peculiar alliance of wind and tide probably explains why
much tropical plankton, which is taken so often in this locality, occurs at no other
points along the coast.

Woods Hole also forms the northern limit of most of the southern boreal
pelagic animals. Many copepods and celenterates, of which Mnemeopsis is a
striking example, occur often in Great Harbor but never farther north along the
coast. Thus, it is clearly evident that Woods Hole is a very unsatisfactory spot to
work out the characteristic pelagic fauna of the north Atlantic coast region, for not
only northern and southern boreal types appear with the littoral plankton at certain
seasons, but the Gulf Stream and other oceanic forms are likely to be carried in at
any time. Again, the swift currents rushing through the passage produce local
peculiarities in the plankton. However, if we desire to study the conditions at
Woods Hole as a special problem and try to understand the conglomeration of
faunas, their interrelations, and the factors governing their appenrane and dis-
appearance it becomes highly interesting and instructive.

SALINITY AND DENSITY

The salinity at Woods Hole normally varies comparatively little throughout
the year. No streams of importance empty into Great Harbor, and as all the
larger rivers of Buzzards Bay are situated at the upper end the salinity of the
southern area is not sufficiently different from that of the sound to have any appre-
ciable effect on the plankton.

Titrations made almost daily from July until October, 1922, during the flood
tide (Table 1) indicate that the water entering Great Harbor is of a slightly lower
salinity than that of Vineyard Sound, found by Bigelow (1915) to be 32.2 per mille
in August, 1913, and by Sumner (1913) to be 32.2 per mille in August, 1906. In 1922
the average salinity at the Fisheries dock for late July and August was 31.57 per
mille, and for September and early October, 31.03 per mille.

PLANKTON OF THE WOODS HOLE REGION 99

TaBLE 1.—Salinity of surface water at Woods Hole from July to October, i922

Degree Degree Degree Degree Degree Degree
Date of Date of Date of Date of Date of Date of

salinity salinity salinity salinity salinity salinity
July 27_.- 31. 53 || Aug. 6_--- 30. 30 || Aug. 16-.- $1.58:|| Aug. 27... 31.85 || Sept. 8___ 31. 22 |} Sept. 24__ 31. 40
July 28___ 31. 46 ||-Aug. 7_.-- 31. 20 |} Aug. 17__- 31. 53 || Aug. 28__- 31. 31 || Sept. 9__- 31.18 || Sept. 27__ 31. 44
July 29___ 31. 62 || Aug. 8_--- 31. 29 |} Aug. 18__- 31. 82 || Aug. 29__- 31.49 |] Sept. 10_- 31. 33 || Sept. 29__ 31. 40
July 30__.| 31.31 || Aug.9_..-| 32.01 |] Aug.19.--| 31.60/| Sept. 1_-- 31. 36 || Sept. 11__ 31. 02 || Sept. 30__ 31. 62
July 31-—- 31.31 || Aug. 10__- 32. 01 || Aug. 20_-- 31.65 || Sept. 2__- 31. 36 || Sept. 12__ 31.06 || Oct. 1__-- 30. 88
Aug, To... 31.31 || Aug. 11_-- luca || eAug. Zee 31. 85 || Sept. 3_-_- 31. 09 || Sept. 13__ 31/15) |} Oct 22... 31.35
Aug.2....| 31.56 || Aug.12._.| 31.82]) Aug.22._.| 31.85 }| Sept.4.--] 30.91 || Sept.14_-/ 31.18] Oct.5_-__] 31.35
Aug. 3--.- 31.71 || Aug. 13_-- 31. 53 || Aug. 23__- 31. 85 || Sept. 5__- 31.18 || Sept. 16_- 31,15 }| Octz10_.- 31.49
Aug. 4... 31. 64 |} Aug. 14__- 31. 60 |} Aug. 24___ 31.65 || Sept. 6__- 31. 04 |} Sept. 17_- 31.06 || Oct. 11___ 31. 20
Aug. 5_..- 31. 46 || Aug. 15_-- 31. 67 || Aug. 25__- 31.71 || Sept. 7__- 31.49

After southerly winds a slight increase in salinity usually can be noted. This
would naturally be expected, for the outlying waters always have a higher salinity—
in the case of the Gulf Stream upwards of 35 per mille. It was to determine to what
extent this influx of ocean waters takes place after storms that the titrations were
made in Great Harbor. They covered the period when most tropical oceanic
animals appear in the plankton. The results showed that very little change takes
place even during hard southerly winds unless they extend over a long period of
time. This is probably due to a dilution resulting from a mixture with the fresher
waters of the southern part of the bay. Marked changes may have occurred in
Vineyard Sound but were not evident farther inland.

On August 6 and 7 a heavy southwest storm took place, reaching its height on
the second day. During this time the wind blew continuously and much Sargassum
was noticed in the sound. A slight increase in salinity from 31.29 to 32.01 per mille
on August 9 and 10, followed by a gradual decline, was the only evidence of outside
water, and this was below the usual average for theSound in August. However,
this again may have represented a mixing of the bay water with that of a higher
salinity than is usually found in the sound.

Hard southerly winds extending over a long period of time replace the waters
of the region to such an extent that the dilution by bay water is hardly noticeable
except after a hard rain or a period of melting snow. This was shown by the
density records during the spring of 1922. Figures 2 and 3 give the daily variation
in the density at Great Harbor, taken by Mr. Hamblin at the Fisheries dock at 12
o’clock noon. As these unfortunately have no relation to the tides, they can only
indicate in a general way the conditions existing at any particular time. Standard
hydrometers, certified by the Bureau of Standards, were used, the error being
probably not greater than + 0.0001.

The density in shallow waters is governed by two factors, temperature and
salinity, the comparative influence of each being clearly shown in Figures 2 and 3.
A comparison of Figures 2 and 3 with Figures 4 and 5 indicates the effect of the
temperature. During the warmest seasons a minimum density is found, and
during the coldest months it reaches its highest pot. Were there no change in
the salinity the curve would rise and fall evenly, corresponding to the rise and fall
in the temperature of the water. The sudden increase or decrease in the curve at
any particular time is due to an increase or decrease in salinity. As previously

100 BULLETIN OF THE BUREAU OF FISHERIES

stated, there are no rivers in the immediate vicinity of Great Harbor, although
melting snow and hard northerly winds cause the sudden appearance of waters
of comparatively low density. Prevailing southerly winds extending over a long
period of time cause high density. In the spring of 1922 (fig. 2), combined with
the usual low temperature, the density almost equaled that of ocean water and
remained that way until the middle of May.

The effect of melting snow shows clearly (fig. 2) in the first week of April, 1922,
and (fig. 3) on January 2 and 3, 1923. On the latter dates a marked increase in the
number of diatoms was also noticeable. The greatest change took place on March 31,
when the density dropped from 1.0260 to 1.0244 in one day. A heavy snowstorm
had occurred on March 30, followed by rain and snow on March 31 and April 1.
The rapid rise took place during a period of constant hard southwesterly winds.
The extreme point reached on April 9 (1.0270), accompanied by a drop of 1° in
temperature, is impossible to explain on the basis of local conditions. Southerly
winds prevailed, but were not unusually strong. Some hydrographical change
beyond the limits of the immediate region must have accounted for it.

It is probable that the salinity plays little or no part in the seasonal distribution
of the planktonic animals of this region. The fresh waters of the upper bay no
doubt form a barrier for the oceanic species and the brackish water forms probably
do not go far out tosea. Such conditions, however, are not found in this immediate
region.

TEMPERATURE

The subject of the temperature at Woods Hole and adjacent regions is so fully
discussed by Sumner in his report that only the particular conditions existing in
Great Harbor during the past two years need be considered here.

Figures 4 and 5 show the variations in the temperature of the air and water
for the years 1922-23, inclusive, to December 31. The figures were obtained from
the records taken daily at 8 a. m. by Mr. Hamblin, of the Bureau of Fisheries.
This hour was selected because it eliminates the temporary midday rise of surface
temperature typical of all shallow water. Figure 6 was compiled by Sumner to
show the mean average temperature of the air and water for a period of five years.
A comparison of this chart with that of the past year shows many important points.
The fact that Sumner’s chart is based on noon records must be considered, although
it probably had little effect on the water curve. The highest point on this curve is
on August 12, when the mean temperature was slightly over 71° F. The highest
point reached in 1922 was 71° F., on August 8. The curve for 1922 agrees well
with that of the average temperature for other years. The lower point of the
latter (30° F.) was reached only once, on February 19. In 1922 the curve fell
below this on two occasions (January 25 and February 17-19), and reached it on
February 4.

During the spring of 1923 very unusual conditions prevailed. The tempera-
ture went below 30° F. on January 29, and never rose above this point until March
14. Throughout this period the temperatures fluctuated between 28.5 and 29°
F., reaching 28° F. on February 24. This unusually cold water, occurring for such
an extended period, accounts for certain peculiarities in the plankton of the present

OCTOBER NOVEMBER. DECEMBER.

EPTEMBER.

30

15

10

20 25 gor 5 10 15

5

~

10

| HHEEEEEELEEU HT HAESHHERREEEUU ETT EEEETEEETLE

20270
1.0265
1.0260
1.0255
1.0250
1.0245
L} 1.0240
1.0235
1.0230
1.0225
1.0220

Beams

Tn TM
tii PEURERUSRUEAUERESDEUUIEUERULLIILLER
PERU RERU EAA DEA EAU ES TUERA EEA ESUEERT EET ERT TEETER I EAIIIII
RET TER ERT EST ERUERA ERI EREEL TELE ail
EEDELEERIERLERUEA U2 LUEERUEAU LEAL ERUELUEAIIEAEAUAMIEAIIE

SERAULERCAIIERATI Da TATIUESAASTSERATIEAHAITISTARIISE0ITI1

SE ET tt

SOA OTT iiee eH niGgiinr

=>

EREEEIERALHEL een
HEDEALFALUDISSUTISIVERA TENT ESITSILEAIESIITEA15> ERITIE100

ESS YESS CT a a CO) GNF eS (SD A CG PE ae SETTITIIITILLLL StH LTL BS ees

Sup atatntianii HE
Ni
i | | SUELIUEEIESIEIS EI

y¥. (Face p. 100.) No. 1

5

oF

8242°—

\tu

1,0270

1,0265

21,0245

21,0235

1,0225

21,0220

JANUARY.

13 1 15 0

35

Eau

FEBRUARY. MARCH. APRIL. MAY. JUNE. JULY. AUGUST. SEPTEMBER. OCTOBER NOVEMBER. DECEMBER.
Yo EES ed AS ie STO RTS 90 AS SMES tO SUS miso SUAS NORINS: 500 11s) ian) aS, S80 es) tO its: iso) lag. 90d 1g) to 3S, 3038) yl 101s 0 IMHO Isieniz0! (9g). i990) ie; no) 1s: 30) as) go 6 Se to ee
foes. es eed ares are | ~~} j- ef Te++ef {-++-{-++-} - — J -- efor} jf — _
Tins anes teste Fes es es (esata Eaasina aes ean
saa neees beeed 1,0270
teen} nee}
el +
i Beeet beeed Saeed ra
he es pony | J++} teh feet cre prveeteees pares} fees ze paca
ioe eaat i | baal based
+] 1.0260
pores} to pote leet p+ joa te ba pete a ay +--+ oy
SINE aa eel
PSE = +] bd baad Panel eel T pasa {fu o2es
Se jena C a pened ea asst f
= ae aoe DS eee Bee t jibes pe ee ESI an Be peoea| ngess based
/ES320 ere 1 1 | 1.0250
5 DS La Dal BSS [ese “| based bee rt =a
[aes [ee aA [eal oe
= it 1.0245
es t Hy 7+
hae ese alee Lee ol bessi besa bee eile Eesiieral
( zzz “I 1.0240
osetia band poset 4 pened bess } | af
ma Fs name eee ee aed ie | = =
cde — peed basa based 1.0235
Bs IE +f .
x pa baal 1,0230
— os nanan + aaa i panne
Lr feed eee pesed 2] + +]
re a + Li} ES eoni baaed feat
=i besed po | cy] pesea 1,0225
‘cane | ‘meee | paaaaal Gene | c| anenendl Tenemel |
[ae I ba =} ~
Aged on 1.0220
8242°_254. (Pace p, 100.) No. 1

Fic, 2—Dingram showing density of water at Woods Hole, Mass,, for each day of the year 1922. Observations made at 12 0’clock noon. Density at the temperature in situ

i SATA NE WOaAM ON AAU ARIA

a to seed . F os
A 4 zt ‘

viernes eins ape ge ae m
vheabeiey ped aaa aoe Ay ate Vo
| 4 / {
petits ‘ 24: ate .

i
Wee Sethi wane fre

i foree F mas
A NTO ji 15
ep vee any ak wry tafe aad

i
swiss el ey WE eal Ba Ta a baie

i

i '
Re ele ye (att | Ca et obese
fools 4 1

fie eee PPI

er ee

care epee dunes

hat ie

I i
‘aden te let Kanone ft oe) iP aee ee re | ede Coen
t H }
> pace tip epbigi = ace ileal Sia ic te Se ee a anal
y Eumuoeee wens | nN Wee Pare eee in| efi
| | i } i t
‘ ‘ Penne oo adnan A tates
}
+ sehen Ba vy ® epee ed flaed
i ee ae ee ee whi4
# pan: aaa. ans wbhis.
' “1
1 a ee Ws
t an ; { i ; 1 ? a,
i yee ey eats easy . as _ pres £ =i
' : awd,
Ei cing serpin cbt cy anche Na Je cen iy Se rea e a

i ;
srh/nds Oh Whandh inineieedad Palade

|

pews ferrepinifeieniana conan ae “1
pitint il dysquaticn 1 bi ppbete £35 Cree |
' | { bial ;
‘ ~ Pi few dd ee ORY whe “t+
to gee } vast | db ed ert
ied or aee oe nad teak a. aicatea dlaal
! } } } L F }
\ : | Laat 4 q
f Won ee Maley
i Al
pe t
{ H
ia tees
{
whe 4 Likes

=

+ ain wey gaye

#3

i
a
»

4
4
;

2"

ides

5

: euetecear

t

ate

t
So. a ae

ih

'

|
}
} :
pa ike Picre sas ld tS

nee

+

iho begs

{
nests

Ee an eee on ee Ca Cee Ses

ae Sati
i

ee
} \ ’
- ee ne ee eee eet
i \ i
} : ae BOE
J

ee Oe ed

'
BOI Bs iy oa aero d = Tbcped
‘ ; +4

rear Rea | tk Joy

1 Sat dae a

: i ;
’ Cae ere hone tees) Los na act oe en 02
bee 4 dumapthaytcee- bat jij a ab wey
f : } i t
ro mines - +4
{
wy 1} . 4 ne horwaw A a
1) t { i
oS ee ee cere ee oer Peon
ti H / oval
ce ae Le aA ae ste: cosets

OCTOBER. NOVEMBER. OECEMBER.

;PTEMBER.

1o

~
°
>

°
&

w
~

1.0270
1.0265
+0260
20255
1.0250
0245
1.0240
2

0230

1.0225
0220

i

IHIIIIIIIE BRTEEEUESZaE

EEELELIELISTEDIEII GIGI TAIEAI Aix: Hil

BAS SRTTTETIETERILH
EUEREEA LALA LELUSDESEELLLRIUIITA Eddies HDI ESEEan HELI

HET ETE PT
&
SADEERINEAIEATIESUEINITA11<S0RERIENIIESIIINA

=

| ST Re
gz

SUETERIEII

EEUDDELUIIEEEUTEREATIIEEAUIIELAL ESAT IED ceSAen Eee ODInaL

—]
SAUTTLELIEAEATLUEA IA TATA DEA TAEAIEAEA ERI .c7 O20 0010200001
ttt a | f petit ia

9

No.

(Face p. 100.)

257.

8242°—

situ

JANUARY

FEBRUARY.

SEPTEMBER. OCTOBER NOVEMBER. DECEMBER.
15 jose ss ets ess es es tess fm St gw sew sss ee et os YI $1 1s 37 45
Sis |e bet = = ~ — — jl...
port poset tenet [
Hl a Balen clia ie
~e rerefefiretisye] poset asad i
sl j + | +—-}oeae] |-—~-}+—+-} sete tee linet t+ = Pattee as t a ean poy 23 pres t
A et SESS eee
SSlbsltete clientele
Eaies feaieeleetle elena i
bees E34 ——
ee re Ste Si a SEES) ese
aa es =e See P= = Sees
esate is = testes Sam =e
eae E teste Pelee Ceabesars near fa es
Cares ial ta Sia eae eae lee
i — etepererteeet a — Ptreet ree}
- anes} stretereet —
ee feteet en eahieresiNet = isseleslea ta seaeest
: ie Soles ee teste ieee
a } Wests) i Tea Fassel ris i noaat
1
eset Saas
a based pees lesed 2 eel Lt
ES Bares! ie ee Peace
etal Bes erie
Steals
ES! Beles ite aie
Fst ea
= Ee 4 Seacinios —
Fiq. 3.—Diagram showing density of the water at Woods Hole, Mass., for each day of the year 1923. Observations made at 12 o'clock noon, Density at the tempersture in sita (Face p. 100.) No, 2

1.0270

1.0265

1.0260

1.0255

1.0250

1.0245

1.0240

1.0235

1.0230

1.0225

1.0220

HOAAM Sm OVARY

pie + of BE
SAAR ey AAMT 2

; eee weenie Oru
it See cand lncaty Se iy apap ae
t 4

Beet gape Cae caida a hil

Ale @ 5

=e vin denon oplitiruiei ibe eth

Ie vine toot eine nds Bike
H

; (+
Airc hSiatehe O Pink ee aes Deh 9 .

eee yt

ie pm On Waly emp ea beh
,
ine ag Ha dn adipaapdes hha
tes heget omebané 6

el Cope gan AA yg washout plenwatucted int
ne es eke | ftw vi oareanetipcaanttia std Seoul |

sir Fave nm fovea Oi ima Rebel al atid hye

avin legeee oe adi eee SM eee ridadatie teeta geht He thannle.
1
-spuyactpordelliersoetoeh Toten a) ‘ sacle booties Vadose wabroah Wilece fm dace iain
' { j
SPO (eo Sk et Many Ree Pe PE Sey ee DDE eee Fee ree! in"
| “oy }

i

‘ opm Soe ree eR ee fale eM hag aC eta aeaeefiens 7 fre ators a Pe aie er
! : ; | j | f | )
§ G+: i Rasorcatrevnvitrt Povn fee prerri fh ung sha big eee eee cy bape ©! ae F
+4, i} . eget ees ee er ee ee ed aa vepeenal etree epve trates

i
re! vert saad Bak

f ~ Wgreve any re
|

) en BWC
Ki : “ Oe ain

ia

estan} Joa SES Se ai hae ee aie

|
ae
Hl: oe ire sr ”

Amare eaewny

- 1 hafey te
os,

‘ieee be tapered ee

"ge

one} : aie, > i" ih
ah epee ES = 7a ment BY vemos
; eve (bE Cet whee ere

sta py ore aap 9 Sibi

bah

Lt f np tie meme erent) my ee eee ee Se nseva ania acl eels 14

| ee: ;
e544 ett) man's "| a | ephy
greg Hh nermenper hae trian 9 Ca Ps Se wipes

od aks sara ek CEN CH 5 etal exh rad

saab or i Be aia cook th ee ; a aces

antl. Hadith sap nbes? Bre Pn alee vert te aide

OCTOBER NOVEMBER. DECEMBER.

:MBER

“0 °0 at ce}

+
ce

a o °
wo
wn fo] x

60°

L} 35°
30°
74

w wo
Ht HIERIELINIAII Us
a ee
iil PECEle SchUesha cues sasaereninik

EEEESSEE085i00—0--—-—-=="=
aot tt ete tt tT tT ft |
ERERESRBNS 2 aaa See

fern —

PPB EEEE T a ]
Billtscsccesssseesss
ae

agen a= ———
BEVIERREAUEAREAEUAGA 02S 0iuns=-SRIEEIRCEIELEIRETIEI00001
| ERS aRREERS
AEE eer EE EE
e==an Gene

aet ans EL
CETTE eS
PEFDLLDPIDMETDENTED anazaassseeifelNMEDIEDIEEIEE ili
EL SCELISASASES sat ETILITLLLELiL
HEE eee
iS Qw a =—

Oe ene eres sansaal ela leatataieaaieasieataila
BUERERL TEN s ce 22222" TERIEEELILIEIIATD EAI OLEATAGnISaTanD
sn2222== i

HUERTESU2013==-sansSEUSRTEATEINZILE ELLE

a a SS eed od BS SS ee Ge eee eS Ss id ET a ES ee ee

SUEEELI EDIE ID222=2-=: UHUTITTIT ITE EE
SETTTEEEY His==2i! HT BUTLULERTSIEITALIELI

y~. (Face p. 100.) No. 3

25

8242°

Fahrenheit, and observations were made at 8 a. m.

BULESSS 220i sane es seta DEAT IESAAGaaA LEAT; LI LSLAnED

bt
[RRESS SERRA Ses AERA ASRS
“no S th 4 % rs ra © ig 8
nt é re) 3 rs) B + g b » Nn

JANUARY. FEBRUARY.

0S) 20 SEAS S98 ES gd 19 Ns OE SY,

SEPTEMBER OCTOBER NOVEMBER. DECEMBER.

To }uNLES fan 20 TUNES dO LS pO IS 30 a5 99 NEES 0 iS Iso l2s | sor) Siu it0,

ws

ts
rove | baobd based RASS Bana beooa = =] >-}——+— a =
eee Leet Salt
paso bees. oes
1nesea based ses bd Poa Pes ae
{ Peeanesa peeed
i x
| [esa]
t+ ~~ —
65° E -
i
|
|
| Teseg [aero]
ea eealemel } f+}
eeeeleee| +t ft
Sal pani eee eee IC] fl
mee tT :
: +
rs. i
= St eeaiaeel as [a
so? f— [ise | epg bese ee [eee
ea | - ni Bead Doped | 800 | Bow cE:
bead sete By
40°
5 | eee |
{ pened boned peeed
l eal prey
ao fall
sot Ft pees
f ++] peeed beeed|
t)
2
| z [ = ce
Ge 19° 19° 19° 19) reel!
10° 10°5°9415° 2°| ar’ at,
q Fio 4.—Diagram showing temperature of air and water at Woods Hole, Mass,, for each day of the year 1922. The less regular line represents air temperature and the more regular line water temperature, Readings are in Fahrenheit, and observations were made at § a. m. 8242°—257. (Face p. 100.) No. 3

wlinta” HOSAM 2 ¥RAU ARTY ALGUSE! YRAUMAL

ae ee

9 Z u ? bo 8 > et 1% a Te See
} eran —_ - nea sce an ENF hrachileane sive Saar wae 5 ae vi
Settee r: ; a wana aan al sla He : = fi ine be a
a es tot ied ' Econ aie I Bee ie eer | ' z pemiareere> UE A Roh
| let tas were § meen fame Gail (inant (hand es ane ee pea e y mittens en ~ tt Lat ay
| hata i 9 PS ERG es. Ca 4 Birtstiter tases A tage
{ f ie nawe 44 +. oo} pm anom't madre ie m f Faia Wasa |e
aes ; iT "I 15 i a
| : sell dhts onst neal i ‘ 138 ft led
waeuee pe sef : oaefe ah
ai ahha (ahi aaecde Hike. Alt |ne iht Oe anak = =p =i oe) ma |
\e oe tn ee
it ut
eres hi ‘ beer La epeenef Hie 8
Veggten ame 4 ey eer OG EES ROS Bl BTN thud ee heh eed
4) i Aan 1 aE si eet bat rottnetb 8:¥ ha APRS Bi DB KF 2 .

aA why. oes BA

oy Toa We +

yea gt pss Be Pe) Da Sece ee hedoatas
oh ahig ova! abs 4 7
ed aE 1] a

—

ee

eet chia aia | bacdeoeiaadalendt

+

mete polecdte hepa gpd tee ee
{

er

{
re

ih wees ea
‘

= . ’
hn le

pas ey » oh mh

ay eran ey

gna’ haan tae tf
Yabba sah ba
aa een

Bf niet acon eilinsine dpe Noche AX ind Sgn cita Sand

“x

Bes alah dow fesgtee PER An oats eee wople ey otanid =a
2

NOVEMBER. DECEMBER.

OCTOBER

BER.

PR. = s 3 Ps 9 wes
3 fA santy
FIDLLATDEREDTIDIALIENIIIIL Ti E 7ST
HEH FHnaTHfesayiiliitesa EH

bbigae SUEUEZEERLAAL? GRU DEeeteccennizii

HH HEH He c dU eee

TERT i BERRI Diao znscse=2 <3! SELILIELIBIEITI

SERIE Ue TTITITETEE

PERUSE TERE ae fetes

jem gag
OUEDTERESELIELEDIGI ULI EDoS22=252e0284
m4 Boone non sss
ReccueCnny
spe
He

SHATAHAAIFAAEHEHEEEETATAD {ALLL TPEFRR

Hoe be seas SeSRR ERE ESE SST eae SS anSoeanoennaa
= SS
SUDLLLEATLELULLULLLIA03==:S 121 EIITIEITA tata ti tatiill
fasta
sranendugutut=sszssaz7 AIIM MEEIEIEIEIENTE
RELILIGI002520018822225525525
Dee eae Hii titi
Ff |
FRMLEETIERI E22 <= S01 iiiii
Za._==
: HEE Seat SHEL
HESS CURSE iaa® Bee still SESRETERLGLIZIAS
NEE ES Ree Se MSS: fceafoeeet = STi SESereaura
ee ee ee ~
fizsce= STH HITE!
ctl
Hi HHP Hii HEE HHH
SEREELLIE Se
i Enna :
“o ° (-y ° ° °
2 8 8 Bite 8 8 Ps 8 R

4

nN

No, 4

(Face p. 100.)

| Fahrenheit and observations were made at 8 a. m.
|
|

JANUARY. FEBRUARY.

SEPTEMBER.
+ 3 70 NASON as BO) eS OTIS ENT 39 Ma Yrs) 30.

NOVEMBER. DECEMBER.

Pea cMees 09 Dm 9 oe nw to
| rT : sen :
at ——
{ 2) ea bees ses Leo et fc Bese
| peaed | rl
2 | teed | row "| ia wea
a Hose boseteees eee
[eal a

* Sees ede

a leet
Cp) eaten 1
as’ {cr

i

t
x

A a al

i oo
zs
# CESS pte :

19°) 12) 12°13) ° © 399
ar.6°he$nh JirXfi2® rs%s°79NABS 17°] Ihe is? 7

Fio, 5—Diagram showing temperature of nfr and water at Woods Hole, Mass. for each day of the year 1923, The less regular line represents air temperature and the more regular line water tempornture, Readings aro in Fahrenheit and observations were made at § a, m,

(Face p. 100.) No, 4

OMAN

JEATA

fea

t

= ©
>
~j<f
i t
+ +t
. ¢
‘3
=)

Jo et-w

— +
So Bat

tobe te

ya tals eerie ct Swot a =
= sag - - oJ

st irs Ek We SU BIE” ;
| Sir ee

; 2 4 =

= — i =

. ; c

; =

} :

E :

; ; =

> bas bar ond oe | :

OCTOBER NOVEMBER. DECEMBER.

SEPTEMBER.

Taher

PLE CEePRe TT

nenass anil |

sl AT

: ey HE eset
i a, s0S-s00nnas |
L i Beall Hn PCEEEEEET A
sHH panied nt rT HET
Litt HOHE : eer HnnLLETTTE
A EEUATUEERULTEAUAEATESERDSGEE002="0ETTNIN LAH
ERRERESREREAIERAEEEIUI pro TTT He
° SERBIA 4a5>s aRAEHRLLESLLLTLLLE Rees Leeese
ae HAMHIMEMHINIEL
NTH a HHLEHHLALIETLLLTEEAL
NETH Still TERLILEETALIIT
TE Hi wet JERLALESEALILIEL
EVERIUEALAL Anse 22 SEEALATERTITIIEL
Ee et
te SVUETEELEEEELEEEELTEEEELEEULEEEEELEUELLETDL

42°—257. (Face p. 100.) No. 5

82

ular line water temperature

70°

70°

65°

46°

JANUARY FEBRUARY.
Pee) ty
(ei iseed Is zl salt
fox: ES}

i ee
i

MARCH. APRIL. MAY. JUNI
JUNE. JULY. AUcusT. SEPTEMBER. OCTOBER NOVEMBER
Goo 0) Oi ee Bae A oe en wo 45 we mS 5 pt Ot res oo wo 6 : ECE BEER:
— — — ~ _ osm ss el ses
SRSIp x ieeed E
1a Bsa 2a eo CI . a Hf:
— = al poss pmeads ee Soe; Need pase an SEUSS oy eae ea
este eee YN =
f ie beged teesa reok: Seana
tL. a Posed Doge BSS In alee SISIRSIo pegel based SS
bo sat L.| a bees = elena poe BSSoN PSI
= i Ets eee ieeieet = : r
| Dae 209 pe eal ie (ae
+ — + -|

F10. 6.—Dlagram showing mean air and water temperatures at Woods Hole, Mass., for each day of the year, 1902 to 1906, Inclusive, The less regalar line represents air temperature and tho more regular line water temperature

§242°_25f.

(Pace p. 100.)

No. 5

ase

30°

25°

29°

ALIAS TA

cae J

HUAAM

HATA

t

~ we t
}
1
1

ee

—_
a

+ St...

cay tiie

*
i

ae
na

oo awe
~ one

i
t
i
j
I
|
‘
i

ee
t

Sotere
ba cas barca
{
mene er 1

caine lima

4

JERR EERE sag rae

Ht ‘al
EEE H

oe a Se ao

Ei EPC EEL ESEES CBEE itd

;
3.8 te ee : ars a>
a ee ee Se eae 2
ee ee Se oer Se 4 ee eee
- 4 1 : + rH Ut en wy =
SF Se J je ea ape tate imminent a
" - a ete oo eee -<¢ 2 «6 is cl |
- $% a ee aie | ; 1s" 2 4
+ +h eRe tae aN >» ¥ ‘
Bi 4 5 4 . t > 4
a i oe ae ah ag oth
: is : ie:
oe oe ae | ¢ 7 t oe |
SS : Ss . : 4
eae Died ls Lae } ;
ap me mr } - = et
fee iG . = ai SB |

fst tS ate ste Cond

¢* k

sens wind ieee E23 :BbES

a ord

-

re i ie

}

1

seats oe
}
|

eet

Bae Wren ere oen

1

fet

)

gui oie duet

vit

PLANKTON OF THE WOODS HOLE REGION 101

spring. The warmest months were July and August; the coldest were January,
February, and March.

The strong currents rushing through the ‘‘Hole” on the flood tide churn the
water to such an extent that any change in the temperature of the air affects the
bottom water as quickly as it does that of the surface. Table 2 gives a series of
temperature observations taken when the temperature was suddenly rising or fall-
ing. These show the result of the mixing of the waters.

TaBLE 2.—Surface and bottom temperatures taken at Woods Hole in 1922-23

“Ovat °C. at °C. at °C. at
Date surface | bottom Date surface | bottom
APU y{20 nl O22 eteeee <a sae scoleccseetcas 20. 2 QO Ze WDecgls, 1022 tena sees eeas aaa ae eee 3
Rr ky Bes A ae 21,2 le 2D eCriS 1922: - 82 oe oes eee cone ee see 1 1
Spr 7 01 77 22 20mm | Mecy1 9s 1022 aoee meen en den ebe eee eesce =015 -0.5
PAI gm SepLO22et a 2 neo aannee 20. 5 SOS O NEAR LO, hoes ane meccenenccdcecduecoeacncs 2 2
Ju Ps TEE, TROP Se eer eee 21.8 2178;||eMlare20 91023 ae sane wee eee nee as Seen -1 -1
ANU, I) PP 22 22 Varo NODS © ee Se ne A) 1.5
sh) S\is G5 LEP ees 22 22 INGSE S22, 80032 Soc cee eee cee vee nec ce 2 2
Beptwelo22e-2- 2.25. 22 elie 21.4 PA Ue SOLER (rch tiles, Ss ht eee enn coe in, Cae Renee (0s 0.3
BODUslOee teers hw nsec sce ceceekacasccess 20. 4 20. 4

GENERAL DISCUSSION OF PLANKTON

Marine plankton at Woods Hole falls naturally into two great groups—the
oceanic and the neritic—each of which has quite distinct characteristics.

The oceanic plankton consists for the most part of adult animals existing
throughout life as a part of the pelagic fauna. The only immature forms normally
occurring are the young of these oceanic species. Occasionally larval animals from
the neritic plankton are blown out in offshore winds, but this does not occur often.
However, it would be impossible to draw a line denoting the boundary between
the two types. There exists a broad intermingling area into which each species
extends to a point where external conditions form a natural barrier. As all forms
are not subject to the same conditions, this wide intermingling zone results.

The neritic plankton, in contrast to the oceanic, consists for the most part of
immature forms which in adult life are not a part of this community. There are,
of course, many truly pelagic animals common to the littoral zone, but these are
usually greatly outnumbered by the temporary intruders, except during the winter
and spring months, when the larval forms reach their minima at the same time
that many copepods and Sagitte have their maxima.

Figure 7 illustrates in a general way the constituent parts of the zooplankton
at Woods Hole at different times during the year. It will be noticed that the
summer and winter plankton are made up of representatives from the same groups.
The great difference in the relative abundance of these in the two seasons will be
discussed later.

The influx of oceanic species occurs both in summer and winter, although
the number of different forms occurring in the colder months is comparatively small.
During the summer, however, swarms often appear. Great masses of Sargassum,
with its many inhabitants—Physalia and other floating forms—often fill the waters
of the Sound after a southeast storm or a continued hard wind. Wheeler noticed

102 BULLETIN OF THE BUREAU OF FISHERIES

that such copepods as Pontella meadit and Anomalocera patersoni appeared in the
tow only after this weather. Dr. H. M. Smith attributes the presence of prac-
tically all of the small tropical fish that are taken each year in the Sound and at
Katama Bay to southerly winds. The floating Sargassum offers shelter for these
animals after they have been passively transported up from the south, and as
the weeds are blown inland the fish accompany them. During the winter months
such winds are fatal, but in summer the broad expanse of water extending from
the coast reaches a temperature as high as that of the stream. Here any forms

Variation in the constituent parts of the plankton at Woods Hole during the year

PLANKTON

Summer Spring and fall Winter

Temporary Constant

Many |

oftlenterates
Permanent Temporary Temporary Permanent

Rei sh

Usual Ocean Larval Benthos, Benthos, Benthos, Few Larval Ocean Usual
neritic forms nekton adults carried earried larvée nekton forms neritic

adults blowm (fish and lar- by cur- by cur- of (fish) blow adults
and lar- in and vae (ad- rents and rents and Benthos in and lare
vae squid) ults on- on float- on float- and free- vae
ly in ing ob- ing ob- swimming
breeding jects jects adults
season)
Fic. 7

carried out of their courses can exist very well until the temperature drops in the
fall, when all perish.

The shallow waters of the immediate region, sheltered as they are by the arm
of Cape Cod, respond very quickly to changes of weather, heating rapidly and cool-
ing suddenly. In summer Buzzards Bay is warmer than Vineyard Sound, the
maximum temperature occurring at the head of the bay, the minimum around
Cuttyhunk. Such conditions continue through August and into October. With
the sudden drop in air temperature the bay water responds immediately and results
in an exact reversal of the conditions found in August. At this time the water of

PLANKTON OF THE WOODS HOLE REGION 103

the bay becomes colder than that of the Sound. That has an important effect on
the plankton of Woods Hole, for as long as the bay remains warmer than outside
waters all animals carried into it will survive and be carried through the passage
during the flood tides. As soon as the temperature drops in the bay the tropical
forms entering it will perish and not return to the Sound. Such a condition is very
noticeable in the fall, when all Gulf Stream forms suddenly disappear from the
plankton of Great Harbor, although they may be taken in abundance in the Sound
or at Katama Bay throughout October and November. The arrival of southern
forms is usually noted as soon at Woods Hole as in the Sound.

In some parts of the eastern Atlantic a great many animals having a double
breeding season are found. These forms appear in the spring and again in the fall.
At Woods Hole examples of this group are limited to one phylum—the Celenterata.
Figure 26 (p. 124) shows clearly that almost all Hydromedusz common to this region
have been taken both in spring and fall, the spring swarm being the largest and
lasting for the longest time. The Scyphomedusz rarely have this double period-
ical appearance. During the past year, however, early ephyre of Aurelia flavidula
were taken in November and again in April. The Ctenophora have been taken in
small numbers throughout the winter, but usually swarm in the fall and early
spring.

I know of no permanent planktonic animals at Woods Hole having this biannual
distribution. Certain copepods are most abundant in the fall or spring, but never
at both seasons. Such forms are also plentiful in the winter and have been included
in the winter plankton. ‘Two species of importance belong to this group— Tortanus
discaudata from December to June and Pseudodiaptomus coronatus from August to
January.

Under normal conditions the zooplankton, although varying considerably in
its constituent parts, is always abundant at Woods Hole. The dominant winter
or Summer form of one year may be totally absent the next, but some other species
usually takes its place. During the winter of 1899 and 1900 Temora longicornis
formed the greater part of the plankton, while in the fall of 1922 and the spring of
1923 hardly an adult specimen was found, the dominant species that year being
Pseudocalanus elongatus. In the winter of 1923 the temperature of the water re-
mained so high that neither of these winter species had appeared by December 29.
Centropages hematus and Acartia bijfilosa constituted the bulk of the collections.
Temperature and weather conditions, no doubt, determine to what extent the north-
ern forms pass south of Cape Cod and erter local waters. Almost no cold-water
species were found in the early winter of 1923. Such diatoms as Rhizosolenia alata,
Skeletonema costatum, and Ditylium brightwelli appeared rarely or not at all, and
even the cod apparently sought deeper water, for no young were taken in surface
collections.

Normal diatom maxima have no noticeable effect on the larger planktonic
forms. When the unusually large swarms of phytoplankton appear, however, the
zooplankton decreases rapidly and may even totally disappear for a time. Such
conditions are often found during the summer maxima of Rhizosolenia semispina.
Usually the winter maxima do not affect the larger forms. In the winter of 1922-23
the phytoplankton and zooplankton were both abundant at the same time. At

104 BULLETIN OF THE BUREAU OF FISHERIES

this time Rhizosolenia alata was the dominant diatom. In 1923-24 Nitzschia serrata
occurred in such abundance that the zooplankton disappeared almost entirely from
November 16 until February 1. During this period top and bottom collections in
the shallow water of the bay and sound yielded nothing but diatoms. The zoo-
plankton was found to be fairly abundant in the deeper waters at the western end
of the Sound. As soon as the diatoms declined in numbers tie larger forms returned
to the shallow water. Figures 8 and 9 show the relative abundance of the zooplank-
ton and phytoplankton in 1922 and 1923.

DIATOMS AND OTHER PLANTS

The diatoms of this locality may be divided into two great groups—the pelagic
and the bottom forms. In certain parts of Great Harbor the bottom diatoms are
very abundant, and often large numbers occur in surface collections after storms or
particularly strong winds. As no fresh-water streams of importance are found in
the vicinity of Woods Hole to carry the various chemicals needed for diatom pro-

Vebe

ves. He

Fiac. 8.—Relative abundance of zooplankton and Fic. 9.—Relative abundance of zooplankton and phyto-
phytoplankton in surface collections from May plankton in 1923. , zooplankton; —-e—, phyto-
to December, 1922. ——_—, zooplankton; plankton
—.— phytoplankton

duction, much essential material must be carried through the ‘‘ Hole” from Buzzards
Bay by the strong currents. For that reason bottom diatoms were found to be
more abundant in eddies and pockets about the entrance of the bay than elsewhere
in Great Harbor and not scattered about evenly on the bottom in shallow water,
as might otherwise be expected.

Together with the dinoflagellates, the pelagic diatoms make up the greater part
of the phytoplankton of the region. On all but two occasions the former were far
outnumbered by the latter. In every haul made during the year with a No. 20 net
diatoms were found. They had regular seasonal variations which were very similar
to those of previous years (figs. 10 and 11). There is a regularity in the quantita-
tive variation as well as in the qualitative. The maximum of one year may be
larger, smaller, earlier, or later than that of another, but the basic characteristics
of the rhythm remain for the most part unchanged. An exception to this rule is

PLANKTON OF THE WOODS HOLE REGION 105

found in the appearance of certain oceanic forms, which will be discussed later.
Extremely unusual physical factors may even eliminate a part of the cycle, but as
soon as normal conditions are restored the progression continues as before. Such
was the case in December, 1923, during an unusually warm period (fig. 11). Al-
though quantitatively the winter diatom maxima remained approximately the same
as in the previous year, qualitatively it was very different (fig. 10).!

To understand the seasonal distribution we must know something of the nature
of the various forms that enter Great Harbor. The individual species of neritic
phytoplankton are much more widely distributed than the zooplankton, and factors
governing their appearance and disappearance are for the most part quite different.
As in the case of land plants, the diatoms are able to form organic substances from

4 & a 2 5 3
4 o> oo Bb B
‘Vode
eS a! late ae ese ee ho eer ia
Ao 7
re +H
Vode
Bienes iniite
8 H : 4 HED HE
° a +E Ht + tt
: ip :
++ i i
hy et HH
Be Ht
at iH He
tt tH F
V.5. 5s ih 2} tf
tea Re Ae
ves,
see rear ais
aH
Ne Be aoa
Fic. 10.—Seasonal distribution of most abundant Fic. 11.—Seasonal distribution of most abundant diatoms
diatoms occurring in surface collections from occurring in surface collections of 1923. Rbhizosolenia ex-
June to December, 1922. Rhizosolenia ex- cluded. , Cheetoceros; —--—, Corethron valdiviea,;
cluded. , Cheetoceros; —.—, Corethron —eco—, Dityliwm brightwelli; -----.-- , Leptocylindrus
valdivie; ——, Ditylium brightwelli; ----- ‘ danicus; —-e —, Nitzschia seriata; —-————, Skeletonema
Leptocylindrus danicus; —_--—=, Nitzschia costatum
Seriata; —-e.oe—=, Skeletonema costatum

the various inorganic chemicals. Together with the littoral marine flora they form
the basic source of food supply in thesea. Since all plankton animals are consumers
and depend solely on the organic materials produced by the plants, the importance
of the diatoms and the necessity for information regarding the sources of their
production can not be overestimated.

Obviously the two fundamental necessities for diatom growth are sunlight and
food material. Secondary factors, such as temperature, salinity, necessary resting
periods, etc., limit the geographical and seasonal distribution of certain species but
do not usually affect the group as a whole. Physical conditions fatal to one species
may be particularly favorable to another. Sunlight limits the vertical range of the
species to the narrow zone penetrable by the light (photic zone). This usually does
not form a limiting factor of production in local waters or littoral plankton in general.

1 See paragraph 1, page 119.

$2407! Son Lo

106 BULLETIN OF THE BUREAU OF FISHERIES

It is of extreme importance in ocean waters, however. Food material is the domi-
nant governing factor for all diatoms. The supply of the substances not common
to all sea water arises from two sources—outwash from the land and the replacing
of chemicals by the breaking down of the organic substances in the sea. For this
reason diatoms are much more abundant in the coastal waters, particularly near
the mouths of large rivers (Table 3). No doubt the large amount of disintegrating
material often found in coastal water (Bigelow, 1914) after a diatom decline forms
an important item in the replacement of essential chemicals such as silicic acid and
nitrates, but in comparison with the source of supply from the land it must be
rather small. Conversely, in the open ocean it probably forms the most important
source in regions where the land areas exert little or no influence. In this respect
the oceanic and coastal conditions are widely different.

TaBLE 3.—Distribution of diatoms in Long Island Sound in early March, 1923. Volume determined
from vertical hauls

Locality Date | Volume | Depth

Meters
Throgs! Neck 2.22 5..<5 Sous 9 ieee Ea fee CATA ee ty 3 ea Ao a ee een
Hempstead Harbor---_-
Matinicock Point_-_--_- :
Cold ‘Spring arbors 222. eee a ele Sat Se si a cee au
Peckstued gen aoa. ou. ios Soe eR OURS eee aya! Sa he SNe 2S Ee re see eee
Pino Creek! Point 22% 2-152 = SSE ae eS EE oP a a Sn ea LE ..-d
Off Bridgeport Harbor--------__ Be

StratiordsPoint: 22-22 - Yao sae ee
Housatonic River, near breakwater
New Haven Harbor

ororer

HeD
SNSORANYNARSE

CUM OO OA HOA CLOT ONO OH ON A OL OT A CHA OL

Stations insideiofiHarborsNo spl Ses eae les See ae ee ee eae eS 8 Oe Re a ee 8
Stations}insideof/HarborsNon2sse a8 = 22 as Se oS eI ek ee ie 6.5
Stations inside of Harbor No. 3_- 6.5
Five-fathom Rock---_______ 7
Peewee 2 ee 10.5
8.5
2
1.5
1
4.5
BY6)
4

In the ocean, where uniform physical conditions often exist to comparatively
deep water, Nathansohn (1909) found that the diatoms are most abundant in localities
where the greatest amount of vertical circulation takes place. (Gran, 1912, gives
maximum abundance often as deep as 50 meters and large numbers at 100 meters.)
Large quantities of organic material are constantly sinking to the deeper water,
and the decomposition of dead plants and animals at these levels sets free the nutri-
tive substances, which are returned to the photic zone in the ascending currents.
In anticyclonic systems like that of the Sargasso Sea, where little or no vertical cir-
culation takes place, the diatoms were found to be very scanty. Nathansohn’s (1909)
theory, no doubt, does not apply to all conditions found in the sea, but remains as
the best explanation of the source of production of oceanic diatoms. Ocean currents,
which themselves change according to the seasons, cause the diversity of high sea
plankton in fixed geographical positions. The occurrence of certain species far
beyond the limits of their natural range is usually attributed to currents.

In the coastal waters an entirely different condition of affairs exists. Over the
deeper parts of the neritic zone plant life usually is limited to a very thin surface

‘PLANKTON OF THE WOODS HOLE REGION 107

layer, which is differentiated from the deeper water masses by a lesser density and
higher temperature. In seasons when there is great outwash from the land the
neritic diatoms often form great swarms. In localities where upwelling or vertical
circulation takes place under these conditions the surface layers, with their flora,
are blown away from the shore and replaced by infertile water drawn up from the
bottom layers. The outwash of this fertile water is very favorable to the offshore
plankton but causes a diminution of diatoms near the coast, the few that were not
carried out having adverse conditions to combat. An inshore wind, on the other
hand, heaps up the surface waters and is conducive to luxuriant plant growth.
Repeated investigations (Gran, 1912; Nathansohn, 1909; Leder, 1917) of this prob-
lem have confirmed the belief that often the rapid appearance and disappearance of
diatom maxima is not so much a biological question as a hydrographical one. Gran
and Nathansohn in 1909 observed, ‘‘ We find an intensive plant life, and conse-
quently also an intensive animal life, everywhere at the surface of the sea where an
influx of water masses takes place, which has not, or at least has not immediately
previous, served as a source of nourishment for phytoplankton.”

Sometimes a diatom society is found in summer in the lower strata, with its
higher density and lower temperature, which was present in the surface waters
earlier in the spring. Such conditions are common among the zooplanktonic forms
and are occasionally found among the diatoms. Miss Ogilvie found the same dia-
toms in the lower strata off the south coast of Ireland in August as were present at
the surface from January to April. This is an indication that certain neritic forms,
which are apparently periodical in their occurrence, might remain as permanent
members of the plankton if conditions of existence were more uniform. ‘This is
interesting in view of the fact that many investigators have considered that a resting
period (spore formation) is a necessary part of the existence of truly littoral species.

It is impossible in local waters accurately to determine the real relationship of
the local conditions of existence and the development of the diatoms, because the
currents often cause variations much greater than those actually due to conditions
of existence. Gran (1912), realizing this, substituted a study of the rate of growth
as a measure of production in place of quantitative chemical analysis of food mate-
rials present in the water. In the vicinity of Woods Hole, where the currents are
unusually strong, the production would have to take place at an extremely high rate
in order to maintain itself were it not for the many ‘‘pockets”’ of quiet water which
are supplied with abundant land outwash. In certain less protected sections of the
coast this may be an important factor in the sudden disappearance of certain species.
As soon as the rate of production declines the species is unable to maintain itself, and
this inability to replace the numbers carried away by the currents may cause the
maxima to disappear long before the food supply is exhausted.

In dealing with the conditions of production it is very important to know just
where the production of floating forms takes place before attempting to explain
their appearance or disappearance. Two theories are now held. One contends
that all production of pelagic neritic diatoms takes place off the coasts, the sudden
swarming in inland rivers and small bays being the result of tides and winds. The
second theory is that production also takes place within certain limits in inland
waters. To be sure, winds may blow quantities of diatoms into open harbors and

108 BULLETIN OF THE BUREAU OF FISHERIES

small bays, but this does not explain the conditions as they are often found. Since
Woods Hole is a particularly unfortunate location to observe the factors of diatom
production, I shall cite results obtained in Long Island Sound in 1922 and 1923.

The first indication of local production was the variation in the species of
pelagic diatoms found in the different harbors and river mouths along one shore.
Had winds carried them there, one would expect to find the same species in all the
harbors. This was not the case except during the greatest swarms, when the
Sound seemed filled with a single species. In succeeding cruises it was noted that
the volume of phytoplankton in the vertical hauls taken at the mouths of rivers and
in harbors connected with inland streams was much greater than that of the mid-
Sound or harbors containing no land outwash. Table 3 (p. 106) shows the centri-
fuged volume, in cubic centimeters, taken with a Hensen medium-sized vertical net
at various points in the Sound. The predominant species was Skeletonema costatum.
A strong west wind had prevailed for several days before the collections were made.
Had the distribution of diatoms resulted from this they should have occurred
most abundantly along the southeast shore near the eastern end. The table clearly
shows that the greatest swarms occurred at the mouths of the rivers and harbors
where the most land drainage is carried into the waters. The salinity is low in all
parts of the Sound, and for that reason the diatoms do not penetrate far into the
mouths of the rivers and harbors. No great tides sweep the Sound at any place
except at the ‘‘race,”’ and even there Galtsoff found that 8 miles is the maximum
distance that the water is carried in a single tide.

Another source of evidence can be found in Peck’s (1896) report on diatom
collections in Buzzards Bay. His stations were laid in two lines, one at right
angles to the other, extending the length and width of the bay. A series of observa-
tions at various points along these courses showed that the greatest abundance of
diatoms occurred at the two inshore stations. The other two ends of the courses
were located in Vineyard Sound and the rapids at Woods Hole, and therefore are
not considered. He concluded from these records that there was a shallow area of
diatoms surrounding all the shores of Buzzards Bay. A glance at Figure 12 will
show that the two inshore points he selected (indicated by A) were near the
mouths of the greatest harbors of the entire Bay. The large rivers at the head of the
Bay empty their waters near Peck’s north station, while the waters of the Acushnet
and Nasketucket Rivers join at the point of his western station. Undoubtedly
Peck would have found his hauls less rich if he had selected spots along the eastern
shore.

A noticeable characteristic of neritic plankton flora is the variety of diatoms
that is usually found in every swarm. One or two species predominate, but the
many other species occurring in smaller numbers make up the so-called ‘diatom
society.”’ Allen, in 1920, made the following statement:

Detailed study of the records has clearly shown the important fact that when there is an
increased production of the most prominent forms there is also increased production of the less
prominent forms and an increase in the number of different forms. Such facts naturally lead to
the assumption that conditions favorable to high productivity of diatoms in the sea affect a large
number of forms in the same way. They also lead to the inference that determination of the

species that shall lead in production is due largely to the biological factors, such as rapid multi-
plication and vigorous development.

PLANKTON OF THE WOODS HOLE REGION 109

As most rules have exceptions, so, too, an exception to this rule is often found
at Woods Hole during the summer months when the oceanic diatom (Rhizosolenia
semispina) occurs in such abundance that almost every other form of animal and
plant life disappears for a time. The occurrence of this interesting species will be
discussed later.

SO) JOI4WNON

Ign20u
ening O

SVHILYU VW
forar 1768100

a

i

eS

if

punog pivAsour, pues Avg spiezzng Jo deyw—'ZI ‘old
QYUVAINIA
0

3904 14vd

The diatom spores are no doubt at all times exceedingly numerous in local
waters and are carried about by the currents and winds. When conditions become
favorable for those already present or those transported to a harbor or river mouth

110 BULLETIN OF THE BUREAU OF FISHERIES

where rich food material has been washed from the land, the spores germinate and
increase rapidly in numbers. The development will continue until the food supply
is exhausted or other unfavorable conditions arise. In discussing spore formation
in diatoms Gran (1912) stated:

When we subsequently find the same species once more in abundance, we have every reason
for surmising that the resting spores on the bottom were the principal source from which these
forms have been derived. Ability to form resting spores must be of the utmost importance for
the existence of the species in coastal waters. The chief difference between coastal seas and the
ocean, so far as hydrographical conditions are concerned, lies in the extreme and rapid changes
in such fundamental conditions of existence as salinity and temperature in coastal waters. Rest-
ing spores, therefore, must be the means by which many species continue in coastal seas, not-
withstanding the fact that there conditions of existence are favorable only for a limited portion
of the year. The Arctic diatoms, for instance, which sometimes are to be found in the plankton
of the Skager-Rak, are very easily affected by a rise in temperature, but their development takes
place during the winter months from February to April, when the temperature is at its minimum.
In the summer they are not to be seen, but their resting spores are then most probably on the
bottom. In the same way a whole series of warmth-loving species pass through the winter as
resting spores and are to be found along our shores only in the warmest months of Summer and
autumn.

As in the case of the littoral pelagic fauna, the winter diatom flora throws an
interesting light on the effect of the arm of Cape Cod on the winter forms in local
waters. In summer the cold waters north of the cape form a barrier for southern
neritic plankton. Samples taken by Bigelow in August, 1922, in Massachusetts
Bay, contained the same diatoms as those which appeared in Woods Hole in greatest
abundance in December. No doubt many of the northern diatoms are carried
south in the summer, but the sudden rise in temperature apparently is sufficient
to cause them to form resting cells or die. The effect of a slight change of tempera-
ture was evident at the end of March, 1923, when the winter forms suddenly disap-
peared. In winter, on the contrary, those carried south find a favorable climate
with a supply of food material that has accumulated since the disappearance of the
summer forms. Together with local winter neritic species they form a maximum
the extent of which depends upon the supply of silicates, nitrates, etc., in the water,
and remain until the food is exhausted or the temperature becomes unfavorable.
In this way the arm of Cape Cod forms a southern barrier for northern littoral plank-
ton only in summer and not at all times, as in the case of many benthonic species.

If this assumption were based upon the neritic diatoms alone, it could hardly
hold, because, combined with the evidence of the existence of diatom spores in all
coastal waters, the factor of temperature alone could explain the condition, and trans-
portation by currents around Cape Cod would not be necessary. However, as the
most abundant species (Rhizosolenza alata) north of the cape in August was a truly
oceanic form and proved to be the first to appear in large numbers at Woods Hole,
I think it justifiable to attribute it to the currents, just as in the case of the northern
copepods appearing about the same time which were certainly transported in that
manner.

None of the so-called “pulses” which Allen observed on the Pacific coast
occurred at Woods Hole in 1922 or 1923. The seasonal curves rose and fell evenly.
On April 3, 1913, Bigelow found the waters of Massachusetts Bay filled with dia-
toms. These were not evenly distributed but appeared as brownish-colored bands

PLANKTON OF THE WOODS HOLE REGION 111

~ alternating with clear areas. It may be that patches like these formed the pulses
of which Allen speaks, for his collections were on the open coast and taken from the
end of a wharf past which the belts of uneven abundance would drift.

The seasonal variation of the diatom maxima and the appearance of oceanic
species in local waters can be understood best by considering the geographical
position of Woods Hole as compared with other areas of the eastern and western
Atlantic. Steuer (1911) found that in general the maxima of the various species,

Se ee oe a

Karajakfjord in Greenland
‘North European Coast
Skager Rak

aarietto (ote ot trteot Mm — |_| ‘lame! tt |r

Fic. 13.—Schematic diagram of the seasonal distribution of the diatom maxima in the northern and southern parts of the eastern
Atlantic. After Steuer

both neritic and oceanic, is closely related to temperature, and thus varies according
to the latitude. It has long been known that on both the European and American
coasts the most luxuriant diatom growth does not take place in the warmest months
even as far north as Norway and Newfoundland.

At Karajakfjord, in Greenland, Vanhéffen (1897) found only one maximum,
from May until the beginning of September. South of this there occurs the typical
spring and fall maxima, which retreat farther and farther from the warmest seasons
as one approaches the Tropics. Steuer (1903) found that this constant succession
of diatom maxima toward the south necessarily leads to the assumption that some-
where in the south there will be a meeting of the two maxima in winter, and this
was found to occur in the Adriatic Sea by Leder (1917), Steuer (1903), Stiasny (1908),
and Gran (1909) (fig. 13). A smaller maximum was also found to occur in June and
July. Conditions on the American coast are surprisingly similar to those of the
eastern Atlantic (fig. 14).

Cr Grae a ea
Bay of Fimay ee eee Es

Massachusetts Bay

Vineyard Sound and
ards Bay

- ——_ =<
ck Se | mg ED HH

Fic. 14.—Schematic diagram of the seasonal distribution of the diatém maxima on the western Atlantic coast

Observations in eastern Canadian waters by Bailey, MacMurrich, and Fritz
show that the greatest maxima occur in the spring and fall. Bigelow (1917) com-
mented on the similarity of the diatom distribution in the Gulf of Maine and that
of the North Sea, Irish Sea, and Skager-Rak. He also found a great maxima in
Massachusetts Bay in April and early May, 1913, and a smaller one in September,
1915, and one in late August, 1922.

The striking effect of the arm of Cape Cod on the plankton is again evident
here, for within 20 miles of latitude of Massachusetts Bay conditions similar to
those of the Mediterranean and Adriatic Seas are found in Buzzards Bay. Here
and throughout the shallow waters south of Cape Cod a rich winter diatom plankton

112 BULLETIN OF THE BUREAU OF FISHERIES

gi gh geen calla,
»
Se ouecier fa. a ee
i Hy z He
VoAs PHY eee
Ht A H
He 4 :
seer te
i ae
at HE PEELE
get sseea toy oH
Ae EAE aE He
EEE E ; H
H + satti
+H FEEEEEET EEE H A
sine He
ScEsTap zt
+ He : rH
HE HEE EE EEE, FEE
Se +H G =e
H He # +H
Ht
HHH
pipes :
cous! at 81 Ht
: HEH i
v.35. H
i:
a HH
nt ane as
Ne

Fic. 15.—Occurrence of Rhizosolenia in surface
collections from May to December, 1922.
—.—, Rhizosolenia semispina; -------= pelt.
setigera; , Rk. shrubsolei;- ———, RF, alata
genuina; 4. &. alata gracillima; 090000000, R,
styliformis; —eoem, KR. fxrwensis

starts usually in November and continues until
March, reaching amaximum in December. This
compares very closely with the maxima found year
after yearat Trieste. Corresponding to the short
summer maximum of that region, asummerswarm
occurs also at Woods Hole, starting usually in
July and remaining until September. A compari-
son of the seasonal distribution and breeding sea-
sons of the zooplankton of the two regions shows
that theconditions at Trieste are of amore south-
erly nature than in this region, although it is
farther north. Its relation to the Mediterranean
makes the reason for this obvious. The summer
maximum is very variable, because the local
neritic species play a minor part, the greater
part consisting of a single oceanic form (Rhizo-
solenua semispina). Obviously, conditions be-
yond the limits of the adjacent coast have much
to do with the appearance of thisform. In 1922
and 1923 it was particularly abundant (figs. 15

5
a

Dece

3

Jule
Sepe
Octe

4

ie ee ne
Vole
A.
Se
VeSe
Ne

Fic. 16.—Occurrence of Rhizosolenia in surface collections of 1923.

, Rhizosolenia semispina; -c-c-c-c-0, Fe. setigera;

———, R. shrubsolei; nome, FR, alata genuina; cccccce, R, styliformis; *+t+++*+, FR, calcar avis; mmneom, K, delicatula

PLANKTON OF THE WOODS HOLE REGION

Actinoptychus undulatus
Asterionella japonica
Bacteriastrum delicatulum
Belleroché& malleus
Biddulphia alterans
Biddulphia favus
Biddulphia biddulphiana.
Biddulphia rhombus
Corethron valdivide
Ditylium brightwelli
Grammatophora marina
Grammatophora serpentina
Guinardia flaccida
Leptocylindrus danicus
Licmophora flavellata
Licmophora lyngbyei
Nitzschia closterium
Nitzschia paradoxa
Nitzschia seriata
Paralia sulcata
Rhabdonema adriaticum
Skeletonema costatum
Stephanopyxis appendiculatus
Strdatella unipunctate,
Synedra gallionii
Synedra undulata

Thalassiothrix longissima
Thalassiothrix nitzschioides

Distephanus speculum
Dictyocha fibula
Cerataulina bergonii

Fic. 17.—Distribution of diatoms and Silicoflagellata in 1922 (excluding Cheetoceros and Rhizosolenia)

113

114 BULLETIN OF THE BUREAU OF FISHERIES

and 16), filling the waters of the bay and the eastern part of Vineyard Sound. The
seasonal distribution of diatoms in Long Island Sound in 1922-23, during the
winter months, was found to be very similar to that of Woods Hole except that the
swarms appeared slightly later. _

Aetinoptyehus undulatus SESE TSOSEE

- Febo
Octe
Nove,
DeCo

i

Apre
May
in June
Jule
Auge
# Sepe

Ee
:

Asterionella japonica

Bacteriastrum delicatulum HH
Bidaulphie alterans He
Biddulphia faves FREE Hy Heese
Biddulphia granulata HE aes ae FE PEE .
Biddulphia biddulphiana EEEPUtor eee cu ecetecaze tetout FEES eEeEE ec Ee eee EEE
Biddulphia vesiculosa He HH He a
Corethron valdiviae HH i rH steae rHEEEEE
Dityliwa brightwelli sshcoa tenptsoslo™”SOGsOatE
Fragilaria crotonensis HH. :

Grammatophora marina
Grammatophora serpentina BHEPEEH Ass
Guinardia flaccida BESSHREEE SESHESESHNESEOBSCE ORE H
Hyalodiscus stelliger : 4 H H
Leptooylindrus danicus sTSEseHESEDL BACEESEREESAGHCHEGHCEOGHGLC
Liomophora flavellata : H HEHE
Liomophora lyngbyei
Melosira borreri

Nitzschia closteriun
Nitzsohia longissima
Nitzschia paradoza ee coEee 4 -
Nitzschia seriata 5 Fer a ao
Paralia suleate ; :
Rhabdonema adriaticum
Skeletonema costatun

Striatelle unipunctata
Synedra gailionii soseeens
Synedra undulata :
Thalassiosira decipiens Ht
Thalassiosira hyalina eeasueiiastiinaatt :
@halassiosira nordenskioldiifs i aad aes d : Ht Hae :
Thalassiothriz frauenfeldii Ho ae + = Seer ceraeaaed
Thalassiothrix longissima Soe sete H
Thalassiothriz nitzschioides
Distephamss speculun : .

Dictyocha fibula
Cerataulina bergonil

Cyclophora tenuis

em
rH
ae

Fic. 18.—Distribution of diatoms and Silicoflagellata in 1923 (excluding Cheetoceros and Rhizosolenia)

The pelagic diatoms of the Woods Hole region may be grouped under three
headings—tychopelagic, oceanic, and neritic. The tychopelagic group is made up
of semi-bottom forms, which often occur in coastal waters in enormous numbers.
They are carried about by the winds and tides, usually without relation to any

PLANKTON OF THE WOODS HOLE REGION 115

particular season (figs. 17 and 18). The following common tychopelagic forms
appeared in the surface collections of the past year: Actinoptychus undulatus, Bid-
dulphia alterans, B. favus, B. granulata, B. biddulphiana, B. rhombus, B. vesiculosa,
Hyalodiscus stelliger, Melosira borrert, and Paralia sulcata. All of these species are
temperate forms.

The oceanic and neritic diatoms that have a distinct periodicity in occurrence
may be grouped according to the latitudes in which they are distributed. This
method, originated by Cleve, has been used by almost all planktonologists.
The various groups into which Cleve divided the characteristic plankton diatoms
are represented at Woods Hole by the following species:

Boreal Arctic_____-________ Chetoceros mitra.
Nitzschia closterium.
Thalassiosira hyalina.
nordenski6éldii.
North Temperate_________- Cheetoceros danicum.
debile.
diadema.
sociale.
teres.
Leptocylindrus danicus.
Licmophora flavellata.
Nitzschia longissima.
Rhizosolenia feroeensis.
setigera.
Skeletonema costatum.
Stephanopyxis appendiculatus.
Synedra gallionii.
Thalassiosira decipiens.
Thalassiothrix nitzschioides.
South Temperate_________- Asterionella japonica.
Bacteriastrum varians.
Cheetoceros cinctum.
contortum.
didymum.
laciniosum.
lorenzianum.
schiittii.
Ditylium brightwelli.
Fragilaria crotonensis.
Grammatophora marina.
serpentina.
Guinardia flaccida.
Nitzschia paradoxa.
Rhabdonema adriaticum.
Rhizosolenia calcar avis.
; delicatula.
shrubsolei.
Striatella unipunctata.
Synedra undulata.
FEYO pI Calta ae alo Bellerochea malleus.
Rhizosolenia calcar avis.
shrubsolei.

IN@RIGIC# 4-5-2 oe

116 BULLETIN OF THE BUREAU OF FISHERIES

BorealpAnctic= 22-222 hv isene Cheetoceros atlanticum.
boreale.
criophilum.
decipiens.

Nitzschia seriata.

Rhizosolenia hebetata (semispina),

Thalassiothrix longissima.

Cheetoceros densum.
peruvianum.
willei.

Rhizosolenia alata f. genuina.

f. gracillima.

Thalassiothrix frauenfeldii.

Tropical.c2. vee eee Chetoceros coarctatum.

peruvianum.

Antarctics.t a2 JU een Corethron valdiviz.

Oceknin tie shempetate.2 25. tee aweews

In 1922 the summer swarm was composed almost entirely of members of the
genus Rhizosolenia (figs. 15 and 16). Rhizosolenia semispina and R. shrubsolei.
appeared about June 15, followed in July by R. setigera. The latter two species
were never as numerous as the former. . semispina increased rapidly until July 5,
when the swarms literally filled the waters of the bay and sound, clogging even the
coarsest plankton nets with a slimy brown ooze. Shortly after this it began to
decline, disappearing about September 9. The 1923 maximum was very similar,
except that the two minor species terminated their season earlier than in the previous
year, while R. semispina declined more slowly, remaining in small numbers through-
out the fall and early winter.

The occurrence of this species at Woods Hole during the summer months is
rather interesting. It is a northern oceanic form, known from both the Arctic and
Antarctic regions, and was found by Ostenfeld (1913) to thrive best in the areas
of the North Atlantic where cold currents seek southward. It is particularly abun-
dant in the spring in the region of the Labrador Current about Newfoundland and
Nova Scotia. Bigelow (1917), in July, 1914, found a large maximum off Marthas
Vineyard at the time when the great swarms appear in local waters. In 1923 the
author found them extending from Cape Cod along the eastern side of Marthas
Vineyard to Nantucket and in Vineyard Sound as far as Menemsha Bight. None
were found at the western end of the Sound or in the waters about No Man’s Land.
This indicates that they enter the region from the northeast, as would be expected.
Miss Ogilvie (1923) found a maximum in July, 1920, off the south coast of Ireland.

The summer maximum at Woods Hole, then, is not wholly dependent upon
local conditions. Either of the two neritic species, Rhizosolenia shrubsolei and R.
setigera might dominate if hydrographical conditions prevented the appearance of
R. semispina. The abundance of the latter species will depend partly on the num-
bers blown into the bay and sound and partly upon the food material present there.
Although it is a northern form it must have an extremely broad temperature range,
because its distribution in Buzzards Bay in 1923 showed conclusively that great
production was taking place there at a time when the temperature was ranging
from 19 to 21° C.

PLANKTON OF THE WOODS HOLE REGION 7

As the numbers of Rhizosolenia semispina diminished in the late summer of
1922, Cheetoceros (fig. 19) increased, but lasted for only a short time. Another

diatom (Corethron valdiviz) then became very
abundant and reached its maximum about Sep-
tember 20 (fig. 10, p. 105). A rapid decline took
place after this, followed by another Cheetoceros
increase. In 1923 Corethron appeared on Sep-
tember 21, reaching its maximum on November 3
(fig. 11, p. 105). All available records for Coreth-
ron show it to have its flowering season in the
fall. In European waters Ostenfeld (1913) re-
ports it to be most abundant in autumn. Ogilvie
(1923) found it abundant on the south coast of
Ireland in July, 1920, and in August and No-
vember, 1921. Fritz (1921) records it from the
Bay of Fundy in October, 1916, and in September
and October, 1917.

The winter maximum at Woods Hole consists
usually of a greater variety of abundant species
than that of the summer. In 1922 many species
appeared suddenly about November 8. As the

Fia. 19.—Occurrence of the more abundant spe-
cies of Chetoceros from May to December,

1922, ammo: , Chextoceros decipiens; ——»+ee—,
C. didymum; —e—, C. laciniosum; ——~,
Cxschuttii 7 ——=——,) CO, sociale

season progressed different forms predominated on different days, but all were
usually abundant. At first Rhizosolenia alata f. genuina, a temperate oceanic

g a ” fe b 4 $s ” a s S 3
oO °o @
pepencugin deg Bi go gg Fg sg &
Ves. fi
Ae
Se
Ve Se
Ne ®
Fia. 20.—Occurrence of the more abundant species of Chetoceros in surface collections of 1923, ----- , Chextoceros decipiens;

, C. didymum; —eee—, C. laciniosum; —e—, C. schittii; —.—, C. sociale

118 BULLETIN OF THE BUREAU OF FISHERIES

species, proved to be the most conspicuous form (fig. 15). Later this was
replaced by Leptocylindrus danicus and Skeletonema costatum. Both of these
species are north temperate neritic forms, which are supposed by Ostenfeld
to exist all the year round on the bottom, being carried up among the plankton
in the flowering season and during high winds. The distribution at Wood Hole
appears to substantiate this very well (figs. 10 and 11, p. 105). The winter flowering

C.boreale
Cecontortum
Cecoarcta tun
Cedanicum
Cedebile
Cedenswm
Cediadema
Cedidymum
Celaciniosun
€elorenzianum
Ceperuvianun
Ceschuttii
Cesociale

Cewillei
Cemitra

C.spenov.
Cecriophilum

C.decipiens

Fic. 21.—Distribution of Chetoceros from June to December, 1922

season is evident, and the scattered occurrence throughout the year can be best
explained by Ostenfeld’s theory. Although very similar to tychopelagic forms,
these two species differ in that they multiply greatly while members of the plankton.
Other abundant members of the 1922-1923 winter society were Ditylium brightwelli,
Thalassiothrix nitzschioides, Rhizosolenia setigera, R. shrubsolei, and Chextoceros
sociale, all of which are neritic species (figs. 10, 11, etc.). Two oceanic forms

PLANKTON OF THE WOODS HOLE REGION 119

(Niteschia seriata and Chetoceros decipiens) were fairly numerous 4t times but
always played a minor rdéle.

As previously stated, unusual physical factors may cause great variation in the
time when the maxima appear as well as in the constituent parts. Such was the
case in the winter of 1923 (figs. 4 and 5, p. 100), when, after an unusually warm
season, although quantitatively the winter diatom maximum was approximately
the same as in the previous year, qualitatively it was very different. Rhizosolenia
alata, the first dominant species to appear in the 1922 swarm, occurred only as
scattering forms in 1923; while Nitzschia seriata, a rather scanty form in 1922, out-
numbered all others during the entire winter maximum by more than 1,000,000: 1
(fig. 11, p. 105). Certain other members of the 1922 maximum, of which Dityliwm
brightwelli is an example, did not appear at all.

e “e ~~
uy
Cy ry @

uge
eDa
cts
OVe
e Ce

é

Feb.
June
Jule

Geatlantieun
C.boreale
Cecinctum
C.coarctatum
G.contortum
Cecriophilum
Codeeipiens

Cedensun
Cediadema

C.didyman

Celaciniosun
C.lorenzianun
C.peruvianun
Ceschuttil
Cosociale
Ceteres
Cewillei
Oespenove

Fig. 22.—Distribution of Chetoceros in 1923

The absence of Phizosolenia alata (excepting scattering forms) might have been
caused by the extremely mild weather of the early winter. However, as it is a com-
mon oceanic species there are many other factors which may have affected it.
Certainly the unusual temperature influenced the neritic forms. During the short
time when the temperature was normal in the early fall (fig. 11) many species—
Chextoceros didymum, Skeletonema, Leptocylindrus, etc. (figs. 21 and 22)—appeared
and began their normal flowering season. When the unusual temperature condi-
tions continued, however, they declined and remained as scattering forms or dis-
appeared; but one species, Rhizosolenia setigera, which has an extremely wide tem-
perature range (fig. 16), apparently thrived with Nitzschia during the period.

Nitzschia servata is an Arctic oceanic species which often appears in large
numbers off the Norwegian coasts. It is very variable in occurrence, being present

120 BULLETIN OF THE BUREAU OF FISHERIES

some years and entirely absent in others. In all areas investigated it forms its
greatest maxima in the spring. In the spring of 1923 it reached its maximum in
January, remaining throughout March. The following winter it arrived slightly
earlier and increased rapidly, swarming early in November (figs. 10 and 11, p. 105).
Throughout the winter season it remained as the most dominant species.

The distribution of diatoms during the past year may have been unusual.
Certainly two seasons’ changes are not sufficient from which to draw conclusions.
However, as all available records for past years seem to indicate similar summer and
winter maxima, it is probable that yearly variations will be in the date of the
appearance of these same species and not so much in the species themselves. Winds
may carry in unusual oceanic species, but these may be considered accidental visi-
tors whose appearance again can not be predicted. The following diatoms appeared
in surface collections of the past year: :

Actinoptychus undulatus (Bailey).

Asterionella japonica, Cleve.

Bacteriastrum varians, Lauder.

Bellerochea malleus (Brightwell).

Biddulphia alterans (Bailey).

B. biddulphiana (Smith).

B. favus (Ehrenberg).

B. granulata, Roper.

B. rhombus (Ehrenberg).

B. vesiculosa (Agardh).

Cerataulina bergonii, Peragallo.

Cheetoceros atlanticum, Cleve.

. boreale, Schiitt.

. cinctum, Gran.

coarctatum, Lauder.

. contortum, Schiitt.

criophilum, Castracane.

. danicum, Cleve.

debile, Cleve.

. decipiens, Cleve.

densum, Cleve.

diadema (Ehrenberg).

didymum, Ehrenberg.

. laciniosum, Schiitt.

lorenzianum, Grunow.

. mitra (Bailey).

peruvianum, Brightwell.

. schiittii, Cleve.

. sociale, Lauder.

. teres, Cleve.

. willei, Gran.

Corethron valdivie, Karsten.

Cyclophora tenuis, Castracane.

Ditylium brightwelli (West).

Fragilaria crotonensis (M. Ed-
wards).

Grammatophora marina, Kitz-
ing.

QeeaCeneocaogeoooneoGaa

G. serpentina, Ehrenberg.

Guinardia flaccida (Castracane).

Hyalodiscus stelliger, Bailey.

Leptocylindrus danicus, Cleve.

Liemophora flavellata, Smith.

L. lyngbyei (Kiitzing).

Melosira borreri, Greville.

Nitzschia closterium, Smith.

N. longissima (Brebisson).

N. paradoxa, Grunow.

N. seriata, Cleve.

Paralia sulcata (Ehrenberg).

Rhabdonema adriaticum, Kiitzing.

Rhizosolenia alata f. genuina
(Gran).

R. alata f. gracillima (Cleve).

R. calear avis, Schultze.

R. delicatula, Cleve.

R. feerceensis, Ostenfeld.

R. hebetata var. semispina
(Hensen).

R. setigera, Brightwell.

R. shrubsolei, Cleve.

R. styliformis, Brightwell.

Skeletonema costatum (Gre-
ville).

Stephanopyxis appendiculatus,
Ehrenberg.

Striatella unipunctata (Lyngbye).

Synedra gallionii, Ehrenberg.

S. undulata (Bailey).

Thalassiosira decipiens (Grunow).

T. hyalina (Grunow).

T. nordenskiéldii, Cleve.

Thalassiothrix frauenfeldii (Gru-
now).

T. longissima, Cleve and Grunow.

T. nitzschioides, Grunow.

PLANKTON OF THE WOODS HOLE REGION 121

After southerly storms during the summer Vineyard Sound is often filled with
floating Sargassum bacciferum (Turner). This is a tropical plant from the Gulf
Stream, which is usually accompanied by a community of pelagic animals. As it
has never been known to reproduce in the region, it is probable that all die in the
fall when the temperature of the water drops. Although not true oceanic plankton,
this plant must be mentioned, for many pelagic forms enter Woods Hole attached
to it. A local species (Sargassum filipendula Agardh) is commonly found attached
to rocks and piles below the low-water mark in the harbor, but this has no relation
to the plankton.

PROTOZOA

The protozoa were omitted in the present investigation, with the exception of

the large forms that at times were numerous enough to form an important part of
the plankton. Unless special methods are used
no real estimate of the abundance of the many g
minute organisms of this phylum can be made.
Lohmann (1911) showed that at least 50 percent a.
of the living forms entering the finest silk nets
available pass through the meshes and escape. It HE
is very possible to grow cultures of protozoa, as_ 5, au uae tt tet
Peck has already done at Woods Hole, but it was sea a ay ee auee teat
not my purpose to create artificial complexes, so ain
that method was not employed. It is of value, 4,5,
however, in obtaining many of the rarer species.

Certain of the larger protozoa were very rai oh
abundant at times, particularly Ceratiwm tripos, yf ape :
Peridinium depressum, P. oceamicum, and several USE IE Marra PIRSA ce Oe
species of the genus Tintinnopsis. The distribu- Protozoa in surface collections from June to
tion of these animals often appears to be very December, piace herr ie alee die

' Dre iy fd eslesinetos waco 5 CW INACTOCETOS. an men a 5 CO, SUSUSS
definitely related to that of the plants. During a coooccc, Peridinium depressum; — - —
heavy-diatom. maximum very few of the larger Tistinnonsls sp; ——- —, Heterophiys sv;

A , —eoe —, Peridinium oceanicum var. oblongum
forms appear, particularly the dinoflagellates. It
may be that as soon as the plants have exhausted their food supply and disappeared
the protozoa that utilize the nitrates and not the silicates increase rapidly. Just
why they should follow immediately after the diatoms is a puzzle, but it is clearly
noticeable and can readily be seen by comparing Figures 15, 16, 23, and 24. Thus,
after the great Rhizosolenia semispina maximum of the summer, Ceratiwm tripos
swarmed, followed closely by C. macroceros and C. fusus in smaller numbers.
These would have reached a maximum earlier, I believe, had it not been for the
influx of Corethron valdivie, which came in September, 1922 and 1923. For that
reason their normal high point was never reached. Throughout November and
December, 1922, they declined as the winter diatom maximum increased, disappear-
ing shortly before the diatoms ceased in April. This may have been caused by the
gradual rise in temperature at that time.

Within three days after the bulk of the diatoms disappeared two species of pro-
tozoa fairly swarmed in the plankton. The most abundant of these was an unidenti-

8242°—25}——3,

Jule
Auge
Sepe
Oct

Nove
Dece

122 BULLETIN OF THE BUREAU OF FISHERIES

fied species of the genus Tintinnopsis, although the other (Peridiniwm depressum)
was also taken in great numbers. Hundreds of the thimblelike cups of Tintinnopsis
could be seen at one time in the field of the microscope. Certain other forms were
noticed at different periods throughout the year, but they never formed an im-
portant part of the plankton.

A second species of Peridinium (Peridinium oceanicum var. oblongum) had a
maximum in the fall of both years. This is a much smaller form than P. depressum
and was never present in such large numbers. In 1922 it appeared on July 9,
reaching its maximum late in August and disappearing about Spetember 15. In
1923, as in the case of almost all the planktonic forms of that season, the period was
later, commencing about September 2 and remaining until November 20.

Ae

Se

VeSo

ui
Re
Fic. 24.—Occurrence of most abundant forms of Protozoa in surface collections of 1923, —-—-—., Ceratium tripos; ------= A
C. macroceros; —-ece—, C. fusus; ——-e—, Peridinium depressum; -e-e-e-e-e-, P. oceanicum var. oblongum; ——_y

Tintinnopsis sp

During the fall maximum of Ceratium the water fairly blazed with light when
disturbed. They caused the net to gleam like a lantern, and often bottom forms
not normally taken at the surface were attracted to it.

An interesting radiolarian (Heterophrys sol) also occurred in the fall. During
September and October, 1922, the numbers gradually increased until they became
exceedingly abundant, often being found in bunches of 20 or 30 specimens. Afterthe
26th of October the number rapidly diminished until November 1, when the last
one was seen. None appeared in the collections of 1923.

Of the Silicoflagellata, Distephanus speculum and Dictyocha fibula occurred as
scattering individuals throughout the year except in the warmest months. Diste-
phanus was most abundant from November, 1922, to March, 1923, and Dictyocha
appeared from September to May. Many Foraminifera appeared, usually after a
storm. These, however, sank quickly to the bottom again and were rarely taken

PLANKTON OF THE WOODS HOLE REGION 123

in surface collections in calm weather. Some six species were distinguished, but
positive identification was impossible because there was not sufficient literature
available at the time.

The following protozoa were identified from the surface collections of 1922-23:

Acineta tuberosa, Ehrenberg. March 4, 1923.
Ceratium fusus (Ehrenberg). See Figures 23 and 24.
C. longipes (Bailey). February to June, 1923.

C. macroceros (Ehrenberg). See Figures 23 and 24.
C. tripos (Miiller). . See Figures 23 and 24.

Dictyocha fibula, Ehrenberg.

Distephanus speculum, Heckel.

Glenodinium compressa, Calkins. March 4, 1923.
Gonyaulax tricantha, Jorgensen. April 21, 1923.
Gymnodinium gracile, Bergh.

Heterophrys sol, Ehrenberg.

Peridinium depressum, Bailey. See Figures 23 and 24.
P. oceanicum var. oblongum, Aurivillius. Figures 23 and 24.
Tintinnopsis davidoffi, Daday. October 14, 1922.
Tintinnopsis sp. See Figures 23 and 24.

Julo1922
Sepe
Jane1923
Fede
Mare
Apre

May

June
Jule
Auge

Auge
Oste
Nove
Dece

Bougainvillia superoiliaris
Gemmaria oladophora

Obelia ep.

Podocoryne fulgurans H +H

Podocoryne carnea Fy EH
Stomotoca apicata i ae Er

Xotopleura ovhracea
Hybocodon prolifer i Anse HH
Lizzia grata

Syncoryne mirabilis

Syncoryne producta i fi ; HHH Feet HH
Turritopsis nutrioula
Bilercertiun campanula
Tiaropsis diademata
Dysmorphosa fulgurans
Bougainvillia carolinensis
Dipurena strangulate

Fig. 25.—Occurrence of Hydromedus in surface collections from June, 1922, to December, 1923
CG@LENTERATA

One hundred and sixty species of ceelenterates were recorded from the Woods
Hole region by Sumner. Of these, 132 were Hydrozoa, 5 were Scyphoza, and 8 were
Ctenophora. Thirty-eight species are listed in the tow records of Vinal N. Edwards
for the years 1893-1907. Figures 25, 26, and 27 show the maximum occurrence
of the more common species, while in Table 4 the rarer forms, together with the
particular dates of appearance, are noted. The records of the more common
Scyphomedusz and ctenophores are also recorded on individual charts.

The diagrams show clearly that there are definite seasons of occurrence for
the various species of ceelenterates. In most cases the species have a long spring
maximum and also a short one in the fall. Such a semiannual appearance is not

124 BULLETIN OF THE BUREAU OF FISHERIES

Syneoryne mirabilis
Hybocodon prolifer
Turritopsis nutricula,

Podosoryne fulgurans

Lizzia grata
Bougainvillia superciliaris f
Bougainvillia carolinensis +
Nemopsis bachet

Tima formosa

Tieropsis diademata
Epenthesis folleata
Obelia spe

Zygodactyla groenlandica

Aglantha digitalis

Liriope scutigera

Fic, 26.—Hydromeduse occurring commonly in surface towings. Maximum occurrence based on records of years 1893

cyanea
Pleurobrachia
Bolina

Mnemeopsis

Beroe LTT tty tt ie

Fic. 27.—Maximum seasonal distribution of Seyphomeduse and Ctenophora, based on records of the years 1893 to 1907. See
individual charts for Aurelia, Pleurobranchia, and Mnemeopsis

PLANKTON OF THE WOODS HOLE REGION 125

common among marine animals. Bougainvillia superciliaris, Hybocodon prolifer,
Nemopsis bachei, Tiaropsis diademata, Podocoryne fulgurans, and Tima formosa are
examples of Hydromedusx having double seasonal distribution. However, hardly
a single species that occurs normally in the spring has not also been taken in small
numbers in the fall. With the exception of Podocoryne carnea none of the summer
visitors have this biannual appearance.

A regular progression of the more common species of Meduse can usually be
noticed in the spring. Hybocodon prolifer appears first, followed closely by Syn-
coryne mirabilis and Lizzia grata. In early July, as these species reach the end of
their season, Podocoryne carnea and P. fulgurans appear, followed in August by
Dipurena strangulata and Bougainvillia carolinensis. The summer and fall species
always occur in smaller numbers than the spring forms. Certain forms appear to
be distributed throughout the year. LEpenthesis folleata has been recorded for
almost every month.

g 8
a " i
ss inte veh GS, Mgeeets we : ee ae
Bes ea eee, OBB Boe
Aurelia
Cyanea
Dactylometra |
Pleurobrachia
Mmemiopsis

Fic, 28.—Occurrence of Scyphomedusz and Ctenophora in surface collections from June, 1922, to December, 1923

Three species of Scyphomeduse are taken frequently in surface collections
(fig. 28). The most common (Aurelia flavidula) appears usually in March, April,
and May, although ephyre have been taken from August to October.

TaBLE 4.—Occurrence of uncommon Hydromeduse in surface towings

Species Date Abundance
Ectopleura ochracea-_____.__.._.._-__-.------------------- PASI 219 el O04 NE nee 2 oR, LUNN Rue oe ee oe to Few.
Corymorphatpendulaz:2. fetes es ie Apre2sandi29: M905 sa ees eee ee eas Many.
Stomotocaispicataz.s 2. eo) os ek ee seek ed occa Apr.i27, May 1, and Aug. 15, 1903.....J0..222.4c.h....-.- Few.
Staurostomailaciniatasi. 2... eee ee! PAT OW L906 aac ie eae a ee _|_ Do.
Butima mira... 2-2 Ms | PAto rl 9022 ee onens Many
Oceanis ilanguida: of) oo es _| May 16 and 17, 1904_ a Do.
Genus Clytia (probably C. bicophora) _- S22 May 16; 1905.2 --..- Few.
Rieemratodes\tennwisuc Lah ee a Septiangd iit; W907 se et ok ce eee _.| Many.
Nelanthaconita so. 20 oo. el ke docu wae actee cou Apr. 24;:25,,and' 30! and May 2,1906. ...-4....2..<c<---2c2 Few.

Every year in Waquoit Bay immense swarms of strobelias and ephyre of
Aurelia appear before the ice leaves. They also occur in varying abundance in all
local protected coves and shallow bays where eel grass (Zostera marina) grows in
abundance. The young apparently rests on the bottom during the ebb tide, rising
with the flood tide. During this period the water is often filled with them, while a
few hours later none may be seen. In the spring of 1923 ephyre were particularly
numerous at Waquoit Bay, although only a single specimen appeared in my collec-
tions from Great Harbor. By April the meduse had increased in size, varying

126 BULLETIN OF THE BUREAU OF FISHERIES

from 1 to 3 inches in diameter. Shortly after this they disappeared. The disap-
pearance probably took place when all strobilization had stopped and the currents
carried the meduse away. Occasionally at a later date swarms of large adults have
been seen in Vineyard Sound or Buzzards Bay. No adults were noted during the
past summer (1923) in local waters, although large swarms of mature Aurelia were
seen on two occasions in neighboring localities—Mount Hope Bay on July 14, and
at the entrance of Oyster Bay in early August.

It is difficult to understand how the planule get back into the harbors (particu-
larly Waquoit Bay) in such large numbers when apparently no adults remain in the
region. The eggs can not be deposited before

Eo» & a es & the meduse leave in the summer because the ani-

< £ 5S 5 = © mals are not mature at the time. I have never

1e95 {| | | | {| | | {| seen a mature specimen in Waquoit Bay. There
SRBl Ltt tt | seem to be but two possibilities—either enough

1896 - ~ - aa adults remain in the bay until the breeding season
eee (is BEL BasSEs (perhaps on the bottom) to repopulate it or the
Tall TTL] Planule are drifted in by the tide. I believe that

isos | | I | Ty {-{_{_] the first assumption is more probable; that is, that
hl | | | | [| sufficient adults remain to restock the waters even

1899 | | | | | though none may be seen at the surface. The dif-
| tt} | | ficulties besetting the second possibility make it

1900 ai almost impossible except under rare conditions
ae a when Vineyard Sound is filled with adult Aurelia
at the correct time. In the first place the medusz

1902 are entirely at the mercy of the winds and tides.
They may be widely scattered in coastal waters

1903 or piled together in great banks, as described by
Hargitt and Agassiz. The latter author consid-

1904 ered that the animals gathered together in the
rene breeding season, but this is not probable. After
storms large numbers of disks, minus lobes and

1a0e tentacles, of both Aurelia and Cyanea are often
found at the surface in local waters. All are de-

1907 stroyed before winter arrives. As the sexes are

Fic. 29.—Oceurrence of Aurelia flavidula during separate in Aurelia it is largely a matter of chance
iuppaca bree trees et whether fertilization takes place at all, because the
adults are likely to be widely separated before reaching sexual maturity. Under
these conditions it would hardly be possible for the species to maintain itself, because
it is apparently beset with more difficulties than the cod and has a proportionately
much smaller number of eggs. Therefore, the few adults that remain in the bays
may serve to maintain the species during seasons when fertilization in the open
waters is impossible, while a fortunate gathering of adults during the breeding
season may sccount for the enormous swarms present in certain years. This
dependence on a chain of circumstances to bring the sexes together at the right time
probably goes far to explain the irregularities in this and allied neritic species.

PLANKTON OF THE WOODS HOLE REGION 127

Cyanea capillata appears commonly in spring and fall, but not in as great
numbers as Aurelia. On April 14, 1923, the first specimen appeared. Throughout
May and early June specimens varying from 10 to 50 mm. could be seen daily at
the surface in Great Harbor, often in large numbers. Alexander Agassiz observed
great numbers of Cyanea at the surface between 4 and 5 a. m. at Provincetown.
“By 7 a. m. all had returned to deeper waters, although not a breath of air had
disturbed the surface.” A variation in abundance was clearly noticeable in local
waters during the past year, but the vertical migration did not affect the whole
group, some specimens occurring at the surface throughout the day. Their numbers
increased rapidly, however, during the flood tide. It may be that Agassiz’s observa-

1898 |
1899
1900

1901

1902
1903
1904 |

1905 |

ce

oa i Ea

ras lg

| Rene

1906 | SEuagnH ae
[il etc es SEC Enemge

a nee

Fia. 30.—Occurrence of Cyanea capillata during successive years, 1893 to 1907

1907

tions could be explained on that basis. Unfortunately no records of the tide were
given.

Dactylometra quinquecirra occurs occasionally in Vineyard Sound and Buzzards
Bay, although in very small numbers. In Narragansett Bay it is usually very
abundant in September and October. On August 8, 1923, a single specimen was
taken in Lackeys Bay, and several days later a few were observed in Vineyard
Sound. George Gray records large numbers taken on several occasions, together
with Salpa democratica-mucronata, off Nonamesset Island at the mouth of Great
Harbor. This species is known to be nocturnal, and for this reason the local appear-
ance may be greater than the records indicate because very little night collecting

128 BULLETIN OF THE BUREAU OF FISHERIES

has been done except from the Fisheries dock. This species appears to prefer the
relatively impure water of bays and rivers, rarely being taken in coastal waters.
Ctenophora present a very difficult problem to anyone attempting to determine
seasonal distribution. They are found scattered throughout the year in many
places. In this region the limits of the seasonal appearance are very definite,
although the abundance varies greatly. Plewrobrachia pileus (figs. 27, 28, and 31)

[JSR
PEPER EEE EEE EEE
He (at rc rel i CCl od Jee —~
Be ee) aaa
SERED Ite In aa

iz
saree ME Sel ae
fo EE tet

Fic. 31.—Occurrence of Pleurobrachia pileus during successive years, 1893 to 1907

appears in late December and remains until the latter part of May. The occur-
rence during 1923 was very scattered. For a few weeks in December, 1922, they
were abundant in all collections and then diminished gradually until February.
From February until April few were seen, but on April 1 many appeared and
remained throughout the month. In certain seasons immense swarms occur.
During the latter part of April, 1895, Mr. Edwards often noted that the nets filled
in a few minutes with these jelly-like organisms.

PLANKTON OF THE WOODS HOLE REGION 129

Mnemiopsis leidyi appears in smaller numbers at Woods Hole. In Long
Island Sound there is a very large fall maximum in August and a large winter
maximum in December and January. They are rarely found in Buzzards Bay in
large numbers, and were taken in only 3 years during the 15 for which the author
has records. During the past year a single specimen appeared on December 11
and three on December 15. Cape Cod is, no doubt, the northern limit of this
species, and its appearance in local waters depends upon the winds. Specimens
taken this spring were stragglers from the winter maximum of more southern
waters. No remnants of the fall maximum found their way into Great Harbor in
1922 (figs. 27, 28, and 32).

Bolina alata has been taken at Woods Hole in September by Mr. Edwards.
Agassiz described it as being one of the commonest species in Massachusetts Bay,
but rare south of Cape Cod. None was seen in Great Harbor during the past year.

Beroe cucumis is usually very rare in this region, although Mr. Gray found it
abundant on one or two occasions in late April and May. It is a northern form

Mare

§

tees CODEC EE
MIOCHIEVIEN I A aA eee eae | ee
1907 PECCCCAIEC CCC ical
1908 [| | HNC eG

Fic. 32.—Occurrence of Mnemiopsis leidyi during successive years, 1893 to 1908

Jane
Feb.
Apre
June
July
AUgo
Sept.
Oct.
Nov.
Dece

whose appearance in local waters is accidental, depending upon strong easterly
winds.
The following ccelenterates appeared during the years 1922 and 1923:

Hydromeduse: Hydromeduse—Continued.
Bougainvillia carolinensis Syncoryne mirabilis, Agassiz.
(McCrady). S. producta, Hargitt.

B. superciliaris, Agassiz,
Dipurena strangulata, McCrady.
Ectopleura ochracea, Agassiz.
Gemmaria cladophora, Agassiz.
Hybovzodon prolifer, Agassiz.
Lizzia grata, Agassiz.
Melicertum campanula, Agassiz.
Obelia sp.

Podocoryne carnea, Sars.

P. fulgurans (Agassiz).
Stomotoca apicata (McCrady).

Tiaropsis diademata, Agassiz.
Turritopsis nutricula, McCrady.

Scyphomeduse:
Aurelia flavidula, Peron and Lesueur.
Cyanea capillata, Eschscholtz.
Dactylometra quinquecirra (Desor).

Ctenophora:
Mnemiopsis leidyi, Agassiz.
Pleurbrachia pileus (Fabricius)

130 BULLETIN OF THE BUREAU OF FISHERIES

Ceelenterates recorded from 1893 to 1907 were:

Hydromeduse: Hydromedusee—Continued.
Aglantha conica, Hargitt. Staurostoma laciniata (Agassiz).
A. digitalis, Miller. Stomotoca apicata (McCrady).
Bougainvillia carolinensis (Mc- Synocoryne mirabilis, Agassiz.
Crady). Tiaropsis diademata, Agassiz.
B. superciliaris, Agassiz. Tima formosa, Agassiz.
Clytia (probably C. bicophora), Turritopsis nutricula, McCrady.
Agassiz. Zygodactyla grcenlandica (Peron and
Corymorpha pendula, Agassiz. Lesueur).

Ectopleura ochracea, Agassiz.
Epenthesis folleata, McCrady.
Eutima mira, McCrady.
Hybocodon prolifer, Agassiz.
Liriope scutigera, McCrady.

Scyphomedus2:
Aurelia flavidula, Peron and Lesueur.
Cyanea capillata, Eschscholtz
Dactylometra quinquecirra (Desor).

Lizzia grata, Agassiz. Ctenophora:

Nemopsis bachei, Agassiz. Beroe cucumis, Fabricius

Obelia sp. Bolina alata, Agassiz.

Oceania languida, Agassiz. Mnemiopsis leidyi, Agassiz.
Podocoryne fulgurans (Agassiz). Pleurobrachia pileus (Fabricius).

Rhegmatodes tenuis, Agassiz.
ANNULATA AND VERMES

The free-swimming annelids may be grouped under three headings—true
pelagic adults, benthonic adults swimming during their breeding season, and the
early larval stages of all marine Polycheta. A fourth group may be added in this
case to include the bottom forms carried by strong currents during storms.

Of the true pelagic annelids only one species occurs frequently in the waters
of Buzzards Bay and Vineyard Sound, although Moore (1903) has described two
other types from this region. Tomopteros helgolandica is taken from December to
April at Woods Hole. During seasons when southerly winds are prevalent they
have been taken in considerable abundance. The greatest number recorded was in
1906, when many specimens were taken almost daily throughout April until May 2.
During the spring of 1923 there were almost no winds from the south, and as a
result oceanic forms have been rare. One specimen of Tomopteros appeared on
February 5, that being the only specimen taken during the year.

Benthonic annelids often appear at the surface in great numbers, particularly
in the evening, during their breeding season. In the groups where the sexual
products are discharged directly into the water the active period is comparatively
short, sometimes lasting less than a week. This occurs in the various species of the
family Nereidex. The adults swarm at certain definite places, usually along sandy
beaches or protected harbors, and literally fill the water with cloudy masses of
eggs and sperm.

From July 20 to 24, 1922, Nereis lumbata swarmed in immense numbers at the
surface in the eel pond. A few were noticed at other spots along the shore, but
none appeared in the daily surface collections. On April 1, 1923, the beach at
Nobska Point was the scene of a swarming of N. virens. On many occasions during
the first two weeks of April ripe males were seen swimming among the Fucus about

PLANKTON OF THE WOODS HOLE REGION 131

the Fisheries dock. In this case, as in the case of NV. limbata, free-swimming larvee
appeared in great numbers in the tow, but few adults were taken. The usual
swarming season for NV. limbata ranges from June to September. A few adults of
N. pelagica were taken during the year, but none of these contained ripe sex products.
The breeding season of this species is in August and September. Platynereis
megalops is also commonly taken at the surface from July to September. Although
the young were taken on several occasions, but one adult appeared in the collections

Ce ie)

eo te nN &

a o th hor > o fe fe
eee Shc 2es2a3

a
e°0 @ 2 8 Bs 3 et g2enog8gs
lay ty SD OG ors? 0 8 wW
a arProsOQk &R mM on WAP FO
Sn ee eo ee es get eee es
»<t DOBAAN & < Sree noa2an
Amphitrite ornata EARS a

|_|
Arabella opalina eee
Autolytus cornutus |_| | ae ~~ ITS
Autolytus ornatus eee oleae ete
Autolytus alexandri Rem ft ~ SARA
Autolytus emertoni
Autolytus varians _
Autolytus longisetosis
Dodecacera concharum
Harmothoe imbricata
Ichthyobdella rapax
Lepidonatus squamatus
Iarval Lesquamatus
Lumbrineris tenuis
Magelona rosea
Nereis limbata
Nereis pelagica
Nereis virens
Nephthys bucera
Odontosyllis lucifera,
Udontosyllis sp.
Paedophylax dispar
Phyllodoce catenula
Phyllodoce gronlandica
Platynereis megalops
Podarke obscura
Spio setosa SLC B
Telepsavus larvae rT pa | eee
Tomopterus helgolandica ha Pete
Unidentified larvae Sts

Fic. 33.—Occasional occurrence of annelids in surface collections from June, 1922, to May, 1923.
@, single specimen taken

of the past two summers. All of the members of this family undergo extensive
physical changes in adapting themselves for pelagic life during the breeding period.
The anterior, nonsexual part remains the same, but in the posterior, sexual region
the parapoda become broad and flat and the chetz increase greatly in length.
In this form the worm is known as Heteronereis and is able to swim very rapidly.

In contrast to the Nereide stand the families Syllide and Hesionide. The
different species of Autolytus carry their eggs and swim about for varying lengths

1382 BULLETIN OF THE BUREAU OF FISHERIES

of time, often occurring for periods of more than four months. For the greater
part of the year they remain attached to hydroids and alge on rocks and piles as
nonsexual individuals. In this form they are not free-swimming and their occasional
appearance in surface collections is accidental. In the breeding season certain of’
these nonsexual individuals develop eggs in the posterior part of the body (posterior
to the gizzard), while others develop sperm. Strobilization then occurs, and sexual
individuals, which immediately become pelagic, are broken off. The females that
break off carry clusters of eggs in a pouch on the ventral side. The stolons are either
male or female, the two sexes never developing from the same stalk. Occasionally
chains of five or six worms, which have not yet separated, may be seen at the surface.
Free-swimming males and male chains are usually more abundant than the females.
Alexander Agassiz fully described this alteration of generations in 1862. The sexual
species of Autolytus are highly phosphorescent and are often extremely numerous
in the tow. ;

Podarke obscura is a very characteristic member of all evening surface collec-
tions of the summer. On calm, dark nights swarms of them appear at the surface
in protected harbors. The first specimen taken in 1922 appeared on July 6, the
last on September 27. In the strong currents about the collecting station the
occurrence of Podarke was more scattering than is usual, although many were
carried into the nets during both day and night. In daylight, however, the number
was much smaller, because at that time the adults seek protection under rocks and
among the Fucus.

Larvel annelids appear in the plankton at all seasons of the year. During
the early spring they form almost the only representatives of the benthos in the
tow. A very small percentage of the species has been worked out, and for that
reason it has been impossible to identify a large number of the larval forms that
were taken during the past year. Larval Nereis were very abundant during April
and May in 1899, 1900, and 1923. These spring forms probably were Nereis
virens. Another large increase in the latter part of October in each of the years
recorded may have been N. limbata, although the date is rather late for this species.
Such conclusions must remain as mere speculation until further data on the breeding
seasons can be obtained. This can readily be realized if one considers that there
are six species of the family Nereidx represented at Woods Hole, and larval Nereidi-
formia have been taken in every month of the year except September.

Two very characteristic larval annelids appear each year in large numbers.
The first occurs in late July and continues throughout October. Fewkes has described
it from Newport as a species of the genus Telepsavus. His identification is doubt-
ful, however, for no adult of the species has been recorded from this section of the
Atlantic coast. In 1922 it appeared first on July 26 and continued to be taken until
October 25. The second larve (Lepidonatus squamatus) appeared first on Decem-
ber 19. Throughout the spring it was taken daily in large numbers. The season
lasted until the last of April. This fact is rather unusual for Sumner records the
breeding season as late April, May, and June. An adult female of this species
taken on February 2, 1923, was filled with ripe eggs. During May and June, 1922,

PLANKTON OF THE WOODS HOLE REGION 133

no larve appeared in the surface collections. From these observations the breed-
ing season is seen to be much more extended than has hitherto been supposed.

Occasionally postlarval forms occur after northeast storms. As these are not
true free-swimming larvex they are listed with adults taken under similar conditions.
During the past year several nonplanktonic annelids have been taken. Certain
of these may swim freely in their breeding season, but the occurrence in the collec-
tions was so scattering that I have not considered it as normal. Dodecacera con-
charum offers a peculiar problem. Scattered specimens, often quite numerous,
varying from 15 to 20 mm. in length, appeared from July 16 to August 15, 1923.
The presence of these immature specimens over such an extended period of time
could hardly have been accidental, and yet Dodecacera is known to be a truly
benthonic annelid.

Comparatively few leeches have been taken from the Woods Hole region.
Sumner records five species, all of which were taken from fish. One species (Jch-
thyobdella rapax) appeared twice in the surface collections of 1922-23—once on
January 20 and once on April 7. Both occurrences were during the breeding season
of the winter flounder (Pseudopleuronectes americanus). Former records give the
summer flounder as its host, but it is highly probable that it will be found on both
species.

The following annelids were taken in 1922-23:

Amphitrite ornata (Leidy). Nereis pelagica, Linneus.
Arabella opalina (Verrill). N. virens, Sars.

Autolytus cornutus, Agassiz. Nephthys bucera, Ehlers.

A. ornatus, Verrill. Odontosyllis lucifera, Verrill.

A. alexandri, Agassiz. O. sp.

A. emertoni, Verrill. Pedophylax dispar, Webster.
A. varians, Verrill. Phyllodoce catenula, Verrill.
A. longisetosis, Agassiz. P. groénlandica, Oersted.
Dodecacera concharum, Oersted. Platynereis megalops (Verrill).
Harmothée imbricata, Malmgren. Podarke obscura, Verrill.
Ichthyobdella rapax, Verrill. Spio setosa, Verrill.
Lepidonatus squamatus, Leach. Telepsavus larvee?

Lumbrineris tenuis, Verrill. Tomopterus helgolandica, Greef.
Magelona rosea, Moore. Unidentified larve of several species.

Nereis limbata, Ehlers.

Sagitta is the only true pelagic representative of the phylum Vermes found in this
region. It usually appears in December and remains until June. In listing the
Sagitte of past years no attempt was made to distinguish between Sagitta elegans
and S. serrodentata. The former is more littoral and northern in its distribution,
while the latter is a southern oceanic form often occurring in the Gulf Stream.
During the spring of the present year (1923) no specimens of S. serrodentata were
taken. This may be explained by the fact that the prevailing winds have been
from the north and comparatively few oceanic forms of any sort have found their
way in. However, since S. serrodentata forms such an unimportant part of the
outside plankton, its presence in the region of Woods Hole is, no doubt, so rare that
the distribution curve of Sagitta for any year can be considered to be the seasonal
variation of S. elegans. A sudden appearance after July and before November

134 BULLETIN OF THE BUREAU OF FISHERIES

would probably follow a southwest wind, and in this case the species might be
S. serrodentata, although deep-water collections off the coast in warm weather often
reveal large numbers of S. elegans. Such a condition may have taken place in August
1903 (see fig. 35). On August 4, 1922, one specimen of S. serrodentata was taken
and another on August 5.

In the 16 years that S. elegans has been recorded, with one or possibly two excep-
tions, none appeared before November or remained after July. The usual time of
appearance is December. In 1899 a few were taken on December 23, and in 1898
many suddenly appeared on December 12. In 1922 two specimens were found on
October 4, one on October 5, two on October 10, and gradually increased from then
until early December, when large numbers appeared. The highest point is usually
reached in February. During this month they swarm.

It is interesting to compare these results with those of Dr. H. B. Bigelow (1914)
in Massachusetts Bay. In late December he found S. elegans in the tow. Through-

So her good gd 008 Bling comanng
4
s
Ve5s
Ne

Fig, 34.—Occurrence of Sagza evegans in surface collections from June, 1922, to Vecember, 1923.
, distribution in 1922; —.. —., distribution in 1923

out January and February the numbers increased until they formed the bulk of the
plankton. Occasionally S. serrodentata was taken, but always S. elegans was by far
the most abundant. When the water began to grow warmer in early March, the
numbers fell off rapidly, so that on March 4 only 12 specimens were taken. The
last Sagittze appeared on April 14. This is merely additional evidence of the
similarity of plankton north and south of Cape Cod in winter.

In March and April, 1923, swarms of S. elegans with ripe eggs were abundant in
Great Harbor. During the latter part of April large numbers of eggs appeared and,
together with the eggs of the mollusk Littorina litorea, made up the greater part of
the tow. On May 2 the first young were observed. These increased rapidly in
number and were very abundant throughout May and June. The last specimen was

PLANKTON OF THE WOODS HOLE REGION 135

taken on July 18, although the numbers had been very small since June 20. In
August, 1923, large numbers of small Sagittz of the spring brood were taken off No
Man’s Land in deep water.

Many species of Platyhelminthes and Nemathelminthes have been recorded from
surface collections, but these have been accidental in occurrence and, with the
exception of certain early larve, do not form a part of the littoral plankton. Most
members of the phylum, excluding internal parasites, live among the marine plants

Jane
Feb.
Mare
Apre
May
June
Jule
AUS.
Octe
Nove
Des

Fic. 35.—Occurrence of Sagitta elegans in surface collections during successive years. Broken
lines indicate scattering occurrence

and detritus on muddy bottoms or on piles. Some forms like the rotifers, of which
a few marine genera occur at Woods Hole, swim about freely, but even these are not
a part of the open-water plankton. Only one rotifer (Syncheta triopthalma Lauter-
born) was observed during the past year, and this was seen but once.

Often in summer planarians appeared, but no attempt was made to identify them.
One species, however (Microstomum davenporti Graff), was taken in the harbor on

136 BULLETIN OF THE BUREAU OF FISHERIES

two occasions, August 5 and 16, and in the Sound on August 18 and September 21.
A single specimen of Nectonema agile Verrill was found among much detritus after
a hard wind on July 11.

MOLLUSCA

Gastropod larve are found throughout the year in all surface collections from
inshore waters. There is considerable doubt as to the percentage of forms whose
early stages are free-swimming. Many species, such as Busycon canaliculatum
(Say) and Buccinum undatum Linnezus, secrete cases in which the young pass
their early stages, emerging in the form of the adults. Others deposit eggs in jelly-
like masses attached to the underside of rocks and on marine plants. Littorina
palliata (Say) is an example of this type. Still other forms, such as L. rudis Maton,
are viviparous. The eggs of all of these are never found floating, and the young
normally do not appear in the plankton. Certain young after emerging from the
egg cases may accidentally be carried along by the currents. This probably explains
the presence of many species taken during the summer and fall.

A fourth group of gastropods no doubt contribute the bulk of the planktonic
larve. This group, of which Littorina litorea and Lacuna vincta are examples,
discharge their eggs directly into the sea water. In these two species each egg is
especially adapted for floating by a surrounding ring of jelly, which gives the appear-
ance of the trench helmets worn by the American soldiers in the late war. This
device serves also as a means of protection. The eggs and free-swimming larve
are found in great numbers from March until July. This is also the breeding
season of Littorina litorea in English waters, according to Tattersall, who made
extended observations upon that species. Lacuna vincta also swarms in February
and March, some eggs having been found as early as December by Sumner. The
eggs of this species may be distinguished by a light greenish tinge. In March of
the present year (1923) great numbers of Littorina eggs appeared daily and in-
creased throughout April. None were found in collections of the previous June.
There was a maximum of eggs identical to those of Littorina in the fall, which the
author has not been able to identify. There can be little doubt that this floating
condition explains the rapid expansion of Littorina litorea after it was once estab-
lished on the western Atlantic coast.

An interesting adaptation to pelagic existence is found in a larval vitrinellid,
the species of which I have been unable to ascertain. Shortly after the nucleus
has formed, a broad shield grows out as an extension of the shell. This shield
appears like the wide brim of a straw hat and enables the larva to float. Later,
as older specimens showed, the shield is lost and the young mollusk sinks to the
bottom. It is an interesting adaptation and has never before, to my knowledge,
been noted.

In the summer of 1922 Dr. Paul Bartsch kindly aided me in identifying the
gastropod larve that appeared during June, July, and August. The many forms
often bear no resemblance to the adults, but are identified by comparing the nuclear
whorls. These never change and offer an excellent means of identification. The
nucleus is now used as a basis for classification among adult mollusks also.

PLANKTON OF THE WOODS HOLE REGION 137

Apparently none of the larger gastropods have free-swimming stages, the bulk
of the summer forms coming from those minute species that live on the floating
Fucus and Sargassum. The following gastropods were. distinguished in surface
collections of 1922:

Littorina litorea (Linneus). Tritonofusus stimpsoni (Morch).
Bittium alternatum, Say. Triphoris nigrocinctus, Stimpson.
Astyris lunata (Say). Lacuna vincta, Montagu.
Skenea planorbis, Fabricius. A vitrinellid genus (?).
A combellid, probably Anachis avara

(Say).

But one Nudibranch mollusk has been recorded by Edwards from surface
collections. This species (Facelina bostoniensis (Couthouy)) appears each spring,
often in large numbers. This year it appeared on January 21 and continued to
be taken until May. Upon examination many females were found to contain
dozens of small larve, which were very similar in form to the adults. Four other
species were represented by single specimens taken during the year—Flysiella
catula (Agassiz), June 1; Doto coronata (Gmelin), September 6; Alderia harvardi-
ensis (Agassiz), March 29; and family Dotonide, November 8.

Clione limacina (Phipps), a pteropod, is often taken in large numbers around
Marthas Vineyard. It is a member of the oceanic plankton and is occasionally
blown into Great Harbor during southern storms. ‘The author has five records
of its appearance in surface collections. The first four—September 10, 1888,
March 20, 1896, April 28, 1911, and May 2, 1911—-were taken by V. N. Edwards;
the fifth, on May 3, 1918, by R. A. Goffin. Another pteropod (Heterofusus retro-
versus (Fleming)) had been recorded once in local waters (Sumner, 1913a). A
single specimen was taken in Great Harbor on January 12, 1924.

The larval Pelecepoda present a most difficult problem to the plankton investi-
gator. The early larve all look alike and can be distinguished, with any degree
of certainty, only by careful measurements. During the summer the author was
able to make few such measurements and for that reason the results are very in-
complete. The late larval stages are more easily distinguished. J. Stafford’s
excellent paper on bivalve larve made the identification of these forms a rather
simple matter. At this stage, however, the bivalves sink to the bottom and are
taken in much smaller numbers.

The most common pelecepods of this region live in the shallow waters of pro-
tected bays and harbors. For that reason they are quickly affected by the increase
in temperature during the spring. The length of time required for the ripening of
the gonads is not known, but many larve of Mytilus edulis were found early in
June, 1922. Later in July larve of a slightly different shape were noted. These
proved to be the young of both Venus mercenaria and Mya arenaria. Many
Pecten larva were taken near Block Island in September, but none appeared at
Woods Hole. Mya, Venus, and Mytilus remained throughout the summer and
until late in the fall. By August 10 Mytilus had almost acquired the adult shape
and appeared less frequently in the collections, although many were taken through-
out November and December. By this time the larve had long since passed the

8242°—25}—_4

188 BULLETIN OF THE BUREAU OF FISHERIES

swimming stage and were carried into the nets by the strong currents. No oyster
larvee were noted during the summer of 1922.

Of the Amphineura one species (Chetoplewra apiculata Carpenter) was taken
on September 23, 1922. The larva was at that time in a late stage of development,
the shell measuring 1.2 mm. in length. However, the free-swimming period had not
ceased, for the little animal continued to float about in a watch glass for several
hours.

Throughout the latter part of May and June the eggs of Loligo pealii Le Sueur
are found in great abundance. Scattering young forms appeared on June 2, 1922,
and increased rapidly until July 11, when the largest number was taken. On clear,
calm days small schools of these little cephalopods could be seen swimming at the
surface in much the same manner as the adults. Such schools were particularly
common about the fish traps, where large numbers of adults are frequently captured.

> ] a & five he 4
het Sa ae: Oe aes, Pe
Vea. Fog 8 eB. 2098 org
z B & & @ & & a
Woke
Ae + 4
A !
B rs : f
S. Shee
=n 3.
# ons gs
: : ete Gon : H Hit 4
VeS. ct : VoSe L
me EE FTE
Ne.
Fia. 36.—Occurrence of larval forms of Loligo Fig. 37.—Occurrence of Phyllopoda in surface col-
pealii in surface collections of 1922 and 1923. lections from May to December, 1922. —..—,
, distribution in 1922; _______, » dise Podon intermedius; , LEvadne norde
tribution in 1923 manni}: =-—-j2225 , E.tergestina

In August there was a decrease in the abundance. ‘This continued throughout that
month and early September. Two specimens were taken in October and one on
November 20. The last occurrence is rather surprising, because no young forms
had been seen since October 18, and then only one specimen was found. In 1899
the last specimen was taken on October 24. In 1923 the season lasted from June 26
until October 16 (fig. 36).

ECHINODERMATA

Practically all of the echinoderms of the Woods Hole region have a free-swim-
ming stage. A few holothurians and one starfish (Henricia sanguino lenta (Miller)),
are viviparous, but these are uncommon forms. In certain years great numbers
of the larve of Asterias have been taken in surface towings. None were found in
collections of 1898-99 nor during 1922, although Asterias is known to breed through-
out the summer months in this region. In 1923 a single brachiolaria of Asterias

PLANKTON OF THE WOODS HOLE REGION 139

appeared on July 16, it being the only specimen taken that year. In Narragansett
Bay the season is usually completed in a few weeks in late June; after that hardly
a ripe adult can be found. As four species of Asterias have been recorded from
Woods Hole it is probable that all do not breed at the same time. This might
account for the extended breeding season.

A specimen of Leptosynapta inherens (Miller), 20 mm. long, was taken on
September 19 after a hard northeast storm. This was not a free-swimming form
and would not normally occur in surface collections.

CRUSTACEA
PHYLLOPODA

Two species of marine Phyllopoda (Podon leuckarti and Evadne nordmanni)
have been recorded from the Atlantic coast of the United States. D. L. MacDonald
records three species from St. Andrews, New Brunswick, two of which (Z. spinifera
Miiller and Podon jfinmarchichus) have never since been taken. As the name of
the original describer does not appear on the list, I am unable to find any other
record of P. finmarchichus. This name is not given in any available literature on the
subject. EH. spinefera is a southern form that has not appeared in this region during
the past year.

Two species of Evadne were taken at Woods Hole in abundance during the
summer of 1922. Hvadne tergestina, new to this region, appeared on May 20, be-
coming very numerous by July 1. During the summer diatom maximum the num-
bers decreased but rose again in September. After that they declined until Novem-
ber, the last being recorded on November 15.

Evadne nordmanni appeared shortly after EF. tergestina, but never became
abundant in the summer months (fig. 37, p. 138). In October they increased and
reached their highest point about November 1, at a time when E. tergestina
was fast disappearing. Throughout December they declined rapidly and disappeared
about January 20. FE. nordmanni is easily distinguished by its pinkish color as
well as its different appendage formula. £. tergestina is usually quite colorless
and very transparent.

Podon intermedius was first recorded from the western Atlantic by MacDonald
at St. Andrews, New Brunswick. This species appeared in the surface collections of
Great Harbor on May 27, 1922, and increased rapidly, reaching a high point in the
last week of June. The numbers declined during the period of the diatom swarms,
but rose again, reaching the peak in the middle of September. Another diatom
maximum in early October reduced the number a second time, but they once more
rose and remained until the last of the month. During November P. intermedius
became scarcer and disappeared about December 15. In general, the season is the
same for the various species. Hvadne nordmanni has the longest occurrence. The
distribution of P. intermedius in 1923 was very similar to that of the previous year,
except that it arrived later (fig. 38).

No specimens of Podon leuckarti (Sars) were taken during the past year, and a
careful search through the collection of 1899 and 1900 failed to show any, although

140 BULLETIN OF THE BUREAU OF FISHERIES

°

Pratt and Sharpe recorded them as occurring in great abundance. No specimens
have been placed in the National Museum, and as Sharpe’s collections were lost I
have been unable to find any identified material. It seems strange, however, that a
species not recorded from the region appeared in such great abundance, while the
common form was absent during those three years.

On July 28, 1923, Podon polyphemoides appeared in the surface tow. No
specimens of this species had been observed in the collections of the previous year
or in 1899 to 1900. The season was very short, lasting less than four weeks. The last
specimen was taken on August 22. At the mouth of New Haven Harbor in Long
Island Sound, August 1 to 3, 1923, swarms of this species were observed. They

Jane}
Fede
Mare
Apre
May
JUNe
Jule
Auge
Sepa
Octe

Ae

Se

VoSe

Ne

Fic. 38.—Occurrence of Phyllopoda in surface collections of 1923. ——, Podon intermedius; —eee—, P. polyphemoides;
—eo, Evadne nordmanni; --e---=-, , E. tergestina

were so numerous that a surface tow of 15 minutes yielded 80c. c. of P. polyphemordes
and almost nothing else.

The following phyllopods appeared in the surface collections of 1922-23: Podon
intermedius Lilljeborg, P. polyphemoides (Leuckart), EHvadne nordmanni Loven,
and F. tergestina Claus.

OSTRACODA

With few exceptions the ostracods are not true planktonic animals. None of
the Woods Hole species belong in the pelagic group, although many appear in sur-
face collections after storms or hard winds, along with particles of sand, Foramini-
fera, and other bottom forms.

Cushman found that, excepting one specimen, all species of the Myodocopa
taken in the survey of Vineyard Sound and Buzzards Bay came from the “‘ Gut of
Canso,” directly across the harbor from the fisheries station.

PLANKTON OF THE WOODS HOLE REGION 141

In the collections of the past year one of this tribe (Cylindroleberis marie)
appeared with greater frequency than any other one species, even though the
Podocopa are much more abundant at certain spots in Great Harbor. This instance
shows how easily wrong conclusions may be made in the study of littoral plankton if
the bottom fauna is not clearly understood. It illustrates, also, an important
point about the fauna of the harbor. The bottom forms dwelling here are so dis-
tributed that they are protected from the rushing currents, although they are able
to derive benefit from the food material carried by these waters. For that reason,
even under unusual conditions, the benthos occurring in surface collections proba-
bly is transported from Buzzards Bay. This is quite evident in the case of amphi-
pods where the distribution is very well understood. .Even the animals of the
“Gut of Canso”’ are carried away rarely, and the ostracods become dislodged only
when the hydroids and Fucus, to which they attach themselves, are torn from their
bases.

The following ostracods were taken in 1922-23: Sarsiella americana Cushman,
Cylindroleberis mariz (Baird), C. zostericola Cushman, Lozoconcha impressa (Baird),
Cythereis emarginata Sars, and genus Cythereis (several species).

COPEPODA

Together with the Phyllopoda and an occasional euphausid or hyperid, the
Copepoda form the only truly pelagic Crustacea of the local plankton. Except in
the seasons of diatom maxima, they are always present in abundance. Farran
found that whenever a species is present in sufficient numbers a distinct periodicity
in its occurrence is noticeable. This is true at Woods Hole. Although copepods
are always present in varying numbers, certain species are continually disappearing
and being replaced by others. The copepods of Great Harbor may be divided
roughly into two great groups—the summer community and the winter community.

The summer forms may arise from three sources: (a) Annual appearance of
local coastal species common to the region, (6) the young of these common forms,
appearing often in large numbers during the breeding season, (c) southern oceanic
forms blown in by winds from the Gulf Stream during the warm weather.

The first of these sources accounts for most of the summer species. These
may again be grouped under two headings: (1) True pelagic species and (2) bottom
forms appearing after hard winds. The most typical summer pelagic species are
Acartia tonsa and Centropages typicus. These form the bulk of the summer copepod
fauna. Later in the fall Pseudodiaptomus coronatus reaches its maximum and
outnumbers all other forms. This, however, is not a true summer species, but
serves as a connection between the warm and cold water copepods. Tortanus dis-
caudata serves in a similar capacity in the spring and early summer. Benthonic
adults of the family Harpacticide are often taken in surface collections. These are
usually found among bottom plants and alge but are capable of swimming quite as
well as the Gymnoplea. The most common summer Copepoda are Acartia tonsa,
Centropages typicus, Pseudodiaptomus coronatus, Labidocera exstiva, Oithona similis,
O. brevicornis, Alteutha depressa, Parategastes sphericus, Amphiascus obscurus,
Ilyopsyllus sarsi, and Dactylopusia vulgaris.

142 BULLETIN OF THE BUREAU OF FISHERIES

The young of the summer copepods never appear in large numbers, as in the
case of winter breeders, and only three species—Acartia tonsa, Pseudodiaptomus.
coronatus, and Centropages typicus—were identified.

The third summer group varies considerably in different seasons. If the
prevailing winds through June, July, and August are from the south, great numbers.

of Gulf Stream forms may appear. Such was the case in 1922, and for that reason
several species new to this coast were

a ae |e eae | taken. The common annual visitors also
Fobs a We Oe eee ees oe
H Vo.
Be
a
Se
Se
ToSe
WS.
Be - 5 fo :

Fic. 39.—Occurrence of species of Acartia in sur-
face collections from June to December, 1922. Fic. 40.—Occurrence of species of Acartiain surface collec-

ACOTLIG LONE OS Fa iexten inal tee , immature A, tions of 1923. —-.—, Acartia tonsa; , A, bifilosa;
LOTS fees SACRE FELOS ees et meng ACL UL S10 ee eo , A. clausii; —-——, A. longiremus

appeared in abundance. The southerly winds did not continue in the fall, however,
and the result was that the usual tropical fish and ccelenterates were not observed at.
Katama Bay and in Vineyard Sound.

| ee OR: Ce aa a No doubt these conditionsa ffected cope-
HE pods as well. As an illustration of this
Ae Microsetella rosea appeared in great
LP py a) poppe agg
Be es =
eine ~. F Be
q.s. i { ae Be

VeSe

Ne s.

Fig. 41.—Occurrence of Pseudodiaptomus coro-
natus and Tortanus discaudata in surface col-
lections from June to December, 1922.
P. coronatus; —.—, 7. discaudata

numbers on September 2 in vertical hauls taken off Block Island. Later during this

month (fig. 46, p. 145) scattering specimens were observed at Woods Hole. Much
larger numbers would probably have been found here if hard south winds had

Fic. 42.—Occurrence of Pseudodiaptomus coronatus and Tor-
tanus discaudata in surface collections of 1923. PB
coronatus; —=—e—, 7. discaudata.

:}

PLANKTON OF THE WOODS HOLE REGION 143

prevailed. The summer forms from the Gulf Stream taken in 1922-23 were
Pontella pennata, P. meadii, Anomalocera patersoni, Microsetella rosea, Setella gracilis,
and Thaumaleus claparedvi.

No distinct division can be made dividing the summer forms from the winter
ones. Figures 40, 42, 44, etc., show clearly how much the seasonal distributions of
the various species overlap each other. Certain forms, such as Centropages hematus,
appear as early as August and remain until May. As the breeding season is in
December and January, they are considered to be true cold-water forms.

The winter copepods may roughly be divided into four groups: (a) Those
northern species that remain in deep water or north of Cape Cod during the sum-
mer, entering this region every winter in great numbers, (b) the young of the winter
species, (c) northern oceanic forms occasionally finding their way in, (d) Har-
pacticide, usually acci-

dental members of the Bea te ee Bd a eae eg

ae

Teds

Fic. 43.—Occurrence of Centropages
in surface collections from June to

December, 1922. , Centro- Be

pages typicus; -----, » C. hematus; Fic. 44.—Occurrence of Centropages in surface collections of 1923.
—e—, C. hematus (immature ———, Centropages typicus; , C. hematus—eco—, C. typicus
forms) (immature form); —-.e—, C, hematus (immature form)

plankton, but in a few cases rising to the surface during the breeding season.

Three copepods are usually characteristic of all winter plankton—Pseudo-
calanus elongatus, Temora longicornis, and Centropages hematus. During the
years 1922 and 1923 almost no specimens of Temora appeared. This is very
unusual, for all samples of past years taken at this season are literally filled with
them. As they appear in the greatest numbers in February, March, and April,
the unusually cold weather of the spring of 1923 (fig. 5, opp. p. 100) may have affected
them as it has many of the other animals. The young of Pseudocalanus and
Centropages became so abundant in January and February that they far out-
numbered the adults, a condition which was never found among summer forms.
A few immature Temora were noted, but their appearance was not common.

Northern species are sometimes plentiful in the waters of Vineyard Sound and
often appear in surface collections in Great Harbor. Calanus finmarchicus is the
most common of these cold-water forms. Metridia lucens, Eurytemora herdmant,
and E. hirundoides were taken often during the spring of 1924. No other northern
copepods to my knowledge have ever been recorded from Woods Hole.

144 BULLETIN OF THE BUREAU OF FISHERIES

Members of the family Harpacticide sometimes appeared during the winter
months. Only one species (Tachidius brevicornis) had a definite free-swimming
period. Egg-bearing females were taken in towings throughout the spring, often
in great abundance. This, apparently, was the only one of the group that had a
pelagic period during the year. Others may have been free-swimming but did not
occur in sufficiently large numbers to indicate it.

wits bred

&e

Se

VeS-

Ne

Fia. 45.—Occurrence of Pseudocalanus elongatus in surface collections from June, 1922, to December, 1923.
-----, distribution of adults in 1922; —- —- , distribution of immature specimens in 1922; ————, dis
tribution of adults in 1923; —-.—, distribution of immature specimens in 1923

The winter forms collected during the past year were as follows:

Pseudocalanus elongatus. Acartia clausii.
Calanus finmarchicus. A. longiremus.
Centropages hematus. A. bifilosa.

Temora longicornis Tortanus discaudata.
Eurytemora herdmani. Microsetella norvegica.
E. hirundoides. Idya furcata.

Metridia lucens. Tachidius brevicornis.

Over 50 species of parasitic copepods have been recorded from Woods Hole.
Often they are taken in surface collections, but they do not normally form a part
of the plankton except in their larval stages. None appeared in 1922. In 1923 a
male Caligus schistonyx was taken. ‘

Three lists of free-swimming copepods have been made for this region. Wheeler
recorded 30 species, but most of these were taken in the vicinity of the Gulf Stream
and are extralimital. Sharpe recorded 60 species in 1911, of which only 23 occurred
at Woods Hole. Twelve others were quoted from Williams’s report on Narragan-
sett Bay, and the remainder were taken from Wheeler’s list. Sumner, in 1911,
compiled 25 (plus 1%) species from the combined data of Wheeler and Sharpe, no
new additions being made.

PLANKTON OF THE WOODS HOLE REGION 145

During the past year 42 species of free-swimming copepods appeared in the
surface collections taken from the end of the Fisheries dock. Of these, 19 belong
to the tribe Gymnoplea and 22 to the tribe Podoplea. In Sharpe’s list 12 species
from this region belong to the Gvmnoplea. The list for this tribe. I believe, is

Ota ee Ler. Ve ef ne Oe * le

Sasa eee es ee Ts 6g UR Ul ee
a Ct)

Bem oS ee Ohiaw os ee Bl oS Oe OR et Ug

Calanus fimmrchious
Pseudocalanus elongatus
Inmature P.elongatus
Paracalanus parvus
Centropages typicus
Centropages hematus
Immature Cehematus
femora longicornis.
Eurytemora herdmani
Eurytemora hirundoides
Metridia lucens
P.coronatus

Labidocera destiva
Pontella meadii
Anomalocera patersoni
Acartia tonsa

Acartia clausii
Acartia longiremis
Acartia bifilosa
Tortanus discaudata
Oithona similis
Wicrosetella rosea
Microsetella norvegica
Setella gracilis
Thaumaleus claparedii
Harpactieus chelifer
Harpacticus umiremis
Altevths depressa fs
Parategastes sphdericus
Idya furcata f
Dactylopusia vulgaris
Tachidius brevicornis
Amphiascus obsourus
Parawestwoodia minuta

Longipedia coronstus
Tlyopsyllus sarsi

Fic. 46.—Occurrence of Copepoda in surface collections from June, 1922, to May, 1928. (Oithona brevicornis

now fairly complete.

is not distinguished from O., similis)

The Podoplea, however, have scarcely been touched and will,

no doubt, yield many more species when carefully studied. Twelve species taken in
1922-23 are new to the Woods Hole region. I have not listed as new any forms
previously recorded from Narragansett Bay.

146 BULLETIN OF THE BUREAU OF FISHERIES

May
June
Jule
AUS 6
Sep.
Octe
Nove
Deca

Alteutha depressa
Anomalocera patersoni
Calanus finmarchicus
Céiligus schistonyx
Barytemora herdmani
Halithalestris croni
Harpacticus chelifer
Iabidosera aestiva
Llyopsyllus sarsi
Oithona brevicornis
Oithona similis
Paracalanus parvus
Parategastes sphaerious
Pontella meadiz
Pontella pemnata
Tachidius brevicornis
femora longicornis

Fic. 47.—Occurrence of certain copepods in surface collections from May to December, 1923

The following Copepoda were taken during 1922-23:

Tribe GYMNOPLEA Tribe PopoPLEA
Family Calanide: Family Cyclopide:
Calanus finmarchicus (Gunnerus). Oithona similis, Claus.
Pseudocalanus elongatus (Beeck). O. brevicornis, Giesbrecht.!
Paracalanus parvus, Claus.! Hermanella sp.
Family Centropagide: Family Harpacticide:
Centropages typicus, Kréyer. Microsetella rosea, Dana.
C. hematus (Lilljeborg). M. norvegica, Boeck.
Temora longicornis (Miiller). Setella gracilis, Dana.!
Eurytemora herdmani, Thompson and Thaumaleus claparedii.
Scott. Harpacticus chelifer (Miiller).
E. hirundoides (Nordquist). Alteutha depressa, Baird.

Harpacticus uniremis, Kroyer.
Parategastes spericus (Claus).
Idya furcata (Baird).
Dactylopusia vulgaris, Sars.
Laophonte sp. A.?

Metridia lucens, Beck.

Pseudodiaptomus coronatus, Williams.
Family Pontellide:

Labidocera xstiva, Wheeler.

Pontella meadii, Wheeler. L. sp. B2
P. pennata, Wilson.* ‘ Tachidius brevicornis (Miiller).
Anomalocera patersoni, Templeton. Amphiascus obscurus, Sars.!
Acartia tonsa, Dana. Parawestwoodia minuta, Claus.!
A. longiremis (Lilljeborg).' Longipedia coronata, Claus.
A. bifilosa, Giesbrecht.! Ilyopsyllus sarsi, Sharpe.
Tortanus discaudata (Thompson and Asellopsis sp.
Scott). Halithalestris croni (Kréyer).
Acartia clausii, Giesbrecht. Family Caligide: Caligus schistonyx, Wilson.

1 New to Woods Hole. 2 Both differing distinctively from L. longicaudata Boeck. ‘New to Woods Hole.

PLANKTON OF THE WOODS HOLE REGION 147
CIRRIPEDIA

At certain seasons of the year barnacle larvee are very abundant in the plank-
ton. In both the nauplius and “‘cypris”’ stages they swim freely, although as a rule
the “‘cyprids”’ settle on the Fucus soon after the metamorphosis and are not taken
in large numbers in surface collections. At such times often thousands can be
taken in a single sweep of a hand net drawn through the Fucus near the water’s
edge. The nauplii of the three species of Balanus are so much alike that even
the most careful identification is often rather uncertain. However, the difference
in the breeding periods makes the identification easy in the field, although in certain
years the seasons of Balanus crenatus and B. eburneus overlap.

Balanus crenatus is not as abundant in the immediate vicinity of Great Harbor
as are the other species of the genus, and for that reason the larve occur in much
smaller numbers in surface collections. The breeding season starts early in June
and generally continues until the middle of July. In 1922 (fig. 48) the first larvee

324968 6428 &€ 2.8 3

Vode
a.
3. af
w3,
Ne
Fig. 48.—Occurrence of barnacle larve in surface Fia. 49.—Occurrence of barnacle larve in surface collections
collections from June to December, 1922. of 1923. , Balanus balanoides nauplii; ------, B.
=——aoom, Balanus balanoides; .---==a0, B. balanoides ‘‘cyprids”’; —-—-——, B. crenatus; ——o—,

crenatus; =e —, Chthamalus stellatus;
B. eburneus; +, Lepas sp.

B. eburneus; +, Chthamalus stellatus

appeared on July 2; the last on July 16. They were abundant on only one day—
July 18. After this a single specimen was taken on July 15 and one on July 16.
It is possible that nauplii may have occurred after August 1, when fairly large num-
bers of B. eburneus suddenly appeared. However, as an interval of 15 days elapsed
between the two periods, the possibility of a stray B. crenatus nauplius being present
would probably be so small that it need not be considered. In 1923 the first speci-
men appeared on June 29 (fig. 49). Scattered nauplii and ‘‘cyprids”’ were taken
until July 23. Off Coney Island, N. Y., swarms of early nauplii (no doubt B.
crenatus) were taken on June 12, 1923.

Balanus eburneus is usually found in August, although the nauplii seldom form
an important part of the surface collections. This may be due to the fact that the
summer plankton is particularly rich and the barnacles, therefore, are greatly
outnumbered. It is certain, however, that they never appear in such swarms as
does B. balanoides. The first nauplii in 1922 (fig. 48) appeared on August 1, scat-

148 BULLETIN OF THE BUREAU OF FISHERIES

tering individuals being taken ‘until November 12, when the last specimen was ob-
served. In 1923 the first nauplii appeared on August 12 (fig. 49).

Balanus balanoides appeared first on December 16, 1922 (fig. 48). By January 1
great numbers filled the tow. An examination of adults at this time showed that
almost every specimen was filled with young and all seemed to be at exactly the
same stage of development. In 1923 the first nauplii appeared on December 7
(fig. 49). On February 8, 1923, the first “‘cypris larve” appeared. These were
at all times far less abundant that the nauplii. Throughout February and March
they continued to appear, declining in April, although a few specimens were found
in every haul. In certain parts of Long Island Sound, on March 5 and 6, the
“‘eypris larve’”’ were exceedingly abundant. The season in 1899 and 1900 coin-
cided exactly with that of 1922 and 1923.

A comparison of this locality with other places along the coast is necessary
in order to understand the relative position of Woods Hole. In Massachusetts
Bay Bigelow found nauplii of Balanus balanoides throughout March and early
April, 1913. Nauplii swarmed off Boon Island on April 5 of the same year. By
' April 9 large numbers of the “cyprids” with few nauplii were observed, while
collections of seven days later revealed only ‘‘cyprids.’”’ These were most numerous
from April 25 to 30, when they formed the bulk of the macroplankton, and con-
tinued to appear as scattered forms until the middle of May, when all had practi-.
cally disappeared.

In early March swarms of Balanus were found in the “‘cypris stage” among
the Fucus along the shores of upper Narragansett Bay. Some were already
attached. In Newport Harbor the author found large numbers of nauplii from
January 25 to 31, 1922. The largest swarm appeared on January 30. On March
4 of the same year “cyprids”’ literally filled the waters in the harbor of Bristol,
R. I. It was not possible to carry on further observation in this locality, so the
duration of the season was not determined. The author has taken nauplii in
upper Narragansett Bay in large numbers in late January.

From these records it appears that the breeding season in Narragansett Bay
and vicinity is somewhat later than at Woods Hole. This may be because the
water responds more quickly to sudden drops in air temperature and retards the
developing eggs. As one goes farther north the season grows later. Thus, Dr.
Bigelow found that the breeding time starts in March in Massachusetts Bay and
terminates quickly, due to the apparent rapid development of the larve. In
Newfoundland the breeding season of this species is in June and July.

Chthamalus stellatus, although quite abundant locally, appeared in very small
numbers in the plankton on only two days in 1922—August 15 and 16. No “‘cyprid
larve”’ were found. In 1923 a single specimen was taken on July 23.

A single nauplius of Lepas appeared in the collections on September 30, 1922.
This larva often occurs locally, although the adults are not real residents of the
region but are blown in by southerly winds and often appear in great numbers on
floating logs and Sargassum. During such seasons the larve of several species
are frequently found.

PLANKTON OF THE WOODS HOLE REGION 149
ARTHROSTRACA,

Twenty-seven species of Amphipoda were taken in surface hauls during the
past year. But one of these (Huthemisto bispinosa), belongs to the pelagic family
Hyperiide. Young specimens were found on five occasions in January. The adults,
which are often parasitic in Aurelia and Cyanea, are usually seen after southerly
winds, when the medusz are blown into the harbor. All other amphipods belong to
the benthos. During the breeding season, however, some species swim at the sur-
face, both in daytime and at night, and are often taken in the tow in large numbers.
Thus, the bottom forms may be divided into three groups, viz: (1) Those that
swim during the breeding season, (2) those that are carried by the currents, and
(3) those forms that for some reason other than the breeding season are attracted
to the surface.

o Py e 3
mye Be. B®
a t i E 3 &
SS
nesta =
Se 7 ‘s5 i “5 sos steal
VS. aah VeS. He -
4 bh
‘: HH
it cH
t PuTiEe HH a4
i ptt
He Ne i
Fic. 50.—Occurrence of amphipods in Fic. 51.—Occurrence of amphipods in surface collections of 1923. Free-
surface collections from June to De- swimming period during breeding season. , Calliopius lxvius-
cember, 1922. Free-swimming pe- culus; —-ese—, C. leviusculus (young); -~------- , DMonoculodes
riod during the breeding season. edwardsi; —-——-—-, Batea secunda; —. —, Gammarus annulatus
EBOteG, “SECUNAG | —«6.c.5 =a,
Monoculodes edwardsi; —. —, Gam-
marus annulatus; ------- , Callio-
pius leviusculus; —-.—, Stenothoé
cypris

In the first group there are two very conspicuous summer breeders. These
can be found in Figure 52, designated by along line. Certain forms, like Caprella,
appear to have such a season, but this is caused by another condition. They live
on hydroids, and as many of these are found floating after every strong wind the
amphipods attached to them will float long after other forms have sunk again to
the bottom. Of the summer forms Batea secunda and Stenothoé cypris are very
noticeable. At times hundreds of specimens were taken in a single haul, many of
the females carrying eggs or early embryos.

On November 6, 1922, Monoculodes edwardsi started breeding (fig. 50). Many
were taken throughout December and on a few occasions in January, the last
occurring on January 21. About the middle of December two other species (Cal-
liopius leviusculus and Gammarus annulatus) suddenly appeared in abundance.
The former often swarmed at the surface in large numbers, and individuals could

°

150 BULLETIN OF THE BUREAU OF FISHERIES

be seen darting about in the water around the Fisheries dock throughout the spring
months. G. annulatus reached its maximum after Calliopius had started to decline,
although the collections of April often contained many specimens of both species.
Verrill records great swarms of Calliopius far out at sea during this season. On
one occasion they were found to be very abundant in the Gulf Stream.

After heavy northeast or southeast storms great numbers of amphipods are
often found in the tow. At such times, however, many species usually appear.
This condition characterizes the group and contrasts it with the first group, where

June
July
Auge
Septe
Octe
NOVe
Dece
Jane
Feb.
Mare
Apre

~
g
Corophium cylindricum
Caprella geometrica
Amphithoe longimana
Amphithoe rubricata
Gemmarus iccusta :
Pontogenia inermis. EEE H 3
Unciola irrorata AH Ht 5 Be

C.emucronatus

Stenethoé cypris H : EEE
Batea secunda :
Elasmopus laevis
Ptilocheirus pinguis
‘Caprella linearis

Paraphoxus spinosus
Ampelisca compressa
Ampelisca spinipes Ht
Byblis serrata ;
Synchelidium spe Hanegueealbaacaneue HH
Jassa marmorata PEELE H H H
Ampelisca macrocephala H+ HHH H
Monoculodes edwardsi H

Calliopius laeviusculus Hy t
Gammarus annulatus dees sscenuseueseadae PH
Grubia compta
Tryphosa pinquis
Euthemisto bispinosa
Euthemisto rubricornis cee

Fic. 52.—Occurrence of amphipods in surface collections from June, 1922, to May, 1923

one or two species make up tne entire amphipod representation. These conditions
are particularly obvious in summer. On July 24, 1922, after a hard northeast
storm, seven species of amphipods and two species of isopods were taken in one
day’s collection. Such heavy offshore winds carry the surface waters out and
cause an upwelling of bottom waters, carrying many of the bottom animals with
them.

The third group appeared onlyin summer. It was made up of the same species
as the second group, but these occurrences were the result of different causes.
‘Throughout the summer and particularly after the great diatom maximum the
water was extremely phosphorescent. At such times the net appeared like a ball

PLANKTON OF THE WOODS HOLE REGION 151

of fire as it swayed back and forth in the current. As the amphipods are positively
phototropic, many, no doubt, are attracted by the light and are drawn into the net.
Another factor as well may influence these collections which were always found more
abundant at night. Experiments have shown that many amphipods rise to the
surface at night and go down in the daylight. If this is true for many of the species,
we should expect to find them more abundant in surface collections taken in the
evening. How much effect this really has upon the plankton hauls I do not know,
but I offer it as a possible explanation. I found no conditions in winter that could

have resulted from such causes. Possibly the evening migrations do not take place

during the cold season.

The following amphipods were taken in surface collections during 1922 and

1923:

Euthemisto bispinosa (Beeck).
Tryphosa pinquis (Boeck).
Paraphoxus spinosus, Holmes.
Ampelisca spinipes, Boeck.

A. macrocephala, Lilljeborg.

A. compressa, Holmes.

Byblis serrata, Smith.
Stenothoé cypris, Holmes.
Monoculodes edwardsi, Holmes.
Calliopius leviusculus (Kroyer).
Pontogenia inermis (Kréyer).
Batea secunda, Holmes.
Gammarus locusta (Linnzeus).
G. annulatus, Smith.

Carinogammarus mucronatus (Say).
Elasmopus levis (Smith).
Ptilocheirus pinquis, Stimpson.
Amphithée rubricata OSs
A. longimana, Smith.

Jassa marmorata, Holmes.
Grubia compta (Smith).
Ericthonius rubricornis, Stimpson.
Corophium cylindricum (Say).
Unciola irrorata, Say.
Synchelidium sp.

Caprella linearis, Linneus.

GC. geometrica, Say.

The Isopoda, with the exception of certain parasites, do not normally form a
part of the plankton. They are most abundant in surface collections in summer.

This is because numbers of Idothea and allied genera are found on floating Sargassum
and Fucus, which, when carried into the nets or forced by them, often deposit many
of their passengers. In winter this condition does not exist and few species are taken.
On one occasion in the spring of 1900 many adult Cirolana concharum appeared in
the tow. No doubt these were floating on a piece of wood or a dead fish which
may have been carried into the net.

The most interesting by far of the isopods taken during the summer were four
minute species of the family Bopyride, which are parasitic on copepods. These
occurred in large numbers at certain times. Two species were found on Acartia
tonsa, one on Centropages typicus, and one on Labidocera xstiva. They were most
abundant from July to October, one specimen appearing unattached on December 20.
None of the winter copepods seemed to be infested. No species have been recorded
from this coast, and as a paper on these forms, now in the course of publication in
England, is not yet completed, it was decided to wait for it before attempting to
identify these isopods.

The following species were taken in 1922-23:

Idothea baltica (Pallas).
I. phosphorea, Harger.
I. metallica, Bose.
Edotea triloba (Say).

Circolana concharum (Stimpson).

Tanais cavolinii, Milne Edwards.
Chiridotea ceca (Say).
Leptochelia savignyi (Kroyer).
Hrichsonella filiformis (Say).
Family Bopyride, four species.

152 BULLETIN OF THE BUREAU OF FISHERIES
CUMACEA

The Cumacea occupy a place in the plankton similar to that of the amphipods.
Large numbers are often taken at the surface during the breeding season, the
females carrying eggs or larve. This particular group differs from the Arthrostraca
in the length of the breeding season. Females of two species (Diastylis sculpta
and Cyclaspis variens) were found carrying eggs at various times between July and
January, although both species were most abundant in September and October.
Females of Oxyurostylis smithi were also found with eggs on October 19. With
the exception of the greater number taken during the breeding season, no particular
time can be given for the occurrence of Cumacea in the plankton. They are found
to be most abundant usually after astorm. D. quadrispinosa, which is reported to be
abundant in this region, was not taken during the past year. The following forms
were taken in 1922-23: Cyclaspis variens Calman, Leptocuma minor Calman,
Oxyurostylis smitht Calman, Diastylis polita Smith, and D. sculpta Sars.

SCHIZOPODA AND STOMATOPODA

The larval stages, and often the adults (Neomysis americana) of the Myside,
at certain times of the year are very characteristic members of the Woods Hole
plankton. The euphausiids, however, are ‘‘outside” forms and appear with other
oceanic plankton only after southwest winds.

The Myside, living among the eelgrass in shallow water, are not true pelagic
animals, but an occasional adult may be carried into the net at any time. Certain
species apparently never swim freely during the breeding season. Heteromysis
formosa and a species of the genus Erythrops (new to the region) are examples of
this type. The former species has been recorded for every month of the year.
Neomysis americana, on the contrary, has a definite pelagic period and swarms
in surface waters from December to April, inclusive. The larve appeared during
the last week of April in 1899 and 1900 and continued in small numbers until
July, the young being liberated in the form of the adults. In 1923 the first adult
appeared on May 17.

Adult euphausiids have been recorded at various times by Edwards, but none
are permanent inhabitants of this region. Their occurrence will be better under-
stood when the distribution of the various species off the coast is more fully worked
out. Five species from the surface collections of 1898, 1899, 1922, and 1923 were
identified. On December 12, 1898, after a hard southwest storm, two Thysanoéssa
inermis and one T. longicaudata were taken. There may be something in the
occurrence of the former species to give a clue to its distribution. Zimmer gives
it a wide range. It is a cold-water form, extending from the Vineyard Sound to
the Gulf of Maine in the North Atlantic, always being found within the 50-fathom
line. Records made to date seem to indicate a northerly migration throughout
the summer months. The specimens recorded from Woods Hole were taken on
December 12. The Albatross found scattered individuals in the deeper parts of
Vineyard Sound in late July and August. Bigelow found it most abundant north
of Cape Ann in early July and on German Bank in August, with minor centers of
abundance off Penobscot Bay and in the northeast corner of the Gulf during the

PLANKTON OF THE WOODS HOLE REGION 153

same month. Just as a northerly movement takes place in summer a southerly
one is noticeable in late fall and winter. More complete data will be necessary to
verify these statements, but it is evident that this species is most likely to be taken
at Woods Hole from late fall until early spring.

The young of Thysanoéssa longicaudata in the late “ cyrtopia’’ stage were com-
paratively abundant from May 10 to June 24, 1899. From this data it would
seem that the adults enter the shallow waters during the breeding season of May
and June. Bigelow found them abundant only in the center of the Gulf of Maine
during the fall. This species, according to Zimmer, is also a cold-water form.
It occurs occasionally in Vineyard Sound and quite frequently out beyond the
Gulf Stream. As the young have never been taken since 1899, it is probable that
the occurrence is not annual, but was due to unnatural conditions. Figure 53
gives the seasonal distribution for that year.

One specimen of Huphausia krohnii was taken on June 22, 1899, and another
on November 9, 1922. Off the Atlantic coast they were taken in abundance in
July and August. This is asouthern species and
may be expected to enter Vineyard Sound in the
summer months. A single specimen of £. tenera Se
appeared on October 30, 1923.

A battered specimen of the genus Thysano- |
poda was taken on June 23, 1922. The condi- BeBs
tion of the carapace made a determination of the Cin. CE ea
species impossible. This was unfortunate be- ea
cause, although three species are recorded from the
western Atlantic, each has been taken on only one
occasion. Thysanopoda xqualis (H. J. Hansen)
was recorded nearest the Woods Hole region. N..

A single specimen of Meganyctiphanes nor- iE
vegica, taken April 25, 1906, was found in the ee ee eee see Seed
surface collections of Mr. Edwards. This is q 18%: (One adult on December 12, 1898)
very common boreal Atlantic species, and it is surprising that more have not been
taken in Great Harbor.

The following Schizopoda were taken in surface collections at Woods Hole:
Thysanoéssa inermis (Kroyer), T. longicaudata Kroyer, Euphausia krohnii Brandt,
Meganyctiphanes norvegica (Sars), Thysanopoda sp., and Euphausia tenera Hansen.

Seven species of stomatopod larve have been recorded from the Woods Hole
region, although but two species of adults occur here. Most of the larve are
Hast Indian forms carried north by the Gulf Stream. The various members of the
order are known to have an extremely long pelagic life with many larval stages.
This, no doubt, accounts for the tropical larve occasionally appearing in Great
Harbor. The larval Squillide are of two forms—the Alima and the Erichthus
form. All the species recorded locally, with the exception of Chloridella, belong
to the latter form.

Adult Chloridella empusa (Say) are rather scarce in the immediate vicinity of
Woods Hole, and for that reason the larve are not abundant in the plankton.

8242°—25 ¢——_5

154 BULLETIN OF THE BUREAU OF FISHERIES

In 1899 a single specimen was taken on August 7.. None were observed in 1922.
Figure 54 shows that the normal season is in August. Edwards’s earliest record
was in 1895, when several specimens were taken in August. His largest captures
were made in 1905, when many appeared on October 21 and 22. Heretofore
adults of Chloridella have been comparatively plentiful, but during the past few
years they have gradually disappeared until they are now very rarely found. This
explains the absence of larve in surface collections of recent years.

The Erichthus larve of Lysiosquilla armata
Smith are among the most common on the south-
ern coast of New England. They are usually
found farther from the coast than Chloridella,
probably because the adults are found in moder-
ately deep water. Chloridella empusa is found on
the muddy bottoms of bays and rivers. Vinal
Edwards took 12 specimens of Lysiosquilla larva
off Gay Head on September 12, 1902. Two speci-
mens were taken in the same locality on August
15 and one on August 25, 1923, in Muskeget

Gl) Sed i (oe i cd cd Ns Joe

rt ttetLrteiei Erichthus larve of two species of the genus
es ne Odontodactylus are recorded by R. P. Bigelow

from this region. One was taken off Nantucket
October 3, 1883, and the other at Woods Hole
August 22, 1876. One of these appears to be the
same as that incorrectly identified by S. I. Smith
(1874) as the larva of Chloridella empusa. His
specimens were taken in Vineyard Sound on
August 11. In 1923 two specimens of Smith’s
species appeared in surface collections from Great
Harbor on August 21 and three on August 22.

Bigelow considers these larve to be West
Indian forms carried north in the Gulf Stream.
Considering conditions existing during the past
summer, this appears to be questionable. In 1922,
when tropical plankton was abundant in local
waters, none werefound. In 1923 no Gulf Stream
plankton or fish were taken, either in Vineyard
Sound or Katama Bay. If these stomatopod larve

della empusa in surface collections of successive ATE from the south, they are apparently the only
years from 1893 to 1907 tropical forms that found their way into shallow
water this year. This seems hardly possible.

On July 17, 1908, Edwards found over 2,000 Erichthus larvee in the stomach
of a small mackerel taken at Woods Hole. Upon examination the author found
them to be the young of the species of Odontodactylus figured by Smith. The
specimens were for the most part entire and were probably found not far from
Great Harbor. As one fish was able to capture more than 2,000, they must have

Fic. 54.—Occurrence of larval forms of Chlori-

PLANKTON OF THE WOODS HOLE REGION 155

been extremely abundant. It is difficult to see how such large numbers could have
remained together in the long journey from the West Indies (where they never
form a very considerable part of the plankton) to our coast and then not be scattered
by the strong winds, which were necessary to blow them in. It is more probable
that they are the young of an unknown species of the genus Odontodactylus inhabit-
ing the deeper waters off the New England coast, possibly beyond the range of

Vee

Ae

Se

VeS

Ne
Fic. 55.—Occurrence of larval Macrura in surface collections of 1922. , Pagurus; eceee, Crago
septemspinosus; ——e —— Palemonetes vulgaris; —-eee—, Naushonia crangonoides; —_e— Hippolyte

zostericola; »eeeeee, Callianassa stimpsoni

Lysiosquilla armata. Two unidentified species of Erichthus larve were taken by
Verrill off Marthas Vineyard in August. One he suggests to be the larva of Pseudo-
squilla ciliata Miers. Both species were no doubt southern forms.

MACRURA

The Macrura form a very important part of the summer plankton. None of
the members of this group are pelagic in adult life except some of the Caridea during
the breeding season, but in all the larve are planktonic.

156 BULLETIN OF THE BUREAU OF FISHERIES

Usually the first larvee to appear in the spring are those of Crago septemspinosus,
but the spring of 1922 was unusually cold and for that reason none occurred during
April. On April 21 several adult females bearing eggs were taken at the surface.
This is characteristic of the species. In Narragansett Bay, on May 7, 1922, great
numbers of adult females bearing eggs, as well as a few young, were taken in surface
collections on a bright sunny day. Bumpus found young forms appearing in March
at Woods Hole, while Thompson observed them as late as September 19. The first
young were seen on February 1 in 1900. After this none were taken until April 3.
From that day on they were abundant, declining in July and August. On October
17 the last specimen was taken. In 1922 the first of this species was noted on May
15, and great numbers were taken throughout July and early August. During the
latter month there was a rapid decline, and none were taken from August 27 until
October 29. On this date four specimens appeared. Scattered individuals were
found in almost every haul until December 13, when a single Crago, 10 mm. long,
occurred. In 1923 the first larvee appeared on May 9 and the last, a specimen 6 mm.
long, was taken on December 13. The maximum was reached early in July. All

Apre
‘June
July
Auge
Sept.

>
|

g
Crago septemspinosus
Palaemonetes vulgaris

Hippolyte zostericola
Homarus americanus
Upogebia affinis
Callianassa stimpsoni

Naushonia crangonoides
Emerita talpoida
Eupagurus spe

Fic. 56.—Occurrence of larval Macrura in surface collections of 1923

available records indicate that the normal season starts early in April, reaches its
maximum in June or July, and usually ends in November.

Palzemonetes vulgaris appears usually much later than Crago (figs. 55 and 56).
Bumpus found females with early eggs on June 20. Throughout July and August
the larve are very abundant, but all breeding ceases by September, according to
Thompson. In 1899 larval Palemonetes appeared suddenly in great numbers in
the tow of June 15. Scattered specimens had been taken for a few days previous.
From June until September 18 young in all stages of development were very abund-
ant. From this date they declined rapidly and had practically disappeared by
September 28, few specimens occurring after this. A single postlarva! individual
was taken on October 31. The first larvee appeared on June 25, 1922. A gradual

‘increase continued until the middle of July, when the maximum abundance was
reached, followed by a gradual decline through August and September, late stages
being taken throughout the month of October. The early larve are rarely found
after the middle of September, however. In 1923 the first specimen was taken on
July 16; the last on August 22.

157

PLANKTON OF THE WOODS HOLE REGION

Hippolyte zostericola was observed first in collections taken in the second week
of July, 1922. Earlier records show that the young may occur at any time after
July 1. The season is much more extended than that of either Crago or Palex-
monetes, for very young specimens are often abundant throughout October. Scat-
tering older larve were taken in November, the last appearing on November 18.
Figure 55 shows the distribution of this species,
which reached its maximum in September, 1922. A
In 1923 four early larvee were taken on July 26.

One late larval stage (4 mm.) appeared on Decem-
ber 13 and one on December 17.

Only three adults, including the type speci- 1895
men, of the rare species Naushonia crangonoides
have been found. Two of these were taken on
the island of Naushon and one on the smaller of
the Weepecket Islands. The distribution is much
broader than has been supposed, however, because
numerous larve appeared in surface collections
from Katama Bay on the seaward side of Marthas
Vineyard. Although the larval forms are never
exceedingly abundant in the surface collections of
Great Harbor, they occurred regularly in small
numbers in almost every tow taken during the
breeding season. The first specimens appeared
on July 8, and the last were taken on September
19,1922. The greatest numbers were found on
July 24, although the average abundance was
higher around August 1 (fig. 55). Figure 56
shows the distribution in 1923.

In spite of the fact that Homarus americanus
breeds in great abundance in all the deeper waters
of the region, larval forms are rarely taken in the
plankton. During the summer of 1922 none were
found in Great Harbor, although a single speci-
men appeared in surface collections from Vineyard
Sound on July 24. As this larva was in rather a
late stage, no doubt it had been clinging to the

Ei |
pele eS SE SL eis ee tel eles:

floating weeds, which were abundant in the net.
The few captures of past years (fig. 57) were, with
one exception, made during June and July. This

Fic. 57.—Occurrence of larval forms of Homarus
americanus in surface collections of successive
years, 1893 to 1907

appears to be the normal maximum season for the species in this region. A
specimen taken on September 12, 1902, probably resulted from heavy winds,
which were prevalent that year. On June 26, 1923, one was found after a hard
southwest wind. A natural conclusion in the matter is that the larval lobsters
under normal conditions do not form a part of the surface plankton but remain

158 BULLETIN OF THE BUREAU OF FISHERIES

near the bottom. Storms and strong currents may carry them to the upper strata,
but the fact that they usually appear on only one or two days a year at the most
indicates that their presence there is not normal.

Certain other Macrura were found occasionally in summer surface collections.
Emerita talpoida was first taken on July 22, 1922, and continued to appear in small
numbers until September 1, when four appeared, the greatest number found on any
one day. Invariably these larvae when placed in a watch glass would cast their
shells and acquire the adult form within 24 hours. It was interesting to observe the
- little creatures as they labored continuously to dig into the glass bottom of the dish.
After a time they would drop exhausted but could be made to resume their activity
by disturbing the surface of the water.

Upogebia affinis was taken twice in August, 1922. In English waters larvee of
this genus are extremely numerous, but such is not the case at Woods Hole. The
adults are not uncommon in this region, and in some years the larval forms may
occur in greater numbers. In 1923 they were fairly abundant. The first specimen
appeared on July 20; the last on October 25. The largest numbers were taken in
early August.

The transparent larve of Callianassa stimpsoni Smith are frequently found in
surface collections. Like Upogebia they are never found in abundance, although
small numbers can usually be taken throughout July and August. In 1922 the first
specimen appeared on July 16. After that scattered individuals were taken until the
middle of August. The first larvee were observed on July 26, 1923. From August
12 to 15 they were unusually numerous but soon declined again. The last specimen
was taken on October 4.

The young of the many species of Paguride found in this region are always
present in large numbers throughout the summer months. They are very similar
to the larve of various Caridea but may be distinguished by the cephalothorax,
which is drawn out in two points on the posterio-ventral margin. M.'T. Thompson
(1903) made a careful study of this group and described the development of the
interesting larvee. The early stages of the various species are almost identical, and
in the case of the two most abundant forms—Pagurus longicarpus Say and P.
annulipes (Stimpson)—it is impossible to distinguish them apart. Thompson
found that P. longicarpus has the longest breeding season, extending from May
until mid-September. Other species with eggs were found at different times
during the summer. On April 8 of the present year (1923) two second-stage larvee
appeared. This is unusually early and far antedates any records for the region.
No other specimens occurred during the month. On May 8, 1922, a single larva
was taken. After this scattering forms appeared until June 1, when they became
very abundant. Together with all other macroplanktonic animals they decreased
during the summer diatom maximum (see figs. 15 and 55). In September the
swarms appeared for a short time but soon declined, the last one disappearing on
November 9. The first Glaucothoé was seen on July 13. After this scattering
forms appeared throughout the summer, although they never were as abundant
as the zoée. Their distribution in 1923 is shown on Figure 53.

PLANKTON OF THE WOODS HOLE REGION 159

The following Macrura were taken in the surface collections for 1922-23:

Crago septemspinosus (Say). Homarus americanus, Milne-Edwards.
Paleemonetes vulgaris (Say). Emerita talpoida (Say).

Hippolyte zostericolaé (Smith). Upogebia affinis (Say).

Naushonia, crangonoides, Kinglsey. Pagurus sp.

Callianassa stimpsoni, Smith.

BRACHYURA

Larval crabs are always present in the summer plankton in large numbers and
form very important food for many fish. As few of the zoéx had been worked out,
they were a source of much trouble until the many forms were finally identified.
The development of the various species will be taken up in a later paper.

All the crabs of this region have free-swimming larval stages, although certain
species are seldom taken in surface collections. The megalops are found in smaller
numbers than the zoéw. Investigation showed that in this stage the young crab
is usually found among the eelgrass and

Fucus. Itcanswimaswellasthezoéa 9... sraratus
but remains closer to the bottom. cancer oreatis
After hard winds large numbers were ©v*24P#5 oveliatus
Callinectes sapidus
3 & é 2 : Carcinides mibnas
pase lie, a lamar hated Libinia spe
Polyonyz macrecheles (Gibbes) Heteroorypta gramlata
Pinnotheres ostreum (Say) . Pelia mtica
Pinnixa sayana Stinpson Neopanope texana say!
ep aisastctpecedts dnc) Polyongs macrosheles
Ovalipes ocellatus (Herbst) Pinniza éhéetopterana
Carcinides maenas (Linnaeus) Pinniza sayana
Planes minutus (Linnaeus) Pinnotheres maculatus
Uca pugnax (Smith) Goa spe
Dea. pugilator (Bosco) yas coarctatus Leach
Fic. 58.—Brachyura occurring rarely in surface Fic. 59.—Occurrence of larval forms of Brachyura in surface
collections of 1922. Eggs of Planes minutes collections of 1923. No observations were made from
(Linnaeus) from an adult taken in the tow August 22 to September 18. Megalops of Hyas coarctatus
were hatched in the laboratory Leach were obtained from Muskeget Channel on August 25

often taken in the nets. The megalops transforms into the “young crab” stage in a
single molt. The ‘‘young crabs” are very rarely found swimming, except in the
species Pinnixa chxtopterana and P. sayana. These have no megalops stage but
change directly from the zoéa into a young crab, which may be compared with
the megalops of other species, for they swim about in much the same manner and are
often very abundant in the plankton.

The zoéz of Uca were rarely taken at the surface. Megalops appeared on only
two occasions, after storms. This seems very strange, because Uca is probably the
most abundant crab found in this region. Hymen reports the zoée as being very
abundant in the surface collections at Beaufort, N. C., at all times during the
summer. I believe that the fiddler crabs of this region have a very short larval
period in which the zoéz as well as the megalops remain at or near the bottom.
Carcinides menas larvee may have similar habits, in this locality at least, for zoés
were taken on only three days in October, 1922, and on one occasion in 1923 (figs.
58 and 59).

160 BULLETIN OF THE BUREAU OF FISHERIES

The females of Pinnotheres maculatus are commensal in Mytilus. The males
swim freely about and were often taken during the breeding season but never after.
Young males in all stages of development were frequently seen swimming. The
young of this species formed one of the most abundant members of the plankton from
July 6 to November 1, 1922 (see fig. 60). P. ostrewm has similar habits but is not as
common as P. maculatus.

The unusual larvee of Polyonyx macrocheles occurred scatteringly from July 26
to October 29, 1922. These peculiar zoée differ from all other forms in the great
length of the rostrum. On July 16, 1892, a sample of towings from Taunton River,
Mass., was found to contain swarms of this species. Hardly anything else appeared.
Faxon found the zoéx swarming at the mouth of Massachusetts Bay in August, 1878.

The adults are exceedingly rare. Agassiz found

Ei 2 Ei é& 3 8 one adult at Newport under a stone, and Doctor
VA. Pe Tennent collected one on Devil’s Foot Island in

a Chetopterus tube. G. Gray reports that
several were found in Chetopterus tubes at Woods
a : Hole in 1922. In 1923 the zoée were taken from
July 20 until August 22 (see fig. 59).

The various species of crabs have definite
breeding seasons, which often overlap each other.
ini, ais Cancer wrroratus appears first, followed closely by
HE Neopanope texani say (fig. 62). In 1922 the first
zoéa of Cancer was observed on May 10. None
ves) Lot ll appeared during April of the present year (1923),

Tete although many females bearing late eggs were
sv TT ERT taken on April 10 in lobster pots. Figures 60,
mT ad ap peut | 61, and 62 show the breeding seasons of the most

_ abundant species taken in 1922, while the scarcer

Fic. 60.—Occurrence of common grapsoid larvee forms appear on Figure 58.

oan Carnation Eph In 1922 another zoéa, almost identical to

that of C. irroratus, was first found on Septem-

ber 8 and continued until October 31. This was undoubtedly C. borealis,

although the megalops were smaller than those of ©. irroratus, a smaller species.

Adult specimens of C. borealis taken at No Man’s Land on August 31, 1923, con-

tained ripe eggs. The first larvee appeared in Great Harbor on October 4; the last
on October 28 (fig. 59).

The following brachyuran larve were taken at Woods Hole in 1922 and 1923.

Ne

Cancer irroratus, Say. Pelia mutica (Gibbes).

C. borealis, Stimpson. Neopanope texana sayi (Smith).
Ovalipes ocellatus (Herbst). Polyonyx macrocheles (Gibbes).
Callinectes sapidus, Rathbun. Pinnotheres ostreum, Say.
Carcinides menas (Linnzus). Pinnixa sayana, Stimpson.
Libinia emarginata, Leach. P. chetopterana, Stimpson.

L. dubia, Milne-Edwards. Pinnotheres maculatus, Say.
Eurypanopeus depressus (Smith). Uca pugnax (Smith).

Planes minutus (Linneus). U. pugilator (Bosc).

Heterocrypta granulata (Gibbes).

PLANKTON OF THE WOODS HOLE REGION 161

PYCNOGONIDA AND XIPHOSURA

Pycnogonids are not pelagic animals, but live on hydroids and among the
alge, occurring in surface collections only when the objects to which they are
attached float into the nets. For this reason they are usually taken during the
summer months. Only one specimen appeared in collections made after October
1. That was on March 29 of the present spring (1923), when a single Pallene
brevirostris Johnston was observed. This species is very abundant in the im-
mediate region of Woods Hole and occurred almost daily during July and August.
Females carrying eggs were found on August 21. On October 1, 1922, a male of
Tanystylum orbiculare Wilson appeared. ‘This was the only member of the species
taken during the past year. A specimen of an unidentified genus new to the

e eo e e e
g fa! 20 2 »
3 Es °
wr KR < wn (o)
b ee) & ee
| Ge) aoe we 8S
VoAo
Ae
Ae
Se
Ss.

VeSe
VS.

Ne Ne —

' Fic. 61.—Occurrence of common larvx of the tribe Fic. 62.—Occurrence of common larve of the tribe
Oxyrhyncha in surface collections of 1922, —-.—, Cyclometopa in surface collections Of 1922. -—__-,
Iibinia emarginata and L. dubia, species not dis- Cancer irroratus; —_—, Neopanope terana sayi;
tinguished; , Pelia mutica —.—, Callinectes sapidus

region appeared in August, 1922, and on July 23, 1923, a single Anoplodactylus
lentus Wilson.

Limulus polyphemus deposits eggs on sandy shores below the low-tide line.
There are not many such spots about the bay in the vicinity of the “Hole,” and
for that reason few young are carried into Great Harbor. In certain localities,
such as the sand flats at Duxbury, Mass., and Cold Spring Harbor, L. I., great
numbers of the young forms in the so-called “trilobite” stage swim about at the
surface. However, heavy shells prevent these animals from being very active
members of the plankton, and consequently they are usually taken only in calm,
shallow water. When disturbed, they become motionless and sink to the bottom,

162 BULLETIN OF THE BUREAU OF FISHERIES

where it is almost impossible to distinguish them from the sand. None were

taken in the surface collections of 1922; in fact, they are recorded only twice by

Edwards in 15 years. In 1899 a few were taken on July 11 and again on July 12.

In 1904 several appeared on August 9. On August 14,1923, a single specimen was

taken. As these are the only times that they have been seen here, it is probable
that they are usually absent in surface collections except after northeast storms,

when specimens may be transported from Buzzards Bay. The specimen taken in

1923 appeared after a hard northeast wind.

CHORDATA

After a storm on July 16, 1922, a postlarval Balanoglossus aurantiacus (Girard)
was taken. On September 9 and 11 of the same year a single acidian larva, 1 mm.
in length, appeared. They were the only representatives of this phylum seen during
the year, excepting the Appendicularia, which at times appeared in great abundance.
There were two species of the latter, one occurring during the summer and fall and

AUSe

p i 3

oe e e e
A, » > °
o. a) ° o
a C2 Ea ~-

Mare
pre

g 1:2
Sm

Ae

V.S.

Ne He
Fic. 63.—Occurrence of Appendicularia in surface collections from June, 1922, to December, 1923, —-. —,
distribution in 1922; , distribution in 1923

the other in winter, the seasons almost overlapping. Both belong to the genus
Oikopleura. On several days in the latter part of July, 1922, single specimens
were noted. In August the number increased until they became very abundant.
Throughout September, October, and November they grew scarcer, rapidly disap-
pearing in December (see fig. 63).

During October and November the “Haus,” characteristic of Oikopleura, was
taken. At times the tow contained hundreds of these pink “‘ Hauser,” each filled
with copepods. One, on October 24, 5 mm. in length, was found to contain exactly
100 copepods; 97 of these were Acartia tonsa, 2 were Centropages hematus, and 1
was Labidocera zestiva. The contents of all had been removed, leaving only the outer
transparent shell. Lohmann found that this ‘‘Haus” was often so delicate that the
most minute organisms, which normally pass through the finest nets, were captured.
In some forms the mesh gradually becomes finer toward one end. Undoubtedly
those taken in my collections were not complete, for the wall mesh was compara-

PLANKTON OF THE WOODS HOLE REGION 168

tively coarse and both ends were broken. Lohmann gives 17 mm. as the average
length of the ‘‘Haus” of Oikopleura albans Loeck. It is difficult to understand
how copepods could be induced to enter such a small opening. Possibly, as in
O. albans, the complete ‘‘Haus” is made up of two compartments—one of coarse
and the other of fine mesh. The currents of water produced by the movement of
the animal’s tail cause microorganisms to collect in the fine mesh. This rich food
center may attract

e
4 ° e ° oO e
the copepods, which gE 2 § e Z 2 a o BR g s é
crowd into the outer apa = rateiVrsls Mats) ico ie

opening. The re- 1893 a |
moval of the soft
partsof thecopepods 1894 eats eas
was no doubt the
work of protozoa. I
have observed them
completely clean out
a decapod megalops yg97 [] | | 1 1 Ij
in two days. The
difficult thing to im- 189g
agine, however, is
how so many cope- 1899
pods could get into
such a small amount 1900
of space. Lohmann
found that a new
“Haus” is secreted
every six hours.
This fact accounts j993 [gm
for the great number
taken. 1904
Only one species
(Orkopleura longi- 1905
cauda (Vogt), listed
by Pratt as Appendi- 1906

cularia longicauda (a SSRs
arc iitias Rg wor te TMi ln elute alindemteste den

corded from the Fia. 64.—Occurrence of Appendicularia during successive years. No record was made
region. Neither after 1904

member of the genus taken this year contains the ‘“‘Kapuze” characteristic of
Pratt’s species. The winter form agrees very closely with, and probably is, O. van-
hoffent Lohmann, while the summer form has many of the characteristics of O. dioica
Fol. At the time lack of sufficient literature prevented a final determination, and
the preserved forms are not in a sufficiently good state of preservation to be iden-
tified positively.

164 BULLETIN OF THE BUREAU OF FISHERIES

In records of past years (fig. 64) the overlapping of fall and winter species
shows very clearly. The winter type often appears as late as April, but disappears
as soon as the temperature of the water rises. In 1894 and 1896 appendicularians
appeared in large numbersin June. This is very unusual and may have been caused
by an influx of Gulf Stream water.

Swarms of Salpa democratica-mucronata Forskil blew ashore at Menemsha
Bight in Vineyard Sound on January 11, 1901. None are recorded from surface
collections in Great Harbor, but they may be expected at any time during that
month after hard southwest winds.

FISH

Ehrenbaum ‘states (in his excellent volume on the “Eier und Larven von
Fischen”) that the young stages of all fish, even those belonging to the bottom
dwellers, are usually true planktonic forms during and often after their larval
period.

From the standpoint of the planktonologist, fish of the Atlantic coast may be
grouped roughly under two headings—those that have pelagic larve and those that
have not. The latter group, of which Opsanus tau (Linnzus) is a striking example,
contains very few members and does not enter into the plankton problem.

The first group is of great importance. A division may again be made here to
separate those fish having pelagic eggs from those having demersal ones. No
relationship exists between the condition of egg laying and the habits of the fish
or between the various species of fish having these habits. Bottom-living forms,
such as Gadus callarias and Tautoga onitis, have pelagic eggs while Clupea harengus,
a surface dweller, has demersal ones. As a rule, most of the larger fish of this
region belong to the group having buoyant eggs, the demersal group being composed
of such small forms as Ammodytes, Pholis, Apeltes, Cyprinodon, Lucania, Fundulus,
and Menidia. As many investigators have shown, special adaptations enable both
types of eggs to have the best possible chance to survive.

In order to overcome the many difficulties besetting pelagic life, fish with
buoyant eggs extrude enormous numbers of ova. ‘These are small, translucent, and
practically invisible against the bright sky, which forms the background. A very
few species have pelagic eggs, which float together in a gelatinous membrane,
often many feet in length. Such a condition is characteristic of Lophius piscatorius.
The incubation period of pelagic eggs is comparatively short, largely governed by
the temperature of the water. The young fish hatch in a very immature con-
dition, and these, too, are translucent except for the eyes and scattering yellow and
black chromatophores. For several days they are quite helpless, and undoubtedly
during this period enormous numbers are destroyed. Later they become very lively,
darting about and feeding ravenously on copepods. It is interesting to note that
the eggs become translucent just before spawning. During development they are
rather opaque, and the yolk is deeply colored.

Demersal eggs are laid in bunches on the sea bottom or attached to plants Bg
fine threads. Here, again, there are special adaptations for fertilization and pro-
tection. Contrasted with the former group, where the females outnumber the
males, McIntosh found that fish of this group are mostly males. This condition

PLANKTON OF THE WOODS HOLE REGION 165

he believed to be necessary, for the milt rises and is likely to be lost before the eggs
can be fertilized. The eggs are usually quite opaque and heavily laden with yolk.
By being grouped in large bunches they are not so easily preyed upon by the bottom-
feeding animals, although no doubt many are lost in this way. The eggs are com-
paratively fewer in number and have a longer incubation period.

Young fish of this group are often just as numerous in surface collections as
those hatching from pelagic eggs, for they usually hatch in a much more advanced
stage, thus greatly reducing the mortality.

Gadus callarias and Pholis gunnellus, characteristic members of the spring plank-
ton, are excellent representatives of these two groups. The former emerge from the

g g

Jule
Sepe
Octe
NOVe
Dece

Tautogolabrus adspersus

Tautoga onitis

Prionotus carolinus
Stenotomus chrysops

Brevoortia tyrannus

Syngnathus fuscus

Spheroides maculatus
Lophopsetta maculata
Merluccius bilinearis

Poronotus triacanthus

‘Menidia menidia notata
Urophycis spe Hae
‘Leptocephalus Elops ? r EEE SHEE
Rhinonems cimbrius HE :

Heplatessoides

Myoxocephalus aeneus ? EHEELESEEEEEEEE EEE EE 4
Microgadus tomcod EEE EE

Fic. 65.—Occurrence of fishes in surface collections from June to Decemner, 1922

egg in a helpless condition and for some time are tossed about at the mercy of the
waves as delicate little transparent larve. (The black chromatophores arrange
themselves in vertical bands and may camouflage the young fish in much the same
way that similar designs served to protect our ships during the late war.) The
other species (Pholis gunnellus) is never found in an entirely helpless condition.
The young, which are much farther advanced than those of the cod when they
appear in surface collections, are always very lively and swim rapidly toward the
light when placed in a glass tray. (The larval cod were always dead when removed
from the nets.) Copepods were always found in the intestines of even the smallest
specimens. ‘This is further evidence of the activity of this species in its very early
pelagic existence. The eggs are laid on the bottom in a compact mass and are
guarded by the adult fish until hatched.

166 BULLETIN OF THE BUREAU OF FISHERIES

In summer the most abundant larve are Tautogolabrus adspersus and Tautoga
onitis. Both have pelagic eggs and appear in June, remaining until August.
During this time the eggs are often very numerous, appearing like masses of minute
bubbles on the surface in the examining dish.

Mr. Edwards took 34 species of larval fish in the 15 years recorded in Figures 67
to 81. During the past year 20 species were identified. Of the summer forms all
but one (leptocephalus of Elops?) are common to this region. The leptocephalus is
not that of an eel but of a true fish, as the tail is well developed and forked. I
have placed it in the genus Elops because that is the only common southern fish
recorded from this region that has a leptocephalus stage.

Of the winter larve all were of species breeding in the region except Gadus cal-
larias. This is a northern species common off southern New England, the adults
of which never enter the immediate region. As the nearest important spawning

se 2

BB aR Be Beet
‘Tautogolabrus adspersus
Tautoga onitis
Prionotus carolims
Stenotomus chrysops
Brevoortia tyrannus
Syngnathus fusocus
Spheroides maculatus
Lophopsetta maculata
Merlucoius bilinearis
Poronotus triacanthus
Menidia menidia notata
Pholis: gunnellus

B28

Jal

Auge
Sep.
Oot.

Gadus callarias
Myoxocephalus seneus ?
Microgadts tomcod
Ammodytes americanus
Poamericanus

Fic. 66.—Occurrence of fishes in surface collections of 1923

grounds are on Nantucket Shoals, the appearance of early larve at Woods Hole
probably results from southerly or easterly currents. Postlarval forms, usually
about 20 mm. in length, find their way into this region and are often taken in May
(fig. 67) in large numbers, depending upon the season. Still later postlarval stages
(40 to 50 mm.) are always present in the shallow coastal waters in May and June.
Many were taken within the boat basin at the Fisheries dock on May 24, 1923, with
a fine net. During this period they are destroyed in large numbers by Loligo pealit.
A school of over 200 of these squid, all about 5 inches in length, seined in Great
Harbor, were found to be feeding entirely upon young cod. Several specimens
were observed with a young fish protruding from the beak and one or more others
held securely in the tentacles.

In 1923 early larval stages of cod appeared in small numbers in the tows of
January, February, and early March. Surface collections made in Vineyard Sound
at various times during this period showed that they were present there also, but
likewise in small numbers. Just what effect the artificial conditions created by
the hatchery had is not known, but probably the 351,000,000 larve liberated during

PLANKTON OF THE WOODS HOLE REGION 167

Anguilla rostrata
Brevoortia tyrannus
Anchovia. brownii
Apeltes quadracus
Microgadus tomcod
Gadus* callarias

Brosmius brosme

Fic. 67,—Maximum seasonal distribution of fishes not common in surface collections, based on records of the
years 1893 to 1907

be > q LS Lote)
1894 4.4. .5..6 5
ALTE | oe ee
is95 ig} B B 4 & 8
ri | ft re93[_] [ [met TTT)
1895 -——— 1694
1897 = 1695 |
1898 Sa ate
ce a ‘Sr weeeeee

1900

1901

1902

aT Bes
Eine eee
as es

ieee eal eons} abe a
Sar see

1993

1904

1905

1906

ee
eae eit
— Say

1907
pa ee

Fia. 68.—Occurrence of Pholis gun-
nellus during successive years, 1893
to 1907

Fic. 69.—Occurrence of Tautogola-
brus adspersus during successive
years, 1893 to 1907

168 BULLETIN OF THE BUREAU OF FISHERIES

the spring less than a mile from the collecting station increased the surface catch
somewhat. It is barely possible that all early specimens taken may have originated
from that source. Their absence in records of past years, combined with their
total absence during the early spring of 1924 (when no larval cod were liberated at
Woods Hole), seems to substantiate this possibility. It must be remembered,
however, that the unusually warm weather existing during the early part of the
past winter apparently prevented the adult fish from spawning on their usual
grounds, and for that reason no eggs were available at the hatchery. The cod
may have sought deep areas so far from the coast that the early forms, which were
never abundant in this region anyway, could not be transported into these waters.
Fishermen reported that they had never known the cod catch to be so small on the
southern New England grounds as during the early winter of 1923. They were
totally unable to supply the Woods Hole station with any spawning fish. The
usual fall school which annually enters Narragansett Bay also failed to appear.
It will be interesting to learn whether the postlarve appear as usual in May, 1924.

In Figure 67 and Table 5 several fish taken rarely in past years are listed.
Most of these are southern forms and occur only in the summer when the Gulf
Stream inhabitants are blown in. Lactophrys trigonus, Cryptacanthodes maculatus
Anarhichas lupus, and Seriola zonata are representatives of this fauna.

TaBLE 5.—Fishes very rarely taken in surface collections

Species Date Abundance
————— eee —
Leptocephalus conger_----______-_-- Janes) LOOSE 8 Le ee eben ees eet Mieco preter One.
Osmerus mordax Ete Feb. 28, 1895; July 14 and 15, 1896; and June 19, 1899___________________. Many.
Poronotus triacanthus.2-_£_ 44-223 Aug. 9 and 10, LBOSE oie 3 Te US STE Be ei 2nd) Ae ce ey See
Lactophrys trigonus..__.....__.._-.- Serntni26,c1803- 3 Po Os be a a eS eae
Cyclopterus lumpus.--._._..__.____- June 20, 1898; June 5, 1908, and June 18, 1907-
Cryptacanthodes maculatus_________ Apr. 22, 23, 25, and 26, B70 / (is eR Aa pines nin nigh pape ang are pays, eel EE yo
Anarhichas lupus’. 1. !__) ee May l and ws 1899, and June 3 and 5e18900) oe eee ee Few
Rhinonemus cimbrius-______.______- Apr. 1, 27, 28, and 30, 1900; June 2, 3, 4, 5, and 6, 1906; and July 17, 1906- Very many.
Lophius piscatoriussc-2 lif 23 June 10, 5-1: ea TE NEL TAD Gal a Se NN UR ET TP SS
Pseudopleuronectes americanus- ---_- AUB ALS ISO7 eo Ne ee Oe ete ee ee eee Tes
Limanda ferruginea?-_---__..._______ Jan. 4, 1908....-.--- 2s ol bo. . eee petite inet) Sal ne
Seriola zonata__. 127-7 PTs ee PATIC NSO, LOO (2s 0 eee 2 ee ee ee ee eee Few
Pomolobus pseudoharengus--_--______ Wuryyies 1806-2 - cco ee eee ene eee Do.

The remaining figures show clearly that the fish have a definite breeding
season within certain limits, usually determined by temperature. Temperature chart
for the spring of 1923 (fig. 5, opp. p. 100) indicates how unusually cold the water
was. The result was that many of the fish, as well as other larval forms, did not
appear. The approximate temperature throughout the breeding season of each
common species may be found by comparing the individual figures with Figures
4,5, and 6. This particular temperature, however, must not be regarded as the
complete governing factor. Atsome time earlier in the year a rise or fall in tempera-
ture caused the ovaries and testes to ripen. When the sex products have com-
pletely matured they will be extruded within certain limits irrespective of tempera-
ture. After this extrusion the immediate temperature then plays its part. Cod
eggs have been made to hatch in 9 days or 64 days by varying the temperature of
the water.

PLANKTON OF THE WOODS HOLE REGION 169

Jane
Fave
More
Apre
May
June
Jule
Auge.
Sdp,
Cote
Ove

Fia 72.—Occurrence of Ammodytes americanus during
successive years, 1893 to 1907

Fia. 70.—Occurrence of Tau-
toga onitis during succes-
Sive years, 1893 to 1907

1893

1894

HA

1895

1896

E
eI
a

IE
alia
a
.
a

1897
1898

1899

ago. [I I

Fic. 71.—Occurrence of Myozocephalus xneus during
successive years, 1893 to 1907

Fic. 73.—Occurrence of Pollachius
virens during successive years, 1893
to 1907

8242°—25+ 6

170 BULLETIN OF THE BUREAU OF FISHERIES

As the salinity of Woods Hole is not noticeably different from the outer waters,
it probably plays no part in the distribution of the larval fish of this region. The
governing factors are, then, temperature and food, if we omit the effect of winds
and currents, which at times may influence the distribution greatly.

The abundance of food has a powerful effect on the lives of the young fish.
Given favorable temperatures, the larvez will develop rapidly and in great numbers
if food is plentiful. If food is scanty, few larval fish will be found. During the
winter and spring months copepods make up practically the entire food of the
young forms. The most abundant copepods during colder weather are usually
Temora longicornis and Pseudocalanus elongatus. No doubt both of these species
contribute equally to the food supply. As the spring of 1923 was exceptionally
cold, Temora did not appear. Scattering forms were taken during the winter, but
never more than three or four specimens on any one day. Out of 200 examinations
of stomach contents of larval fish this spring not a single Temora was found.
Pseudocalanus elongatus and Centropages hematus were very plentiful, particularly
the former, and these constituted their food, the bright red color of the Pseudo-
calanus showing clearly through the thin walls of all the young fish collected.

The summer fishes have a much greater variety of food and for the most part
do not limit their menu to Copepoda, although Acartia tonsa and Centropages typicus
are eaten in great numbers. A young puffer (Spheroides maculatus), 3.5 mm. in
length, examined on June 28, 1922, was found to contain 12 Littorina litorea, 9
Venus mercenaria, and 2 Acartia tonsa. Often a young fish was taken with a large
copepod or phyllopod protruding from its mouth.

The relationship of the larval fish to its food supply is therefore very close,
and one must determine it accurately in order to understand the distribution of
a species. Such a study was attempted at the Plymouth laboratory by Doctor
Lebour, who obtained some interesting results. More extended observations will
be necessary before the relationship of the many factors can be clearly understood.

The following forms were taken in surface collections of 1922-23:

Tautogolabrus adspersus)(Walbaum).._-__.__1 017 or PT Ee ee Cunner.
Tautogaonitis (innseus)2j. tb ke oo te ee te ere es Et ene Tautog.
Prionotus, carolinus, (Linneeus) = ben = ee ee ee Sea Robin.
Stenotomusichrysops.(Lanneeus)22 =...) ete Sa ee ee en ee Scup.
Brevoortia tyrannus;(uatrobe)s=. “2 Sd eee ee eee Menhaden.
Syngnathustuscus! Storer. @ cee se eo es ee Pipefish.
Spheroides maculatus (Bloch and Schneider) -____________-__---------------- Puffer.
Hippoglossoides platessoides (Fabricius) ___________------_------------------ Sand dab.
Merluccius bilinearis:(Mitchill) 22 a ae ST OY ye Seas Whiting.
Poronotus tricanthus: (Reck)2 a ee et oe tae, ea Butterfish.
Menidia menidia'notata*(Mitchill) (22) 0.2 2 2 SE ea oe ee ee Eee Silversides.
Pholis gunnellus; (inns) ooo eee foe | a ot ee Rock eel.
Urophycis.22) 258240) 05 0) ue Vener ou 2 Eee tt ee Hake.
Gadus callarias, Linnzeus2 ot 208s oo as ape Cod
Leptocephalus, Elops?.

Microgadusitomcods(Walbaum) 222222. 22 2a se ae eee ree ea ees Tomcod.
Myoxocephalus/zeneus;(Mitehill) jo 22-2 2 ae ee eee eee Sculpin.
ophopsettaimaculata c( Mite lill) ses sage ee ee ne eee Window-pane.
Rhinonemus,cimbrins: (ain seus) 55 oe es ee en PT a Uae ane Rockling.

Ammodytes;americanus: De Kayey 8. _ = 2s aa NE Spat at eae eee Sand launce.

PLANKTON OF THE WOODS HOLE REGION

| yl eee ope
PTT TTT ee re yt
ST. -

Fic. 74.—Occurrence of Urophycis sp. during successive years,
1893 to 1907

| soma! | |
PS ee eee eae b sh

‘Fic. 76.—Occurrence of Clupea harengus during successive years, 1893 to

1907

Fia. 75.—Occurrence of Steno-
tomus chrysops during suc-
cessive years, 1893 to 1907

asia ee
Sore =
aa | |_|
Be |
Bis |
EEeS |_|
EERE |_|
BARA a
gy] | [|
Heeaa |
| | pt =
oe |
1898 fel | TT TT
BREDBESaS an
reso |_| | | sume! | TT TTT
Sprereeretce
1900
HES Ess
am | | al
ag
ceaeae |
| {|
Lier)

Fic. 77.—Occurrence of Syngnathus
fuscus during successive years, 1893
to 1907

ays

172 BULLETIN OF THE BUREAU OF FISHERIES

The following fish were taken in surface collections of 1893 to 1907:

Anguilla rostrata (Lesueur). Pholis gunnellus (Linnzus).
Leptocephalus conger (Linnzus). Cryptacanthodes maculatus Storer.
Clupea harengus Linnzus. ; Anarhichas lupus Linneus.

Brevoortia tyrannus (Latrobe). Prionotus carolinus (Linneus).
Anchovia brownii (Gmelin). Pollachius virens (Linnzus).

Osmerus mordax (Mitchill). Microgadus tomcod (Walbaum).

Apeltes quadracus (Mitchill). Gadus callarias Linneus.

Syngnathus fuscus Storer. Urophycis sp.

Menidia menidia notata (Mitchill). Rhinoneumus cimbrius (Linnzus).
Ammodytes americanus De Kay. Brosmius brosme (Miller).

Poronotus tricanthus (Peck). Lophius piscatorius Linneus.
Stenotomus chrysops (Linnzus). Pseudopleuronectes americanus (Walbaum).
Tautogolabrus adspersus (Walbaum). Hippoglossoides platessoides (Fabricius).
Tautoga onitis (Linnzus). Lophopsetta maculata (Mitchill).
Lactophrys trigonis (Linnzus). Pomolobus pseudoharengus (Wilson).
Spheroides maculatus (Bloch and Schneider). Seriola zonata (Mitchill).
Myoxocephalus zneus (Mitchill)?. Limanda ferruginea (Storer) ?.

Cyclopteras lumpus Linnzus.

GENERAL CONCLUSIONS

I shall not attempt to summarize all the conclusions arrived at during the
past year. For the most part these have been taken up under the various subjects
and in the discussion on plankton. The following are some of the more general
conclusions concerning the nature of the plankton and the physical factors affecting
its distribution, resulting from a 2-year study of the Woods Hole pelagic fauna:

1. Woods Hole is an excellent location for the study of the seasonal distribution
of plankton.

2. It is impossible to investigate diurnal distribution in Great Harbor. The
current rushing through the passage during the flood tide mixes the water so com-
pletely that the distribution of plankton remains the same at all times. The entire
body of water is affected at the same time, even during periods of sudden heating
or rapid cooling of the air.

3. No great amount of fresh water enters Woods Hole. The salinity averages
about 31.5. For this reason titrations are of importance in determining the
presence of ocean water.

4. As in the case of the benthonic animals, the plankton of this region is made
up of a complex of faunas. It forms the northern limit of many southern species,
the southern limit of many northern species, and a pocket where oceanic animals
blown in by strong southerly winds are deposited.

5. The tropical species appear gradually in Great Harbor in the summer, but
stop suddenly in the fall. This is because the temperature of the water in Buzzards
Bay rises higher than that of the coastal waters im summer but responds quickly to
the falling temperature of the air and by fall becomes much colder. Animals
carried into this region in summer survive, but in the fall the lower temperature
proves fatal and few live to be carried through the passage back into the deeper
waters. However, members of this group may be taken throughout the fall in
Vineyard Sound, where the decline in temperature is not so rapid.

PLANKTON OF THE WOODS HOLE REGION es

Fic. 80.—Occurrence of
Hippoglossoides plates-
soides during succes-
sive years, 1893 to 1907

Fic. 78.—Occurrence of Prionotus car-
olinus during successive years,
1893 to 1907

a e e e e e e

Sa aes Aue Bore oly ve

$156.6 8..8°°8 8:3

we93{_| [| | TTT TT TT Tad

mim |

1s94 || |
|

, (2 Sooo eedeeeee ee
Fig. 79.—Occurrence of
Spheroides maculatus a207 HRREE SERRE e eT BS

during — successive Fic. 81.—Occurrence of Menidia menidia notate
years, 1893 to 1907 during successive years, 1893 to 1907

aiy(a! BULLETIN OF THE BUREAU OF FISHERIES

6. The arm of Cape Cod forms a permanent northern barrier for the southern
neritic plankton but only a summer barrier for northern pelagic species.

7. The proportion of benthonic animals occurring in the plankton of this
region is much greater than that found in normal littoral plankton. After north-
east storms Buzzards Bay types predominate; after southerly storms Vineyard
Sound types are most plentiful. This is particularly noticeable in the case of
amphipods.

8. No correct impression of the relative abundance of the local benthonic
fauna can be obtained from surface collections unless the distribution of each of
these animals in the bay and sound is completely understood.

9. A distinct periodicity in the occurrence of all the common animals of the
region is clearly noticeable. The succession of species remains almost the same
each year, the only variation being in the time of their appearance and disappear-
ance.

10. The planktonic animals of the region, with one exception, may be placed
in two general groups—the summer community and the winter community. The
celenterates are the exception. For the most part these have a long spring maxi-
mum and a short one in the fall.

11. The pelagic diatoms exert a very great influence on the zooplankton. When
the greatest maxima appear most of the zooplanktonic forms disappear. There
are possibly two reasons for this. First, the common species having these swarm-
ing periods do not form the food of the zooplankton so far as I have been able to
determine. During the maxima of the larger diatoms the smaller members of this
group which are eaten by pelagic animals disappear, causing a scarcity in the
food supply. This may account for the similarity in the time of disappearance of
the larger forms and the small diatoms. Second, the great numbers of the diatoms
filling the water apparently cause conditions unfavorable for animal life of any
sort. The macroplankton seems to be literally choked out. This, however, is
hardly probable, and is offered merely as a possible explanation.

12. Conditions favoring the increased production of one species of diatoms
are also favorable for many others, provided that one does not become so abundant
that almost all others disappear. The summer maximum often exemplifies the
latter condition. The winter swarm usually consists of many species in which
various forms predominate at different times.

13. My observations on the distribution of pelagic diatoms lead me to disagree
with the theory that all production takes place in deeper waters off the coast, the
species occurring in the littoral waters being the result of winds and tides. Such
factors no doubt account for the distribution of the various species, but the quan-
titative distribution can not always be explained on that basis.

All evidence points clearly to the fact that great production of floating diatoms
takes place at the mouths of rivers where the largest amount of drainage from the
land is emptied into the coastal waters. Peck’s observations in Buzzards Bay also
indicate that the greatest swarms are found where the greatest outwash from the
land occurs. Buzzards Bay is a great reservoir in which pelagic diatoms accumu-
late and multiply, and as a result the swarms carried into Great Harbor are often
exceedingly large.

PLANKTON OF THE WOODS HOLE REGION N75

14. Temperature is the dominant factor in governing the seasonal distribution
of all local pelagic animals. It also determines whether oceanic species entering
the region shall perish at once or live long enough to become an important factor
in the local fauna. Three general conditions cause the appearance of the pelagic
animals—winds, tides, and the food supply. Salinity forms barriers in some locali-
ties, but not at Woods Hole. Once introduced into the region, the organisms remain
until the temperature becomes unfavorable or the food supply is exhausted, and
then they must leave or perish. Food is also an important factor in causing the
disappearance of a species during a period of favorable physical conditions. This
is probably the limiting factor of the summer diatom season. Temperature governs
the breeding seasons of all planktonic and benthonic animals of this region. The
temperature prevailing at the time of the extrusion of the eggs is not often the im-
portant factor, for the eggs are usually thrown off as soon as ripe, provided the
conditions are not too unfavorable. After the eggs have been deposited in the
waters the existing temperature plays a part in determining whether the incubation
period will be long or short. The determination of an early or a late breeding
season, then, depends upon the temperature at some previous date when a warming
or cooling of the water started the development of the sex products. This fact
must be considered when interpreting the appearance of certain larve in the
plankton.

15. Reactions to changes of temperacure are fur cue uivse pact more evident
among planktonic animals than among benthonic forms. Bottom dwellers, par-
ticularly sessile forms, in order to maintain themselves must be able to withstand a
great range of temperature. Unusually low temperatures often kill large numbers,
but as a rule both the larve and adults are extremely hardy. This is not true in the
case of planktonic forms. Certain species, such as Calanus jfinmarchicus, although
preferring cold water, are able to stand sudden rising or falling temperatures and
appear to survive as well in water of 22° C.as at 0° C. Most pelagic animals, how-
ever, particularly the phytoplankton, disappear as soon as the temperature condi-
tions become unfavorable.

16. The annual distribution of the diatom maxima of the American coast is
very similar to that of the eastern Atlantic waters in that the seasons of the greatest
swarms retreat farther and farther from the warmest months as one approaches the
Tropics. A similarity in the seasonal variation in European and American waters
of the same latitude is particularly noticeable, conditions at Woods Hole correspond-
ing to those in the Adriatic Sea. The great effect of the arm of Cape Cod on the
local plankton is again evident here, for within 20 miles of Massachusetts Bay,
with conditions similar to the Norwegian Sea, conditions comparable to those of the
Mediterranean and Adriatic Seas are found in Buzzards Bay.

17. The distribution of the plankton of the western Atlantic coast is little
understood, and the number of animals new to the region taken during the past
year indicates that most of the eastern Atlantic coast pelagic species probably will
be found here also.

176 BULLETIN OF THE BUREAU OF FISHERIES

BIBLIOGRAPHY
ALLEN, W. E.

1921. Some work on marine phytoplankton in 1919. Transactions, American Microscopical

Society, Vol. XL, No. 4, 1921, pp. 177-181. Menasha, Wis.
Agassiz,’ ALEXANDER.

1862. On alternate generation in annelids, and the embryology of Autolytus cornutus.
Boston Journal of Natural History, Vol. VII, No. III, 1862, pp. 384-409. Boston.

1865. North American Acalephe. No. II, Illustrated catalogue, Museum of Comparative
Zoology at Harvard College, 1865, 234 pp., figs. 1-360. In Memoirs, Museum of
Comparative Zoology at Harvard College, Vol. I, 1864-65. Cambridge.

Baitrey, L. W.

1917. Notes on the phytoplankton of the Bay of Fundy and Passamaquoddy Bay. Con-
tributions to Canadian Biology, 1915-16 (1917). Supplement to the 6th Annual
Report of the Department of Naval Service, Fisheries Branch, pp. 93-107. Ottawa.

BicELow, Henry B.

1914. Explorations in the Gulf of Maine, July and August, 1912, by the United States
Fisheries schooner Grampus. Oceanography and notes on plankton. Bulletin,
Museum of Comparative Zoology at Harvard College, Vol. LVIII, No. 2, pp.
29-147, figs. 1-38, pls. 1-9. Cambridge.

1914a. Oceanography and plankton of Massachusetts Bay and adjacent waters, November,
1912-May, 1913. Jbid., No. 10, 1914, pp. 385-419, figs. 1-7, 1 pl. Cambridge.

1915. Exploration of the coast water between Nova Scotia and Chesapeake Bay, July
and August, 1913, by the United States Fisheries schooner Grampus. Oceanography
and plankton. Jbid., Vol. LIX, No. 4, 1915, pp. 149-359, figs. 1-82, 2 pls. Cam-
bridge.

1917. Explorations of the coast water between Cape Cod and Halifax in 1914 and 1915,
by the United States Fisheries schooner Grampus. Oceanography and plankton.
Ibid., Vol. LXI, No. 8, 1917, pp. 161-357, figs. 1-100, 2 pls. Cambridge.

BIGELOW, RoBERT PAYNE.

1895. Scientific results of explorations by the United States Fish Commission steamer
Albatross. No. XXXII. Report on the Crustacea of the order Stomatopoda
collected by the steamer Albatross between 1885 and 1891, and on other specimens
in the United States National Museum. Proceedings, United States National
Museum, Vol. XVII, 1894 (1895), pp. 489-550, figs. 1-28, Pls. XX-X XII. Wash-
ington.

Branpt, K.

1905. On the production and the conditions of production in the sea. Appendix D, Rap-
ports et Procés-Verbaux, Conseil Permanent International pour l’Exploration de la
Mer, Vol. III, 1905, pp. 1-12. Copenhagen.

Bumpus, H. C.

1898. The breeding of animals at Woods Hole during the month of March, 1898. Science,

New Series, Vol. VII, No. 171, April 8, pp. 485-487. New York.
CASTRACANE DEGLI ANTELMINELLI, FRANCESCO.

1886. Report on the Diatomacee collected by His Majesty’s ship Challenger during the
years 1873-1876. Report on the scientific results of the voyage of His Majesty’s
ship Challenger during the years 1873-76. Botany, Vol. II, 177 pp., 30 pls.,
1886. London.

CusHMAN, JOSEPH AUGUSTINE.

1918. The Foraminifera of the Atlantic Ocean. Part 1. Astrorhizide. United States
National Museum Bulletin 104, Part 1, 1918, pp. 1-111, pls. 1-39. Washington.

1920. The Foraminifera of the Atlantic Ocean. Part2. Lituolide. United States National
Museum Bulletin 104, Part 2, 1920, pp. 1-111, pls. 1-18. Washington.

1922. The Foraminifera of the Atlantic Ocean. Part 8. Textulariide. United States
National Museum Bulletin 104, Part 3, 1922, pp. 1-149, pls. 1-26. Washington.

PLANKTON OF THE WOODS HOLE REGION 177

CusHMAN, JosEPH AUGUsTINE—Continued.
1923. The Foraminifera of the Atlantic Ocean. Part 4. Lagenide. United States National
Museum Bulletin 104, Part 4, 1923, pp, 1-228, pls. 1-42. Washington.
1924. The Foraminifera of the Atlantic Ocean. Part 5. . Chilostomellide and Globigerinide.
United States National Museum Bulletin 104, Part 5, 1924, pp. 1-55, pls. 1-8.

Washington.
EXRRENBAUM, E.

1905-1909. ier und Larven von Fischen. I Band, Nordisches Plankton, 413 pp., illus.
Kiel und Leipzig.
Farren, G. P.

1920. On the local and seasonal distribution of the pelagic Copepoda of the southwest of
Ireland. Publications de Circonstance, No. 73, Conseil Permanent International
pour l’Exploration de la Mer. 30 pp. Copenhagen.

Paxon, WALTER.

1879. On some young stages in the development of Hippa, Porcellana, and Pinnixia. Bul-
letin, Museum of Comparative Zoology at Harvard College, Vol. V, No. 11, 1878-
79, pp. 253-268, Pls. I-V. Cambridge.

Fawkes, J. WALTER.

1885. Studies from the Newport Marine Zoological Laboratory. XIII. On the develop-
ment of certain worm larve. Bulletin, Museum of Comparative Zoology at
Harvard College, Vol. XI, No. 9, 1883-85, pp. 167-208, Pls. I-VIII. Cambridge.

Fritz, Cuara W.
1921. Plankton diatoms, their distribution and bathymetric range in St. Andrews waters.
Contributions to Canadian Biology, 1918-20 (1921), pp. 49-62, Pls. I-III. Ottawa.
Gran, H. H.
1912. Pelagic plant life. Chapter VI, The Depths of the Ocean, by Murray and Hjort,
‘ pp. 307-386. London.
Gran, H. H., and ALEXANDER NAaTHANSOEN.

1909. Beitraige zur Biologie des Planktons. II. Vertikalzirkulation und Planktonmaxima
im Mittelmeer. Internationale Revue der gesampten Hydrobiologie und Hydro-
graphie. Band II, Nr. 4 and 5, October, 1909, pp. 580-632, figs. 1-29. Leipzig.

Haraitt, CHARLES W.

1905. The Meduse of the Woods Hole region. Bulletin, United States Bureau of Fisheries,

Vol. XXIV, 1904 (1905), pp. 21-79, Pls. I-VII. Washington.
Hyman, O. W.

1920. The development of Gelasimus after hatching. Journal of Morphology, vol. 33, No.

2, March 20, 1920, pp. 485-501, pls. 1-12. Philadelphia.
Lesour, Maris V.

1920. The food of young fish. No. III. Journal, Marine Biological Association of the
United Kingdom, New Series, Vol. XII, No. 2, July, 1920, pp. 261-324. Plymouth,
England.

Leper, H.

1917. Einige Beobachtungen itiber das Winterplankton im Triester Golf (1914). Interna-
tionale Revue der gesampten Hydrobiologie und Hydrographie, Band VIII, Nr. 1,
March, 1917, pp. 1-21. Leipzig.

Loumann, H.

1899. Appendicularien. Band II, E. c., Ergebnisse der Plankton-Expedition der Hum-
boldt-Stiftung. Kiel und Leipzig.

1911. Uber das Nannoplankton und die Zentrifugierung kleinster Wasserproben zur Gewin-
nung desselben in lebendem Zustande. Internationale Revue der gesampten Hy-
drobiologie und Hydrographie, Band IV, Nr. 1 und 2, April, 1911, pp. 1-38, Pls.
LIV. Leipzig.

MacDonatp, D. L.

1912. Onacollection of Crustacea made at St. Andrews, New Brunswick. Contributions to

Canadian Biology, 1906-1910 (1912), pp. 83-84. Ottawa.

178 BULLETIN OF THE BUREAU OF FISHERIES

McIntosu, WILLIAM CARMICHAEL, and ARTHUR THOMAS MASTERMAN.
1897. The life-histories of the British marine food-fishes. 516 pp., illus., 1897. London.
McMoraraicg, J. PLayrair.

1917. The winter plankton in the neighborhood of St. Andrews, 1914-15. RIN 8'3 to

Canadian Biology, 1915-16 (1917), pp. 1-8. Ottawa.
Meap, A. D.

1898. The breeding of animals at Woods Hole during the month of April, 1898. Science,

New Series, Vol. VII, No. 177, pp. 702-704, May 20. New York.
Moors, J. P.

1903. Some pelagic Polycheta new to the Woods Hole fauna. Proceedings, American
Academy of Natural Sciences of Philadelphia, November, 1903, pp. 793-801,
Pl. LV.

Oaitvig, HELEN S.

1923. Microplankton of the south coast of Ireland. Part 2, Rapports et Procés-Verbaux
des Reunions, Vol. XXIX, Conseil Permanent International pour l’Exploration de
la Mer, Vol. XXIX, 1923, pp. 30-71. Copenhagen.

OsTENFELD, C. H.

1913. Bacillariales (diatoms). Bulletin Trimestriel, Conseil Permanent International pour

)’Exploration de la Mer, Part 3, 1913, pp. 403-508. Copenhagen.
Peck, James I.

1896. The sources of marine food. Bulletin, United States Fish Commission, Vol. XV,

1895 (1896), pp. 351-368, pls. 64-71. Washington.
Pratt, Henry SHERRING.

1916. A manual of the common invertebrate animals, exclusive of insects. 737 pp., illus.,

1916. A.C. McClurg & Co., Chicago. :
SHARPE, RicHarD W.

1911. Notes on the marine Copepoda and Cladocera of Woods Hole and adjacent regions,
including a synopsis of the genera of the Harpacticoida. Proceedings, United
States National Museum, Vol. 38, 1911, pp. 405-436, figs. 1-20. Washington.

STAFFORD, J.

1912.. On the recognition of bivalve larve in plankton collections. Contributions to Cana-

dian Biology, 1906-1910 (1912), pp. 221-242, pls. XXTI-XXIV. Ottawa.
Strever, ADOLF.

1903. Beobachtungen iiber das Plankton des Triester Golfes im Jahre 1902. Zoologischer
Anzeiger, X XVII Band, Nr. 5, December, 1903, pp. 145-148. Leipzig.

1911. Leitfaden der Planktonkunde. 382 pp., illus., 1911. B.G. Teubner, Leipzig und
Berlin.

Strasny, Gustav.
_ 1908. Beobachtungen tiber die marine Fauna des Triester Golfes im Jahre 1907. Zoolo-
gischer Anzeiger, XX XII Band, Nr. 25, April 14, 1908, pp. 748-752. Leipzig.

1910. Beobachtungen iiber die marine Fauna des Triester Golfes im Jahre 1909. Zoolo-
gischer Anzeiger, XX XV Band, Nr. 19. April 12, 1910, pp. 583-587. Leipzig.

1911. Beobachtungen iiber die marine Fauna des Triester Golfes wihrend des Jahres 1910.
Zoologischer Anzeiger, XX XVII Band, Nr. 25, May 30, 1911, pp. 517-522. Leipzig.

Sumner, Francis B., Raymonp C. Osspurn, and Leon J. Cour.

1913. A biological survey of the waters of Woods Hole and vicinity. Part I, Sec. I—Phy-
sical and zoological. Bulletin, United States Bureau of Fisheries, Vol. XX XI,
1911 (1913), Part I, pp. 1-442, charts 1-227. Washington.

1913a. A biological survey of the waters of Woods Hole and vicinity. Part II, Sec. ITI—
A catalogue of the marine fauna. Jbid., Part II, pp. 544-794. Washington.

TATTERSALL, W. M.

1920. Notes on the breeding habits and life history of the periwinkle. Department of
Agriculture and Technical Instruction for Ireland, Fisheries Branch, Scientific
Investigations, 1920, No. I, 11 pp., 1 pl. Dublin.

PLANKTON OF THE WOODS HOLE REGION 179

Tuomeson, Miuuett T.

1899. The breeding of animals at Woods Hole during the month of September, 1898.
Science, New Series, Vol. IX, No. 225, April 21, pp. 581-583. New York.

1904. The metamorphoses of the hermit crab. Proceedings, Boston Society of Natural .
History, Vol. XXXI, 1904, pp. 147-209, pls. 1-10. Boston.

VANHOFFEN, ERNST.

1897. Die Fauna und Flora Grénlands. Erster Teil, Gronland-Expedition der Gesellschaft

fir Erdkunde zu Berlin, Band II, 1897, pp. 1-383, pls. 1-8. Berlin.
VERRILL, A. E.

1873. Report upon the invertebrate animals of Vineyard Sound and the adjacent waters,
with an account of the physical characters of the region. Report, United States
Commissioner of Fish and Fisheries, 1871-72 (1873), pp. 295-778, pls. Wash-
ington.

1923. Crustacea of Bermuda. Schizopoda, Cumacea, Stomatopoda, and Phyllocarida.
Transactions, Connecticut Academy of Arts and Sciences, Vol. 26, October 1923,
pp. 183-211, Pls. XLIX-LVI. New Haven.

WHEELER, Wituiam Morton.

1901. The free-swimming copepods of the Woods Hole region. Bulletin, United States

Fish Commission, Vol. XIX, 1899 (1901), pp. 157-192, figs. 1-30. Washington.
Wriagut, RAaMseEy.

1907. The plankton of eastern Nova Scotia waters. An account of floating organisms
upon which young food-fishes mainly subsist. Contributions to Canadian Biology,
1902-05 (1907). Thirty-ninth Annual Report, Department of Marine and Fish-
eries, Fisheries Branch, pp. 1-19, Pls. I-VII. Ottawa.

ZIMMER, CARL.

1909. VI. Die nordischen Schizopoden. Jn Nordisches Plankton, herausgegeben von Prof.
Dr. K. Brandt und Prof. Dr. Apstein in Kiel. Zwélfte Lieferung, 1909, pp. 1-178,
figs. 1-384. Kiel und Leipzig.

&

DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES.
AN INVESTIGATION OF ENZYMES IN FISHES, WITH COM-
PARATIVE STUDIES ON THOSE OF AMPHIBIANS, REP-
TILES. AND MAMMALS

&
By WALTER A. KENYON

Contribution from the Zoological Laboratory, University of Wisconsin

Bad
CONTENTS

Page Page

Introduction.2.5£....5.-4.2..22-----_- 181 | Results and discussion—Continued.
Purpose of the investigation__-_-_--_ 182 Tryptic digestion.___-___-________ 188
Animals used in this work__-_----- 182 Ereptic digestion.__._____________ 190
Description of methods_-_---------- 183 Carbohydrate-splitting enzymes____ _ 192
Results and discussion______------------ 184 IAM YIASCL Seen ee eae Soa 192
Reaction in the digestive tract__--- 184 Inverting enzymes___-----------_- 195
Peptic digestion___.-------------- 184 | General discussion_-_-._-------------- 197
Esophagus_...2...--.=------- 184 Conclusions. 22_ 22h. 0 2-- LL ek 198
Stomach<-.22220.2--2222251- 185 || Bibliography-=2_--.-2o-.l5 522-2. .L oie 199

INTRODUCTION

The literature on digestion in poikilothermal vertebrates shows a need for
further investigation, especially by systematic comparisons of the digestive enzymes
in representatives throughout the vertebrate series. It would be of value to know if
differences in digestive enzymes have occurred in the course of evolution of groups
of higher vertebrates from more primitive ones, with the migration of aquatic
vertebrates to land habitats, and with the metabolic changes necessitated by the
transformation from a poikilothermal to a homoiothermal condition. It would also
be of interest to know more of relations of digestive enzymes to different types of
alimentary structure and the adaptations of digestive enzymes to differences in food
habits.

Comprehensive reviews of previous works have been published by Yung (1899),
Sullivan (1907), and Biedermann (1911). Since a review of early literature would
be largely a repetition of the essence of these reviews, the reader who desires further
information in regard to early studies on digestion in fishes is referred to these
articles.

It is obvious in the records of previous investigators that there has been much
difference of opinion in regard to digestion in fishes. Much of the inconsistency in
results is doubtless due to the fact that in most of the investigations careful quanti-
tative methods with well-regulated hydrogen ion concentrations have not been

181

182 BULLETIN OF THE BUREAU OF FISHERIES

employed. In fact, the greater part of the studies made upon digestion in fishes
was made before the development of improved quantitative methods and especially
before the development of accurate methods for hydrogen ion regulation. The
study of digestive enzymes in amphibians and reptiles has been almost wholly
neglected.

The only record that the writer has been able to find of any attempt to compare
the rate of digestion in different groups of poikilothermal vertebrates is that of
Riddle (1909), which, unfortunately, yielded no conclusive results of a comparative
nature.

Recently Bodansky and Rose (1922) published a preliminary study on digestion
in certain elasmobranchs and teleosts, employing well-controlled hydrogen ion
concentrations. Using Dernby’s (1918) method of liquefaction of gelatin as criterion
for proteolytic digestion, they found that the optimum pH for gelatin digestion by
fish pepsin was about 3. This is in agreement with the optimum for mammalian
pepsin. These investigators found coagulated egg albumin to be very slowly
digested by the proteolytic enzymes of fishes. From extracts of the pyloric caeca of
the red snapper (Lutjanus aya) they obtained trypsin, pepsin, rennin, amylopsin,
and lips (weak). Invertase was present to a very slight extent; inulinase, maltase,
and lactase were altogether absent.

PURPOSE OF THE INVESTIGATION

The purpose of the investigations described in the present paper has been
twofold: (1) To make a comparative study by means of reliable qualitative and
quantitative methods of the rate of digestion by the more important digestive
enzymes in representative fishes, amphibians, and reptiles, with less extensive
studies on those of mammals; and (2) to extend the present knowledge of the
presence, character, and distribution of proteolytic and carbohydrate-splitting
enzymes in the animals studied.

The work has extended over a period of two years. Many of the experiments
upon fishes were performed while the writer was employed by the United States
Bureau of Fisheries during the summers of 1922 and 1923. The writer is especially
indebted to Dr. A. S. Pearse and to Dr. H. C. Bradley, of the University of Wiscon-
sin. for valuable suggestions and criticisms.

ANIMALS USED IN THIS WORK

The following fishes have been used in this investigation: Bluegill, Lepomis
incisor (Cuvier and Valenciennes); carp, Cyprinus carpio Linneus; black crappie,
Pomozxis sparoides (Lacépéde) ; pickerel, Hsox lucius Linneus; perch, Perca flavescens
(Mitchill); sucker, Catostomus commersoniit (Lacépéde); and white bass, Roccus
chrysops (Rafinesque). Of these the pickerel, as a representative of carnivorous
fishes, and the carp, that is mainly vegetarian in its diet, have been the most thor-
oughly studied. The principal amphibian studied was the mud puppy, Necturus
maculosus Rafinesque. Among the reptiles the bull snake, Pituophis sayi (Schlegel) ;
snapping turtle, Chelydra serpentina (Linneus); and the painted turtle, Chrysemys
belli Gray and C. cinera (Bonnaterre), which intergrade in this locality, were used.
The dog was selected as a representative mammal.

DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES 183

DESCRIPTION OF METHODS

The reaction in the esophagus, stomach, and intestine of various animals was
determined approximately by means of indicator papers as soon as possible after
killing the animal.

For digestion studies the fresh alimentary tract was thoroughly washed and
cleaned from traces of pancreatic tissue and mesentery. When the stomach con-
tained substances in process of digestion, it was ligated at both ends to prevent
passage of its contents into the esophagus and intestine while being cleaned. Then
the tract was cut into the divisions to be studied, each division opened by a longi-
tudinal slit, and the contents thoroughly washed out with the view of preventing
any contamination of mucosa of one region with enzymes from other sources.
After cleaning the mucosa or other digestive tissue was removed, weighed (moist),
and triturated in a mortar with the aid of crushed glass when necessary. The finely
ground tissue was made into a brei containing 10 per cent toluol, 40 per cent tissue
for proteolytic enzymes or 25 per cent tissue for carbohydrate-splitting enzymes,
with distilled water to 100 per cent. The brei was allowed to extract two to three
days at room temperature and was strained through two thicknesses of cheesecloth.

Coagulated egg white was used as substrate for pepsin and trypsin. It was
prepared in considerable quantity and preserved in 10 per cent toluol. In making
this substrate whites of eggs were repeatedly strained through several thicknesses
of cheesecloth. To five parts of egg white four parts of water were added and the
ege white coagulated by heat while being stirred constantly. One part of toluol
was added and the mass stirred to uniform consistency. This made a very finely
divided product, of 50 per cent egg white (moist weight), which could be readily
pipetted.

In the experiments on peptic and tryptic digestion the digests were made up
to contain 12.5 c. c. of brei and 20 c. c. of coagulated egg white per 50 ¢. c. total.
The approximate pH value of each digest was determined by paper indicators.
For peptic digestion 0.2 N hydrochloric acid was added until the point was reached
where congo red was just turned blue, a pH of about 3. In 50. c. digests about
15 c. c. hydrochloric acid were required. For tryptic and ereptic digestion tests
0.2 N sodium carbonate was added until phenolphthalein gave the first pinkish tint,
at a pH of about 8.4. In the alkaline digests 2 to 4 c. c. of toluol were added in
_ addition to that contained in the brei and egg white to prevent any bacterial action.
Casein was chosen as substrate for ereptic digestion, since this protein is readily
split by erepsin, while most other native proteins, such as egg albumin, are not.
In the ereptic digestion tests 2 grams of powdered casein were used per 50 c. c.
digests.

The progress of digestion was determined for pepsin, trypsin, and erepsin
in terms of both the initial cleavage rate (tyrosine production), as shown by the
colorimetric method of Folin and Denis (1912), and the amino acid production by
the Sorenson formol titration method. The rate of digestion by the carbohydrate-
splitting enzymes was determined in terms of the reducing power in milligrams
of glucose per c. c. digest by Benedict’s micro method for quantitative determina-
tion of sugar. In studying the rate of starch digestion the iodine test. was also
employed. Careful controls were made by the use of boiled extracts.

184 BULLETIN OF THE BUREAU OF FISHERIES

RESULTS AND DISCUSSION
REACTION IN THE DIGESTIVE TRACT

The reaction in the stomach and intestine was found to be exceedingly variable
even among the animals of the same species, apparently depending upon the func-
tional state of the organ. In the stomachs of all groups of poikilothermal verte-
brates the reaction varied from about neutrality to an acidity sufficient to turn
congo red purple, a pH of about 4. Generally the acidity was higher when the
stomach contained food, subsiding to near neutrality when empty. The greatest
acidity was observed in a bull snake, the stomach of which contained the remains
of two mice, still largely intact, which had been devoured two days before killing.
Here the hydrogen ion concentration was almost pH 38, while no acidity could be
demonstrated in the stomach of a bull snake that had been fasting. In the stomachs
of garter snakes that had swallowed frogs one to three hours before being opened
a similar region of acidity was observed around the frogs. These findings agree
with the early statements relating to fish. Tiedemann and Gmelin (1827) and
Decker (1887) observed that contents of full stomachs of fish would always redden
blue litmus, while the reaction at other times was neutral, slightly acid, or even
very slightly alkaline.

The contents of the intestine also exhibited similar reaction in different classes
of vertebrates. The reaction was in most cases about neutral, sometimes devia-
ting to the acid side and sometimes to the alkaline, more frequently being slightly
alkaline. The highest alkalinity, pH 8 to pH 8.4, was frequently observed in the
small intestine of turtles. In the carp, which has no stomach but a long, compara-
tively undifferentiated alimentary canal, the reaction was in most cases slightly
alkaline or neutral, though occasionally slightly acid. Variations in reaction of
different parts of the tract of this fish were common. Acid reaction was observed
most frequently in the posterior end of the intestine. Such acidity may be due
to bacterial fermentation, which would perhaps take place to the greatest extent
in this region, since it contains food materials that have remained longest in the
intestine. It may be that the carp, an animal of rather sluggish habits, which stuffs
its entire alimentary canal with vegetation, muddy alge, minute insect larve,
etc., depends more or less upon bacterial action as a material aid in the digestion
of its food.

PEPTIC DIGESTION
ESOPHAGUS

Extracts of the esophageal mucosa of the pickerel, snapping turtle, and bull
snake produced no increase in tyrosine or amino acid when allowed to act upon
coagulated egg white at a pH of 3 for eight days. Absence of pepsin from the eso-
phagus of these is in agreement with the known facts of their histological structure.
Oppel (1897) cited observations of different workers upon numerous species of
fishes, including the pickerel, and made the generalization that no glands are
present in the esophagus of fishes. In sturgeons, however, gastric glands extend
up into the esophagus (Kingsley 1917). In the amphibians glands are known to
exist in the esophagus of Proteus anguineus, Necturus maculosus, and Rana, occurring
most abundantly in the region of the esophagus nearest the stomach (Oppel, 1897).

DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES 185

Among the reptiles glands have not been found in the esophagus of Sauria and
Ophidia, but have been described in the lower portion of the esophagus of certain
Chelonia and Crocodilia (Oppel, 1897).

Decker (1887), using the disappearance of fibrin in dilute hydrochloric-acid
extract as his criterion of peptic digestion, reported “peptic digestion” in the
esophagus of fishes as of general occurrence. Biedermann (1911) believed that
Decker mistook the simple action of acid upon fibrin for peptic digestion. It may
be that Decker failed to thoroughly clean the esophagus from pepsin-laden fluids
backing up from the stomach. It seems to the writer also possible that what Decker
thought was peptic digestion may have been digestion by autolytic enzymes, which
act in sufficiently dilute acid but are practically, inhibited at a hydrogen-ion concen-
tration of a pH 3 to 2.5, at which pepsin acts most rapidly. Observations that have
been reported of small fishes digesting within the esophagus of larger fishes (such as
those reported by Spallanzani, 1785) probably indicate merely digestion by gastric
juice flowing forward into the esophagus.

STOMACH

Table 1 is a comparative record of results obtained for the rate of peptic digestion
upon coagulated egg albumin by extracts from representatives of different classes of
vertebrates. Increase of tyrosine is taken as the chief criterion for initial cleavage.
While amino-acid production in gastric digestion is too small to be measured with
any high degree of accuracy, it is interesting to note that there is slight increase in
amino acid in each digest and to compare the amount produced here with that
produced in tryptic and ereptic digestion (Tables 3 to 6).

TaBLE 1.—Rate of peptic digestion in different groups of vertebrates

[Parallel digests with boiled extracts were run for each animal. Only two controls are included in the table to save space, since
all controls produced no tyrosine or amino acid. Room temperature averaged 23 to 26° C. Each digest consisted of the
following proportions: 12.5c. c. extract of stomach mucosa (except carp) +20 ¢c. c. coagulated egg albumin+0.2 N HCL (about
15 c. c.) to approximately pH 3 to 50 c. ec. tota}]

pov Peace: C.c.0.2 N aminoacid
Milligrams tyrosine in 2 ¢. ¢c. filtrate in 10. ¢. filtrate
Animal ieee
nima er O 5

aaeta Days : Net Days Net

gain, gain,
0 1 2 4 8 8 days 0 8 8 days

Fish:

Oar es ore tocol tcelgbccseetcososs control_- 3 | 0.087 | 0.087 | 0.087 | 0.087 | 0.087 | 0.000 | 0.100 | 0.100 | 0.000
Alimentary tract to first bend ...--._------------ 3} .079] .119] .120} .120] .120] .041] .083 | .133 - 050
Posterior to first bend_-.-.--------------------1- 2; .067] .125 144] .144] .150] .083 | .125] .225 - 100
Bluegill 1 . 154 . 154 . 154 154 . 154 . 000 - 100 - 100 - 000
mn eD Teen ik) legen oak 1 154] .418) .424 |-------| .467] .313 |] .100] .450 . 350
@rappie sess hes ee ke ee eon S .143 | .400] .446/ .446{ .472] .329] .150] .500 - 350

Pickerell® 27 vice 2k!

Amphibia: Necturus 5131 | 1435 | 1444] 1462 | 1468] 1337/ 1125] 2350] 1225

Reptile:

Painted Tirta — ee NTE ee ot 7 1 . 091 - 461 476 . 493 ~ 505 . 414 - 100 - 400 - 300
Snapping turtle.2.222 22220523 28 A 2 . 088 -428 | .456 - 470 . 526 . 438 - 100 325 . 225
Bullisnake ie pees a 2 122 A73ilseecaes 506 506 . 384 100 525 425
Mammals Dogs -.s2-s22s.n5) co ee et oe 1 222 46T oo eae 500 500 | .278 200 500 300

With the exception of the carp (the only animal used having no well-formed
stomach) all the animals showed a remarkably uniform rate of digestion. Since
the carp differs so widely in anatomical structure and activity, it will be discussed

186 BULLETIN OF THE BUREAU OF FISHERIES

by itself later. Individual variations among different animals of the same species
were not great. With the exception of the carp, also, there seems to be little
change in the rate of peptic digestion correlated with differences in food habits.
Among the fishes the pickerel, which feeds almost entirely upon fishes, does not
show an outstandingly higher rate of peptic digestion than the bluegill and crappie,
the diet of which includes a larger proportion of vegetable matter, more insects,
and small crustaceans. In 18 bluegills examined Pearse (1921) found that plants
and alge constituted 24 per cent of the total volume of food in the digestive tracts.

Coagulated egg albumin seems to be equally well digested by the enzymes of
fish, amphibians, reptiles, and mammals. Several workers, however, have re-
ported that coagulated egg albumin is only very slowly digested by fish pepsin
(Decker, 1887, for pike, cited by Biedermann; Yung, 1899, for selachians; and
Bodansky and Rose, 1922). Bodansky and Rose (1922) stated that the slow
digestion of coagulated egg albumin by fish pepsin was not entirely due to the
small surface exposed, for finely divided albumin was not digested with an ap-
preciably greater rapidity. ‘‘Furthermore, a commercial pepsin solution capable.
of producing an effect on fish meat similar in extent to that exerted by fish pepsin
digested coagulated egg albumin quite readily.” Their criterion for the extent of
digestion of the coagulated egg albumin, or just how finely the egg albumin was
divided, was not stated. Obviously a comparison between the activity of extracts
of gastric mucosa and the activity of commercial pepsin would not be as satisfactory
as the comparison between the activity of uniformly prepared extracts from the
gastric mucosa of the animals.

The only previous attempt to compare the rate of digestion in representatives
of different classes of poikilothermal vertebrates is that of Riddle (1909), who tried
to determine the rate of gastric digestion in living fishes, amphibians, and reptiles
by comparing the rate of disappearance of coagulated egg white in Mett’s tubes
introduced into the stomach. Riddle concluded that there was a progressive loss
in digestive power in ascending the vertebrate scale from fishes to reptiles. How-
ever, in view of the extremely wide differences which Riddle obtained in the same
species under the same conditions, and the lack of fairly constant results for each
class of animals studied, such a conclusion seems highly speculative, and the method
used an unsatisfactory one for reliable comparisons. The results obtained in the
present investigation definitely show that in evolution from fishes to reptiles there
has been no loss in digestive power. On the contrary, the digestive power in the
reptiles used, if different, seems to be a trifle greater than in other groups. ‘The
results in Table 1 indicate that in poikilothermal vertebrates with well-developed
stomachs and in the dog ! equal amounts of stomach mucosa possess approximately
the same digestive power.

The carp, from which outstandingly different results were obtained, differs
in anatomical structure from fishes of other groups in the absence of a glandular
stomach and in the presence of an extensive hepatopancreas. The bile duct enters
the alimentary tract immediately behind the pharynx, and the entire alimentary
canal, which is about 4 feet in length in a carp weighing 2 kilograms, has the appear-
ance and histological structure of an intestine. It is wholly devoid of gastric
glands (Oppel, 1897).

1 In preparing the extract from gastric mucosa of the dog a larger proportion of the pyloric region than of the cardiac region
was used, which may account for the slightly lower rate of tyrosine production in this animal.

‘

DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES 187

In digests with extracts from the anterior segment and with those from the
entire posterior part of the alimentary canal production of tyrosine was very small,
and it seems doubtful whether it was peptic in nature. Since practically all of the
increase in tyrosine occurred during the early part of experiments, the apparent
peptic digestion in the carp may have been due to the action of autolytic enzymes
for a short time before they were entirely inhibited by the hydrogen-ion concen-
tration of pH 3, which was approximated by the use of congo red indicator paper.
In studies on autolysis of liver and kidney, Bradley (1922) observed that the maxi-
mum digestion took place at pH 4.0 to 4.5 (obtained by adding 25 to 40 ¢. c. of
0.2 N hydrochloric acid per 250 ¢. c. of brei), and that digestion by autolytic en-
zymes was practically at a standstill at a pH of 3 to 2.5 (obtained by the addition
of 75 to 100 c. ec. of 0.2 N hydrochloric acid per 250 c. c. of brei). In the writer’s
experiments about 15 c. c. of 0.2 N hydrochloric acid were added per 50 c. c. digests,
which would be equivalent to 75 c. c. for 250 c. c. of brei. It would seem possible,
therefore, should autolytic enzymes of similar character to those of liver and kidney
be present in the mucosa of the digestive tract, that they might be able to exert
a slight effect before their complete inhibition. That the digestion by extracts
from the alimentary tract of the carp was not peptic seems almost certain from the
fact that in nearly all the digests (cf. also Table 2) no digestion to an appreciable
extent took place after the first day, though plenty of egg albumin was present
and conditions were near the optimum for peptic digestion.

Table 2 shows differences in rates of peptic digestion at 37° C. and at room
temperature. Except for differences in temperature the tests were made exactly
asin Table 1. With the exception of the pickerel, where no appreciable difference
is shown in the rate of digestion at 24 and 37° C., peptic digestion was uniformly
more rapid at 37° than at room temperature. Moreover, about the same pro-
portionate increase in tyrosine (initial cleavage) at 37° C. over that at room
temperature was produced in amphibians, reptiles, and the mammals.

TaBLE 2.—Peptic digestion at room temperature and at 37° C.
[Digests were prepared exactly as in Table 1, with initial pH 3]

C. c. 0.2 N amino
Milligrams tyrosine in 2c. ¢. filtrate acid in 10 «@ «@
Tem- filtrate
Animal Pate.
in de- Days Days
reas sf Net tf Net
[22 Oa cand Ln +421 a by
0 1 2 4 g |Sdays| 9g g |8days
Fish: 7
{ 23 | 0.078 | 0.118 | 0.118 | 0.118 | 0.118 | 0.040 | 0.05] 0.15] 0.10
Oe “l\ 37] .078]) £100} .105] £105) .111 | .033] .05] 115] 210
: { 24] .085} .340] .417]| 441] 2441] 1356] 10} :30] [20
Pickerel__..---------------+--+---+---+------+-- 37| .085| .339] .374] .431] .431] .346 10] .25 15
th D3) |se182) (jet 435 | Hee 465| .476] .294] .15] .45 30
Amphibia: Necturus....-.-...-.-------------------- { 37| .160| .500 |---__.. 500} .500] .340; .15] 155 40
Reptile:
» : 1 23] .101} .428] .453] .463] .463] .362| .05] .25 20
Snapping turtle--------------------------------- 387 | .101| .470] .538] .541] 1541] .440 05 30 25
ort 23) .143} .500] .535) .535| .535] .392] .10] .55 45
Bull snake No, 1.-----------------+--27--------- 37| .160| .527] .582] .582] .582] .422] .10] 165 55
D311) 2100!) 455! | 0 eae 476 | .476 | .376| .10] .50] 40
Bullisnake Nos 2. 2-22 own nnn nee wn enon emcee 37] .089| .500 |---_-- .556 | .556| .467/ .10] 60] .50
‘ 23 | .222] .467|-..__.-] .500|} .500] .278] .20] .50 30
Mammal: Dog..-----..---------2-----n----2---00--= 37 | .222] .500| .547| 1547] .556| .334| .20] ‘60 40

27864°—25——2

188 BULLETIN OF THE BUREAU OF FISHERIES

TRYPTIC DIGESTION

Tables 3 and 4 show data obtained for tryptic digestion by extracts from the
pancreas in different animals and from the hepatopancreas of the carp. The
results are typical for tryptic digestion, a comparatively large amount of amino
acids being split off in addition to a large amount of tyrosine, while in peptic digestion
amino acids were split off only to a very slight extent (cf. Tables 1 and 2). Since
the experiments were carried on in definitely alkaline media having an initial pH
above 8, peptic and autolytic digestion would be practically ruled out. That the
digestion of egg white by extracts from the pancreas was truly tryptic was further
shown in separate tests by the fact that as the hydrogen-ion concentration was
increased to pH 7 and higher the rate of digestion rapidly decreased, and vice versa.

TaBLe 3.—Tryptic digestion in different groups of vertebrates

[Except for Necturus ! and bull snake,? digests were made up of the proportion: 12.5 c. c. pancreas extract+20 c. c. coagulated egg
al ers Ta N sodium carbonate to approximately pH 8.4+toluol (10 per cent of total)+distilled water to 50 c. c. total
volume ;

Aver- Milligrams tyrosine in 2c. c. filtrate C.c.0.2 N amino acid in 10 ¢. c. filtrate
age
tem-
Animal pera- Days Days
eS ee) Net
gain

*90| 1.00] 1.10] .75

Pee eer ee 2.25} 1,90
1.35 |------ 2.10} 1.85
1.85 | 2.00 | 2.30] 1.60

1 Necturus digest contained only 5 c. c. of extract, thus representing only 40 per cent of the amount of pancreas used in the
first five tests. :

2 Bullsnake digest contained only 4 c. c. of extract, representing about 32 per cent of the amount of pancreas used in the first
five tests. It was necessary to work with a much smaller quantity in Necturus and snake because of the very small amount of
pancreas obtainable.

TaBLE 4.—Tryptic digestion at room temperature and at 37° C.

[Digests were prepared exactly as in Table 3, with an initial pH approximately 8.4]

Milligrams tyrosine in 2 ¢. c. filtrate O. ¢. 0.2 N amino acid in 10 c. c.

filtrate ‘
Tem-
pera-
Animal _ture Days Days
in de-
grees Net Net
gain gain
0 1 2 4 8 1 2 4

Do-.--------------------------- 37| 1143 | 1211 | 1235 | .385| 1455| 312] 130] .55| :65| .90/1.10] .80
93 | .313 | .645| .704| .681| .681| .368| .70] 1.70] 1.85] 2.00}2.30] 1.60

Dog (pancreas) - - -.----------------- 37 | 1313] 16951 .741| .695| .695| .382| .70| 2,10] 230] 2.40]240]| 1.70
, 23 | 1190] :303 |------- 400| 1435 | .245| :30| .50.|_----- 75| .90| .60

Necturus (pancreas)!_.-.------------ 371 160 |) -.,476 |sasanue 527 | .527| .367| .30| .70|_-.--- 1.10 | 1.40] 1.10

1 Necturus digest contained only 5 c. c. of extract.

DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES 189

In the carp the rate of tryptic digestion for a given amount of hepatopancreas
was low as compared with that in the true pancreas, such as is found in Necturus,
turtle, or dog. However, if the relative size of the hepatopancreas is taken into
consideration, this organ would probably contain an even greater amount of trypsin
for the size of the animal than does a typical pancreas in other animals. The pan-
creas of the pickerel, like that in many other teleosts, is a diffused organ, so that the
pancreatic tissue used was more or less mixed with fatty and connective tissue.
The well-defined pancreas in the amphibians and reptiles studied gave a high rate
of tryptic activity. Since the amount of pancreas available from bull snakes and
Necturus was very small, it was necessary to use less than half the proportion of
tissue in the digests for them. Considering the comparatively rapid rate of amino
acid and tyrosine production in the pancreatic digests of these two animals, it is
probable that the capacity of a given weight of pancreas in the bull snake and
Necturus for tryptic digestion is equal to that of the turtles and the dog.

From Table 4, showing the results of parallel experiments at room temperature
and at 37° C., it may be seen that tryptic digestion was much more rapid at 37°
than at 23° C. for all the animals tested. The higher temperature seemed to produce
a slightly greater rate of increase in tryptic digestion in the carp than in Necturus
and the dog.

Experiments were carried on in the same way as with the pancreas, using
extracts of mucosa from different regions of the digestive tract to find out whether
or not trypsin is secreted in other organs than the pancreas. No tryptic digestion
was obtained in extracts from the stomach mucosa of various animals, including
the pickerel, white bass, Necturus, and snapping turtle, nor from the intestinal
mucosa of Necturus, turtle, bull snake, and dog. There was slight evidence of
tryptic digestion (increase in tyrosine and amino acid) in the intestine of the pickerel
and in the caeca of the crappie. Repeated experiments using extracts from the
mucosa of the anterior, middle, and posterior regions of the alimentary tract of
the carp all showed no increase in amino acid or tyrosine in 8 days. Bile from the
carp also produced no digestion of coagulated egg albumin in digests at a pH of
approximately 8.4, 7, and 3. Homburger (1877, cited by Biedermann), however,
reported that in carp aqueous extracts of hepatopancreas, intestinal mucosa, and
bile itself digested fibrin in neutral or alkaline but not in acid solution. Krukenberg
in the same year (1877, cited by Biedermann) reported the indubitable existence of
trypsin formation in the intestinal mucosa of many teleosts and especially in the
carp. If true trypsin were secreted in the intestinal mucosa of the carp, digests
made up in exactly the same way as those which show strong tryptic digestion for
the pancreas should likewise produce tyrosine and amino acid. In the pickerel,
where some tryptic digestion by intestinal mucosa extracts was observed, the
amount of amino acid and tyrosine was exceedingly small as compared with diges-
tion by the pancreatic extracts. It seems probable, therefore, that the intestinal
mucosa plays no part of digestive importance in the secretion of trypsin in animals
having a well-defined pancreas, and that in fishes with a diffused pancreas a trace
of tryptic digestion by extracts from the intestinal wall may be due to small ramifica-
tions of pancreatic tissue embedded within the wall.

190 BULLETIN OF THE BUREAU OF FISHERIES

EREPTIC DIGESTION

Up to the present time the general occurrence of erepsin in the intestinal
mucosa of lower vertebrates has not been demonstrated. In fact, most of the
investigations that have been made of digestive enzymes in lower vertebrates
were made before the discovery of erepsin by Conheim (1901). The only record
of ereptic digestion in poikilothermal vertebrates which the writer has been able
to find was a brief statement by Kriger (1905) that an extract could be obtained
from the small intestine of Gadus morrhua, which showed characteristic erepsin-
like activity.

Tables 5 and 6 show the results obtained for ereptic digestion by extracts of
intestinal mucosa from representatives of different groups of vertebrates. In
Table 6 parallel experiments with digests at room temperature and at 37° were
made. In Table 5 all experiments were made at room temperature. Since erepsin
digests casein but not other native proteins, with the exception of histones and
protamines, casein was used as a substrate. The digests were all made alkaline
with 0.2 N sodium carbonate to an initial pH as low as 8.4, thus preventing action
by autolytic enzymes and pepsin. Parallel digests with coagulated egg albumin
were used as controls against any tryptic digestion. It will be noted that in this
case, as in tryptic action, the amino acid increase was comparatively high.

The data presented in Table 5 show that erepsin is present in the intestinal
mucosa of the representatives of all classes of poikilothermal vertebrates in as
great an abundance as in the intestinal mucosa of the dog. In the carp ereptic
digestion was greatest in extracts from the anterior end of the alimentary tract,
gradually decreasing posteriorly. In the pickerel a definite though small increase
in both amino acid and tyrosine in the control digest probably indicates a trace of
tryptic digestion. This may have been a factor in the high rate of casein cleavage
for the pickerel. Where the animal possessed a well defined small and large in-
testine, digests using extracts of mucosa from the small intestine only are included
in the tables. Extracts from the mucosa of the large intestine of the snapping
turtle, an animal in which the large and the small intestine are clearly defined,
also produced a high rate of ereptic digestion. Erepsin has likewise been reported
present in considerable quantity in the large as well as small intestine of the rabbit
(Glaessner, 1910).

DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES 191

TaBLe 5.—Ereptic digestion in different groups of vertebrates

[Digests = 12.5c.c. extract of intestinal mucosa + 2grams casein + distilled water + 0.2 N sodium carbonate to approximately pH
8.4 + toluol (10 per cent of total) + distilled water to 50 c.c. Controls have 20 c. c. coagulated egg albumin substituted

for casein]
cers Milligrams tyrosine in 2c. c. filtrate Ceca smite acid AOC:
age
: tem-
Animal satis Days Days
in de- Net Net
gain gain
Ss 1 2 4 8 Colt t ol eeulerd es
Fish:
Carp, to first bendfcontrol------ 22 | 0.087 | 0.087 | 0.087 | 0.087 | 0.087 | 0.000 | 0.10} 0.15] 0.15] 0.15] 0.15) 0.05
of intestine_-__--- experiment__ 22) .097) .238| .271] .402] .513] .416] .10] .35] .50] .70) 1.10] 1.00
Carp, middle of in-fcontrol_--_-- 22} .067| .067 . 067 . 067 . 067 -000} .15] .15 Bali} 15 Pp ts) . 00
testirie-<22 2 5-2-2 experiment_- 22 . 073 -200} .205| . 285 ~ 334 . 261 15} .20) .35|] .40] .65 50
Carp, posterior re-fcontrol-_--_-- 22,| 057 | «057 | °..057. |; «057 | ..057 | -.000|} 10") 10, .10) -10] .10 - 00
gion of intestine-\experiment_- 22) .057| .190] .267] .254 SUS) 4258))' +s TOL 3-22 220|escce- - 50 - 40
control_____- 23°) 2073: 2073 4 2073) 080.) 210 t 088)) 15) 216: 1-2). os. . 30 15
Pickerel-_-_---.----- experiment__ 23 | .067 -200] .325] .541 .909 ; .842] .10] .10] .35 -65 | 1.20 1.10
experiment__ 23 . 057 173 313 | .455| .752 . 695 7 0 fl eee .40 | .70] 1.30 1.15
control____-- 21} .057| .057| .057} .057] .057 | .000] .10) .10}) .10) .10} .10 .00
Amphibia: Necturus -jexperiment.- 211 5059) |i 51824)" 23541 9.'334 |) 2476) 2317] -.10')12_--__- 50] .80] 1.00 . 90
experiment_- 2331 074) 196 fi eeess2 1818:)' '.885:) . SEL) . 10) 240) j25<..- - 80 | 1.00 . 90
Reptile:
control__...- 26 -100] .105 . 105 . 105 20D) {005) |" LD Ieee ce BLOn | eeoeee . 20 05
Snapping turtle__-jexperiment__ 26) .173| .532| .633] .892| .985} .812| .20|------ nO} |ceeaee 1.35] 1.15
experiment_-_ 23 . 156 . 524 . 851 . 943 980 | .824 ~25| .72 .95 | 1.10] 1.70 1,45
control. __-_-- QU SORT te S000 00N fe O0RF O50) 1) 000") 10122. 0) |) LON 10 . 00
Bull snake_------- experiment_- 20 057) 2208) .225) .200)) 0313) 256) «10 jos... -30|] .40}) .70 . 60
ee ey a COT VES oe eccee 294] . 371 5 a oe eee ee sails Bt eri) . 60
: control. ___-- 2 o12)| 12 112 .112 SPAY) oe .30] .30} .30] .30] .30 . 00
Mammal: Dog-.--.-- experiment..| 23) .112] .222] .282| .334] .392]) .280] .30] .45] .55| .80] 1.30] 1.00
TaBLe 6.—E#reptic digestion at room temperature and at 37° C,
[Digests were prepared exactly as in Table 5, with an initial pH approximately 8.4]
Milligrams tyrosine in 2 c. c. filtrate C.¢. 0.2 N euoecd in 10 ¢. ¢.
Tem-
f pera-
Animal : gate Days Days
grees Net Net
gain gain
0 il 2 4 8 0 a 2 4 8
Fish:
Gar 22 | 0.073 | 0.200 | 0.205 | 0.285 | 0.334 | 0.261 | 0.15 | 0.20] 0.35] 0.40 | 0.65 0. 50
US aici aaa 37] .07 222| .220| .417| .448| .375] .15| .30|_.---- 75|1.05| .90
Pickerel 23 057 163 .313 455 . 752 695 Ls) Sees 40 70 | 1.30 1,15
inet ores aie alae aaa 37 057 371 | .444 667 | .895 839 15: |e ses 70) 1.25] 1.50] 1.40
aeueee 23 . 074 ~ 196s 22524 . 313 . 885 .3ll lLOs|" 340) |Eeses2 .80 | 1.00 O
ASIDES OATES Se Bee { 37] cor | ase (oo 500| .556| 1482] :10| .75|...... 120] 1.45] 1535
Reptile:
q 23 089 149 209 325 508 419 20)|e2Ses. 30 50 | 1.00
Snapping turtle....-.----.----- | coe | a08 |-1a8e | ese | leer + rea | cap fot 60 | 1.05 | 1.40! 1.20
23 OMG pees wee ee sees 294] .371 300 UO: | seen] ences, 50] .70 . 60
SURO eae eres ethos 37010 07.16 |Remaisa ceeei 556 | .667) .596| .10|_.----|_.---- 110} 150] 1.40
Do 21 057 103 125} .200] .313 256 LOM se-522 30; .40 70 . 60
aS aa gag a ape af Oe thes a ° abt 541 484 LON Goce) S51 6.7011) 100 . 90
: ql 282 | .33 . 392 280 30 45 55 . 80} 1.30 1.00
Mammal: Dog------.-----1----+--- 37| .112| 1267] :385| [500] :654| 542] [30] [75] 1.00] 1.50] 1.80] 1.50
1

The series of parallel experiments at room temperature and at 37° C. shows
that in all cases the rate of digestion was much more rapid at 37° C. than at 19 to
23° C. For all the animals the increase in rate of digestion at the higher tempera-

192 BULLETIN OF THE BUREAU OF FISHERIES

ture was in about the same proportion. It may be concluded from these experi-
ments that erepsin like that occurring in the intestinal mucosa of mammals occurs
in the intestinal mucosa of fishes, amphibians, and reptiles.

CARBOHYDRATE-SPLITTING ENZYMES
AMYLASE

The results of all the digestion tests with carbohydrate-splitting enzymes are
given in terms of the reducing power in milligrams of glucose per c. c. of digest.
Table 7 shows the rate of starch digestion by extracts of esophageal and stomach
mucosa from pickerel, bull snake, and snapping turtle, and of stomach mucosa from
the crappie. While slight hydrolysis of starch took place in all these experiments,
especially in the esophagus of the pickerel, its rate was too low to indicate the
presence of amylase in sufficient quantities to be of digestive significance.

TaBLE 7.—Amylase

[50 c. c. digests were used containing boiled starch to 1 per cent and toluol to 5 per cent of total volume. Controls contained
boiled extracts]

Milligrams glucose per ¢. ¢.

digest Net gain
ze o C0) pa | fodine reg
escription of ex- at 2 or
tract Days Days days
0 1 2 4 1 2 4
Esophagus:
PickereliNoal control____._- 2 1. 08 1. 08 1.09 1.11 0. 00 0. 01 0. 03 Blue.
8 Me eos ae sere aS experiment ss 2} 1.08] 1.79] 263] 3.12 .71| 1.55] 2.04 Brick red.
. + control __.._- 2 T0632 seo 106) B=2 2e-2= . 00 600)[- 2 eee2 ue.
Pickerel No. 2 -------------- experiment__ Pid kpos titi Mees gees D453 ead ol a kewl 1440 [i ooeeaes 0
ora control____-_- 2 a A V3 jo ee O04) eto os | Shc Oe Do
Pickerel No. 3 -------------- cae 2 T0312 ees LSD 1h | Soe eRe 21 89| Sees aa ae
a control___-_-_- 6 1024 |28ee2s-5 1. 02 ANO2N oss aceon . 00 {0}
Bull snake ------------------ experiment. Pl fee cbs(095 Cece Testy fies vgs (sy) eames “09 14] Do
: control... .-- 2 1.15 Ea Vig parka 1,15 SOON pessoa 00 Do
- Seis turtle --.---------- Seer ay Oy Wars bap Wage] Reel TY yes gy 1.82 Bb toy) ipa 67| Do
tomach:
Crappie Heute Pek oe 2 1. 20 1. 20 1. 20 1. 20 00 . 00 00 Do
ret te ecilaeiatae experiment.- 2) 20M) 621433) aN 5OnI a atOhI7 23 . 39 97| Do
: 7 control____.- 2 nBaul 111 pEnbl 1.11 00 . 00 00 Do
Pickerel No. 1 -------------- ae 2 14 NT 1.30 1.47 06 ; Re 36 ae
A control__-___- 2 ISOBR ate ose TOS Tie ee ea ee HO0i}2=t2e es to)
Pickerel No. 2 --..---------- experiment__ 2 1.030) te 2 428 TS 34y eee |e mol g| s=se cea Do
control____-- 2 TAOZA ee 1.02 02) 2k oe 60 00 Do
Bull snake --..-------------- cee a Laci 7 Bean Fe ap 07 13) Bo
A control -__--- 2 115 L155 e 2 52s 1.15 J00\ ee ee ee (0)
Snapping turtle ---------.--- Hemeab eerie 2 115 igs oles 1.73 10s) 222 ee 58 Do

In the intestinal mucosa of representative fishes, amphibians, and reptiles, and
in the caeca of the crappie (Table 8) a considerably larger amount of amylase was
present. However, 2 c. c. of extract did not convert all of 50 ¢. c. of 1 per cent
cooked starch in four days, a much slower rate than that of the extremely rapid
digestion by extract from pancreatic tissues. These data indicate the presence of
a greater amount of amylase in the intestinal mucosa of the animals whose diet in-
cludes a considerable amount of vegetable matter, as the carp and turtle, and a less

DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES 193

amount in animals such as the bull snake, whose diet consists almost wholly of
animal food.
TaBLE 8.—Amylase

[50 c. c. digests were used containing boiled starch to 1 per cent and toluol to 5 per cent of total volume. Controls contain boiled

extract. In iodine test, — = blue, no color change; + = some color change (reddish or brownish); +--+ = colorless, no starch.]
Milligrams glucose per c. c. .
digest Net gain
CC Cs a ae pee eY
Description of ex- at end o
w tract Days Days experiment
0 1 2 4 1 2 4
Intestine:

Carp !—
Sections vce cc ecs ewes an control__ 2 1.11 1.11 111 111 0. 00 0. 00 0. 00 _
MeCHOM Tes) eee cn eee experiment__ 2 1.11 2. 44 3. 85 4.76 1. 33 2.74 3. 65 +
Section ge eu ee cat ceee do-__- 2 111 2.00 3. 12 4.35 . 89 2.01 3. 24 +
BOCHON deco. see Ceo uel l ok do-__-- 2 ae | 1. 36 2. 38 3. 33 25 1, 27 2, 22 +
ection 4: ios. 26e asc -Beee cs = do-_.-- 2 Tico ee 2. 94 Oazol|eeeee ees 1. 83 2.01 +
i . ‘ control __-_-- 2 1.15 1.15 i 316s 1.15 00 00 00 -
Carp, middle of intestine... Lecco eg g 2| 115] 238| 286] 3.57] 1.23] 171] 242 af
Sucker, anterior third of in- qcontzel Bake oe 2 1.11 1.11 1.11 111 . 00 . 00 . 00 -
testine ices eee ooo oe experiment __ 2 1.11 1.47 2.17 2.78 36 1. 06 1. 67 +
control ..__-- 2 1: O00 cae -
Necturus- - --.--------------- experiment fi op ts 1 48 aaa +
control _-._.. 6 1. OO Seo -
Bull snake.------------------ experiment... 6] 1 ‘37 |"-"1506 At
control ____._ 2 1.00 1.00 1.00 1.00 - 00 . 00 . 00 —
< experiment __ 2 1.00 2.38 3.33 4.00 1. 38 2. 33 3. 00 +
Snapping turtle...------.--.- control ...___ nee th fee Tet pee Lil} .00 00| 00 as
experiment: i 2 Liat 2586) |aeeeenss 5. 55 5, seeaeee| 4.44 +
= . control -____- 1 1. 36 1. 36 1. 36 1. 36 - 00 . 00 -
Caeca: Crappie. ---------------~ Aecheunione: é 1| 1.36) 227| 278) 3.85) ‘91| 142] 2149 iE

1 Equal length sections, anterior to posterior, of entire alimentary tract.

With the exception of the pickerel, in which the pancreas is not well defined, all
pancreatic digests produced exceedingly rapid digestion (Table 9). In nearly all
cases 2 c. c. of extract converted 50 c. c. of 1 per cent starch solution to reducing sugar
within one day, and the greater part of it was reduced within a few minutes or hours.
The iodine tests likewise indicated rapid disappearance of the starch, generally
becoming colorless within a few hours. With 0.5 c. c. of extract of carp hepato-
pancreas in 50 ¢.c. of 1 per cent starch, the iodine test indicated the total disappear-
ance of starch, and Benedict’s sugar test indicated an increase in reducing sugar
of 5.57 to 6.17 milligrams per cubic centimeter in four hours (Table 10). With the
same amount of extract of turtle pancreas per 50 c. c. digest the iodine starch
test was almost colorless, and the reducing sugar was increased 5.15 milligrams per
cubic centimeter in one hour (Table 11). In order to determine if amylase from the
turtle’s pancreas acted upon starch more rapidly at 37° C. than at room tempera-
ture, 1 c. c. of extract diluted 10 times was used in 50 c. c. of 1 per cent solution,
which would be the equivalent of extract from 1 gram of tissue to 4,000 c. c. of 1
per cent starch (Table 12). In one hour the digest at 24° C. had gained 2.01
milligrams of glucose per cubic centimeter, while the other at 37° C. had gained
2.22 milligrams of glucose per cubic centimeter, indicating more rapid digestion
at 37° C.

194 BULLETIN OF THE BUREAU OF FISHERIES

In comparing starch digestion in the carp and the pickerel, it should be noted
that the carp, which is largely a vegeterian, possesses amylase in tremendous amounts
in the hepatopancreas and to a less extent throughout the intestinal mucosa. The
pickerel, on the contrary, which ordinarily does not eat plant foods except what it
may take in ‘‘second hand” within the digestive tract of its prey, possesses only a
negligible quantity of amylase, having little in the pancreas, esophagus, intestine,
and practically none in the stomach.

TaBLE 9.—Amylase

[50 c. c. digests were used containing boiled starch to 1 per cent and toluol to 5 per cent of total volume. Controls contain boiled

extract. In iodine test, — = blue, no color change; + = some color change (reddish or brownish); +-++ = colorless]
Milligrams glucose per ec. c. .
digest Net gain
ee C..¢) jlaass ina a ae LOCLUneKtest
Dearne ee Dass Days = authperend at
0 1 2 4 1 2 4
Eee 1 2 a 17 0. 00 0.0
A control. ---L- nial Ie } .00 | —
Pickerel No. 1--------------- lesperiisnt. 2) 1i7| 208 1.21). a,-774) Se
+ control ____-- se 17 py en She 00 fesseenes -
Pickerel No. 2.-------------- fel een 2 Deri7a| Petes ee 1430i| 5 ee +
. control___---- 2 1.19 1.19 -
Pickerel No. 3--------------- {conte ers 2 1h ap 1. 92 +
A controls. ==" up | J a -
a Pickerel No. 4--------------- deeneR ent re 2 fet) | a Be +
epatopancreas:
Garninion control___--- 2 1.14 apo -
ge Mee Or experiment. 2 1, 1") “10: 00 10. 00 ++(1 day).
control__- <= Lil Ll -
Carp No. 2..-.--------------- experiment_- 2| 1.11] 11:10] 11:10 ++(1 day).
control 2e-==— 1 1.20 1. 20 700) |2--2e ae -
Sucker_..---.---.------------ {operant 1] 1.20] 6.25 5.05 |e ais ++(1 day)s
Pancreas: ei ‘ : ' on
control__.--.- aga sb eee Sees © 00} | acc sete ee -
Necturus..-.---------------- eater ei DAP ASAE TGS heen em |BEaNe eaeA B14 |e ae ae ++(1day)s
Bull snake eontrol.-- 2. 2 1.09 1.09 PS 09)'|- 4202 225 - 00 300) |te. 2223 =
aaah a dal al als experiment-- 2 1.09 5. 88 Coo eeceeccal. stivo 5.16 |.-------] ++

TaBLE 10.—Amylase in hepatopancreas of carp

[Digest=0.5 c. c. extract in 50 c. ce. total (1 per cent starch, 5 per cent toluol). Control contained 0.5 c. c. boiled extract]

Milligrams glucose
per c. c. digest
Pees Net gain | Iodine test
Description Hours in 4 hours| in 4 hours
0 4
control___-_.-- 1. 25 1,25 0.00 | Blue.
@arp None o eos 2 2. ec ae ee a ee et Se 28 eee Oe ee ee experiment 1_- 1,25 7.42 6.17 | Colorless.
experiment 2_- 1/25 7.42 6.17 Do.
eontroless- 2. n= 1.10 1.10 -00 | Blue.
@arpiNos2. 622.22 22-2.-0.- 83 Ae. hare ee i eee experiment 1-__ 1.10 6. 67 5.57 | Colorless.
experiment 2__ 1.10 6. 67 6. 57 Do.

DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES 195

TaBLE 11.—Amzylase in pancreas of snapping turtle

[Digest=0.5 c. c. extract in 50 c. ce. total (1 per cent starch, 5 per cent toluol)]

Minutes a
Net gain
Description aR RE SEE aC UR aa mao ( R821 1]
0 5 10 15 30 60 AUS
Milligrams glucose per cubic centimeter digest - - Lil _ 2.86 4.00 5. 00 5. 55 6. 25 5. 14
Woloriodine test {2222.2 es ee Blue. | Lighter | Lavender. |....------ Very light | Nearly%ol-
blue. blue. orless.

TaBLe 12.—Amylase in pancreas of painted turtle

[Digestion at 24° C. and at 37°C. The (25 per cent tissue) extract was diluted to one-tenth strength. Digests=1c. c. diluted
extract in 50c. c. total (1 per cent starch, 5 per cent toluol). Equivalent to the extract from 1 gram of pancreas to 4,000 c. c.
1 per cent starch]

Minutes
atase Net gainin
Description Gomme’
0 30 60

Temperature 24° C.:

Reducing power in milligrams glucose per cubic centimeter digest .__.._---- algal 2. 08 Oy le, 2.01

Wolomotiiodine' tests 22 Soi foes occ nc <2 olen lotcct conn fecson bo bsesc secs Deep blue. Blue. | Purplish.
Temperature 37° C.: ;

Reducing power in milligrams glucose per cubic centimeter digest _-________ 1.11 2.13 3. 33 2, 22

Wolomofiodine test’) 24 2. o.oo a ak ee eh 58 Seon sh soe scecdonee Deep blue. Blue. | Purplish.

INVERTING ENZYMES

Tables 13 and 14 show tests made for the presence of invertase. With the
exception of the bluegill, the caeca of which were used, and the pickerel, of which
the entire digestive tract was tested, investigation for invertase was limited to the
intestine. In the carp different regions of the entire alimentary canal, which shows
characteristics of an intestine, were tested. In the bluegill some invertase was
found in both the caeca and the intestine. No invertase was found in the small
intestine of Necturus, the bull snake, or the large intestine of the snapping turtle.
It was present in a very small amount in the intestine of the pickerel. The carp,
bluegill, and snapping turtle each possessed a relatively large amount of invertase.
It is an interesting fact that these three animals include a considerable amount of
vegetable food in their diet. The other three animals in which invertase was absent
or present only as a slight trace feed almost wholly on animal food.

196 BULLETIN OF THE BUREAU OF FISHERIES

TABLE 13.—Invertase

[50 c. ec. digests were used containing sucrose solution to 1 per cent and toluol to 5 per cent of total volume. Controls contain
boiled extract]

Milligrams glucose per ec. c. di-

gest Net gain
wieg Cc:
Description of ex- Days Days
5 tract
0 1 2 4 1 2 4
- 5 control.23. 2 eb eer 1.11 1.11 0.00} 0.00 0. 00
Caeca: Bluegill_.-.-.------------------------- heteere Bil Bonet a seas 23 1.96\|| 2:63)| 0.0 .cee .85| 1.52
Intestine: ci 3

. control. oct VEG Ty A se Se eed 111 bE br WR . 00 - 00:
Bluegill. .-.------------------------------ {eee iione! Bila wE Tes ohne 2. 50)... 'B:8B)|seaweuse 139) 2.74

Carp (entire tract in 4 sections)—
Section: Uren. oer ee we ene naoee control 2 105 1.05 1.05 1.05 -00 .00 . 00:
SeCtloniletes sa cceoe wee eee 2 1.05 1. 59 2. 96 3. 33 . 54 1,51 2. 28
Section 2___- 2 1.05 1. 52 1.85 2. 44 47 - 80 1.39
Section 3__-_ 2 1.05 1.41 1. 67 2. 22 36 62 1.17
Section 4c. =2-. 22 a-as coco cececcncesesetesee 2 1.05 a BG, 1.19 1.19 .12 14 14

; A F control. ____- 2 1.10 1.10 1.10 1.10 .00 .00
Carp (middle of intestine)_------.---.---- feet 2} 110] 1.52] 200| 250] [42] [90] 1.40
control _ ____- 3 ub 1508) |e ee soo BUY OE ene
3 experiment_-_ 1 VA Na epee Spe oped mae 03) Sees
Pickerel__-------------------------------- control. ...-- oi tt 1. ra anne mee newer
experiment- - 2 1 1 00 21 32 85
control. ----- 2 MOS | See eee 00 .00
Necturus...---.-------------------------- (eee: z 2 O5' | ss een .00 00
control _-___- A : 95 . 00 .00 - 00
F i vd 1 . 5 -

Snapping turtle. .—--—---------e-ce-eoeen- contro) 1 a | Lon [Lah
experiment-_- 2 1 . 85 (Alcala 2.74
Snepping turtle (large intestine)... fexperment=.| 2) koa) noe {l77} Loe] leo fc} Zo
control _-_-_- 2 D500 i|eooeoeee 1.00 1.00 00 - 00 - 00:
Bullisnake S22 eae bee ees ee ees experiment_- 2 T2004 /(S22 sees 1.00 1.00 . 00 . 00 - 00
DevEce. dotit 6 L502 PEE 1.02 1.02 . 00 .00 - 00

TaBLE 14.—Invertase. Entire alimentary tract of pickerel

[50 c. c. digests were used containing sucrose solution to 1 per cent and toluol to 5 per cent of total volume. Controls contain
. boiled extract]

Milligrams glucose per ce. c. digest Net gain
D wage C. ce. of
escription epee Days Days
0 1 2 4 1 2 4
control ____-- 2 ak) ere |
Esophagus - ---------------------------------- {ear z 2 ile} 03316 ae ewe
Hn control. ---_- 2 1,15 1,15
Stomach. -.---------------------------------- experiment_. 2| 115] 116
eae cane 2 1.08 1.08
Do...-------------+---2-2--2-2-2--------- experiment_- 2} 1.08] 1.08
A control - --__- 2 1.15 1.15
Intestine (small) -.---------------------------- peed ere - a es Oe es ee Bs
D ieee bedete 2 ANOS swe eee s
Derren rene eae Sooo aaa ao aaa experiment. id : 1. at
Rectum, “large intestine”_-._-.-.------------ AGRE eee Fal herrea ymin

Table 14 shows the results of tests for invertase in extracts of mucosa through-
out the digestive tract of the pickerel, showing the presence of invertase only inthe
intestine and here in very small amount.

DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES 197

A few tests were made for the presence of maltase and lactase. Maltase was
found in great abundance in the hepatopancreas but not at all in the small intestine
of the carp. It was present in small amounts in the small intestine of the snapping
turtle. Tests for lactase in the intestine of the carp, Necturus, and the snapping
turtle all yielded negative results.

GENERAL DISCUSSION

The relationship of nutritive processes to all the physiological activities of
organisms makes the importance of obtaining a knowledge of the digestive enzymes
in all groups of animals apparent. Extensive comparative studies of digestive
enzymes in representative animals throughout the animal series would be of great
value. In his comprehensive review of the work done in the physiology of digestion,
Biedermann (1911, p. 1049) made the following statement in regard to the lack of
knowledge of digestive processes even in the vertebrates:

Wenn man von den karnivoren und omnivoren Saugetieren absieht, unsere Kentnisse der
Ernaihbrungsphysiologie hier kaum minder dirftig und liickenhaft sind als bei den Wirbellosen.

The data obtained show that apparently little change in the general character
of enzymes or their rate of activity has occurred in the evolution of amphibians,
reptiles, and mammals from primitive types. The rate of peptic digestion of coagu-
lated egg albumin per given weight of moist stomach mucosa is remarkably uniform
for the representatives of all classes of vertebrates studied. This substrate is equally
well digested by the enzymes of fishes, amphibians, reptiles, and mammals. The
reaction of the stomach in each group of poikilothermal vertebrates is variable,
usually being acid when the stomach contains food and nearly neutral when the
stomach is empty.

A close correlation exists between alimentary structure and the distribution
of digestive enzymes. The data for the carp (which is a fish without a stomach
or gastric glands) indicate that no pepsin is secreted by any part of the alimentary
tract, and that the initial cleavage of proteins in this animal is largely by the trypsin
from the hepatopancreas. Production of pepsin seems to be limited almost univer-
sally to a stomach mucosa possessing gastric glands. In a few cases, as in the frog,
pepsin production has been demonstrated in the esophagus. In this animal, how-
ever, there is no sharp line of demarcation between the esophagus and the stomach.
Extracts of esophageal mucosa from the pickerel, bull snake, and snapping turtle
were tested for pepsin and none was found. While there was indication of a trace
of pepsin in the intestinal mucosa of the pickerel and the turtle, it was not found in
the intestine of other animals studied, and in all cases the presence of trypsin in
any considerable quantity was limited to the pancreas (hepatopancreas in the
carp).

The general occurrence of erepsin in the mucosa of the intestine in poikilother-
mal vertebrates is shown for the first time. This enzyme is present in as great or
greater abundance in the intestinal mucosa of fishes, amphibians, and reptiles as in
the intestinal mucosa of the dog.

Digestion experiments were carried on mainly at room temperature. For
pepsin, trypsin, and erepsin parallel tests for representatives of each group of the

198 BULLETIN OF THE BUREAU OF FISHERIES

poikilothermal animals and the dog were made at 37° C. With one exception
(viz, peptic digestion in the pickerel, which will be further investigated) digestion
was more rapid at 37° C. than at room temperature.

Certain adaptive features are shown in the relation of digestive enzymes to
food habits. Those animals which include much plant food in their diet show a
striking difference in the amount of carbohydrate-splitting enzymes from those
that are wholly carnivorous. In the former amylase and invertase are present in
‘much greater abundance than in strictly carnivorous animals. For example, the
carp possesses a massive hepatopancreas containing large quantities of amylase,
while extracts obtainable from the small diffused pancreas of the pickerel digest
starch very slowly. Furthermore, the mucosa throughout the entire alimentary
tract in the carp possesses amylase in considerably greater amount than is found
in the mucosa of the alimentary tract in the pickerel. The bull snake, which is
entirely carnivorous, possesses no invertase and little amylase as compared with
that present in the painted and snapping turtles, which often include vegetable
food in their diets.

CONCLUSIONS

1. The reaction of the stomach in each group of poikilothermal vertebrates
studied is variable, usually being acid when the stomach contains food and nearly
neutral when the stomach is empty.

2. No trace of pepsin was found in extracts of esophagus mucosa of pickerel,
bull snake, and snapping turtle.

3. The rate of peptic digestion of coagulated egg albumin per given weight of
moist stomach mucosa is remarkably uniform for the representatives of all classes
of vertebrates studied.

4, With one possible exception, which the writer hopes to investigate further,
peptic digestion in all classes of vertebrates was more rapid at 37° C. than at room
temperature. Tryptic and ereptic digestion were always more rapid at 37° C. than
at room temperature.

5. Erepsin is present in as great abundance in the intestinal mucosa of fishes,
amphibians, and reptiles as in the intestinal mucosa of the dog.

6. Coagulated egg albumin is apparently digested equally well by the enzymes
of fishes, amphibians, reptiles, and mammals.

7. Amylase is generally present in very small quantity in extracts of mucosa
from the entire alimentary tract of fishes, amphibians, and reptiles. It is present
in great abundance in extracts from the hepatopancreas of the carp and from the
pancreas of Necturus, painted turtle, and snapping turtle.

8. Invertase occurs in the intestinal mucosa of the carp, blue gill, painted
turtle, and snapping turtle, and to a slight extent in the pickerel. There was no
evidence of invertase in the intestine of the bull snake and Necturus. Lactase was
not found in extracts of the intestine of the carp, Necturus, or snapping turtle, the
only animals in which lactase was sought.

DIGESTIVE ENZYMES IN POIKILOTHERMAL VERTEBRATES 199

BIBLIOGRAPHY

BIEDERMANN, W.
1911. Winterstein’s Handbuch der vergleichenden Physiologie. Band 2, 1911, pp. 1049-
1361. Jena.
Bopansky, Meyer, and Wiuuiam C. Ross.
1922. Comparative studies of digestion. II. Digestion in elasmobranchs and teleosts.
The American Journal of Physiology, Vol. 62, No. 3, November 1, 1922, pp. 482-
487. Baltimore.
Braptey, H. C.
1922. Studies of autolysis. VIII. The nature of autolytic enzymes. The Journal of
Biological Chemistry, Vol. LII, No. 2, June, 1922, pp. 467-484. Baltimore.
CoHNHEIM, OTTO.
1901. Die Umwandlung des Eiweiss durch die Darmwand. Hoppe-Seyler’s Zeitschrift
fiir Physiologische Chemie, Band 33, 1901, pp. 451-465. Strassburg.
Dpan, BASHFORD.
1916-1923. A bibliography of fishes. Vols. I-III. Digestion in fishes, Vol. III, p. 403.
New York.
Decksir, FRIEDRICH.
1887. Zur Physiologie des Fischdarmes.. Jn Festschrift fiir A. von KOlliker, zur Feier
seines siebenzigsten Geburtstage. 444 pp., 17 pls. Leipzig. [Cited by Bieder-
mann (1911).]
Dernspy, Karu Gustav.
1918. A study on autolysis of animal tissues. The Journal of Biological Chemistry, Vol.
XXXV, 1918, pp. 179-219. Baltimore.
Epincer, Lupwia.
1879-1880. Notiz, betreffend den Magen von Tropidonatus natrix. Archiv fiir mikro-
skopische Anatomie, Band 17, 1879-1880, p. 212. Bonn. [Cited by Bieder-
mann (1911).]
Foun, Orro, and W. Dernis.
1912. On phosphotungstic-phosphomolybdic compounds as color reagents. The Journal
of Biological Chemistry, Vol. XII, 1912, pp. 239-243. Baltimore.
GLAESSNER, Karu, and ALIcE STAUBER.
1910. Beziehungen zwischen Trypsin und Erepsin. Biochemische Zeitschrift, Band 25,
1910, pp. 204-214. Berlin.
Homporcer, L.
1877. Zur Verdauung der Fische. Centralblatt fiir die medicinischen Wissenschaften,
No. 31, August, 1877, pp. 561-562. Berlin. [Cited by Biedermann (1911).]
Kinestey, JoHN STERLING.
1917. Comparative anatomy of vertebrates. Second edition, p. 240. Illus. Philadelphia.
Kriicer, ALBERT.
1905. Untersuchungen tiber das Pankreas der Knochenfische. Wissenschaftliche Meerese
untersuchungen, Neue Folge, Band 8, 1905, pp. 59-80, Taf. I-II. Kiel.
KRUKENBERG, C. Fr. W.
1878. Versuche zur vergleichenden Physiologie der Verdauung mit besonderer Beriick-
sichtigung der Verhialtnisse bei den Fischen. Untersuchungen aus dem physi-
ologischen Institute der Universitit Heidelberg, Band 1, 1878, pp. 327-340.
[Cited by Biedermann (1911) and Yung (1899).]
1882. Zur Verdauung bei den Fischen. Untersuchungen aus den physiologischen Institute
der Universitit Heidelberg, Band 2, 1878-1882 (1882), pp. 385-401. [Cited by
Biedermann (1911) and Oppel (1896) .]
OpprreL, ALBERT.
1896. Lehrbuch der vergleichenden mikroskopischen Anatomie der Wirbelthiere. Erster
Teil, der Magen, pp. 33, 41, and 98. Jena.
1897. Lehrbuch der vergleichenden mikroskopischen Anatomie der Wirbelthiere. Zweiter
Teil, Schlund und Darm, pp. 37 and 51. Jena.

200 BULLETIN OF THE BUREAU OF FISHERIES

Pearse, A. S.
1921. Distribution and food of the fishes of Green Lake, Wis., in summer. Bulletin, U. 8.
Bureau of Fisheries, Vol. XX XVII, 1919-20 (1922), pp. 253-272. Washington.
Rippue, Oscar.
1909. The rate of digestion in cold-blooded vertebrates.—The influence of season and
temperature. The American Journal of Physiology, Vol. XXIV, 1909, pp. 447-458.
Boston.
SPALLANZANI, L.
1785. Versuche iiber das Verdauungsgeschift des Menschen und verschiedener Tierarten,
nebst einigen Bemerkungen des Herrn Senebier. Ubersetzt von Chr. Fr. Mich-
aelis. Leipzig. [Cited by Biedermann (1911).]
Sutuivan, MicHaEL X.
1907. The physiology of the digestive tract of elasmobranchs. Bulletin, U. S. Bureau of
Fisheries, Vol. X XVII, 1907 (1908), pp. 3-27. Washington.
TIEDEMANN, F., and L. GMgLIN.
1827. Recherches expérimentales sur la digestion considérée dans les quatre classes d’ani-
maux vertébrés. Traduit de lVallemand par Jourdan. 2 vols. 1827. Paris.
Yune, EMILE.
1899. Recherches sur la digestion des poissons. Archives de Zoologie Experimentale,
Series 3, Tome 7, 1899, pp. 121-201. Paris.
VON SwIiEcickI, HELIODOR.
1876. Untersuchung iiber die Bildung und Ausscheidung des Pepsins bei den Batrachiern.
Pfliiger’s Archiv fiir die gesammte Physiologie des Menschen und der Thiere, Band
13, 1876, pp. 444-452. Bonn.

&

GROWTH AND AGE AT MATURITY OF THE PACIFIC
RAZOR CLAM, SILIQUA PATULA (DIXON)

am

By F. W. WEYMOUTH, H. C. McMILLIN, and H. B. HOLMES

&
CONTENTS
Page Page
maproduction® 0.22 901 ‘The razor clam—Continued.
Hiheuracon clain: #6 se 902 Growth of the adult___------------ 216
Anatomy 204 Ring method of age determina-

a re ae pea Fa aoe Pm are i . 9
Digestive system__________---- 207 eit re aac eee 216
Nervous system______________- 207 Growth in different localities__ 218

Hocomotions. ii. ~~~. 207 Maturity iE areas ee 224
Greenlee. ey Ae te 2 208 Agevat maturity, --=-—---=----= 224
Time of spawning ___________- 208 Size at maturity____-_---_-_-- 227
Relation of water temperature Relation of age and length to =
tO;SOaWuIng onset 209 maturity Sanaa eas ae rin
Larval development and growth of Quantitative HANSEN S
young 213 Maturity in other forms______~ 230

a ee ee ee -_ 92

Length of larval life as com- ne aa a ; STMiants eee tee ne
pared with other mollusks___ 213 TN catch-----~---------+----- Bee
Time of setting and growth of Mirae tiene aaa oat eae FN ia cesT Pe Be a
young_____________________. 214 Bibhography2i 2312 Ul ot ee 235
Mortality of young____________ 215 ‘
INTRODUCTION

In the spring of 1923 the senior author was engaged by the Bureau of
Fisheries to undertake the investigation of the razor-clam fishery of Alaska. The
following summer, accompanied by Mr. Holmes, he spent 8 weeks in the field,
visiting the Cordova beds and those near Chisik Island, Cook Inlet. In March,
1924, Mr. McMillin, who had been working on a similar problem for the State of
Washington, entered the service of the bureau, and during the following summer
accompanied the senior author for 10 weeks in the field, examining the. beds at
Cordova and those in the vicinity of Kukak Bay; the Washington beds were
visited on the return trip.

The data collected was worked up at Stanford University by Mr. Holmes in
the fall of 1923 and by Mr. McMillin during the fall and winter of 1924-25, under
the supervision of the senior author. All general conclusions are the results of
discussion and agreement of the three authors.

201

202 BULLETIN OF THE BUREAU OF FISHERIES

The study of growth and maturity forms the main part of this paper. These
matters are not only of scientific interest but are of importance on account of the
bearing which they have on the measures of conservation that have been put into
effect or may be recommended for the future. On this account the paper will be
of more than usual interest to the men engaged in the clam industry of Alaska,
and it has seemed desirable to include a discussion of certain matters which would
not ordinarily be included in a strictly scientific paper, but which experience has
shown will be of interest to the nonscientific reader and will lead him to a better
understanding of the more technical portions.

We wish to acknowledge the advice and helpful criticism of Dr. Willis H. Rich
during the progress of the work.

THE RAZOR CLAM

The razor clam, Stligua patula (Dixon), occurs in commercial quantities from
near the mouth of the Columbia River to the Aleutian Islands. Canning has been
carried on at Warrenton, Oreg., along the whole coast of Washington, on Van-
couver and Graham Islands in Canada, and at Cordova, in Cook Inlet, and Shelikof
Straits, Alaska. This range possibly may be extended by the opening of canneries
in southwestern Alaska along the Alaska Peninsula, It is essential to determine
whether the variations encountered in this 2,800 miles of coast are of specific value.

The razor clam (Stliqgua patula) was first described by Dixon in 1788 from
specimens found near Coal Harbor, Cook Inlet, Alaska. Conrad found shells near
the mouth of the Columbia River in 1838, and described them as a separate
species—Stliqua. (Solen) nuttallii. Later this was changed in rank to a variety
of the original Séliqgua patula. Other species and varieties exist, which, however,
need not be considered here. <A description of the two mentioned above, taken
from Dall (1899, p. 109), is as follows:

3. Siliqua patula Dixon, 1788, Okhotsk Sea; the southern border of Bering Sea, and the
Gulf of Alaska to Sitka.

Described from Cook Inlet, Alaska. * * * Targe, with the submedian beaks and
straight rib. The following are discriminable varieties, but apparently connected by grada-
tions with the typical S. patula.

4. Siliqua (patula var.) alta Broderip and Sowerby, 1829; Bering Sea and Strait. * * ¥*,

5. Siliqua (patula var.) nuttallii Conrad, 1838, Lituya Bay, Alaska, south to Oregon, and
California as far as Monterey.

* * * The shell is very straight, brilliantly polished, narrower than the typical S.
patula and with a much more oblique rib. * * *

A full copy of the original description is given by Oldroyd (1924, p. 58)
without comment. The distinction given rests on the direction of the rib, which is
said to be “ straight ” in “ patula”’ and “ more oblique ” in “ var. nuttallii,” and on
the proportional width of the shell, which is alleged to be broader in “ patula”
than in “var. nuttalli.”

It is not difficult to separate Alaskan and Washington shells by their general
appearance, but these differences are, we think, not of specific rank. The clams
from the northern waters grow more slowly; the annual rings of growth on the
shell, plainly visible upon even superficial examination, are thus more numerous

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 203

and closely placed. While the number and placing of the rings may distinguish
shells from various localities, these, as we shall show, are the result of environ-
ment and not specific differences. It is also found that clams on different sections
of one beach grow at different rates. Specimens from places only a few feet apart
but separated by some obstruction, such as the Grays Harbor jetty, may show
decidedly different shell markings, and the growth curve constructed from shell
measurements reflects the distinction.

The weight of shells from different beds also varies according to the rate of
growth. As shown in Figure 18, the number of years necessary for clams to

47

45

41

Width (Percent of Length)

39

3S

/ 3 7 9 ul 13 1S 17 19

Age in Years

Fic. 1.—Showing increase in relative width with age. The solid line shows the median width at
each age and the broken lines the 10th and 90th percentiles, between which are included 80 per
cent of the widths. (Compilation of data derived from all sources.)

reach a certain size varies with the beds. Each year a new layer is added to the
inner surface of the shell. In clams that have not reached their full size a thick-
ened portion of the new layer extends beyond the margin, thereby increasing the
size. The additions on some old clams increase only the thickness of the shell.
Thus there is a relationship between the age of the shell and its thickness, and in
comparisons both age and size must be considered.

The direction of the rib is another point of alleged difference between the
two varieties. It is charactized as “straight” in “ patula” and “oblique” in
“var. nuttalli.” The rib is somewhat more oblique in young shells than in old ones,

204 BULLETIN OF THE BUREAU OF FISHERIES

and as Alaskan shells, on the whole, are older than those from the Columbia River
district, it seems probable that “ var. nuttalléi”’,was described from younger shells
than were used by Dixon for the original description. The direction of the rib
varies between shells, but so far as can be determined there is no variation that
is peculiar to or confined to one bed.

In the growth of the clam both length and width increase, but not in con-
stant ratio. There is a variation between individuals of the same age and an in-
crease in average width with age. The average relation of length to width (in
percentages) at each age is plotted in Figure 1. The variation at each age is ap-
parent; for example, at 3 years of age the width of eight out of every ten shells lay
between 37.5 and 42 per cent of the length. If sufficient records are used to give
a reliable average, it is seen that with increasing age the average width rises from
about 38 to 43 per cent of the length. This change of relative width appears to be
more closely correlated with age than with size. In comparing specimens of the
same age from various localities we have been unable to find constant differences
of significant size as compared with individual variation.

From what has been said it is clear that, in the description quoted above, the
influences of age and relative size have not been considered, and that clams from
the ranges of these two varieties do not differ significantly in the characters of di-
rection of rib or of width of shell nor, so far as can be determined, in other fea-
tures apart from those directly dependent upon environment. It will be noticed
that “ var. alta” is not included in this discussion. It is unsuited to canning and
therefore is not considered in this paper. A thorough study of the shell and the
soft parts shows it to be a separate species, distinct from S. patula.

The writers feel, after visiting beds on all parts of the coast and examining
many thousands of shells, that the above views are sound. Excluding valid species,
such as “ var. alta” and S. lucida, which do not form a part of the commercial
catch, we are dealing with a single form. Therefore, comparisons of its reaction
to features of environment on different parts of the coast are valid. A more de-
tailed consideration of these and other points of biological interest will be given
in a later paper.

ANATOMY

Everyone who handles clams has noticed some of the more conspicuous parts
and, since the mollusk is so different in structure from ourselves and other mam-
mals, has wondered what the various organs are and how they work. The follow-
ing brief sketch is intended to answer some of the most common of these questions
and to give a glimpse of the plan of organization of a type of animal strange
to most of us. A number of features of the anatomy of the razor clam merit
more attention than is given here and may be treated later.

The clam, like other members of one group of mollusks, is protected by two
similar shells or valves (hence the term bivalves), which are formed on the right
and left sides of the animal, the back or dorsal side being that where the valves
are joined together. This shell, which is so striking a feature of these animals,
is a product of the mantle, a soft structure also characteristic of this group.

Buuu. U. S. B. F., 1925. (Doc. 984.)

Fic. 2.—Razor clam, Siliqua patula (Dixon), side view

Fic. 3.—Razor clam, Siliqgua patula (Dixon), ventral view

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 205

a F p M Mf T S Is

Fic. 4.—Razor clam. View of left side, with shell and mantle removed. Aa, anterior adductor. Zs,
exhalent siphon. fF, foot. Js, inhalent siphon. G, gill. H, hinge. M, mantle. Mf, mantle fold.
P, palp. Pa, posterior adductor. S, siphon. Sh, shell. 7, papille. U, umbo

cumveceicde
se hleny: oie nee

“SL

Is

Fic. 5.—Same as Figure 4, but with left gills and palps removed, and siphon, foot, and visceral
mass also removed. Ap, anal papilla. O, crystalline style. Hs, exhalent siphon. Ga, pedal gang-
lion. He, heart. J, intestine. Js, inhalent siphon. JL, liver. Mo, mouth. Pr, posterior rectractor.

St, stomach

206 BULLETIN OF THE BUREAU OF FISHERIES

The mantle covers the animal as the flyleaves cover the body of a book, and by its
activity secretes the valves of the shells, which thus come to occupy the position
of the covers of the book. If we think of a book with limp leather covers, which
bend over to meet each other around the edges and imagine the flyleaves doing
the same thing, the picture is a very complete one.

Shells found on the beach or in the cannery shell piles lie flat, as an open
book on a table. They are held in this position by the black hinge at the back.
In life the valves are closed, to protect the animal, by the contraction of two large
muscles, one toward each end of the shell. The muscles are called the anterior
and posterior adductors, the latter being that nearer the siphon.

In the simplest form of mantle the edges are free except on the back, where
the hinge is located, corresponding to the arrangement in a book, and the sea
water may enter the cavity inclosed by the mantle at almost any place; this is the
condition, for instance, in the oyster. In the razor clam, however, the edges of
the mantle are fused, not only along the back, where the valves are joined together,
but for about one-half of the remainder, chiefly at the posterior end. Three open-
ings are left. The largest of these extends from the back around the anterior end
and about halfway along the lower margin. Through this large slit the foot can
be extended, and the hole is guarded by finely branched papille extending like
ruffles along the margin (see figs. 4 and 5). When these are brought together,
the sand is prevented from entering. At the posterior end the mantle is developed
into a long double tube surrounding the two other openings of the mantle cavity.
These form the siphon or “neck”, which, contrary to the implication of the com-
mon name, is not at the anterior or head end but at the posterior or hinder end.
We thus have, when the clam is feeding quietly in the sand, a fairly large cavity,
in which hangs the main body of the clam, closed except for the double tube reach-
ing up through the sand to the water abave.

The mantle is lined with cells, most of which are covered with cilia—fine,
microscopic, hairlike projections, but, unlike hair, always in vigorous motion.
They lash sharply in one direction and return slowly, thus driving along any
water that touches the mantle, like a myriad of tiny oars. The cilia on the mantle
and gills all beat according to a plan which causes a current to flow out of the tube
nearer the back, which is thus called the exhalent siphon, and in at the other, or
inhalent siphon.

The water that passes through the mantle cavity supplies the clam with
food and oxygen. The food consists of a great many microscopic plants and

-animals that live in the sea water. They are passed by the cilia across the gills
(commonly called “livers” by those engaged in the industry) and the palps to
the mouth, which lies just above and anterior to the foot near the anterior
adductor. As the water passes through the gills it comes in close contact with
the blood. An interchange takes place, similar to that in our own lungs in which
oxygen is taken up by the blood and carbon dioxide given off. The waste
products from the intestine are carried out by the water as it leaves the body
through the exhalent siphon,

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 207
DIGESTIVE SYSTEM

The mouth of the razor clam has the shape of a flattened funnel and leads
directly into a slender wsophagus. The csophagus opens into the stomach,
which lies directly under the hinge and in this clam is comparatively large; it
is surrounded by a dark “liver,” which secretes digestive juices into it through
wide open ducts. The posterior part of the stomach is lined with cartilaginous
tissue, from the bottom of which a long diverticulum extends downward to the
lower margin of the foot. This is filled with a clear, gelatinous rod called
the “crystalline style,” containing a starch-digesting enzyme similar to that of
the saliva.

The intestine leads downward from the middle of the stomach and is coiled
below the liver, from whence it follows the posterior margin of the foot upward
through the end of the liver. Here it turns abruptly backward, passes through
the heart, over the posterior adductor, and ends in a fleshy papilla at the base
of the exhalent siphon. There are two dark-colored kidneys, one on either side
near the heart cavity.

NERVOUS SYSTEM

The nervous system of the razor clam is very simple, consisting of three pairs
of ganglia. The cerebral ganglia lie on either side of the mouth near the lower
surface of the anterior adductor muscle. The visceral ganglia occupy a somewhat
similar position on the posterior adductor muscle; they are more closely united
than are the cerebral. The pedal ganglia are closely fused and are embedded in
the muscles of the foot. These pairs of ganglia are connected by nerves. There
are no special sense organs except a pair of minute statocysts, or balancing organs,
lying on either side of the pedal ganglia.

LOCOMOTION

One of the most interesting things about the razor clam is its unusual ability
to move through the sand. The foot is quite different in shape from that found
in the soft-shell and butter clams, being elongated and nearly cylindrical instead
of flattened. It is a very effective burrowing organ, and, unlike many others,
the razor clam retains throughout life a high power of active movement. In
burrowing, the foot is extended from the shell; the tip, which is pointed, is
thrust through the sand, and, when fully extended, the tip expands, forming an
effective anchor. The foot is then retracted, drawing the body toward this
anchored portion. Repetition of this action gives the clam a rapid movement.

The mechanism for the retraction of the foot is easily) demonstrated. The
bulk of the elongated foot consists of longitudinal fibers. These are connected
with the shell by two specialized retractor muscles (anterior and posterior), one
of which les near each of the large adductor muscles which close the she. The
combined contraction of these shortens the foot with considerable force.

The method by which the foot is extended is less well understood. Clams taken
out of the sand very often discharge a large amount of fluid from the tip of the

42886—25—_2

208 BULLETIN OF THE BUREAU OF FISHERIES

foot, apparently through a rupture of the body wall. It is generally assumed that
the contraction of the circular fibers of the foot and visceral mass forces this fluid
to extend the foot by hydraulic pressure.

SPAWNING

In most mollusks the sexes are separate; in a few, however, both eggs and
sperm are produced by the same individual. In the razor clam, as in most forms,
all observations have shown each individual to have germ cells of one kind only.
The reproductive material is cast into the water at spawning time, and there fer-
tilization and further development take place.

Only by examination of the gonads is it possible to tell the sex of the clam.
There are no superficial characters that betray the sex. If the contents of the
gonads are spread on a glass slide or on a scalpel, a marked difference between sexes
is apparent. The ova have a granular appearance, in contrast to the viscous homo-
geneous mass in which the sperm is found. The sexes of the entirely immature
clams can be determined only by microscopic examination of tissue, which was not
carried out in this case while in the field. A large number of clams in more ad-
vanced stages of maturity were examined and the sex recorded. This was done
for two reasons—to determine the stage of maturity and to make sure that both
sexes were equally represented. As far as has been determined both sexes pass
through similar stages in an equal period of time, and from the standpoint of
maturity and growth they do not differ, hence, in the following calculations the
sexes have been combined.

TIME OF SPAWNING

Data were obtained from the 1923 and 1924 spawning seasons on the Wash-
ington coast (McMillin, 1923). These two spawnings differed somewhat and may
represent two extremes. As described in the above reference, the 1923 spawning
took place on May 30, 31, and June 1. It was practically simultaneous, and the
major portion was completed in a very short time. As a result of this spawning
a heavy set of young was produced, the average per square foot on the whole beach
numbering more than 1,400. The spawning of 1924 was different, apparently be-
cause the temperature during May was variable and the average remained low.
Spawning started on a rise of temperature on June 12, the rate of discharge of
eggs was reduced by lowered temperature on June 14, and after that spawning
continued slowly for two weeks. Although the action was slow, all observations
showed the clams to be acting in unison, and at any one time all seemed to be in
the same condition. The result of this year’s spawning was negligible, no speci-
mens representing this class having since been found.

In 1923 the spawning on the Oregon coast was not followed. On the North
Beach section of Washington, extending from Willapa Harbor to the Columbia
River, spawning started on May 19. On the section beween Willapa and Grays
Harbors it started on May 24, and on the Copalis section, as already described,
on May 30. In each of these sectors the whole population of the beach spawned
at the same time. No data were collected for the Queets and Kalalock beds of

Buuu. U. S. B. F., 1925. (Doe. 984.)

Fic. 6.—Small clams, Copalis beach

Fic. 7.—Digging on the “river spit,’ Swickshak beach

. ‘
~
* i
4 4
¢ .
a

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 209

northwestern Washington. There is no available information on spawning on the
Vancouver Island and Graham Island beds in Canada.

While visiting the canneries and beaches of Alaska, as many observations as
possible were made on the condition of the clams at various times. During the
season several cannery men preserved specimens in formalin at stated intervals.
These specimens were collected and examined further to determine the amount of
reproductive material in the gonads during each part of the season. In this way
it is possible to fix approximately the spawning season for each section from which
specimens were obtained.

Spawning records for the vicinity of Cordova are very incomplete. The ob-
servations made on preserved material and the examination of clams on the
beds indicate that spawning started soon after July 1, 1924. On July 13 but 2
per cent of the clams showed they had recently spawned out, 10 per cent had full
gonads of such consistency as to indicate no active spawning, and the remainder
(88 per cent) showed various degrees of incomplete spawning. On August 15,
when over 400 specimens were examined, but 2 per cent carried any appreciable
amount of spawn, 28 per cent showed the presence of a small amount of repro-
ductive material, and 70 per cent were completely spawned out.

The two records cited above were taken over a month apart. They give a fair
example of what is found to exist generally in Alaska. There is no short com-
mon spawning season, as is found on the Washington beaches, but rather a gradual
changing over in the condition of the clams, from those in midsummer with large,
full gonads to the thin, spent ones in early fall. The process takes over a month,
although it is probable that any particular individual may cast its entire spawn
more quickly.

Similar conditions are found to exist elsewhere in Alaska. Near Chisik Island
in Cook Inlet observations were made during the summer of 1923. The clams ap-
peared fullest, or “fat,” in late July. Evidently spawning started between J uly
25 and 30. After August 1 it proceeded more rapidly and continued as long as
specimens were obtained. On September 25 a few showed the presence of a very
small amount of reproductive material, but it is doubtful if any spawn was being
cast.

The figures from Swickshak Beach for 1923 and 1924 show that the clams de-
velop reproductive material rapidly in early May. By July 1 they are quite full,
and they start spawning between July 20 and 25, but proceed most rapidly after
August 1. They appear to spawn out somewhat more rapidly than on either the
Cook Inlet or Cordova beaches. Before September 1 spawning was completed.

RELATION OF WATER TEMPERATURE TO SPAWNING

From the excellent work of Dr. Thurlow C. Nelson (1921) it is apparent that
the water temperature is closely correlated to the spawning of the oyster. A
critical temperature of 21° C. (70° F.) has been found, below which oysters will
not spawn. Above 21° C. spawning will take place more rapidly as the tem-
perature increases. It is further stated that due to continued cold weather, where
the water has failed to reach the critical temperature for any length of time, the
eggs will not be cast, but rather be resorbed by the oyster.

210 BULLETIN OF THE BUREAU OF FISHERIES

From the meager data available it appears that temperature influences the
spawning of the razor clam very much as it does the oyster. During the winter
the gonads are very small; they consist of a series of tubules lying among the
muscle bundles of the foot and about the intestine. After the first rise in tem-
perature in the spring the gonads increase in size. They gradually distend the
visceral mass until they form about 30 per cent of the total body weight and cause
a decided gape in the shell. The eggs and sperm are contained in follicles within
the tubules. Shortly before spawning it appears that the follicle walls are rup-
tured or disappear, and the reproductive products are then free within the tubules.

While more data would be desirable, available records indicate that the tem-
perature of the water over the beds is closely related to the act of spawning. It is
known that a high temperature is accompanied by rapid spawning of the clam,
and a drop in temperature arrests progress. During the season of 1923 the

ae

60 Legend
e@ Temperature -1923
° ” 1924
s1 ron |
e
v »
S ps5 §
v %
sc $ 2
ra Qa
Hae 3
i v
x
50 10

April ae | June

Fic. 8.—Showing water temperatures at Copalis during the seasons of 1923 and 1924. The rise
of temperature that immediately preceded spawning in both cases is indicated between the
square dots (@—M)
temperature of the water over the beds at Copalis rose steadily from 8° C. on
March 27 to 11.5° C. on May 20. During this time there has been a constant
increase in the size of the gonads but no active spawning. Between the taking of
the temperature at 6 a. m. on May 29 and a similar time the following day there
was a rise in temperature from 11.4 to 18.1° C. (See fig. 8.) Spawning started
quickly and with great vigor. There was no appreciable drop in temperature, and
spawning was nearly complete in two days. Some eggs remained in the gonads,
but they were few in number compared with the great mass that had been cast.
The following year (1924) spawning started 14 days later. The temperature
of the inshore water, as well as that at the Columbia River Lightship off the mouth
of the Columbia River, was more variable than that of the previous year. Again
spawning started immediately following a rise in temperature, which in this case
was from 12.8 to 15.3° C. Two days later the temperature dropped sharply to
12.6° C., accompanied by a greatly reduced rate of spawning. Less spawn was cast

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM Q11

in 10 days during this period of low temperature than during the 2 days of the
preceding year. From all indications spawning may have been suspended entirely
during part of the time. No attempt was made to correlate the rate of spawning
with the number of larve in the water, which Nelson has shown to be possible with
the oyster. The condition of the clams was determined by opening a large number
and making direct observations on the contents of the gonads.

Temperatures taken in Alaska near the spawning time were very similar to
those on the Washington beds. The open beach near Cordova had a temperature
of 13.9° C. on July 16, which was shortly after the onset of spawning. The pro-
tected waters of Orca Inlet and adjoining bays from which claims were taken
varied in temperature from 12.8 to 14.5° C. The temperature records from
Swickshak Beach showed an average of 13.5° C, at spawning time. ;

These figures suggest a critical temperature of about 138° C. (55.5° F.)
for the razor clam, as compared with 21.0° C. (70° F.) for the oyster. The
summer temperature of the Alaskan waters is quite high in all cases, except where
influenced by glaciers or glacier-fed rivers. The variation from day to day in
the Ketchikan records is less striking than on the Washington coast. For
instance, the variation at Ketchikan for the month of July was from 13.5 to
16.0° C., and for August, 18.0 to 17.0° C. In contrast to this, similar figures for
the Washington coast were as follows: July, 10.6 to 17.2° C.; August, 11.1 to
17.2° C. On two occasions during the summer of 1923 the temperature of the
shore water at Copalis dropped more than 6° (centigrade) over night. ‘This
cold water was accompanied by a fog which was largely confined to the beach.
Crab fishermen outside the harbor declared it was clear a short distance offshore,
and personal observation showed the fog to extend not more than half a mile inland.

A few hours after the cold water was noticeable on the beach a peculiar
collection of material was cast up by the surf. It contained a great many sponges,
snails and snail shells, worm tubes, cast crab shells of several varieties, and a great
mass of other detritus, which ordinarily rest lightly on the bottom. This indi-
cated the presence of a current of comparatively high velocity moving landward
over the bottom, which caused an upwelling of cold water along the shore. It was
evidently of small extent, as it affected not over 20 miles of coast line.

It was suggested by McEwen, in a personal communication, that this up-
welling was not similar to one described by him (McEwen, 1912) from the Cali-
fornia coast but, rather, it was caused by some local condition of the air currents,
While the Ketchikan temperature record does not show the sharp transient drops
that might be expected from such an upwelling, this would not prove that up-
wellings do not occur in this region. Their effect would be most marked on the
outer coast; as, for example, on the west side of Prince of Wales Island the
‘circulation of the tide among the many islands of the Alexander Archipelago
would tend to obscure fluctuations of the inland waters about Ketchikan.

As has been shown by J. Nelson (1917), oysters acclimated by living for
many years in the Gulf of St. Lawrence have a reduced critical temperature of
20° C. (68° F.). It has been further shown by T. C. Nelson (1921) that oysters
moved from New England to New Jersey spawn two weeks earlier than do the

912 BULLETIN OF THE BUREAU OF FISHERIES

native oysters.
native strain.

The critical temperature for the razor clam of Alaska is probably not lower
than for those on the Washington beach, as the slight difference in summer tem-
perature probably insures an opportunity for spawning along the whole coast
under similar conditions, The temperature of the inshore waters, from Pismo,

Calif. (the southern limit of Siliqua patula), to the end of the Alaskan Peninsula,
is quite uniform during the summer.

Those moved north from Virginia spawn later than the

Temperature
s
Temperature °C

45

ric. 9.—Showing the mean (average) monthly temperatures for Washington beach and Ketchi-

kan (Ketchikan temperatures furnished by U. S. Coast and Geodetic Survey). @ — @, Wash-
ington beach; O—O , Ketchikan

The temperature record for Ketchikan is the only complete set for Alaskan
waters. Comparison of these with records taken over the clam beds indicate that
they are representative of conditions that exist generally over the whole region.
The more southern waters reach what might be called a normal temperature
earlier than do those farther north, and they also coo] later in the year. For
instance, the Washington temperature was above 13° C. for twice as long as was
the temperature at Ketchikan during the 1923 season. (See fig. 9.) The mean
temperature for each month showed a difference in the maxima of 1.8° C., while

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 213

the actual difference was illustrated by the reading on August 5, which was
Washington, 17.2° C., and Ketchikan, 17° C.

LARVAL DEVELOPMENT AND GROWTH OF YOUNG

The spawning season of 1923 presented an excellent opportunity for the study
of larve on the Washington coast. Spawning took place in a very short time, and
the razor clam larva was the predominant form taken in plankton tows made
during the time when they were swimming. Three weeks after spawning the
larval stage appeared in the tows. At that age the shell is transparent and can
be seen only when shattered by slight pressure. One month after spawning a
great number of what proved to be half-grown larve were found in the surf at
high tide. In tows taken in 4 feet of water at the same time many similar larve
and some larger ones were found. More were taken at high tide than at low tide.
This may have been due to the flood tide raising the larve out of the surface
layer of the sand and carrying them in the surf during the remainder of the flood
and part of the ebb of the tide. As the larve increased in size more were found
in the sand, and at the same time the numbers in the tows decreased. All] clams
were considered as larve until they had opaque shells covered with a brown
periostracum, and the shape had become markedly elliptical.

There is no evidence of extensive migration of adult clams. In fact, observa-
tions point to a very limited range for any individual. The distribution occurs
mainly during the larval stages. As the period during which the young are in the
swimming stage amounts to about eight weeks, the excellent set resulting from one
spawning would indicate that the larvee are largely confined to the sand. If the
eggs and larve were free for this length of time they would probably be so widely
scattered that the original beds would not be heavily stocked. The eggs sink quite
rapidly and are not easily raised by surf action. They are probably discharged
at the surface of the sand. If they sink as soon as they have left the body of the
mother and resist the action of the surf, which would tend to scatter them, they
would be fertilized in great numbers in the surface layer of the sand. As they also
spend a large part of their larval life in the sand, it is not surprising that the re-
sulting set in one place is enormous. However, the young clams do swim and the
eggs are doubtless moved about, but apparently not to the extent that might be ex-
pected from an eight weeks’ larval period.

LENGTH OF LARVAL LIFE AS COMPARED WITH OTHER MOLLUSKS

The length of the larval life of the razor clam seems quite long when compared
with that of other mollusks. It has been found by Nelson (1921) that the eastern
oyster (Ostrea elongata) has a free-swimming stage lasting from 14 to 18 days.
This was determined by direct observation under known conditions of salinity and
temperature.

According to Kellogg (1910) the soft clam (Mya arenaria) swims as a larva
for from three to six days. The general observations from which this indetermi-
nate period was inferred are not very satisfactory. A number of difficulties are en-

214 BULLETIN OF THE BUREAU OF FISHERIES

countered. Attempts to rear them artificially were unsatisfactory, and because of
the extended breeding season and the numerous similar young of other mollusks the
direct observations of the larve were not decisive.

Belding (1910 and 1912) concluded that the quahog (Venus mercenaria) re-
mained a free-swimming larva from a week to 12 days, while the pecten (Pecten
erridans) passed through the corresponding stage in about a week. It was con-
cluded by Field (1922) that the common mussel (Mytilus edulis) remained in
the free-swimming stage about two months when living under natural conditions.
He compared the specimens taken in nature with those raised in an aquarium and
arrived at quite definite conclusions.

This length of time agrees more closely with the larval period of the razor clam.
The larvee differ in one respect, however, as the mussel is actually swimming
during the greater part, if not all, of the period, while the razor clam is more
inclined to remain in one place. The mussels are scattered over a great area and
are dispersed by this method. The habitat of the razor clam is more limited, and
the larvee do not scatter to a like degree.

TIME OF SETTING AND GROWTH OF YOUNG

Kight weeks after spawning large clam larve visible to the unaided eye were
found in a quiet lagoon. None of this age was taken with the plankton net in
the surf, although repeated attempts to do so were made at several places. It is
probable that the majority had already passed through this stage. The larval
shell was now dense enough to hinder observation of the organs, and the posterior
end had a marked tendency toward the elongation which is so marked in the adult
clam. Prominent gills were present, but no velum was seen.

While under observation the live specimens would extend the characteristic
long foot from the shell. Five days later small clams in the adult stage were found
in the surface of the sand. They averaged 0.25 centimeter in length and were the
forerunners of a very heavy set. Distribution was slightly uneven, although they
were everywhere abundant, averaging approximately 1,450 per square foot. At
this size they were found in the surface layer of the sand, and if a small hole was
dug in the wet sand it quickly filled with water in which there were a number of
small clams. If the water in this depression were given a circular motion with the
hand, a number of small clams, as well as sand from the edge, were dislodged. The
clams tended to be whirled to the center, but as they are very active, even when
being carried quite rapidly over the surface of the sand, they will catch their foot
in the sand and disappear from sight.

At intervals large numbers of small clams were taken by screening the sand.
These were measured and the median size determined. (See fig. 10.) During
the month of August the small clams increased in average size from 0.25 to 0.88
centimeter. The following month they grew to 1.39 centimeters in average length.
A marked decrease in the rate of growth was found in October, during which
month an increase of only 0.24 centimeter was shown. At this time the average
length was 1.68 centimeters, and no further growth was observed.

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 215
MORTALITY OF YOUNG

The preceding paragraphs give some idea of the great number of small clams
that may settle in the sand in late summer as the result of a successful spawning
year. The set is not equally heavy in consecutive years. Some years the young are
very numerous, while at other times it is difficult to collect specimens enough for an
examination. The abundant set of 1923 gave an excellent opportunity to study the
mortality as well as the growth of these small clams during the first season.

On August 1 the young clams, now only 0.25 centimeter in average length
(about 1% inch), were found on the beach in great numbers. From four counts the

125 3.2 1600

1900

100
24 1200
~
&
1000.) &
a 2.0 a3
< : AS
&
plas é
ss s 800.) 2
= es 1.6 g
a i. ~
S = S
© ry
%
L050 4 ie 600} °
a
es
a2
§
=
0.8 400.
o2s = 4 10" Percentile
04 200.
0.00 0.0} —— ae) 0

August September October November December January February
1923 1924

Fie. 10.—Showing the growth and mortality of small clams on the Copalis beach of Washington
in 1923. Solid lines, read from scale on the left, indicate growth, showing rapid increase
in size during August, soon after settling in the sand, and no increase during the winter.
The broken line, referred to scale on the right, shows the decrease in numbers of small clams
per square foot. The sudden drop in early December was due to a storm

average number per square foot was found to be 1,451. In determining the distri-
bution the clams were taken only from the top layer of the sand about 2 inches in
thickness. The shifting of the sand may account for some errors, as Thompson
(1919) has shown that the action of the surf on exposed sand beaches continually
changes the topography of the beach. This was illustrated by the counts of Sep-
tember 1 and 9, in which 986 clams per square foot were found on the first date,
while 1,122 were found on the same area nine days later.

As indicated by the figures just given, there was a steady decrease in the number
of clams per unit area. By late fall they had been reduced to a third of the original

42886—25——3

216 BULLETIN OF THE BUREAU OF FISHERIES

number. On December 4 there was an average of 494 per square foot. On that date
a heavy storm strikingly changed the topography of the beach. An extraordinarily
heavy surf dragged the sand from the high beach out into deep water. Certain
sections of the beach were lowered 4 or 5 feet. While it is natural to assume that a
great number of the small clams perished, undoubtedly many were carried out into
deep water where their survival is uncertain.

A census of the beach immediately after the storm showed very few clams.
Some places were entirely barren, but an examination of the entire beach gave an
average of 18 per square foot. During the winter there was a lighter mortality,
leaving about 16 per square foot when examined the following February. Of these
survivers very few, if any, were left the following summer. Thus we see that the
number of small clams resulting from one spawning is no criterion of the extent
to which that year class will affect the commercial catch. The records given above
show the total loss of a very heavy set; therefore, a successful spawning does not
necessarily indicate that the population of the beds will be increased by it.

The fate of clams removed to the area below low tide is questionable. Small
clams are not found in water deeper than 12 feet, and probably only stragglers
among the mature clams are found far from the intertidal zone. In the instance
cited above a great many may have been removed to the area exposed at the lowest
tides only or bordering these areas. They would then supplement the number al-
ready there, and might even spread out and return to the higher levels, where they
might be taken by the diggers.

The State of California is attempting by a yearly census to find the abundance
of the annual set and to follow the survival of the young of the Pismo clam from
year to year. This is done by taking all clams from a trench crossing the entire
beach. The process is repeated at several places in order to obtain a reliable
average. The clams are then sorted as to age, and the number of each class for
each meter of trench is recorded. The result of the last census is given by Her-
rington in a paper which will appear in an early number of the California Fish
and Game Bulletin (1925). A similar system of annual surveys of the Washington
beach, supplemented with the examination of the cannery shell piles, would give
an index of the beach population. The census of the set each year, with an analysis
of its subsequent effect upon the commercial catch, would make it possible to
determine the result of fishing, protective measures, and to foretell with some
degree of accuracy the amount of the commercial catch.

GROWTH OF THE ADULT
RING METHOD OF AGE DETERMINATION

The size at various ages is a matter of very great practical importance in
various questions, such as the determination of age at sexual maturity and the
efficiency of protective legislation. Since the absence of direct observation which
can fix the age has made is necessary to use extensively the indirect determination
by “annual rings,” a short statement of this method will be desirable.

Buti. U. 8S. B. F., 1925. (Doc. 984.)

Fic. 11.—Clam in a burrow, ventral view Fic. 12.—Native digger, Snug Harbor

Buu. U.

S. B. F., 1925. (Doce. 984.)

Fic. 13.—Clam digger’s equipment, Swickshak beach

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 217

The method is that which has long been unquestioningly applied to trees, which,
in their structure, show evidences of the number of seasons through which they
have passed. The annual nature of the “rings” visible in the shell of the clam
has been clearly proved by the senior author (Weymouth, 1923) in the case of
the Pismo clam. The same method has already been applied to the razor clam
(McMillin, 1923). In brief, it rests upon the fact that the clam does not grow
at a uniform rate, growth being rapid in summer and slow or absent in winter.
This is illustrated by the results obtained by following the young of the Pismo
clam through two years (Weymouth, 1923) and the small razor clams through
one year.

Figure 10 (reproduced from McMillin, 1923) shows the absence of growth
during the winter. The possible reasons for this, chief among which appears to
be temperature, are discussed in detail in the papers referred to and need not be
considered further here. For our purpose it is sufficient to note that the rings are
formed during the season of slow growth. This is confirmed by many observa-
tions on the razor clam, in that the relation of the rings to the growing margin
of the shell in clams small enough to show appreciable annual growth is always
what might be expected for a winter formation of the ring. They lie close to the
margin in the spring, farther away in the summer, and farthest in the fall.

This is further supported by the study of size frequencies. During the winter
all clams were taken from a certain area of beach and their total lengths recorded.
By tabulating these in size classes a frequency table was obtained from which a
curve was constructed (McMillin, 1923; fig. 8). This curve showed modes the lo-
cation of which corresponded to the lengths of the annual rings for the second and
third years. After the third year the overlapping in size of the different year
classes obliterates the modes of each class.

The annual ring is a structural feature of the shell and not a superficial mark.
In the broken shell a layer marking the surface of the shell at the time the ring was
formed can be seen extending from the ring on the present surface through the en-
tire shell. A white line remains on the underside of the periostracum when it is
peeled off. ‘his is found to mark the location of the annual ring. Some feature of
the growth about the time of the formation of the ring causes the periostracum to
become fused with the calcified portion of the shell, thus giving emphasis to its
location. From these facts it is clear that a ring of definite character is formed
once and only once each vear and is therefore a definite indication of age.

The difficulties in applying this method are of three kinds. (1) Injury to the
edge of the shell in summer may retard the growth and cause a check that might
be confused with the annual ring. (2) A first and sometimes a second ring is
difficult to determine on clams 12 or more years of age, as these early rings are
formed on very thin shells and may be removed by erosion. (3) The crowding of
the annual rings near the edge of the shell, due to the very small increment of
growth, makes the location of the last rings in old shells indefinite. From the
above causes some records may be in error, but the number and magnitude of these
are not large, and because of the great number of observations the general results
can not be appreciably affected.

218 BULLETIN OF THE BUREAU OF FISHERIES
GROWTH IN DIFFERENT LOCALITIES

Many shells of various localities have been examined and the total length of
each ring determined, thus giving a great mass of records of lengths at known ages.
All measurements were made with one caliper by one or two persons. The caliper
was graduated in millimeters, and by the use of a vernier scale could be read to
tenths of millimeters. While all measurements were made and recorded in tenths
of millimeters, the last place was not used in calculations. A few measurements
were made in the field, but the majority were made in the laboratory. There is

Length in In.
ow
Length in Cm.

Ace in Years

Fic. 14.—Showing thé growth of clams from three localities. From this the average size for any
age can be read. For example, the lengths for 3 years of age are as follows: Cordova, 4.61
centimeters (1% inches) ; Swickshak, 6.41 centimeters (2% inches); and Washington, 10.87
centimeters (414 inches). The legend of the abscissa “age in years”, is used for convenience.
The age rings form first at %4 year of age, and annually thereafter. Therefore, the actual age
is % year less than indicated on the graph

usually a change in density or pigmentation of the shell at the annual ring. The
use of light to make these zones more apparent was found to be very helpful. If
a shell is exposed to direct sunlight or a strong arc or incandescent light is placed
under it, the annual rings show up in such a way that they can be measured with-
out difficulty.

From measurements made in this manner growth curves for clams from
various beaches have been constructed. Three curves presented here (fig. 14) were
made from clams taken from beds at Copalis, Wash., and Cordova and Swickshak,
Alaska. The data are given in tables 1 to 4, inclusive.

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 219

TaBuE 1.—Copalis: Adult razor clams, total length at time of formation of each ring, based on ring
measurements

Ring number Ring number

Length, in centi- Length, in centi-
meters, mid-value of |\—7——~ >, 7 || meters, mid-value of

class 1/9/34] 6] 6 | zjcslo9 class 1/2/3]/4!/5!/6|7]8|9

Leoe|soealeenetates Es oh ee eee ete A Re
Had 7 ae Ve (eae Ve ae | Hee
sud a eae) aces (a (Oe Oe oie (Rs

Bip Oo>
Ssso

S8sss

S8s8ss

bese |seealh 6. | okalpoa le

7 | 20
Sere E eal 2! [hee 119) [ele

1] 8

4

1

Soon
ss

t

H

'

eee eed Le el Feary Peal soee|ezeelzcee! 9] 6
Sas eee pee ee Me ee ee Bees ae ee 6| 7
west bexatiose(cele|oaca ee eee socs|eese|bo<e|se5- 1} 2

a
an
id
bo
NPR CN d

sseelseeclecsc|2se~|bees al

220 BULLETIN OF THE BUREAU OF FISHERIES

TaBLe 2.—Cordova: Adult razor clams, total length at time of formation of each ring, based on ring
measurements

Length, in Ring number Length, in Ring number
centimeters, centimeters,
mid-value mid-value
of class A} 2 is | 4h ee | 72) 8 9h) 10 | 11 of class 1|/2;/;3)]4/]5/] 6) 7/7] 8 | 9%!) 10] 11

oOnow-
ooooo

ooooo

ooocoe

CONNwWH ONnwH OLUmMwH
sooso

onawe
ooooo

SOUS SS Re 2 OS NIN NINN: eee Oo Soe 2

Sow onowe
oo

oooo

rie) em (a

|

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 921

TaBLE 3.—Swickshak: Adult razor clams, total length at time of formation of each ring, based on
ring measurements

Length, in Ring number Length, in Ring number
centimeters, |) 0) 2b 2 ht Ee est || centimeters,
mid-value C mid-value |
of class 1/2/3)4/5]6]7/] 89 }10/11|12)13/14 of class 11293)4)5 , 677°) 8 | 9110) 11) 12 Bi 14
|
|
1
ie
1.
al eee Pe
ate aa5|—==|=5=
210M soa. /2525| 225 bY] es ee ae Pelee
DST OE Se age een Sa ey 65). -_|-22 son | eel oe
DBO et Open e (|| len ees ee Pees ee
PAGAL sepia gel ae O6 |e aes. gee |e eee
DNS Sins eee pea 65) Tyes2 S22 Sales
| ee ee en) Pa 4 tae ie ews) sont ae a
Celt Saag a 42 eae
BUS See eee ee of |se eles
ere eee eee ee 31) 4/._-
3:90 8e eee ce bl] ol. 11} 9)... pero ee es
C5). ee ees eee (Nos Hee Bs ee ee ree
COR 0 a a ee CH dee ee ee
CDE SA ee ee (tee ios Pare ee eae
EAT). as ee eae 26|2 22 PA Rese
A OO wee. ook Pe lite hos 8 saen|zaseee
5. sat
5. 2
5. 1
5. cae ee
5. Ce
6. Sisco)
6. Al scalsee
6 1D ees ee
6. 3) lj 1
6. 6} 2) 1
7.10 Beate | ao aio! 6) Deze
7.30 seas. |46] ee, ae Lee
7.50 eet | a2e| 0) 8} 2) 1
TAS ep ae ene --}<--}. 20).28 aires)
Cee eee ee ---| 25] 3 Bi) 2jze=
BO beste Wet so ---}| 13] 20 2i|)
CS HSSk Uae ge a eee OWS 3 cee
£5 (od) pee a te aL ea MLO} bel aaslecelca cles ooslceeteac|sce aoalae
TaBLe 4.—Growth in different localities
Years Copalis | Cordova ples Years Copalis | Cordova re
2. 04 0. 48 0. 38 14. 18 12.57 14.19
8. 61 2. 38 2.70 14. 50 13. 08 14. 63
10. 87 4.68 GATS | OR ee ere ees Mec a Leen 13. 60 14, 94
12. 04 7.73 (M77 Be Se gees oe Be eee a 2 se 14.15 15. 25
12. 81 9. 84 ns FSe Se a) Fe ee Pe | eas ee or 15. 61
13. 40 11. 40 74) |S aE FUSE ed Bs Ae ee eA Pa oka 16.12
13. 84 12. 03 aS or (018 | i ee ee ea ek ee Cy (Oe OES a 15. 96

922 BULLETIN OF THE BUREAU OF FISHERIES

The Washington curve shows the fastest growth, followed by Swickshak and
Cordova, in the order named. The first point on each curve represents the average
size of the young clams from the respective beds during the first winter. The
Washington clams reach an average size of 2 centimeters, in contrast to 14 centi-
meter for Alaska. This is due to the later spawning in Alaska and the earlier
suspension of growth, due to the shorter season, which is about one-half as long
as for the southern beds. The young clams of Alaska are very difficult to find
during their first year. Plankton tows taken at Cordova on August 17, 1924,
showed a large number of half-grown larvee, which could hardly have settled soon

Length in In.
a
Length in Cm.

1 2 a 4 5 6 7 8 9 10 “4 42 43 id is

Age in Years

Fic. 15.—Swickshak clams. Graph showing the average length of each age (solid line) and the
average age of each length (broken Jine). ‘The latter is irregular in the first three years, due
to the necessity of using one year as an interval of time. Center line shows the median or
age, and the lines above and below show the 90th and 10th percentiles, respectively

enough to make appreciable growth before the water temperature dropped below
the point where growth is possible.

While there is not a great difference in the final size of clams from various
beds, there is a marked variation in the rate of growth during the first few years.
As has already been stated, clams from the Washington beds reach maturity at
about 10 centimeters in two years, while those from Swickshak and Cordova do
not reach a similar size before three and four years, respectively.. The legal size
of 11.44 centimeters (41% inches) is reached at the following ages: Washington,
3.5 years; Swickshak, 5 years; and Cordova, 6.38 years. The growing season in
Alaska, as already mentioned, is roughly one-half as long as in Washington.

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 223

However, more definite data may show a closer correlation between the rate of
growth and the length of growing season of the different beds, which in turn may
be confined to clams below a certain size. The maximum for the Washington
clams is about 12 years, while those from Alaska may reach an age of 17 or 18
years, but the large number of specimens necessary for an average to extend the
growth curves to those ages are not available.

The ordinary growth curves, samples of which have been presented in Figure
14, are constructed from a table in which the average (median used) length of
each age is calculated. With the same data used to calculate the values for the
Swickshak growth curve the process was reversed, and the average age of each
length class was determined (fig. 15 and Table 5) and a growth curve constructed
from these figures (Weymouth, McMillin, and Rich, 1925). As the annual interval
is very much larger than is convenient to use, the points at the lower end of the
curve are irregular. The range was too small to calculate the tenth and ninetieth
percentiles until the fifth year. Up to the age of five years the two curves might
well be considered as coinciding; then the second “regression” (the average age
for each length) tends to fall below the first, but later it rises and crosses it at
about nine years, after which there is a distinct separation of the two curves,
This shows that the correlation between age and size is less as the clams increase
in age. In younger specimens size is a fair indication of age. A 2-year-old and
a 4-year-old clam are greatly different in length, but the difference between a
9-year-old and an 18-year-old clam may amount to very little, with the pessibility
of the younger being the larger.

TaBLe 5.—Swickshak beach

Length on age Age on length
Percentiles Percentiles
Age in years Length in centimeters
10 50 90 10 50 90

es a a a 0. 24 0. 38 OSTOUIN Ube ccececeeecuscadetoonacnschloccanensan WOO Wiseeeae ees
PF oS AS ee ee 1. 94 2.70 On02)|||)2.02 = seven sencececeasssccseena|EoonsaceS= 2.00) | eee eee
Doe Sean ee I Ata Ue 4.70 6.41 PRED al | EE aoe OS Se ee ee | fee 2-100), |3. Sasa eee
fh ae ee 8.13 9. 28 LOH O45 | sae bh ot oe Siete dee eee ee eee ak 2400) | Glen
Oba mee mne oO 2 Se ected 10. 30 11. 49 P2IGS i Osa aa cee eee oo feces coe aceteses|-cenewoces ZHOU NE Zoe eaee
GRE Beye s Fa ovis ke i iS 11. 92 12. 74 Fa (3) | Sa Se A ee eae ae ae) eee ee Oe 3.000 | 3s
ene eae ces mbes ee ea o8 12. 74 13. 70 LAVOE REGS oeaweees 2a os ee eae ee Soe es 3002s cacea fe
Se ee see ee a 13. 27 14.19 VSS) 825222 88 oe ees Se ak et SHOU IE -steb EL SS
Dope meee test aiecaeeesso-cuebens 13. 57 14. 63 TOSOONN RO Olan enaen2efectusveseueeeseseltecaccevaa hye ee
1 aes seers eee ee ae eee 13. 94 14. 94 15298! (Obs eot Doser dycee ee eed oa |b Seer 4/05. | 2--.22
Dba Baia ee Bho oo. scene’ 14. 29 15. 25 TGS 28 se ce eee Pee ene stb) 41604) Sesa- ees
LQ ee ee see eee 14. 53 15. 61 UGH6 Us| | 125s ces cesses soce see cawee ee 4.09 5. 03 6. 04
DOES AOE Bid ssh sous wes ahe 2 15.19 16. 12 MOY 78 Bi 5s sss ee a does be ae oe des 4.85 5. 94 7. 25
Nee ee secaeas ss ole ssekeacesctee|seceueeeoe TO%96) |e 1 bs eee ee Spee ee 5.77 7.12 9.15

1 las ee aes eee ee ee ee ee 6. 91 8. 83 10. 97

16752 22653. 221 ewy geist ts 8. 08 10. 03 11. 94

WpOsdcb So ccccseeeee ee secceee se 10. 03 11. 25 12.78

224 BULLETIN OF THE BUREAU OF FISHERIES

MATURITY

AGE AT MATURITY

During the summer of 1924 a large amount of fresh material was examined to
determine the age and size at maturity. In some localities the specimens were
taken from boxes on the cannery floors soon after they arrived from the beds; at
other times they were dug especially for this purpose and examined on the beach.
As the length of the clams used for this study varied from 7 to 12 centimeters (214
to 414 inches) the shells were very thin. The mantle was split along the mid-
ventral line, and the shell broken back to expose the interior. When the mantle is

Percent

Age in Years Age in Years.
Cordova Swickshak
Legend: immature, i 2”° stage of immaturity, fl mature.

Fic, 16.—Showing the coming into maturity of small clams from Swickshak and Cordova, with
reference to age. All are immature at 2 years of age and change gradually, to become com-
pletely mature at 5 and 6 years of age, respectively. Some pass through an intermediate
stage the year prior to full maturity

opened, the larger part of the body is seen to be composed of the foot, or “ digger,”
which contains the gonads and in mature specimens before spawning is distended
with a great mass of reproductive material. In immature clams the posterior part
of the foot appears translucent, and through its walls the outline of the digestive
tract can be seen. After a number of examinations a marked difference, apart from
size, can be noticed between clams that have spawned out and those that have never
formed reproductive material. The mature clams never regain the translucent
appearance and some become very dark soon after spawning.

In localities where maturity comes on at from 3 to 5 or even 6 years of age, there
is a gradual changing over from the immature to the mature state. This makes a
class between the obviously immature and the mature that can be distinguished

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 225

accurately. For the sake of simplicity this intermediate stage has been called the
“second stage of immaturity.” It was first thought that the second stage of im-
maturity, as here explained, was a transient phase preceding complete maturity
by a few weeks. Careful study, however, showed that this intermediate stage per-
sists after the mature clams have spawned. Small egg strings or sperm-bearing
masses are present in unchanged condition in late summer. Female clams have
small elongate eggs, which do not change during the season. For these reasons it
appears that those individuals passing through this stage remain in the same condi-
tion for the year prior to maturity. It is not known whether the reproductive
products are held over the following winter or are resorbed.

Assuming that the stage of maturity is accurately determined in all cases (282
clams examined), the facts seem complicated. For the sake of clearness Figure 16
has been prepared from the data given in Tables 6 and 7. From it the process of
coming into maturity is shown from the standpoint of age of the specimens. At
Cordova all of the specimens examined in their second year were immature. In
the next class a few (5.3 per cent) had become completely mature, and the same
number had developed only partially—that is, had reached the second stage of
immaturity. The following year class showed a few still immature and the mature
ones of the former year still present, with their number increased by the addition
of those in the second stage of immaturity of the former year. It is also shown
that about 80 per cent of the fourth-year class passed out of the immature state,
going into the two advanced divisions in the proportion of about 3 to 5, the smaller
number becoming mature and the remainder advancing to the second stage of im-
maturity. No immature clams were found in the fifth-year class; 80 per cent were
mature and 20 per cent were in the second stage of immaturity. This last figure is
larger than the number of immature of the preceding year, but does not necessarily
show that some individuals remained in the second stage of immaturity more than
one year.

TABLE 6.—Swickshak beach: Size of clams in different stages of maturity

Length in centimeters 3 4 5

Im Im: M Im Im:2 M Im Im: M

926 BULLETIN OF THE BUREAU OF FISHERIES

TaBLE 7.—Cordova: Size of clams in different stages of maturity

Age
Length in centimeters 3 4 5 6
Im Im; M Im Im; M _ Imz M M

Medians

OTS TED eb, LALO 8.60 9.50} 10.75 8. 80 9.67| 10.39 9.67| 10.50
Medians (total) 8.70 "9.87

It must be remembered that cach year class here used had a separate origin.
The annual variation that exists may account for the apparently large number of
5-year clams that were not fully mature. Jn the vicinity of Cordova clams are
taken from several beds, upon which the rate of growth is not equal, and the

Percent Mature

Age in Years

Fic. 17.—Showing the percentage of all clams mature at each age. For example, at Cordova

of all clams 4 years of age 36.5 per cent were mature, while at Swickshak 85 per cent were
mature at the same age

selection of some of the material used may have caused this discrepancy. All
clams more than 5 years old are completely mature and spawn every year.
Figures obtained from 308 clams taken on Swickshak Beach gave more regular
results. All 2-year-old clams were immature, but the following year over 40 per
cent had passed out of this stage, 15 per cent becoming completely mature. In the
fourth-year class 7 per cent were immature, while those in the second stage of

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 227

immaturity numbered about 17 per cent. A great majority in the fourth year
(over 75 per cent) and all specimens that were older were completely mature.

On the Washington beach no such gradual changing over is found. After the
clams pass through their second winter they begin to develop reproductive ma-
terial, and at 2 years of age spawning takes place. Doubtless there are some that
do not spawn at this age, but none such have been found. The amount of the
spawn cast is not large, but it does not differ in appearance from the products of
the larger clams.

Thus it is seen that on some beds maturity comes on over a period of years.
In such cases the age at which 50 per cent of the clams are mature (fig. 17) is
taken as the age of maturity, and as thus determined is found to be 3.2 years at

Percent mature

Length in Cm.

Fic. 18.—Showing the percentage mature at each interval of length

Swickshak, 4.2 years at Cordova, and 2 years at Washington. To summarize,
about one-third of the clams at Cordova and one-half of those at Swickshak and
all on the Washington beach become mature without passing through an inter-
mediate stage. Since the clams at Cordova are slower in growth than those at
Swickshak or Washington, it appears probable that more of them would pass
through the second stage of immaturity, and this is found to be the case. The
bearing of size upon maturity will be taken up in the following section of this
report.
SIZE AT MATURITY

The relation of size to maturity has been treated in the same manner as age
(fig. 18). The percentage of each stage of maturity in each class interval of
length was calculated and plotted. At Cordova no mature clams of less than 9

228 BULLETIN OF THE BUREAU OF FISHERIES

centimeters were found, and all clams more than 1014 centimeters were mature.
The size at which 50 per cent were mature (9.75 centimeters) was taken as the
size at maturity. At Swickshak clams are first mature at 814 centimeters, and
all above 1114 centimeters are mature, maturity (50 per cent) being at 9.91 centi-
meters. All clams from the Washington beach spawn when they are 2 years old.

Length ih Cm.

Age in Years
Fic. 19.—Showing portions of three growth curves and points at which 50 per cent, by age,
and 50 per cent, by size, are mature on each bed (points W, S, and C), For the Swickshak
material, Sio, S, and So represent the points at which 10, 50, and 90 per cent, respectively,
by both age and size, become mature

Since all mature at the same age, the median length of these at spawning time
has been taken as the size at maturity.

Thus in order of size the lengths at maturity on the different beaches are as
follows: Cordova, 9.75 centimeters; Swickshak, 9.91 centimeters; Washington,
10.30 centimeters. At the rate at which the Washington clams were growing at

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 229

this age (1.54 centimeters in three months) this difference amounts to one month’s
growth.

It must be remembered, in comparing these lengths with possible size limits.
that at the above lengths but 50 per cent are mature. A recommended size limit is
intended to provide opportunity for all individuals to spawn at least once.

RELATION OF AGE AND LENGTH TO MATURITY

The comparison of the results of the last two sections throws an interesting
light on the process of maturing on the various beds of the Pacific coast. From a
study of Table 8 and Figure 19 we see that there is a wide difference in age at ma-
turity, but there is a close agreement in size. While the differences in size are small
in comparison with those of age, an examination of the data shows them to be ord-
erly and significant. Clams from beds where growth is slow, while older at ma-
turity, are also smaller. The Washington clams, which show the most rapid
growth, are the youngest but are largest at maturity; conversely, those at Cordova
which show the slowest growth are the oldest but the smallest.

QUANTITATIVE MREASURES

A quantitative measure of the respective influences of age and length may be
obtained by analysis of the figures given in Table 8. Between the youngest and
the oldest there is a difference of 2.2 years, or, compared with the average of the
three ages (3.1 years), a variation of 71 per cent. In contrast with this we have a
variation of 0.55 centimeter in length, which, compared with the average length
(10 centimeters), gives but 5.5 per cent. ‘Therefore, in the process of maturing
an increase in size of 1 per cent accompanies the same sexual development as an in-
crease in age of 13 per cent.

TaBLe 8.—Age and length at maturity

Maturity
Locality
Age Length
NYY CSD NTA a a PS Ea nN at Oc een oD ee Ee RE 2.0 10. 30
SWICK Shia lc eemne rade recent Win UW Ae oe yee eee ee a eh 3.2 9. 91
WONG Oy meee enn neeetierna vase M ae Maye Oot) Ts ee SY Se a Pe a ea ee 4,2 9. 75

The data of Table 8, plotted in Figure 19 as line W—S—C, show the influence
of both age and length on maturity. If age alone determined maturity, this line
would be perpendicular and clams from all beds, while varying in size, would
mature at one average age. If size alone determined maturity, the line would be
horizontal. The age would vary, but the size would remain constant. From this
line another ratio may be obtained by comparing the average relation of age to
length and this same relation in respect to maturity. For example, from Figure
14 (p. 218), on the Swickshak curve near the size of maturity (3 to 4 years of age),
one year’s growth is equal to 2.89 centimeters, or, using this proportion, 0.346 of

930 BULLETIN OF THE BUREAU OF FISHERIES.

a year equal 1 centimeter in length. Like figures obtained from Washington and
Cordova are given in Table 9. From the maturity line (fig. 19, line W—S-C)
it is found that a variation of 2.25 years corresponds to a variation of 0.55 centi-
meter, which gives the following values: 4.10 years equal 1 centimeter; 1 year
equals 0.244 centimeter.

TaBLE 9.—Relation of rate of increase of age and length at maturity

Value of Value of
Locality 1 year 1 centi-

in centi- meter in
meters years
Washington: 20.2 cccceccnclssenestonseae oeeagen suman see cetoees sae cen ate eee oe. eee eee 2. 26 0. 442
Swickshak 2.4 2:22. 2-225,225-2-5 ee nen de be ee ee ee a i ee ae 2. 89 . 346
COPdOVG ccc ec nasncssacncousaescconteUses sss ssencaencat scans soe See ne ne ee ee een oa Re 2.17 . 460
(IAVOlIARZCL. oot sascnsdnsscena sabcoscascaasee eee ase e reas aoe Be one ee ee ee nee See 2. 44 416

The figures given immediately above are comparable with the averages of
Table 9. In growth, therefore, one year corresponds to 2.44 centimeters, but in
maturity one year corresponds to 0.244 centimeter. The influence of a year in
growth is thus ten times as great as in the process of maturing; or, from the stand-
point of size, 1 centimeter corresponds to 0.416 year in age, but in maturity 1 cen-
timeter corresponds to 4.10 years. The influence of a centimeter in growth is thus
one-tenth as great as in the process of maturing. The ratio (1 to 10) thus obtained
compares very favorably with that above (1 to 13), when the differences in methods
are taken into account,

MATURITY IN OTHER FORMS

The relation of age and length to maturity given for the razor clam does not
hold in all forms. The work of Thompson (1914) on the halibut shows that these
vertebrates mature at nearly the same average age in different localities, while the
average size varies quite widely. However, there is a difference in average age at
maturity which, although slight, may be accounted for by the difference in average
size. For example, fish from Hecate Straits mature at an average of 12 years of
age and 40.7 inches in length, while those from Kodiak Island Banks mature at an
average of 12.3 years and 35.4 inches. Thus, there is a variation in age of 0.3 year
in 12, or 2.5 per cent, and in size of 5.3 inches in 35.4, or 15 per cent. The con-
clusion is, then, that a difference of 15 per cent in size at maturity between these
two banks is accompanied by a difference of 2.5 per cent in age, which shows in
this form that age and length influence maturity in the ratio of 6 to 1, age being
the more important.

We offer no explanation for the maturing of halibut according to age and of
clams according to size. Figures for other species are not available. The work
of Baldwin (1921) and Crampton (1908) give data on the size and age of boys
and girls at maturity, which further analysis may show to be similar to one of the
above cases.

Buti. U. S. B. F., 1925. (Doc. 984.)

Fic. 20.—Rising tide, Snug Harbor beach

Fic. 21.—Digging marked area, Copalis beach

Buty. U.S. B. F., 1925. (Doc. 984.)

Fic. 22.—Weigh screen, near Cordova

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 931

BEACHES

No attempts have been made by the writers to explore areas that were not being
exploited at the time the work was carried on in the field. It is impossible to
find the unit population of any unit area, and in most cases it is difficult to get
anything but a very rough idea of the size of the clam-producing beds. On
undepleted beds the same area can be dug over on several consecutive days without
apparent decrease in the day’s catch, a fact which makes an accurate census im-
possible. Thus indications of depletion can not come from either a census or a
survey of the beach, but, as we shall later see, show up quickly in the average
size of clams in the commercial catch and in the average “dig” per man.

The clam-producing areas of the Washington and Oregon beaches can readily
be measured, The beach is uniform in width, and the area exposed varies with
the tide. The Alaska beds, on the other hand, are variable. Those in the vicinity
of Cordova have been built up by sand brought down by the Copper River. They
are made up of an intricate series of bars and channels which change their loca-
tion at irregular intervals. In fact, the whole topography of the clam-producing
area changes from year to year. Beds that are under several feet of water at the
lowest tide of one season may be accessible to digging the next, and some high
bars may be entirely submerged.

Swickshak beach changes less by shifting of bars and channels. The bottom
slopes gradually, and the offshore water is very shallow, giving a wide intertidal
zone. Since this beach is so flat, the height of tide governs the area exposed; on
one spit it was estimated that a change of 1 foot in the tide was accompanied by
a variation of 40 acres in the area of beach exposed. Owing to this, only rough
estimates can be made of the areas,

The condition of the Snug Harbor beds, lying between Chisik Island and Har-
riet Point on the west side of Cook Inlet, is more like that at Swickshak than at
Cordova. The tidal range over these beds is greater than elsewhere. Since the
beach slopes very gradually, it is wide, the tide going out nearly 3 miles in one
place. It was estimated that about 114 square miles of beach were being dug
during the summer of 1923. While this bed is not constantly being changed by
the shifting of bars and channels, as at Cordova, portions of it are seriously
menaced at times by deposits of glacial silt, which covers the beds and destroys
the clams.

COMMERCIAL CATCH

As we have mentioned before, overfishing shows up in two ways—first, by the
small size of clams taken, and, second, by the reduced average catch per man. On
new beds, such as Swickshak in 1923, nearly all the clams taken were more than
5 inches in length. The same is true of the early digging at Cordova and for many
years on the Washington beaches. On the northern beds the large clams appear to
be very inactive, sometimes remaining with the siphon extended until dug. There-
fore the large clams, on the whole, are more easily found and taken than the small
ones, It takes as long to dig a small clam as a large one, and the result of taking

232 BULLETIN OF THE BUREAU OF FISHERIES

them is apparent in the reduction of the average daily catch. On Washington
beds, during 1924, the commercial diggers were able to get fair “digs ” on but three
tides of each “run” (two weeks), unless storms interfered, in which case the
whole run was lost. The average size of the clams was small, being less than 41%
inches. A large part of the catch had never spawned. Many were lost through the
screens at the weigh sheds, and still more were thrown out by the cleaners because
of their small size. Although the enemies of conservation explained the reduced
pack as the result of an “off year,” such a condition points to serious depletion.
The same was true at Cordova in 1924. The diggers took every clam that
showed, regardless of size, and the cannerymen, in their turn, purchased anything
that was offered. Many of the clams were too small to be used to advantage and
would have doubled in weight by another year. Only serious results can be ex-

60000

50000

40000

30000

20000 I i
plo N Le ee

iain iae

Fic. 23.—Showing the total clam pack of Alaska to date. Data from the “ Pacific Fisherman”

Number of Cases

pected from such treatment. Since the beginning of canning in Alaska in 1916
Cordova has produced between 80 and 90 per cent of the entire clam pack. The
history of the pack is instructive (see fig. 23). After the opening of the beds in
1916 production leaped to approximately 50,000 cases in 1917 and 1918, and then
rapidly dropped off, until only 1,600 cases were packed in 1921. This was in part
caused by economic conditions, but the chief factor was the scarcity of clams, a
number of operators claiming that the beds were completely exhausted and that it
was unprofitable to attempt to pack. The demand for canned clams, however, led
to exploration, and new beds were found adjacent to Cordova. of such extent as to
more than double the clam-producing area. Exploration to the westward at this
time also led to the opening of Snug Harbor and later the beaches near Kukak Bay.

With the increased supply of clams the pack at Cordova again rose to 27,000
cases in 1923. The pack for 1924 was in excess of this, or about 43,000 cases. This

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 233

expansion has been obtained by the entry into the field of one new cannery and a
material increase in the number of diggers. The fact that in 1924 twice the area of
beach and more than twice the number of diggers produced but little more than
four-fifths of the number of cases packed in 1917 is in itself clear evidence that the
beds have been overdug. If more proof is needed, it ‘is furnished by the distinctly
smaller average size of clams taken and the admittedly larger proportion of under-
sized clams. The clams naturally reach as large a size at Cordova as elsewhere.
At Swickshak, in 1923, a clam less than 5 inches long seldom reached the cannery.
We have only to contrast this situation with that ‘at Cordova to realize how the
intensive digging, by taking the older and larger clams, has reduced the average
size.

If the entire pack of Cordova had been uniformly distributed over all the years
from 1916 to 1924, inclusive, this average pack would have amounted to 23,000
cases. From the present state of the beach it is evident that 23,000 cases per year
is in excess of what the beds can support. Perhaps 15,000 cases might be taken
annually without serious depletion, possibly put 10,000, but the data at hand is
insufficient to determine this.

It is clear from our investigations that a yield approaching, area for area, that
of the Washington beaches is not to be expected of those in Alaska, This can
not be too strongly emphasized, as estimates or expectations based on the produc-
tion of the Washington beaches, with which most of the operators and diggers
are familiar, must inevitably lead to disappointment and disaster. In the first
place, the set of young on the Alaskan beaches has never equaled or even ap-
proached that on the southern beds. As has been stated, at Copalis in the fall of
1923 the small clams averaged over 1,400 per square foot. During 1923 and 1924
a careful search for the young on all the producing Alaska beds did not reveal a
single location where a square foot would regularly yield even one small clam,
and only a few of the diggers have even noticed these young, although acquainted
-with them in the south.

Secondly, the growth is far slower in Alaska. We have shown that a length of
414 inches is reached in about three years at Copalis but in six years at Cordova.
Because of this the recuperative power of the northern beds will be far less and
the effects of protection will be slower in appearing. For instance, in a sample
of 152 clams over 4% inches in length, taken at one of the canneries during the
present season (1924), 109 were hatched prior to the beginning of digging in
1916; of the remainder 23 were hatched in 1916. In other words, 87 per cent of
the clams now being taken were in the beds by the summer in which canning
started. Even if we include the undersized clams packed in the last few years,
well over 90 per cent of the pack has been of clams already living when canning
started, and the course of the industry would have been little altered if all spawn-
ing had ceased at that time. Thus, to an alarming and unrecognized degree, the
industry has been consuming its capital. When the reduced spawning, resulting
from the reduced number of spawners, begins to make itself felt the pack, even
though materially reduced, can only be maintained by the ruinous process of
drawing upon that part of the capital which has never given interest—the young
before they have spawned.

234 BULLETIN OF THE BUREAU OF FISHERIES

It is therefore clear that the taking of clams must be drastically reduced or the
industry will vanish before the prohibitive cost of obtaining the few scattered
remaining clams. If this occurs, it must be remembered that its restoration is a
long process. The first generation would not be numerous enough to sustain a
commercial fishery, and for all of them to spawn would require six years, although
part of them spawn at four or five years. For the second generation to reach
even the minimum size of 414 inches would require another six years. Thus no
relief could be had before 12 years, and a material increase would of course
require several years more.

In the face of this situation it may prove that the present size limit is insuffi-
cient to maintain the beds. It is desirable that the clam industry be made a perma-
nent one, and the present size limit should be given a fair trial before additional and
more complicated regulations are imposed. Should this fail to reduce the pack to a
level that the existing beds can support, other measures remain. The size limit
might be increased to 5 inches. ‘This would practically close some of the beds in
which the average size has been greatly reduced by intensive digging, and thus
automatically afford protection where most needed.

It has been claimed by some that any size limit would be unsatisfactory. First,
the size of the clam could not be known before it is dug, and the thin shells of the
undersized clams probably would be broken by the digging operation and the clam
eaten by gulls if left on the beach. Second, a size limit would be difficult of en-
forcement. The writers have considered both of these arguments and fail to find
that either is cogent. If the individual “ digs” of the men are examined, it will be
found that many consist entirely of large or at least legal-sized clams, showing that
it is quite possible to avoid the small clams and still make good “ digs,” for it will
be found that these men usually have more clams than the average. The present
regulation permits a small percentage of undersized clams, which would take care
of those cases where small clams are accidently crushed in digging. Observation of
many clams has shown that broken shells are repaired to a far greater extent than
might be expected. From experience on every beach the writers feel sure that any
person taking more than 1 per cent of undersized clams is either a new and totally
inexperienced digger or does so by intention and not by accident. It requires no
argument to show that, with the clams in the shell on the cannery floor for periods
up to 24 hours, inspection of sizes would be easy.

Another method would be to close certain districts for a year or a term of years.
This would be more difficult of enforcement and might affect some canneries more
than others, but it would be effective if the necessity arose. A closed season has been
proposed on the assumption that it would reduce the pack. It is open to the objec-
tion of discrimination, such a limitation of season falling more heavily upon the
small local firms, at present operating over a long season, than upon the larger
canneries, which now operate during a short season only. The same season, also,
would not apply to all districts, and further investigation would be needed to regu-
late this, which would result in confusion. It is possible that such a regulation
would only result in more intensive fishing during a shorter season and would bring
about no reduction in the total pack.

Buuu. U. 8. B. F., 1925. (Doce. 984.)

Fic. 24.—Cannery at Kukak Bay

Fic. 25.—Cannery at Snug Harbor, Chisik Island

Buuu. U. 8. B. F., 1925. (Doe. 984.)

) ( :

4
x
i
a
i
5
?
i
‘

ee

Fic. 27.—Unloading clams, Snug Harbor cannery

GROWTH AND AGE AT MATURITY OF THE PACIFIC RAZOR CLAM 235

SUMMARY

The present paper deals with the Pacific razor clam (Stliqua patula), a form
of commercial importance along the entire northwest coast from Oregon to the
Alaskan Peninsula. This is considered to be a single species throughout this range,
the observed variations being due to environmental conditions.

Spawning on the Washington coast takes place between the middle of May and
the middle of June, while in Alaska the spawning season is less sharply marked
and occurs from the first of July to the middle of August. Spawning apparently
takes place on a sharp rise of water temperature at about 13° C. (55.5° F.).

The larval stage endures about eight weeks, during the greater part of which
time the young are in the surface sand. At the end of this period they appear
in the intertidal zone as small clams, differing little from the adult in structure
or habits. The “set” of these young is sometimes enormous on the Washington
coast, but is never so heavy in Alaska. The mortality of the young during the
first two years may be very heavy. A yearly census for the purpose of determining
the success of each spawning year and its future effect on the commercial catch is
recommended.

"The growth of the adult on various beaches has been determined from data
obtained by a study of the annual rings. Norms are presented to show the rela-
tive rates of growth at the three places where the industry is most important.

The size and age at maturity has been determined by a study of the young
clams. The size at maturity on all beds is in close agreement, but the age varies
greatly. The relative influence of age and size on maturity has been determined
mathematically and compared with other known forms.

The only feasible criteria of overfishing are, first, the reduction in the average
size of clams in the commercial catch and, second, the reduction in the average
daily “digs” per man. A size limit has been proposed which, by protecting the
small, immature clams, would prevent serious depletion of the beds now greatly
overdug.

BIBLIOGRAPHY

Batpwin, Brrp T.

1921. The physical growth of children from birth to maturity. University of Iowa Studies

in Child Welfare, Vol. I, No. 1, pp. 1-411. University of Iowa Press.
Betpine, Davin L.

1910. A report upon the scallop fishery of Massachusetts, including the habits, life history
of Pecten irradians, its rate of growth, and other facts of economic value. The
Commonwealth of Massachusetts. Special report of the Commissioners on Fish-
eries and Game, 150 pp., 40 pls. Boston.

1912. Quahog and oyster fisheries of Massachusetts. The Commonwealth of Massachusetts.
Special Report of the Commissioners on Fisheries and Game, pp. 1-134, figs. 1-69.
Boston.

936 BULLETIN OF THE BUREAU OF FISHERIES

CRAMPTON, C. WARD.
1908. Physiological age. A fundamental principle. I. American Physical Education
Review, Vol. XIII, No. 3, March, 1908, pp. 141-145. II Jbid., No. 4, April, 1908, pp.
214-227. III. Ibid., No. 5, May, 1908, pp. 268-288. IV. Ibid., No. 6, June, 1908, pp.
345-858.
Dat, WILLIAM H.
1899. Synopsis of the Solenidz of North America and the Antilles. Proceedings, United
States National Museum, Vol. XXII, 1900, pp. 107-112 [see p. 109]. Washington.
FIELD, IRVING A.
1922. Biology and economic value of the sea mussel (Mytilus edulis). Bulletin, United
States Bureau of Fisheries, Vol. XXX VIII, 1921-22 (1923), pp. 127-259 [see p.
200], figs. 100-230. Washington.
KELLOGG, JAMES LAWRENCE.
1910. Shellfish industries. 3861 pp. Henry Holt and Co., New York.
McEwen, Grorce F, |
1912. The distribution of ocean temperatures along the west coast of North America, de-
duced from Ekman’s theory of the upwelling of cold water from the adjacent
ocean depths. Internationale Revue der Gesampten Hydrobiologie und Hydro-
graphie, Band V, 1912, pp. 248-286. Leipzig.
McMiItuin, Harvey C.
1928. Life history and growth of the razor clam. Jn 84th Annual Report, Supervisor of
Fisheries of the State of Washington, 1923-24 (1925). iy
NELSON, JULIUS.
1917. An investigation of oyster propagation in Richmond Bay, P. HE. I., during 1915. Con-
tributions to Canadian Biology, 1915-16 (1917), pp. 53-78. Ottawa.
NELSON, THURLOW C.
1921. Aids to successful oyster culture. In New Jersey Agricultural College Experiment
Station Bulletin No. 351. 59 pp., illus.
OxpDROYD, IDA SHEPARD.
1924. Marine shells of Puget Sound and vicinity. Publications, Puget Sound Biological
Station of the University of Washington, Vol. 4, March, 1924, pp. 1-272 [see p. 58].
Seattle.
THOMPSON, WILLIAM IF.
1914. <A preliminary report on the life history of the halibut. Report of the Commissioner
of Fisheries, Province of British Columbia, 1914 (1915), p. N 76. Victoria.
THOMPSON, WILL F., and Juria Betrt THOMPSON.
1919. The spawning of the grunion (Leuwresthes tenuis). Fish Bulletin No. 8, California
Fish and Game Commission, pp. 1-29 [see pp. 15 and 16]. Sacramento.
WEYMOUTH, FRANK W. ;
1923. Life history and growth of the Pismo clam (Tivela stultorum Mawe). Fish Bulletin
No. 7, California Fish and Game Commission, pp. 1-120. Sacramento.
WeymoutTH, IF. W., H. C. McMiniin, and Witiis H. RIcuH.
1925. The regression of age on size, a neglected aspect of growth. Proceedings, Society for
Experimental Biology and Medicine. [In press.]

&

FISHES OF THE REPUBLIC OF EL SALVADOR, CENTRAL

AMERICA
&

By SAMUEL F. HILDEBRAND

Ichthyologist, U. S. Bureau of Fisheries

Bod
CONTENTS
Page
Imtroductions..2.2-.-2=-2--=- =.=... 238 | Part II.—Annotated list of the marine
Acknowledgments-------------------- 241 fishes collected at the ports of
Explanatory notes__------------------ 242 Triunfo and Cutuco___--______-____-
Part I.—Descriptive catalogue of the Cluppeid settee tae ee
fishes occurring in the fresh waters of Opisthonema libertatis —_____-
WiSalvador. 22-2 2-52.22 ---4-_~--= 243 npraullid eae eee eee
Whanacinidwe= 2225522 a es ere 244 Stolephorus exiguus___-__------
Astyanax fasciatus zneus- - --- 244 Stolephorus panamensis-_--_---
Reeboides salvadoris, sp. nov_-- 246 Stolephorus rastralis__________
PA TC ts Sate fs Li) hs Ses LE eh oe 248 Stolephorus brevirostris__—_-_-_
Galeichthys guatemalensis___-- 249 PACELLI Goes eee Se ees ee eee
Arius taylori, sp. nov____------ 250 Thyrinops pachylepis_________
Pimelodidse:_ 2222.22... ...----- 251 I Rvifeat bho rae ee ae oeeae
Rhamdia guatemalensis_ ------ 252 Mugil curema_____-___-----_--
Cyprinodontid#________._.__-_-_- 253 Polynemid#____...--------------
Profundulus punctatus___----- 253 Polynemus approximans_--__-__
Poeeciliide___._._.______-_-__-_-_- 254 C@arancid S2 ep ese tee eee
Mollienesia sphenops- - ------- 255 Caranx hippos_-_-------------
Priapichthys letonai, sp.nov--. 258 Oligoplites mundus___________
Priapichthys fosteri, sp. nov _-- 260 Oligoplites saurus__-_--_-_-_-
Anablepid se. 222 ne he eee 261 Centropomide_______-___-________
Anableps dovii_____---------- 262 Centropomus pectinatus_____-
Atherinids 4.22222. 55-5-5.-526e 263 Centropomus robalito_________
Thyrina guija, sp. nov_._------ 264 Epinephelidg-__._.-._-------__-.-
Mulgilidees = 225.20 ol ek 265 Epinephelus analogus________-
Mugil cephalus__----_------- 266 lbinbyich Nae Boe ee eee ee
Agonostomus monticola_-_____- 267 Lutianus argentiventris ______-
Centropomide___________________ 268 Lutianus novemfasciatus _____
Centropomus nigrescens _-_-__- 269 Pomadasid 6a 2a. eee ee oe
Centropomus robalito___------ 270 Orthopristis chaleeus-_-____-_--
Centropomus pectinatus_-____- 271 Pomadasis panamensis_ _--_---_-
Cichlids 2vs.2e 82222 ee 272 Anisotremus dovii____________
Cichlasoma nigrofasciatum__-. 273 Anisotremus pacifici__________
Cichlasoma macracanthus- ---- 274 Gerrid eves ee nee ee
Cichlasoma meeki, sp. nov_-____ 275 Gerres peruvianus___________-
Cichlasoma trimaculatum - -__- 277 Ephippide_________1____________
Cichlasoma motaguense --_--_-_- 279 Cheetodipterus zonatus________
leotridss 242 4. Sean ee Se 281 Gobild sae eee
Gobiomorus maculatus__-_-_--_- 281 Gobionellus sagittula_________

Page

283
283
283
283
283
284
284
284
284
284
284
284
284
284
285
285
285
285
285
285
285
285
485
286
286
286
286
286
286
287
287
287
287
287
287
287
287

938 BULLETIN OF THE BUREAU OF FISHERIES

INTRODUCTION

This report is based upon collections made in El Salvador during the last
half of January and the first half of February, 1924, by Fred J. Foster, superin-
tendent of the United States Bureau of Fisheries station at Neosho, Mo., and
the writer. This visit to El Salvador was made in response to a request of the
Government of that country to the Government of the United States for the services
of one or more specialists to examine principally the fresh waters to determine
the status of the fisheries and to study their needs. The United States Commis-
sioner of Fisheries thereupon detailed Mr. Foster and the author to proceed to El
Salvador for that purpose. A report embodying the economic phase of these
studies has been prepared by Mr. Foster and the author; it is sufficient to state
here that fish are very scarce in most fresh waters and that the quantity taken
annually appears to be very small. Two brief collecting trips were made to salt
water, and there, too, fish appeared to be rather scarce. It is well understood,
however, that the abundance of fish in salt water usually varies greatly with the
season, and the results obtained from the two short visits made are not sufficient
to form the basis for any conclusions.

The scarcity of fresh-water fishes may be due mainly to the absence in the
past of measures of protection, and in certain waters probably to the liberation,
through earth disturbances, of poisonous gases. Fishing has been carried on at
all seasons of the year, and such destructive methods as the use of dynamite and
plant poisons appear to have been used frequently. Fishes of nearly all sizes are
taken and marketed. Individuals 75 millimeters in length, belonging to species
that reach a length of 300 millimeters and more, are taken and sold in the markets.
Even the top minnows (Peciliide, locally known as ‘‘chimbolas’’) are sold both
in the fresh and dried state.

El Salvador is the smallest of the Central American Republics, having an area
of 34,126 square kilometers, or about equal in size to the State of New Jersey,
and it is the only one of these republics which is wholly within the Pacific drainage
and which does not also border on the Atlantic Ocean. The climate, of course, is
tropical, but the country is largely mountainous, and a considerable variation in
temperature, according to altitude, exists. This country, like most tropical coun-
tries, has a wet and a dry season. The wet season extends from about May to
November, and the dry season occupies the balance of the year. The wet season
is referred to by the inhabitants as winter and the dry season as summer.

The Republic, from an agricultural standpoint, is rich, and, exclusive of Haiti,
it is said to be the most densely populated country of the Western Hemisphere.
The great mass of the people are Indians, who live a simple life, have little, and
whose wants seem to be satisfied easily. Spanish, the State language, is spoken
everywhere.

El Salvador, although small, is well supplied with fresh-water lakes and streams.
The Rio Lempa, a comparatively large stream, lies mostly within its borders.
Lake Guija, with an area of 300 square kilometers, lies on its border, partly in El
Salvador and partly in Guatemala. It appears to be quite rough a large part of
the time, and plant growths were found only in protected coves. Soundings taken

Buuu. U. S. B. F., 1925.

Fic. 2.—Natives fishing with a cast net in Lake Hopango

Fic. 3.—View from Lake Coatepeque, showing buildings along the shore and a steep mountain side in the
background, with zigzag trail

Fic. 4.—Indian fishing in deep water from a balza on Lake Coatepeque

FISHES OF EL SALVADOR 239

offshore varied from 4.5 to 16 meters. Lake Ilopango, with an area of 70 square
kilometers, and Lake Coatepeque, with an area of 30 square kilometers, are beau-
tiful, deep, clear bodies of water having considerable vegetation along the shores,
some of the plants growing at a depth of 7 meters or more. The plants consist
largely of Chara, Ceratophyllum, and Naias, the plant growing at the greatest
depth being Naias marina. Lake Hopango has an outlet, but Lake Coatepeque
has none, and the water in the latter has a slightly salty taste, the sodium chloride
content, according to an analysis posted in a local hotel, being 0.4465 grams per
liter. The maximum depth of Lake lopango was not obtained, as the lake was too
rough at the time of our visit to make vertical soundings with the apparatus at hand.

The greatest depth found in Lake Coatepeque was 83 meters. Hook and line
fishing from the native raft, the “balza,” is carried on in this lake to a depth as
great as 30 meters, and the species taken at the greatest depth is Cichlasoma mota-
guense. In Lake Ilopango food fishes are scarcer than in any other lake visited.
This may be due to overfishing stimulated by its proximity to San Salvador, where
fishes command a fancy price.

Lake Olomega is another lake of considereble size, having an area of 20 square
kilometers. ‘This lake is shallow, having an average depth of scarcely 2 meters,
and the water is turbid. Vegetation is abundant along the shores where they are
not too rocky. The country also has quite a number of smaller lakes, among which
Metapan,! Ahuachapan, Chalchuapa, Zapotitan, and Chanmico were visited.
Among these lakes Metapan and Ahuachapan are shallow and afford some of the
best fishing found in El Salvador. Chalchuapa and Chanmico are quite deep and
food fishes are scarce, while top minnows are very abundant. Lake Zapotitan
covers considerable territory, but it can scarcely be considered more than a swamp
that is so densely overgrown with vegetation that a boat can not be used.

The Rio Lempa, the largest river in El Salvador, was visited at Suchitoto and
San Marcos, and collections were made at each place. Collections were made in two
of its tributaries also, namely, in the Rio del Desague near Lake Guija, of which it
forms the outlet, and in the Rio Sucio at Sitio del Nifio. In the Rio San Miguel, a
much smaller stream than the Lempa, a collection was made at San Miguel. In
the Rio de Paz basin collections were made in the Rio Molino near Ahuachapan
and in the Rio Pampe near Chalchuapa. One day was devoted to collecting in the
small mountain streams near San Salvador, some of which empty into Lake
Tlopango and some into the Rio Lempa.

All the streams of El Salvador that were seen are rocky, having rapids and
comparatively large, deep stretches between the rapids. Sand and gravel are found
only occasionally, and collecting with drag nets is difficult. The natives use cast
nets both in the streams and in the lakes. Hook and line fishing also is engaged in
in some localities.

The temperature of the water at the surface in the different lakes and streams
ranges from about 74 to 83° F., the average being close to 78°. The difference
between the bottom and the surface temperatures in the lakes is small. In Lake

1In the vicinity of the city of Metapan is located a group of lakes known as ‘‘ Laguna de Metapan,” which are all connected
during the rainy season but which become separated during dry weather. The collections reported in this paper were made in
the small division of the Laguna de Metapan located nearest the city of Metapan,

940 BULLETIN OF THE BUREAU OF FISHERIES

Coatepeque, for example, at a depth of 83 meters at 8.20 a. m., the temperature
was 72.5° F., while at the surface at the same time and place it was 74°; in Lake
Ilopango, at a depth of 15 meters, the temperature was 80° F., at 2.30 p. m., and
at the surface at the same time and place it was 81°; and in Lake Guija, at a depth
of 16 meters, the temperature at 11 a. m. was 74.5° F., while at the surface it
was 76.5°.

Fish, due chiefly to their scarcity, sell very high in comparison with other
foods, bringing a better price than fresh fish do in our inland markets. The fish,
furthermore, are usually small and far from the choice supply seen in our markets.
Many fish are marketed in the dry state, having been sun cured without the addi-
tion of salt or any preservative. This stock, too, finds ready sale. The species
sold chiefly in the dried state are the catfishes, locally known as ‘‘bagre’’; the top
minnows, ‘‘chimbolas’’; and the characins, ‘“sardina” and “‘plateada.”

No collecting worthy of the name had previously been done in El Salvador.
Records of only two species of fresh-water fishes, the four-eye or ‘‘ quatro-ojo”
(Anableps dovir) and the top minnow or ‘‘chimbola”’ ( Mollienesia sphenops, recorded
as Pecilia salvatoris) were found. Only a small amount of collecting likewise
appears to have been done on the Pacific slope of Guatemala, Honduras, and
Nicaragua. It is therefore not surprising that several of the species taken appar-
ently are new. ‘The Pacific slope in El Salvador, as in most of Central America,
is narrow and the rivers are short. A large number of species, therefore, is not to
be expected. Six marine species have been included with the fresh-water fishes
because they appear to be regular visitors. One species of marine catfish, locally
known as the ‘‘bagre’’ (Galeichthys guatemalensis), is one of the common food
fishes found in nearly all the fresh waters, except in the lakes having no outlet.
This fish, although to the writer’s knowledge not previously recorded from fresh
water, has become so much fresh water in its habits in El Salvador that one is led
to believe it may be able to maintain itself there. It is understood that this fish,
in order to reach Lake Olomega from the sea, would have to scale falls of consid-
erable size situated in the Rio San Miguel. None of the other marine species,
such as Mugil cephalus, Centropomus nigrescens, C. robalito, and C. pectinatus, which.
ascend the Rio Lempa far above salt water, appear to be able to ascend beyond
the falls, as none was taken by us in the Rio San Miguel above the falls or in Lake
Olomega.

The accompanying table shows where the different species of fresh-water fishes
were taken in El Salvador, whether their range extends north or south of this
country, or both, and whether they are common to both the Atlantic and Pacific
drainages or only to the Pacific. It will be seen that most of the fresh-water species
inhabit both lakes and streams. The table also shows that several species reach
their southernmost range of distribution, so far as known to date, in El Salvador,
while every species known from more southern countries also ranges northward
of El Salvador.

Buuu. U. S. B. F., 1925. (Doc. 985.)

Fig. 5.—Lake Zapotitan. Nearly the entire lake is overgrown with dense vegetation

Fra, 6,—Collecting fish in E] Salvador

Burts, Uo. Beaks 1925.

Fic. 7.—Rio San Miguel at San Miguel, during the dry season

Fic. 8—Harbor of E] Triunfo, A smoking voleano, the San Miguel, is in the background

FISHES OF EL SALVADOR 241

TaBLeE 1.—Localities in which collections were made in El Salvador and the species taken in each

{The range of the species, as to whether extending north or south of El Salvador, or both, and also whether extending into both
the Atlantic and Pacific drainages or the Pacific only, is indicated]

© > | sty
ad Tribu: |e & = » |e |Salt and
> gs ° =
Lakes 2 eats Z I aisle a betes Range
3 | de Paz ga a | 8/3 |é | water
q 28 | a |S |e
Elests a Stls,|2 (2 5 ale
tlhe Bw (ROSnu| am DO +
Species ie &5|~ 9|* (45) ,8 youn een =e aialas
lg IAD] Oo Sig Aleu|g 3 cs |. 80 ° ° ae
Hecsalecoai ne BIS Se eg8ols | sia is s|/S8jaa
al BIB) sig FQ Sig zlomle a} & 3 |o/sS
A SsisiS S| slolalie# S'e\ a'ol's 8/5 8/18 la SiG loe%
Si21alB)/8/8| 2) 2] Sls SSsiasis 8” aolole Sie >| 3 lao
A; o =| =] gq] O] w Ing As e+! = y=} So oF =
Sials/Si\Sialjsjalg]|s|> a] S314 (AJA | sl) s/S/s8/3!\a las
Bi/S|Z/S/S/S/S/Bls] slo loxisllodig |sigle |2/8/5/2/2) 818°
S|SIAIO/OlOIMS/alSl a le Sw feta |S RI |e l/Ool12Z1|8/B la l=
| |
1. Astyanax fasciatus |
Sons ee es elle Bree ee ele BS | | DS IK Sees) sees |G |G | eee eee: |bOS
2. Roeboidessalvadoris | |
SDyNOVics-<-2s-25- SG leanles Z S lh.4 es). [ae Sadie OR eee oe ee eae
3. Galeichthys guate- | |
malensis -- -_------ YS Pee ea ee eer peel Peed We. pee ee Jesea|aoa=|=ecc|oe-|| Xi) 26, |aoe2|sesuoase|| O°) XK 122 || eee
APATIUSItAVIOLsp;NoVs|- =| 2.|2--1.--|-2-|2--|_-.|£-cH 2/2223) 22 foe Se... Be el OM Saree] ome ee al a ee eee eee
5. Rhamdia guatema- | | H | }
Jensiseckiey. 228 >, aly. ea aed ara Pe eS et hal acre ah ees] eee Gl ese Wea fe ceae| bae soe ere ----| KM Wexee|sane|aanta| ow
| | | |
6. Profundulus punc- | peer
ULL OS ae ie Peers Ee fap z coetaee! Raat oye= PS es peers teers oe ed beg eee 4 reer a |
7. Mollienesia sphe- | | |
MOPSo= = 2s---s-.--- > SMS ip-4| | Posi | PeS | Pees | Palen | BS eoS oes We Setl Pe ata eet | ean ns eo (Gohl Xa | Been ee
8. Priapichthys leto- | |
nai sp. NOV--_------ DGG ee [eae |e |e WG cee eee |G eee = ne || eX |e tox Be Pret aol as ee
OE OSteDN SDs 0 Vass—|aes|ece| G22 |e 8 |Senleew eee esl |oens| een el eney see |aa aa Re aaa Ci leena|aeea| eee gees |e
LOwpeAMableps!do vile ===. |G |b s2}o- 25 [ee fea eee SK eee le aa! XX |ee-cl | XK | OX Do | | See es |e ee
11. Thyrina guija sp. |
NOW. s3-2205-he-2- x a = eto Wo llececlsesclucsclacaall, X |\ ON, ladeclecacleccel=aasl|acce! OG [ensel bene
12. Mugil cephalus- -_--|_- oe 2 [et eee | eee | eee ae |X mt SSg Geese | eee |e
13. Agonostomus mon- | |
GicOlae se non oes Ve ake || a Lee Pa bale Fe Co Uae (ge Renee ee ceed Up Kcr Leer Ceeeeeetl > ed call ce EN F<
14. Centropomus __ni- | | | |
grescens.___---____ Ee |Saed | eee acme [cena | con | oes | | eee ee ene | ie | eee | eee eee |G |e | Coe Sone | CA SCH eeee| OM ao
{
15. C. robalito_..------- Pentece|sentese|esceee (ooo) oo | sec]. safeosa|a == |eace|So-|ececls-25] 0G) 285 XK, Jecael Xo] XM lane-| Xe | See
16. C. pectinatus-_-_.... Spee amet sey A Peat ae pt Pe eas nal Ds gem ey | Semel pe eal (erie | Fee i in Ean a a oe, ais Gat al le a esd |
17. Cichlasoma nigro- |
fasciatum.__--.--- XS |X =| ae |e |p| oer (a eee [eel SE S| eed ees | See Ee Se |e
18. C. macrocanthus----}--_]--_} |---|--- ---|---|---|---|----|----|----]----|----]----]----]----|----/----]--=- Dee XS loses
FOLC. -MeCkiSH. OV so2| 6) XM ioeel Mle dt elect cae Jonas ealeee cloacal, [acne | MX leccctecds eels tal eee (ee
20%. © strimaculatum: ==2|)X |< |223|-2o}-= ee le KX cae pees |b soe oes aaa Ne (OO |S eos Sect |G eae [eae wee
Ziew® Nota UensOves = |< [a= Geel XN | Mi bea|Pea| Beal leeceleoen|ceas|eene|eeceleee.|C ecole ceslaena|eece DS eae eee pee 4
22. Gobiomorus macu- |
latus 22s. 224 22 222-2 seulesdlescleee|eee 222|se2|se.|2ou[25-4]een2 eee see ees cece IN| OG |oees aa ac|eeanl Om eae me [em are
ACKNOWLEDGMENTS

The writer wishes gratefully to acknowledge the financial as well as other
help given by the officers of the Salvadorian Government. Thanks particularly are
due to Sefor Dr. Don Marcos A. Letona, Subsecretario de Fomento y Agricultura;
Seftor Hector David Castro, chargé d’affaires of the Legation of EK] Salvador, Wash-
ington, D. C.; Mr. Frederic W. Taylor, Director General de Agricultura; Senor
José Antonio Sasso, chief clerk de Fomento y Agricultura; Seth H. Dyer; and
Sefior Carlos A. Siri, who served as secretary and interpreter throughout the
expedition. The writer also wishes to acknowledge the many courtesies and the help
given by Messrs. W. I. Mullins and J. H. Clegg, officers of the International Railway
Co. of Central America. Messrs. Max Kohlmeyer and John Bocker, business men of
San Miguel, also rendered helpful assistance. To my colleague, Fred J. Foster,
superintendent of the United States fisheries station, Neosho, Mo., who was with
the author throughout the expedition, much credit is due for the success of the work

942 BULLETIN OF THE BUREAU OF FISHERIES

and the completeness of the collection upon which the present report is based.
Irving L. Towers, scientific assistant, Bureau of Fisheries, deserves credit for much
valuable assistance rendered in the preparation of the report. The illustrations of
fishes are from photographs made by Herbert F. Prytherch, scientific assistant,
Bureau of Fisheries, and retouched by an artist. The plants mentioned in the text
were identified by Paul C. Stanley, of the National Museum.

EXPLANATORY NOTES

The usual abbreviations used in systematic ichthyology by most recent
authors has been followed. For example, the expression ‘“‘head 3.3 to 4.1; depth
4.3 to 6.2” signifies that the length of the head, measured from the tip of the upper
jaw to the bony margin of the opercle (unless otherwise specified), is contained 3.3
to 4.1 in the “standard length’’—that is, in the distance from the end of the snout
to the base of the caudal fin—and the greatest depth of the body is contained 4.3
to 6.2 times in the standard length. In giving the number of spines and rays con-
tained in a fin the spines are designated by Roman numerals and the soft rays by
Arabic numerals. For example, ‘‘D. VII-I, 14; A. III, 8” signifies that the dorsal
fins are separate and that the first consists of 7 spines and the second of 1 spine
and 14 soft rays, and that the anal is continuous, being composed of 3 spines and
8 soft rays. If, in the case of the dorsal fin, the spines and rays had all been con-
nected so as to form a single continuous fin, the result would have been written
thus: D. VIII, 14. The number of scales given in the description, unless otherwise
stated, is the number of oblique rows, running upward and backward, that occur
just above the lateral line. ‘This series is counted from the upper anterior angle
of the gill opening to the last large scale on the base of the caudal.

In the arrangement of the families, Jordan’s “A Classification of Fishes’’
(Stan. Univ. Pub., Univ. Ser. III, No. 2, 1923, pp. 79-243) was followed.

In order to render the catalogue more useful for ready identification keys to
the families, genera, and species have been introduced. Since the species occurring
in the waters of El Salvador are few, the keys are all short and no difficulty in using
them should be experienced. No attempt has been made in the keys to show the
natural relationship of the various groups. In using the keys first determine to
which of the major groups the species in hand belongs, then take up the regular
order of letters under that group. If the characters of the specimen do not agree
with those under the single letter, look under the double letter, ignoring all inter-
vening matter.

Following is a list of the species which appear as new in the present paper:

Page Page
Reeboides salvadoris__________-------- 246.) Priapichthys fosteri_.. .-___-. 2-524 260
PATI S ebay] © riley een ae epee mee te 250))| Dhiyrina, cul) as sees en 2 otek ees 264

PriapichshiyseletOM alba y= == aes 258 | Cichlasoma meeki_______.-.___-_____- 275

Part I.—DESCRIPTIVE CATALOGUE OF THE FISHES OCCURRING
IN THE FRESH WATERS OF EL SALVADOR

All of the fishes of the fresh waters of E] Salvador belong to the class Pisces
(fishes) and to the superorder Teleoster (the bony fishes), which possess a bony
skeleton, a well-developed skull, a single gill opening on each side, at least two
nostrils that are not median, and well-developed fins.

KEY TO THE FAMILIES OF FISHES OCCURRING IN THE FRESH WATERS OF EL SALVADOR ?

a. Scales wanting, body mostly covered with smooth skin; mouth and nostrils with several pairs
of barbels or whiskers.

b. Nostrils close together, neither with a barbel; palatine teeth present ______ Artidx, p. 248.
bb. Nostrils remote from each other, the posterior one without a barbel; teeth wanting on
BAU CS mee yates 2 ee ee yee oS eee Sa See ee Pimelodide, p. 251.

aa. Body scaled; no barbels about the nostrils or mouth.
c. Fins without spines.
d. Adipose fin present; head naked; lateral line present__--_----__-_- Characinide, p. 244.
dd. Adipose fin wanting; head partly scaled; lateral line wanting.
e. Eye divided into an upper and lower portion by a dark-colored membrane in the
cornea; anal fin in the male modified into a thick scaly conical copulatory organ
with an orifice at its extremity__.____._---_--------_--------- Anablepide, p. 261.
ee. Eye normal, not as above; anal fin in the male normal or produced, but not forming
a conical organ with an orifice.
f. Anal fin modified, the anterior rays more or less produced, forming an intromittent
organ, but not enveloped in skin and not hollow as in Anableps__Peciliidx, p. 254.
fee An a letimi normale Oty SeKCS ire aaa ee ee Cyprinodontide, p. 253.
cc. Fins with spines.
g- Body elongate, not very deep; dorsal fins 2, separate.
h. Ventral fins abdominal; lateral line usually absent, never complete.
i. First dorsal with 3 to 9 flexible spines; anal fin with a single weak

0) 1 OY eg et we ae Atherinide, p. 263.
wz. First dorsal with 4 rather strong stiff spines; anal with 3 strong spines
(sometimes 2 in very young)-_------------_---------- Mugilidx, p. 265.

hh. Ventral fins thoracic.
j. Lateral line complete, continued to end of caudal fin; anal with 3 spines,

the second very strong; caudal fin forked________ Centropomide, p. 268.
jj. Lateral line wanting; anal fin with a single weak spine; caudal fin
TOUNC CEs a eae] Mae ee eee eae Sa ae Eleotridx, p. 281.

gg. Body deep, compressed; lateral line interrupted under base of dorsal, reappear-
ing lower down on side; dorsal fin single, composed of spines and rays

Fe pa eA pa tre oe ee ie Ae pee ae ete ar eye eed BER el Cichlide, p. 272.

2? This key is intended to identify only the species of these families which occur in El Salvador, as some of the characters men-
tioned are not true of the families as a whole.

243

244 BULLETIN OF THE BUREAU OF FISHERIES

Glass" EISCES
Order HETEROGNATHI
Family I. CHARACINIDAZ

The Characins

Body variously shaped, usually more or less compressed; upper jaw mesially
formed by premaxillaries, laterally by maxillaries; teeth various; scales present,
usually cycloid, wanting on head; dorsal fin small, without spines; adipose fin

usually present.
KEY TO GENERA

a. Teeth in upper jaw in 2 series, none of them directed forward; anal fin of moderate length;
usually composed of less than 30 rays -__--------------------------- Astyanaz, p. 244.
aa. Teeth in upper jaw more or less definitely in 3 series, the outer series composed of strong
teeth situated on outer margin of jaw and directed forward; anal fin very long, usually
composed ‘of: 45:orimore Vays 2/2 eee oe se See ee ree Reboides, p. 246.

1. Genus ASTYANAX Baird and Girard

Astyanazr Baird and Girard, Proc., Ac. Nat. Sci., Phila., VIT 1854 (1856), 26 (type Astyanaz argentatus Baird and Girard).

Body more or less elongate, compressed, the depth usually more than 2 in length;
second suborbital narrow, leaving a naked triangular area below suture between
first and second suborbitals; premaxillaries with 2 series of teeth, the first with
several teeth on each side, the second equal or graduated, usually 10, sometimes 8
in number; lower jaw with strong teeth anteriorly, usually abruptly smaller, conical
ones on sides; maxillary with a few or no teeth; gill rakers setiform; lateral line
complete; no predorsal spine; adipose fin present; anal fin of moderate length,
usually of fewer than 30 rays; caudal fin naked.

1. Astyanax fasciatus zneus (Giinther)

PLATEADA; SARDINA

Tetragonopterus xneus Giinther, Proc., Zool. Soc. London, 1860, 319 (Oxaca, Mexico).
Astyanaz fasciatus eneus Eigenmann, Memoir., Mus. Comp. Zoél., XLIII, Part 3, 1921, 306. (A complete synonomy is given
here by Eigenmann, to which the reader is referred for further references.)

Head 3.5 to 5.13; depth 2.3 to 4.05; D. 11; A. 25 to 31; scales 33 to 38.

Body compressed, variable in depth and thickness; profile almost straight over
the head, convex from nape to dorsal; caudal peduncle strongly compressed,
broadly rounded underneath, its depth 1.85 to 2.55 in head; head small; snout
blunt, 3.25 to 4.1 in head; eye 2.75 to 4.3; interorbital 2.45 to 3.1; mouth moderate,
terminal; maxillary reaching to or slightly past anterior margin of eye, 2.5 to 3 in
head; teeth strong, mostly tri- to multicuspid, in 2 series on the premaxillaries, the
outer series with 8, the inner with 10 teeth, 2 teeth on each maxillary, 8 or 9 large
teeth anteriorly on the lower jaw and abruptly smaller ones laterally; gill rakers
short, 11 to 14 on the lower limb of the first arch; lateral line complete, slightly
decurved; scales rather large, 10 to 12 rows crossing the back in front of dorsal, 10

FISHES OF EL SALVADOR 245

or 11 between the dorsal and adipose, 6 or 7 from origin of dorsal to lateral line;
dorsal elevated anteriorly, its origin equidistant from tip of snout and base of caudal
or more usually a little nearer the former; adipose over the posterior rays of the
anal; caudal fin broadly forked; anal fin rather long, its origin at or slightly behind
vertical from end of dorsal base, usually about equidistant from margin of opercle
and base of caudal; ventral fins moderate, inserted a little in advance of dorsal;
pectorals short, 1.2 to 1.35 in head.

Color in life silvery with bluish green above; sides with an ill-defined bright
silvery streak, most evident posteriorly; a dark shoulder spot, and frequently an
indication of a second black spot behind it; dorsal olivaceous; caudal greenish,
with an elongate black spot at its base extending to the end of the median rays;
anal pink to deep red; ventrals usually more or less reddish; pectorals yellowish to
red; upper part of iris red to reddish yellow.

Many specimens of this species, ranging in length from 20 to 135 millimeters,
were preserved. This fish is very common in some localities but absent in others.
It is especially abundant in Lake Olomega, where large quantities are taken and
used for food. It was not taken in Lakes Ahuachapan, Chalchuapa, Coatepeque,
Chanmico, and Ilopango. [Especiallylarge and fine specimens were taken in the Rio
Pampe, a tributary of the Rio de Paz, near Chalchuapa. The individuals taken
here are somewhat more slender, and they have a less distinct lateral band than the
fish taken elsewhere. ‘This fish mhabits the deeper parts of rocky and swiftly-
flowing streams as well as quiet and shallow water.

The alimentary canal in this species is shorter than the body, and the peritoneum
is black. The food found in 5 stomachs examined consisted of water fleas, copepods,
water beetles, and fragments of fish. The sexual organs in the specimens dissected
were collapsed, showing that spawning was not taking place during the period
(January and February) when the specimens were collected.

A. fasciatus is anextremely variable species and many forms have been described.
The specimens from El Salvador appear to be nearest the form which Kigenmann,
in his monumental work, ‘‘ The American Characide,”’ has designated as A. fasciatus
zneus. Itwas pointed out in the preceding paragraph, however, that the specimens
from the Rio de Paz basin differ somewhat from those taken elsewhere in El Salvador.
Eigenmann, in speaking of the many variations, says: ‘‘ Whether we call these forms
species, varieties, or do not recognize them as worthy of name, the fact remains that
different rivers are inhabited by individuals that in the aggregate differ from the
individuals of another river, * * * that we have here a series of species in the
making as the result of segregation.”” It would appear that at last two species are
“in the making” in El Salvador.

This form, according to Eigenmann, ranges from Mexico to Panama. The
specimens from El] Salvador are from the Rio Pampe at Chalchuapa, Lake Guija,
Lake Metapan, Lake Zapotitan, Rio Sucio at Sitio del Nitto, Rio Lempa at Suchitoto
and San Marcos, Rio San Miguel at San Miguel, and Lake Olomega.

246 BULLETIN OF THE BUREAU OF FISHERIES

2. Genus RO@BOIDES Gunther

Reboides Giinther, Cat. Fish., Brit. Mus., V, 1864, 345 (type Anacyrtus guatemalensis Giinther).

Body compressed; dorsal profile concave at nape in adult; teeth mostly pointed,
in 3 more or less definite series in upper jaw, the outer ones of both jaws directed
forward; shoulder girdle with a large spine; lateral line straight, complete; anal
fin very long, with 45 or more rays; adipose fin well developed.

2. Roeboides salvadoris sp. nov.
PLATEADO; SARDINA; ALMA SecA; ULUMINA

Type no. 87215, U.S.N.M. Length, 91mm. Rio Sucio, Sitio del Nifio.

Head 3.35 to 4.25; depth 2.55 to 3.4; D. 11; A. 46 to 50; scales 15 or 16—72 to
82; pores in lateral line variable, 53 to 73.

Body strongly compressed; the dorsal region elevated; profile deeply con-
cave at occiput in the adult, not at all concave in young of 40 millimeters and
less in length; head comparatively small; snout blunt, 3.25 to 3.9 in head; eye small,
3.2 to 3.8 in head or 11.5 to 15 in body; interorbital 3.2 to 3.8 in head; mouth large;
maxillary notably broader than preorbital, reaching anterior margin of pupil to
middle of eye, 2.1 to 2.5 in head; teeth moderate, 4 with broad conical bases,
directed forward on margin of snout, 2 to 4 similar ones at sides on the maxillary
and 2 on the margin of mandible; outer margin of maxillary with pointed teeth;
premaxillary with irregular teeth, irregularly placed, some of the inner ones with
small basal cusps; mandibular teeth all pointed, the anterior ones the largest, none
of them definitely tricuspid; gill rankers short, 8 or 9 on the lower limb of the
first arch; shoulder girdle with a spine, pointed both anteriorly and posteriorly,
reaching nearly to base of pectoral; lateral line complete, nearly straight; scales
very small; dorsal fin moderately elevated anteriorly, its origin equidistant from
tip of snout and base of caudal or more usually a little nearer the former; the
elevated portion measured from the origin of the fin to the tip of the longest rays
3.45 to 4.3 in body; adipose fin moderate, situated over the posterior part of anal;
caudal fin deeply forked; anal fin very long, its origin equidistant from tip of snout
and base of last anal ray or more usually a little nearer the former; ventral fins
rather short, reaching opposite the fourth to the sixth anal ray, 5.25 to 6.4 in body;
pectorals moderate, failing to reach the origin of anal, 1.3 to 1.45 in head or 4.6
to 5.7 in body; vertebrae 12+22.

Color in life silvery, slightly olivaceous on back; sides with a silvery band,
most definite posteriorly; base of caudal with dark punctulations on a silvery
background, forming a dark caudal spot; a smaller dark spot just above the lateral
line and posterior to the vertical from base of ventrals, deeper than long, and
variable in intensity, but not entirely wanting on any of the specimens at hand;
upper parts of head dark; fins all greenish.

Many specimens of this species, ranging from 35 to 105 millimeters in length,
were preserved. This fish is not universally distributed in the waters of El Salvador.
It is abundant among vegetation in Lake Guija, apparently rather rare in Lake

FISHES OF EL SALVADOR 247

Metapan, and common in the Rio Sucio at Sitio del Nifio and in the Rio Lempa at
Suchitoto. Elsewhere it was not seen.

The sexual organs in the fishes examined were in a collapsed condition, showing
that spawning was not taking place during the period (January and February)
when the specimens were collected. The air bladder is large and the alimentary
canal is short. The contents of 4 stomachs examined consisted of fragments of
insects and the scales of fish, which apparently remained after the rest of the fish
had been digested.

The specimens from E] Salvador evidently represent a species distinct from those
of Panama. Material from other Central American countries is not available to
the author for comparison. Giinther, the discoverer of R. guatemalensis, considered
his specimens from the Pacific slope of Guatemala identical with others obtained in
the Rio Chagresin Panama. Authors, generally, had recognized but a single species
from Central America. Meek and Hildebrand (Pub., Field Mus. Nat. Hist., Zool.
Ser., X, 1916, pp. 291-3), however, found that the specimens from the Pacific slope
of Panama were different from those of the Atlantic slope, and they described the
former as new, giving it the name occidentalis. The Atlantic specimens were con-
sidered as R. guatemalensis of Ginther, which is probably correct, for Ginther,
when describing the species, had some young from the Pacific slope of Guatemala
and adult specimens from the Rio Chagres, Panama. Giinther’s description appears
to have been based upon the adult fish, for he says: “‘ Back elevated, the upper profile
of the head and nape forming an S-shaped curve.’”’ This appears never to be true
of the young of the genus. The Atlantic representatives from Panama, therefore,
may for the present, at least, be considered the true guatemalensis, regardless of
whether or not they are identical with specimens from Guatemala, which seems
highly doubtful.

The El Salvador specimens differ from both the Panama species with which a
direct comparison of specimens has been made, in having a smaller eye, shorter
fins, in color markings, and in other minor respects. The number of scales in a
lateral series appears to be nearly identical in the present species and in R. occiden-
talis. 'The number of gill rakers, however, is more nearly that of R. guatemalensis.
In the position of the dorsal and anal the El Salvador specimens appear to be
intermediate.

The diameter of the eye in the body in 12 specimens of the present species
varying from 60 to 82 millimeters in standard length, ranges from 11.8 to 13.5,
the average being 13.35. In an identical series of guatemalensis the range is from
9.4 to 13.1, average 11.63; in occidentalis the range is 10.03 to 13.3 and the average
11.64. The length of the fins in the body for 20 specimens from E] Salvador, rang-
ing in standard length from 28 to 86 millimeters, are as follows: Pectoral, 4.5 to
5.8, average 5.32; ventral 5.1 to 6.43, average 5.81; dorsal 3.44 to 4.3, average 3.94.
The results for a similar series of R. guatemalensis are as follows: Pectoral 4.55 to
5.55, average 4.92; ventral 4.6 to 5.8, average 5.21; dorsal 3.2 to 3.66, average
3.51. For occidentalis, also using a like series of specimens, the following results
were obtained: Pectoral 4.2 to 4.94, average 4.63; ventral 4.03 to 5.05, average
4.68; dorsal 3.1 to 3.5, average 3.32. The Salvador specimens, therefore, represent

948 BULLETIN OF THE BUREAU OF FISHERIES

what Dr. C. H. Eigenmann, in several of his recent works, has called a “statistical
species.’

In color the El Salvador specimens differ from R. guatemalensis in having no
black in the lateral band, and in having a more or less distinct dark spot on the
sides, which, however, is much smaller than the one in occidentalis.

Fowler (Proc., Ac. Nat.Sci., Phila., LX XV, 1923 (1924), p.25) recently described
Reboides bouchellei from Great Falls, Pis Pis River, of the Atlantic slope of Nicara-
gua. According to the description, this species would appear to be close to the
present one. It appears to differ as follows, however: (1) In the shorter maxillary,
which is described as reaching only to the eye, while in the Salvador fish it reaches
well beyond this point. (2) The preorbital, in comparison with the maxillary,
is broader, being described as equal in width to the maxillary. In the specimens at
hand it is always notably narrower. (3) Fowler counts 52 rays in the anal fin of
his type. The largest number found in 25 specimens from E] Salvador is 50. It is
possible, therefore, that at least the average number for Nicaraguan specimens is
higher. (4) The pectorals in #. boucheller are described as reaching the anal and
to be contained 1.66 in head. In the specimens at hand they do not reach the origin
of the anal, yet they are contained only 1.3 to 1.45 in head. (5) The dark spot on
the side situated just above the lateral line is described for R. bouchellet as being
‘“‘midway in the predorsal length.’’ In the El Salvador specimens it is only about
an eye’s diameter in advance of origin of dorsal. (6) The bright silvery lateral
band, very evident in all alcoholic specimens from E] Salvador, is not mentioned for
R. boucheller.

The specimens from E] Salvador were collected in lakes Guija, Metapan, and
in the Rio Sucio at Sitio del Nino, and in the Rio Lempa at Suchitoto.

Order NEMATOGNATHI
Family II. ARIIDAE
The Sea Catfishes

Body elongate; head broad, depressed; nostrils close together, neither with a
barbel, the posterior with a valve; palatine teeth present; skin naked; dorsal fin
present, short, situated above or in front of ventrals; adipose fin present; caudal

fin lunate or forked.
KEY TO THE GENERA

a. Teeth all pointed; present on the jaws, vomer, and palatines___-------- Galeichthys, p. 248.
aa. Teeth wanting on the vomer; those on the palatines broad and blunt__---_-- Arius, p. 250.

3. Genus GALEICHTHYS Cuvier and Valenciennes

Galeichthys Cuvier and Valenciennes, Hist. Nat. Poiss., XV, 1840, 28, (type Galeichthys feliceps Cuvier and Valenciennes).

This genus may be recognized by the presence of pointed teeth on the jaws,
vomer, and palatines. The palatine teeth are in small or moderate patches and
do not have a backward extension. A single species belonging to this saltwater
genus is present in most fresh waters of EK] Salvador.

Bou. U. 8. B. F., 1925. (Doc. 9

Fic. 9.—Roeboides salvadoris sp. nov. From the type. Length, 91 millimeters

SLOPOULTTTIUT OVE ‘YQSueT ‘odA} oy} wool ‘Aou ‘ds 210jfid) sniup—'Ol ‘DI

(E86 90d) “SZ6I “A “A ‘8S “N “Ting

FISHES OF EL SALVADOR 249

3. Galeichthys guatemalensis (Ginther)

BAGRE

Arius guatemalensis Giinther, Cat. Fish. Brit. Mus., V, 1864, 145 (Guatemala).

Tachisurus guatemalensis Kigenmann and Eigenmann, Proc., Cal. Ac. Sci., 2ser., I, 1888, 43, and Occ. Pap., Cal. Ac. Sci., 1890, 81.

Galeichthys guatemalensis Jordan and Evermann, Bull., U. 8. Nat. Mus., XLVII, 1898, 2778; Regan, Biol. Cent. Amer., Pisces,
1907, 123.

Head 3.3 to 4.1; depth 4.3 to 6.2; D. I, 7; A. 17 to 19.

Body about as broad as deep at origin of dorsal, moderately compressed pos-
teriorly; caudal peduncle rather long, its depth 3.65 to 4.8 in head; head broad,
depressed, flat above; snout very broad, its length 2.75 to 3.7 in head; eye 4.5 to 8;
interorbital 1.8 to 2.3; mouth very broad; upper jaw strongly projecting; teeth
pointed, the band in upper jaw continuous, the one in lower jaw separated at
symphysis, extending laterally well past the angle of mouth and ending in a sharp
point; vomerine patches well separated from each other and from the larger palatine
patch by a line and slight constriction; maxillary barbel scarcely reaching margin
of opercle in adult, past base of pectoral in small specimens (150 millimeters and
less in length); gill rakers, 11; distance from snout to dorsal 2.4 to 3 in length;
upper surface of head granular in adult, almost smooth in young, occipital process
a little longer than broad; fontanel scarcely produced in a groove, appearing as an
elongate pit slightly behind a straight line drawn between the posterior margin of
the eyes; dorsal fin small, its origin usually about equidistant from tip of snout
and origin of adipose, the spine about three-fourths the length of the longest rays,
1.7 to 2.1 in head; adipose fin small, its base 3.1 to 4 in head; caudal fin deeply
forked, the upper lobe the longer, pointed; anal fin moderate, its origin under or a
little posterior to that of the adipose, its base 1.75 to 2.8 in head; ventral fins shorter
than pectoral, thickened, and somewhat longer in the female than in the male,
usually inserted a little nearer end of anal base than base of pectoral spine; pectoral
spine with barbs on the inner edge, 1.6 to 1.8 in head.

Color, bluish dusky above; sides silvery; pale underneath; vertical fins all
dusky, the anal sometimes partly pale; paired fins black on inner sides.

Thirteen specimens, ranging from 88 to 395 millimeters in length, were pre-
served. The largest individual taken measured 560 millimeters in length, which,
according to local fishermen, is about the maximum size attained. Measurements
and proportions were obtained from this fish and are included in the description.
This fish, which is known as “‘bagre” throughout the Republic, although belonging
to a salt-water genus of catfishes, appears to have become fresh water in its habits,
at least in E] Salvador, where it is one of the important commercial species caught
in the rivers and in some of the lakes. It was not taken in any of the lakes having
no outlet (unless the statement that Lake Metapan has an underground connection
with Lake Guija should prove to be erroneous), and consequently it does not appear
to have become definitely landlocked, the distribution apparently being such that
it can retreat to the sea at will. Falls of considerable size are present in the Rio
San Miguel, which receives the water from Lake Olomega and carries it to the sea,
and no other marine species were seen in the lake nor in that river above the falls.
It may be, however, that this catfish is able to ascend the falls and that it reaches

250 BULLETIN OF THE BUREAU OF FISHERIES

the lake from the sea. The species is said to be nocturnal in its habits, hiding
among rocks or elsewhere during the day and coming out at night to feed. It is
more abundant in Lake Olomega than in any other locality where collections
were made. It does not appear to be esteemed there as a food fish, however, for
the native fishermen did not care to save the fish not desired for specimens. Large
numbers, however, are sun-dried and shipped. Several bundles, weighing 100
pounds or more, inclosed in nets, were seen at the local railway station.

Three large females taken in the Rio Lempa at Suchitoto on February 5 were
examined and found to contain eggs provided with large, red, blood vessels and
measuring about 10 millimeters in diameter. The eggs were still attached to the
ovary by prominent stalks, and since it is known that marine catfishes produce very
large eggs it is improbable that the spawning period was immediately at hand,
although it might be expected to occur within a month or so. Two large specimens
examined had fed on crabs and crawfish, while two smaller ones had eaten fish,
snails, insects, and insect larvee.

This species ranges from Mazatlan, Mexico, to Panama. The specimens at
hand were taken in Lake Guija, Lake Metapan, Rio Lempa at Suchitoto and San
Marcos, and in Lake Olomega.

4. Genus ARIUS Cuvier and Valenciennes

Arius Cuvier and Valenciennes, Hist. Nat. Poiss., XV, 1840, 53, (type Pimelodus arius Hamilton).

This genus of catfishes is characterized by the absence of teeth on the vomer
and by the broad blunt teeth on the palatines, which are in small or moderate-sized
patches and without backward projections. A single apparently undescribed
species of this salt-water genus was taken in strictly fresh water. It is therefore
included here with the fresh-water species.

4. Arius taylori sp. nov.
BaGRE

Type No. 87224, U.S.N.M.; length 360 mm.; Rio Lempa, San Marcos.

Head 3.8; depth 4.6; D. I, 7; A. 23.

Body moderately elongate, somewhat deeper than broad at origin of dorsal,
posteriorly compressed; caudal peduncle rather slender, its depth 3.25 in head;
head not very broad, flat above; snout moderate, 3.25 in head; eye large, lateral,
5.5; interorbital 2.25; mouth not excessively broad, its width scarcely equal to
interorbital space; teeth in the upper jaw villiform, divided by a narrow line, not
pointed posteriorly and not reaching angle of mouth, each half about twice as long
as broad; teeth on lower jaw all pointed except the posterior one at middle of jaw,
these blunt like the palatine teeth, the band well separated at symphysis, tapering
and pointed posteriorly, and extending well past angle of mouth, its greatest width
less than one-third the length of each patch; vomerine teeth wanting; palatine
teeth blunt, in small patches, placed far apart, the distance between them about
equal to half the length of the band on upper jaw; maxillary barbel reaching to
margin of opercle; gill rakers 11; distance from snout to dorsal 2.7 in length; top

FISHES OF EL SALVADOR 951

of head only feebly granular; occipital process scarcely as broad as long, scarcely
keeled; an elongate pit on snout and a fontanel groove reaching from posterior part
of interorbital nearly to occipital process; dorsal fin rather small, its origin about
equidistant from tip of snout and origin of adipose, the spine 1.6 in head; adipose
fin rather large, its base 3.1 in head; caudal fin long, deeply forked, both lobes pointed,
the posterior margin definitely V shaped, the upper lobe the longest, a little longer
than head; anal fin moderate, its origin about an eye’s diameter in advance of the
adipose, the length of its base 1.55 in head; ventral fins rather short, failing to reach
the origin of anal, inserted about equidistant from base of pectoral spine and middle
of anal base; pectoral spine short, 1.6 in head.

Color in alcohol dark brownish above, sides silvery, pale underneath, the
fins all slightly dusky.

A single specimen, 360 millimeters long, was preserved. The species was not
distinguished from G. guatemalensis in the field, therefore others may have been
captured but not preserved.

The species appears to be related to A. furthw, known from Panama. The
El Salvador specimen, however, has a much longer and more deeply forked caudal
fin, the lobes being pointed and equal to or longer than the head, and the outer
margin is definitely V shaped. In A. furthw the lobes of the caudal are not pointed
and are notably shorter than the head, and the outer margin of the fin is broadly
Vshaped. The eye in the El Salvador fish is larger, the fins are longer, the palatine
patches of teeth are much smaller and farther apart, and the band of teeth on the
lower jaw is narrower.

The specimen in hand was taken in strictly fresh water at San Marcos, where
the bridge of the International Railroad of Central America crosses the Rio Lempa,
about 30 kilometers from the sea. The species is named for Frederic W. Taylor,
director general of agriculture for the Government of El Salvador, under whose
immediate direction the investigation was made.

Family III]. PIMELODIDA

Body elongate, compressed posteriorly ; head broad; mouth terminal or slightly
inferior; barbels, 6; nostrils remote from each other, the posterior one without a
barbel; teeth wanting on palatines; adipose fin present, usually long.

5. Genus RHAMDIA Bleeker

Pteronotus Swainson, Nat. Hist. Class. Fish., II, 1839, 309 (type Heterobranchus sextentaculatus Agassiz).

Rhamdia Bleeker, Verhand. Natuurk. Vereen. Nederl. Indie, IV, 1859, 197 (sp.); Bleecker, Nederl. Tijdschr. Dierk., I, 1863, 101
(type Pimelodus quelen Quoy and Gaimard).

Pimelenotus Gill, Ann. Lyc. Nat. Hist., N. Y., VI, 1858, 391 (type Pimelenotus vilsoni Gill).

Body elongate; jaws with villiform teeth; no teeth on vomer or palatines;
nostrils remote from each other; barbels, 6; no nasal barbel; occipital process small
or wanting, never reaching dorsal plate; eye with free orbital margin; dorsal fin
with 1 slender spine and 5 to 8 branched rays; adipose fin long, adnate to the back.
A single species of this genus was taken in El Salvador.

42885—257 2

D5) BULLETIN OF THE BUREAU OF FISHERIES

5. Rhamdia guatemalensis (Giinther)

FILIN

Pimelodus guatemalensis Giinther, Cat. Fish. Brit. Mus., V, 1864, 122 (‘‘Huamuchal,”’ Pacific slope of Guatemala).
Rhamdia wagneri Jordan and Evermann, Bull., U. S. Nat. Mus., XLVII, 1896, 152; Regan, Biol. Cent. Amer, Pisces, 1907, 132.

Head 3.65 to 4.25; depth 4.4 to 6.25; D. I, 6; A. 11 to 13.

Body elongate, compressed posteriorly; caudal peduncle strongly compressed,
its depth 2 to 2.45 in head; head depressed, its depth about three-fourths its width;
snout broad, its length 2.55 to 2.8 in head; eye 5.3 to 6.2; interorbital 2.95 to 3.6;
mouth broad, almost wholly transverse, its width equal to length of snout and half
of eye; teeth small, in rather broad villiform bands in each jaw; maxillary barbel
variable in length, apparently always reaching past the origin of the adipose and
occasionally nearly to the end of this fin; dorsal fin rather small, round, the spine
about three-fourths the length of the longest rays, 2.4 to 3.3 in head, origin of fin
equidistant from tip of snout and vertical from origin of anal or a little nearer
the latter; adipose fin long, its base 2.4 to 2.7 in body; caudal fin deeply forked,
the median rays scarcely half the length of the longest; anal fin with convex mar-
gin, its origin a little anterior to middle of base of adipose; ventral fins short, in-
serted at vertical from end of dorsal base; pectorals small, the spine strong, without
definite barbs on the inner margin, its length 2.2 to 3 in head.

Color in life plain dark green, usually with a dark lateral band; dorsal, adipose,
caudal, and anal dusky; the other fins paler.

This fish is represented by 9 specimens ranging from 90 to 185 millimeters
in length. The species does not appear to be abundant enough anywhere to be of
much commercial importance. The few individuals that were seen were quite
small, none of them exceeding 185 millimeters in length. This catfish inhabits
both lakes and streams, and it is said to be nocturnal in its habits, hiding during
the day and coming out to feed principally at night. It is taken chiefly with hook
and line.

The contents of the stomachs examined consisted of the remains of fish, frag-
ments of insects, and strands of algz. The ovary of a fish 185 millimeters long,
taken in Lake Guija on January 25, was fairly well developed. Its length was
48 millimeters, and it contained several thousand eggs approximately 0.5 milli-
meters in diameter. It seems probable that this fish would have spawned within
about one month.

This catfish is recorded from western Guatemala, and according to Regan it
also occurs in southern Mexico and in British Honduras. The El Salvador speci-
mens are from Lake Guija, Rio del Desague, Lake Metapan, Rio Sucio at Sitio
del Nifio, Rio Lempa at Suchitoto, and Lake Olomega. It was not seen elsewhere.

FISHES OF EL SALVADOR 253

Order CYPRINODONTES
Family IV. CYPRINODONTIDA
The Killifishes

Body elongate, compressed posteriorly; head more or less flattened above;
mouth small; premaxillaries strongly protractile; teeth present on both jaws;
lateral line wanting; dorsal fin single; anal fin not modified in male; species ovi-
parous.

6. Genus PROFUNDULUS Hubbs

Profundulus Hubbs, Misc. Pub., Mus. Zodl., Univ. Mich., No. 18, 1924, 12 (type Fundulus punctatus Giinther).

Body rather.robust; head somewhat depressed; dorsal and anal fins relatively
long and low and inserted far posteriorly; anal fin lower in the adult male than
in the adult female; the oviduct not produced and not extending on the first ray
of the anal fin, as in Fundulus. A single species was taken in El Salvador.

6. Profundulus punctatus (Giinther)

CHIMBOLA

Fundulus punctatus Giinther, Cat. Fish., Brit. Mus., VI, 1866, 320,and Trans., Zoél. Soe., London, VI, 1868, 482, Pl. LX XXIV,
fig. 5 (Chiapas, Guatemala); Regan, Biol. Cent. Amer., Pisces, 1907, 78.

Fundulus guatemalensis Giinther, Cat. Fish., Brit. Mus., VI, 1866, 321, and Proc., Zod!. Soe., London, VI, 1868, 482, Pl. LX XXIV,
figs. 3 and 4 (Lakes Duenas and Amatlan, and Rio Guacalate).

Head 3 to 3.7; depth 2.9 to 3.8; D. 12 or 13; A. 13 or 14; scales 31 to 33.

Body rather robust, compressed; caudal peduncle strongly compressed, its
depth 1.75 to 2.15 in head; head somewhat depressed; snout broad, 3 to 3.5 in
head; interorbital 2 to 2.4; mouth transverse; the lower jaw slightly projecting;
teeth in bands, pointed, the outer ones enlarged; scales firm, cycloid, 10 or 11
rows between the dorsal and anal, many of the scales along the upper part of sides
with pits; dorsal small, its origin usually about equidistant from the posterior
margin of the eye and the end of the caudal; caudal fin broadly rounded; anal
fin somewhat longer than the dorsal, its origin a little behind that of the dorsal;
ventral fins small, about as long as snout and half the eye, reaching nearly or quite
to the vent; pectoral fins broad, 1.45 to 1.75 in head.

Color brownish green above, pale underneath; a dark blotch above and be-
hind base of pectoral; most specimens with a dark lateral band posteriorly, this
band wanting in some specimens and broken up into spots in others; occasionally
with a series of pale spots above and below the lateral band; a dark vertebral
band; the scales on upper part of sides and on base of caudal frequently with
dark spots; fins mostly yellowish in life; the dorsal and caudal dusky in spirits,
the former with indications of a dark bar at base; the other fins usually all pale,
the anal occasionally slightly dusky and with a white margin.

This species is represented by 26 specimens, ranging from 25 to 77 millimeters
in length. It was taken only in Rio Molino, a rather small tributary to the Rio de
Paz. This stream was visited late on the evening of January 27, and only a small
stretch, in the vicinity of Ahuachapan, was seen. That portion was too rocky to

254 BULLETIN OF THE BUREAU OF FISHERIES

permit of the operation of a drag net. Three charges of dynamite produced only
the species here described and, according to a native who was questioned, the
stream in that vicinity contains no other fish.

One of the large females examined had large roe, the ovary containing 30
eggs measuring 2.25 millimeters in diameter and an equal number approximately
half that size. The food in the 3 stomachs that were examined consisted mainly of
unicellular alge, but fragments of worms and winged insects also were present.

This fish is known from the Pacific slope of southern Mexico and Guatemala,
and it is now for the first time recorded from El Salvador. According to Giinther
(1866) and subsequent authors this species also occurs in ‘‘western Ecuador,”’
but this quite evidently is a mistake, for Meek (Pub., Field Mus. Nat. Hist., Zodl.
Ser., X, 1914) did not get it in Costa Rica; Meek and Hildebrand (Pub., Field
Mus. Nat. Hist., Zodl. Ser., X, 1916) did not take it in Panama; and Eigenmann
(Memoir., Carnegie Mus., [X, 1922) does not record it in ‘‘ The Fishes of Western
South America.’ The specimens from El Salvador were taken in the Rio Molino,
tributary to the Rio de Paz, near Ahuachapan. It was not found in the basins
of the Rio Lempa and the Rio San Miguel, indicating that the species may reach
the southermost limits of its distribution in the basin of the Rio de Paz.

Family V. PQ@:CILIIDAE
The Top Minnows

Body elongate, compressed posteriorly; head more or less depressed above;
anal fin in the male modified, the anterior rays becoming more or less united and
usually produced, forming an organ for the transmission of the sperms to the genital
opening of the female; species small; viviparous.

Many genera belonging to this family have been described by Regan, Eigen-
mann, Henn, and recently by Hubbs, based mainly upon the minute structure of
the modified anal fins of the males. The number of genera that have been proposed,
particularly in consideration of Hubbs’s recent additions (Misc. Pub., Mus. Zodl.,
Univ. Mich., No. 18, 1924, pp. 5-11), is rapidly approaching the total number of
species recognized. Geiser (Amer. Midl. Nat., VIII, 1923, 175-188, with 18 figs.),
in his studies of the minute structure of the modified anal fins (“‘ Gonopod’’) of the
Gambusia of the United States, which, by most recent authors, have been considered
all identical as to species, has shown that three and possibly four divisions may be
made. If the microscopic structure of the intromittent organ is used as a generic
character, each species, as described and figured by Geiser (it would appear from the
work of Hubbs, at least), should also constitute a genus. The use of the structure
of this organ alone, as a generic character, obviously results in too many divisions,
and the genus loses its value as a convenience in classification.

The author is of the opinion that the intromittent organs could be grouped as
to general type or gross structure and thus serve a useful purpose, in combination
with other characters, in defining genera; as, for example, in Mollienesia the modified
portion of the fin is not greatly produced and it has a membranous covering (some-
times referred to as a hood), which covers or shields the anterior part of the organ,

FISHES OF EL SALVADOR 255

but which is distally free. In Gambusia, on the other hand, the intromittent organ
is notably produced and the membranous hood is wanting. The characters of the
intromittent organ in these genera are supported by the comparatively long, con-
voluted intestine in Mollienesia, while Gambusia has a short intestine, which is not
convoluted. The minute structure of the intromittent organ would then become
of specific value only, and it is believed that less confusion would result than by
retaining ‘and adding to the already too numerous very closely related ‘‘ genera.”’

Some of the species of this family—as, for example, the Gambusia of the United
States—have been found to be of great value as eradicators of mosquito larve.
Limited observations made in the field, as well as the examination of stomach
contents, indicate that all the species of this family occurring in El Salvador are of
value as destroyers of mosquitoes.

KEY TO THE GENERA

a. Intestinal canal longer than the total length of the body, usually more or less convoluted;
dorsal fin in adult male very high; the intromittent organ not much produced, shorter than
head, anteriorly shielded by a short membranous hood_______________- Mollienesia, p. 255.

aa. Intestinal canal not longer than the length of body and not convoluted; dorsal fin in males
not much, if any higher than in females; the intromittent organ greatly produced, longer
than head, and without a membranous hood_-__--__-_---_----------- Priapichthys, p. 258.

7. Genus MOLLIENESIA Le Sueur

Mollienesia Le Sueur, Jour., Ac. Nat. Sci., Phila., II, 1821, 3 (type Mollienesia latipinna Le Sueur).

Body elongate, rather robust; head more or less depressed; mouth transverse;
teeth in the jaws in bands, the outer ones movable; dorsal fin elongate in the adult
male, much higher than in the female; intromittent organ short, not as long as
head, anteriorly shielded by a membranous hood; intestinal canal convoluted and
longer than the total length of the fish. A single variable and rather widely dis-
tributed species was taken in the waters of E] Salvador.

7. Mollienesia sphenops (Cuvier and Valenciennes)

CHIMBOLA

Pacilia sphenops Cuvier and Valenciennes, Hist. Nat. Poiss,, XVIII, 1846, 130, Pl. DX XVI, fig. 2 (Vera Cruz, Mexico); Regan,
Biol. Cent. Amer., Pisces, 1907, 102, Pl. XIII, figs. 1-7.
Xiphophorus gilli Kner and Steindachner, Abhandl. k. Bayer, Ak. Wiss. Miinchen, X, 1864, 25, Pl. IV, fig. 1 (Rio Chagres,

Panama).
Platypecilus mentalis Gill, Proc., Ac. Nat. Sci., Phila., 1876, 335 (Panama).
Pecilia boucardi Steindachner, Sitzb. k. Ak. Wiss., Wien, LX XVI, 1878, 386, Pl. III, fig. 263 (Colon, Panama).
Pecilia salvatoris Regan, Ann. and Mag. Nat. Hist., 7 ser., XIX, 1907, 65, and Biol. Cent. Amer., Pisces, 1907, 104, Pl. XIV,

figs. 2 and 3 (‘‘Salvador’’).
Platypecilus tropicus Meek, Pub., Field Col. Mus., Zodl. Ser., VII, 1907, 146 (Turrialba, Costa Rica).
Pecilia tenuis Meek, Pub., Field Col. Mus., Zodl. Ser., VII, 1907, 147 (Tiribi; San Jose and Rio Maria Aguilar, Costa Rica).
Mollienesia sphenops tropica Meek, Pub., Field Mus. Nat. Hist., Zo6l. Ser., X, 1914, 116 (Costa Rica).

Head 8 to 4.25; depth 2.75 to 3.65; D. 9 to 11; A. 8 to 10; scales 25 to 29.

Body rather robust, variable, compressed; caudal peduncle strongly compressed,
its depth 1.1 to 1.85 in head; head depressed, flat above, broader than deep over
middle of eyes; snout broad, its length 2.8 to 3.6 in head; eye 3 to 4.4; interorbital
1.65 to 2.35; mouth transverse, directed slightly upward; lower jaw a little the
longer; teeth in bands, the outer ones movable, enlarged, slightly broadened, curved

256 BULLETIN OF THE BUREAU OF FISHERIES

inward, and well separated from the inner, minute, villiform ones; scales cycloid,
enlarged ones present on head and snout, 11 or 12 in median series from oeciput to
dorsal, 8 complete longitudinal rows between dorsal and anal; dorsal fin with con-
vex margin, occasionally nearly straight, the origin in females and young a little
in advance of the anal and a little nearer the base of caudal than posterior margin
of eye; dorsal much higher in the adult male and situated slightly farther forward,
quite as near the eye as the base of caudal; caudul fin strongly convex‘ to nearly
straight; anal fin small, its origin in female slightly nearer base of caudal than eye,
farther forward in adult males, usually an eye’s diameter nearer tip of snout than
base of caudal, and modified into a copulatory organ, but not greatly produced,
always shorter than head, the produced portion anteriorly with a sort of membra-
nous hood, which is free distally, the third, fourth, and fifth rays produced, 2
branches each of the anterior and median produced rays reaching the end of the
organ, the branches of the anterior ray with small antrorse hooks at tip, the pos-
terior branch of the median produced ray with 8 or 9 prominent spurlike hooks
below the apex of the organ, the posterior produced ray not quite reaching the
apex, both branches coterminal, each bearing a small posterior hook; ventral fins
in the female moderate, inserted about equidistant from margin of opercle and end
of anal base, reaching vent in large examples, to origin of anal in small ones, the
third ray somewhat produced in the male and reaching beyond the base of fin;
pectoral fins moderate, 1 to 1.8 in head.

Color variable, fresh females 54 millimeters long, dark olivaceous above; lower
part of sides silvery with bluish reflections; pale underneath; scales on sides with
indications of rusty spots; dorsal red with black spots near base; caudal greenish
with faint elongate dark spots; anal and ventrals plain; pectorals slightly greenish.
Color of male, 37 millimeters long, identical except for more red and larger and
more pronounced black spots on dorsal fin. Some specimens plain without spots;
others with prominent dark spots along the rows of scales on median part of sides;
still others with pronounced brick-red spots (pale in spirits) along the rows of
scales on sides; an occasional specimen with irregular dark blotches on sides;
many specimens with plain to faint pale crossbars; a dark caudal spot frequently
present; dorsal fin usually spotted with black, the spots varying in number, size,
and intensity; caudal fin frequently plain but oftener with elongate dark spots.

Many specimens of this species, ranging from 10 to 120 millimeters in length,
were preserved. It is the most abundant and the most universally distributed of
all the fishes of the Republic. It was taken in every locality except in the Rio
Molino, in which collections were made in fresh water. It was found in lakes,
swamps, and streams, and it does not shun currents to the extent that most top
minnows do, but it evidently finds the environment of the quieter waters more
congenial, as it is found there in greatest abundance. ‘This fish occurs mainly in
shallow water, but it is also found along the shores in water several feet deep. It
is abundant among vegetation and frequently, also, along rocky shores. This
species was found to be especially abundant in Lakes Chanmico, Chalchuapa, and
Coatepeque. In Lake Guija it was common along the shores but less numerous
than its smaller, spotted relative, Priapichthys letonai. In Lake Olomega JM.
sphenops was less abundant than in any other lake visited, notwithstanding that

poylusvur Apjeois ‘sdouayds visauayjoyy apeur oyq Jo uy [wuR oy} Jo aed [eIsiq— TL ‘SLT

(G86 90d) “SZ6l “A “d ‘8 “NO “TIng

Buuu. U. S. B. F., 1925. (Doe. 985.)

Fic. 12.—Priapichthys letonai sp. nov. Upper figure is the male. From the type. Length, 42
millimeters. Lower figure is the female. From a specimen 58 millimeters long

Fic, 13.—Distal part of the anal fin of the male Priapichthys letonai sp. nov., greatly magnified

FISHES OF EL SALVADOR 257

the conditions for its development appeared to be ideal. Water birds, however,
are extremely abundant on this lake and the fish probably are unable to protect
themselves against these enemies.

The size attained by this species varies greatly in the different waters. The
largest fish seen, ranging upward to 120 millimeters in length, were taken in small
ponds at El Angel. Many large individuals were also taken in Lakes Guija,
Coatepeque, and Llopango. In Lakes Chanmico and Zapotitan, on the other
hand, the usual size is rather small, the fishes ranging upward to only 55 millimeters
in length. In this species, as in most top minnows, the males are smaller than the
females, and sexual maturity is reached at a smaller size. Males 35 millimeters
long appear to be fully mature, while the smallest mature female, among a limited
number examined, was 60 millimeters. The males, as usual among the fishes of
this group, are greatly in the minority. A lot of 245 fish gathered up at random
after a discharge of dynamite on Lake Chanmico contained 80 males.

A comparatively small number of the fish collected appear to be gravid,
indicating that the period (January and February) during which the collections
were made was not the principal spawning season. Only a single stage of devel-
opment of the egg or embryo was found in any one ovary, indicating that broods
of young probably are not born in such rapid succession as in some of the other
species of top minnows. ‘The largest number of embryos found was 82, which
were taken from the ovary of a fish 90 millimeters long.

The intestinal canal is about one-third longer than the total length of the fish.
The contents of the stomachs of 24 specimens examined consisted mainly of
disorganized masses, containing a relatively large amount of sand and some
plant fragments, fragments of insects, minute ova (probably insect eggs), and in
a few instances small quantities of Kntomostraca. The species is used for
mosquito control in Nicaragua by Sanitary Engineer F. E. Hulse, of the Inter-
national Health Board, apparently with success, but it is believed from the field
observations made, and from the examination of stomach contents, that its
relatives, Priapichthys letonai and P. foster, are more efficient agents for the
control of the malaria-transmitting mosquitoes, Anopheles.

This fish, as here understood, is a widely distributed, variable species, several
nominal species having been defined. Much variation exists with respect to the
depth of the body, shape of the dorsal and caudal fins, and especially with respect
to color. These characters, however, appear to intergrade. Careful measure-
ments and scale and fin counts were made of 40 specimens, selected to include all
the extremes of variations. The minute structure of the intromittent organ, which
is regarded as a very helpful character in determining species, was critically studied
in all the extreme color phases represented among the El Salvador specimens,
and this organ also was compared with others taken from specimens from Nicaragua
and Panama and found to be identical. The range, as here understood, includes
both slopes from southern Mexico to northern Colombia. This is one of a very
few species of fishes that have previously been recorded from the fresh waters of
El Salvador. Our specimens were collected in the following localities: Lake Guija,
Rio del Desague, Lake Metapan, Lake Chalchuapa, Rio Pampe near Chalchuapa,
Lake Ahuachapan, Lake Coatepeque, Lake Chanmica, Rio Sucio near Sitio del

958 BULLETIN OF THE BUREAU OF FISHERIES

Nifio, Lake Zapotitan, ponds at El Angel, Lake Ilopango, small streams near San
Salvador, Rio Lempa at Suchitoto and San Marcos, Rio San Miguel at San Miguel,
and Lake Olomega.

8. Genus PRIAPICHTHYS Regan

Priapichthys Regan, Proc., Zoél. Soc. London, 1913, 991 (type Gambusia annectens Regan).

Body elongate, head depressed above; mouth transverse; teeth in the jaws in
narrow bands, the outer ones scarcely movable; dorsal fin in the adult male scarcely
higher than in the female; intromittent organ greatly produced, longer than the
head; alimentary canal not exceeding the length of the body and without convolu-
tions. Two species, both apparently new, appear to belong to this genus.

KEY TO THE SPECIES

a. Dorsal fin in the female usually equidistant from the tip of snout and end of caudal, or slightly
nearer the former; two branches of the median produced ray of the intromittent organ
bearing spur-like hooks; median line of sides with a row of round black spots scarcely as
large as the pupil, varying in number from 3 to 10_______________ letonai, sp. nov., p. 258.

aa. Dorsal fin in the female, usually an eye’s diameter nearer the end of caudal than tip of snout;

only one branch of the median produced ray of the intromittent organ with spur-like hooks;
sides with 6°to 10 dark crossbars2_2.-.=--.-b4-2-—9 2.22. 3222_2_ fosteri, sp. nov., p. 260.

8. Priapichthys letonai sp. nov.
CHIMBOLA

Type No. 87251, U.S.N.M.; male, length 42 mm., Rio San Miguel, San Miguel.

Head 3.1 to 4.3; depth 3.15 to 4.05; D. 8 or 9; A. 9 or 10; scales 28 to 30.

Body moderately compressed anteriorly; caudal peduncle strongly compressed,
its depth 1.35 to 2.35 in head; profile nearly straight from snout to nape, slightly
convex from nape to dorsal; head rather broad, flat above; snout short, broad,
2.75 to 4.65 in head; eye 2.5 to 4.25; interorbital 2.2 to 3.28; mouth small, the cleft
transverse; teeth loosely attached, the outer series slightly broadened, curved
inward, and well separated from the very minute inner teeth; scales cycloid, present
on head and snout and on base of caudal, each scale with 11 to 20 radii, average of
30 scales taken from 10 specimens ranging in length from 24 to 63 millimeters, 13.7;
origin of dorsal in female a little in advance of middle of anal base, equidistant
from tip of snout and end of caudal, or a little nearer the former; dorsal placed
further forward in the adult male, about half the length of head, nearer tip of
snout than end of caudal; caudal fin broad, gently convex; anal fin in females and
young similar to dorsal, its origin a little nearer base of caudal than posterior
margin of eye; anal fin in adult males inserted far in advance of dorsal, greatly
produced, much longer than the head, reaching nearly to base of caudal in sexually
mature but small males, proportionately shorter in larger specimens, the third,
fourth, and fifth rays all of about equal length, the branches greatly crowded distally
and forming a compound curve, the apex being directed forward, two of the branches
of the median produced ray bearing about 12 spur-like hooks on their posterior
margins; ventrals reaching vent in small females, proportionately shorter in large

FISHES OF EL SALVADOR , 259

specimens, reaching past origin of anal in males; pectoral fins moderate, 1 to 1.85
in head; vertebre about 15+17; alimentary canal about as long as the body; peri-
toneum black.

Color of the sexes similar; olivaceous above; lower part of sides silvery; median
line of sides with a row of round black spots scarcely as large as pupil, varying in
number from 3 to 10, present even in the mature embryo (wanting only on a single
large female); a faint dark vertebral band; a sharp dark median line from anal to
caudal; base of anal sometimes surrounded by black in large females; fins all un-
spotted; dorsal, caudal, and anal olivaceous; other fins plain translucent.

Many specimens of this species, ranging in length from 10 to 85 millimeters,
were preserved. This fish inhabits both lakes and streams, but it was not taken in
the tributaries of the Rio de Paz nor in Lakes Chalchuapa, Ahuachapan, Coatepeque,
Chanmico, and Hlopango. Neither was it obtained in the Rio Lempa at San Marcos,
notwithstanding that the species appeared to be common in that stream about 200
kilometers further upstream at the village of Suchitoto.

This fish appears to grow much larger in some localities than in others. In
Lake Zapotitan, for example, the largest male and female obtained were, respec-
tively, 20 and 27 millimeters in length, yet fully mature, as shown by the com-
pletely developed intromittent organ of the male and by the presence of large
embryos in the ovaries of the female. Specimens of this size from some of the other
localities at which collections were made are clearly sexually immature. The
largest fish seen, ranging a little more than 80 millimeters in length, were taken in
the Rio San Miguel at San Miguel. The males are much smaller than the female
and they occurred rather sparingly in the catches made.

Many of the specimens in the collection (taken during January and February)
are In spawning condition, the ovary usually containing large eggs in two distinct
stages of development, in addition to smaller ova. An ovary taken from a specimen
67 millimeters long, for example, contained 26 well-developed embryos and 30
apparently fully mature eggs, measuring about 1.5 millimeters in diameter, and
minute ova. The young fish when born is from 8 to 10 millimeters long, and well
developed. The body is covered with dark punctulations, which usually are con-
centrated in about three places on the median line of the sides, later forming the
dark lateral spots which are characteristic of the species.

Stomachs examined contained insects, fragments of fresh-water sponge, alge,
leaf fragments, small ova (probably insect eggs), and inorganic matter consisting
largely of sand. Field observations would indicate that the species probably is of
considerable value as an eradicator of mosquito larve.

The specimens were obtained at the following localities: Lake Guija, Rio del
Desague at Lake Guija, Lake Metapan, Lake Zapotitan, Rio Sucio at Sitio del
Nifio, Rio Acelhuate at San Salvador, Rio Lempa at Suchitoto, Rio San Miguel at
San Miguel, and Lake Olomega.

°260 BULLETIN OF THE BUREAU OF FISHERIES

9. Priapichthys fosteri sp. nov.
CHIMBOLA

Type No. 87263, U.S.N.M.; male, length 38 mm., Rio Lempa, San Marcos.

Head, 2.95 to 5.2; depth, 3.25 to 4.8; D., 8; A., 9 or 10; scales, 27 to 30.

Body moderately compressed anteriorly; caudal peduncle deep and strongly
compressed, 1.4 to 1.8 in head; profile straight from snout to nape, gently convex
from nape to dorsal; head broad, flat above; snout short, broad, 3.2 to 4.25 in
head; eye, 2.7 to 3.7; interorbital, 1.85 to 2.9; mouth small, the cleft transverse;
teeth loosely attached, the outer series slightly broadened, curved inward, and
well separated from the very minute inner teeth; scales cycloid, present on head
and snout and on base of caudal, each scale with 6 to 11 radii; average of 30 scales
taken from 10 specimens, ranging from 23 to 61 millimeters in length, 8.86+;
origin of dorsal in female a little in advance of middle of anal, occasionally equi-
distant from the tip of snout and the end of caudal, but more usually an eye’s
diameter nearer the latter; dorsal placed further forward in adult male, about
an eye’s diameter nearer tip of snout than end of caudal; caudal fin broadly
rounded; anal fin in females and young similar to the dorsal, its origin about equi-
distant from tip of snout and base of caudal; anal fin in adult males inserted far
in advance of dorsal, greatly produced, much longer than head, failing to reach
base of caudal by a distance equal to the length of snout and eye, the third, fourth,
and fifth rays all of about equal length, the branches greatly crowded distally
and forming a compound curve, the apex being directed forward, the posterior
branch of the median produced ray bearing about 16 spurlike hooks on its pos-
terior margin; ventral fins reaching origin of anal in young, scarcely to vent in
adult females, past the origin of anal in males; pectoral fins moderate, 1 to 1.35
in head; vertebrae 15+17; alimentary canal scarcely as long as body; peritoneum
black.

Color of the sexes similar; upper parts greenish; lower parts silvery; sides
with from 6 to 10 dark crossbars; a narrow vertebral line; a sharp dark median
line from anal to caudal; ventral fins yellowish; other fins all slightly olivaceous.

This species is represented by 98 specimens, ranging from 35 to 80 millimeters
in length. It was taken only in the Rio Lempa at San Marcos and at Suchitoto
in quiet water, and 2 specimens were secured in brackish water in the estuary at
El Triunfo. The males in this species, as is usual for the family, are much smaller
than the females and fewer in number. ;

Many of the specimens collected (during January and February) were in
spawning condition. Embryos and eggs of several sizes usually were present in
one ovary. The ovary from a female 80 millimeters long, for example, contained
44 well-developed embryos, 42 “eyed” eggs, and 44 large, probably mature, eggs.
Two other ovaries contained, respectively, 7 large embryos and 20 eggs in the
“eyed” stage in addition to smaller ova.

The stomachs examined contained principally fragments of insects and vege-
table débris. The species is quite probably of value as an eradicator of mosquito
larve.

Buu. U. 8S. B. F., 1925. (Doc. 985.)

Fic. 14.—Priapichthys fosteri sp. nov. Upper figure is the male. From the type. Length,
38 millimeters. Lower figure is the female. From a specimen 50 millimeters long

Fic. 15.—Distal part of the anal fin of the male Priapichthys fosteri sp. nov., greatly magnified

FISHES OF EL SALVADOR ; 261

This fish differs notably in color from the preceding one, but in structure
the two are very similar. The dorsal fin, however, appears to be inserted slightly
further backward in the female in the present
species, usually being an eye’s diameter nearer ¢
the end of the caudal than the tip of the
snout, instead of being equi-distant or nearer ,
the snout. Thescales, on an average, possess
fewer radii in the present species, as shown
by Figure 16, and in the distal part of the
intromittent organ of the male, which only ?
one branch of the produced rays bears spurlike 2
hooks, while in the preceding species two Sa
of the branches have hooks. This species *
also appears to be related to Priapichthys “|
panamensis, Meek and Hildebrand (Pub.,
Field Mus. Nat. Hist., Vol. X, 1916, p. 322), *
from which it differs in the relative position *
of the dorsal and anal fins, the shape of the +
caudal, and in the shape and structure of
the produced portion of the anal fin. -.

The specimens were obtained in the Rio Eee PEERS, Sah Deine

Sree : : in the graph is based on the average number of radii on
Lempa, at San Mar cos and Suchitoto, and three scales from one specimen. Thescales were taken

: : : q below the dorsal fin—one above, one in, and one below
in brackish water at El Triunfo. DOMME Tete ee? Le eee ee

m20~«25~«O 95 50. 65 6 65

40
Length of Fish in min

Family VI. ANABLEPID/Z

Cuatro-ojos

Body elongate, depressed anteriorly, compressed posteriorly; head broad;
supraorbital rims much raised; eye divided into an upper and a lower portion by a
dark-colored transverse membrane in the cornea; mouth mostly transverse; pre-
maxillaries protractile; teeth in each jaw in a villiform band; scales small or of
moderate size; dorsal and anal fins short; anal fin of the male modified into a thick
scaly conical organ with an orifice at its extremity. This family consists of a single
genus.

9. Genus ANABLEPS Scopoli

Anableps Scopoli, Introd. Hist. Nat., 1777, 450 (type Cobitis anableps Linneus).

The characters of the genus are included in the family description. The eye
in these fish is divided into an upper and a lower half by a dark horizontal partition
in the cornea. The upper half of the eye is higher than the rest of the head, is
usually exposed above the surface of the water, and evidently is for seeing in the
air, while the lower portion is for use in the water. It is from this singular structure
and function of the eye that the name ‘‘Cuatro-ojo,” or four-eye, has originated.
A single species is known from Central American waters.

962 BULLETIN OF THE BUREAU OF FISHERIES
10. Anableps dovii Gill

Four-EYE; CUATRO-OJO

Anableps dowei Gill, Proc., Ac. Nat. Sci., Phila., 1861"(1862), 4 (Panama; where it quite certainly does not occur).
Anableps dovii Regan, Biol. Cent. Amer., Pisces, 1907, 108.

Head 3.6 to 4.3; depth 4.4 to 7.5; D. 8 to 10; A. 10, rarely 9; scales 62 to 73.

Body very elongate, notably depressed anteriorly, compressed posteriorly;
depth of caudal peduncle 2.2 to 3 in head; head strongly depressed, flat above, its
depth between eyes about half its width; snout broad, its length 2.9 to 4 in head;
eye rather large, the cornea with a somewhat thickened, black, longitudinal bar in
the middle, dividing the eye into an upper and a lower half, its diameter, 3.6 to 5.9
in head; interorbital deeply concave, 3.3 to 5.55 in head; mouth broad, almost
wholly transverse; premaxillaries broad, excessively protractile; maxillary rather
broad, reaching anterior margin of eye; teeth small, in a broad villiform band in
each jaw; scales rather small, enlarged on head; dorsal small, situated posteriorly,
its origin at least an eye’s diameter behind vertical from end of anal base; caudal
fin rather broadly rounded; anal fin similar to the dorsal, its origin about equi-
distant from margin of opercle and tip of caudal in adult, much further forward in

Fic. 17.—Anableps dovii Gill

young, modified into an intromittent organ in the male, the anterior rays enveloped
in loose scaly skin, extending to the end of the longest rays, where there is a perfora-
tion; ventral fins short, reaching to or a little past the vent in the young, failing to
reach this point in the adult, pectoral fins moderate, 1.25 to 1.6 in head.

Color of a fresh specimen, 115 millimeters long, dark greenish above; white
underneath; sides with a white lateral band about half as wide as eye, bordered
above and below by black; back crossed by several indistinct black bars; eye
with a median black stripe, the upper half dark, the lower half white; dorsal dark
green; caudal somewhat lighter green, with black on distal parts; anal, ventrals,
and pectorals light greenish, the pectorals with black on lower half. In larger
examples the lateral band is yellow and it is broken up into spots anteriorly.

This species is represented by 36 specimens, ranging from 50 to 235 millimeters
in length. It is common in some localities but scarce or wanting in others. It
inhabits both lakes and streams. In the streams it is partial to the quieter places.
This species is especially common in Lake Guija and in the outlet, the Rio del
Desague. It is also very common in the Rio Lempa in the vicinity of Suchitoto.
A single fish was seen and captured in salt water at the wharf at Cutuco. The
fish swim at the surface with the upper half—that is, the portion above a dark
longitudinal band of the modified superior eye—above water, and they are usually

FISHES OF EL SALVADOR 263

seen in small schools. They are rather shy, and when frightened they sometimes
make leaps of a meter or so above the water, or, more usually, they lift the head
and body above the surface, leaving only the tail in the water. In this way they
make rapid headway, and repeated efforts at surrounding a school with a 9-meter
seine failed.

The four-eye apparently is never used for food, although it reaches a much
larger size than several other species that are eaten. The largest fish seen by the
collectors was 235 millimeters long, but according to the natives at Lake Guija
the species attains a length of at least 300 millimeters. The male appears to be
smaller than the female, as no males occurred among the larger individuals examined.
The largest male seen was 180 millimeters long.

The species is viviparous, and the anal fin in the male serves as a copulatory
organ. As the fish develops and reaches sexual maturity the anterior rays of the
fin become enveloped in scaly skin, which extends to the end of the longest rays,
forming a sort of tube, and having an orifice distally. A thin membranous recep-
tacle lies in the abdominal cavity at the base of the anal, which is entered by the
posterior end of the testes. A tube extends into the modified anal fin from this
receptacle. This tube les either right or left of the eveloped fin rays, through
which the seminal fluid is conveyed to the distal orifice. The ovary is single, and
the number of young produced at one time appears to be rather small. One ovary
contained 6 embryos, each approximately 11 millimeters long; another inclosed
only two embryos, each about 14 millimeters long.

This fish appears to feed on alge and small entomostracans, insects, and other
animals found among these plants at the surface of the water. The contents of
the stomachs examined and the habit of surface feeding and swimming indicate
that this species may be of value as an eradicator of mosquito larvee.

This fish is known from Southern Mexico, Guatemala, and El Salvador. The
type locality given by Gill, who first described the species, is the “‘ Pacific coast of
Panama,” where the species quite certainly does not occur. It is more probable
that the single specimen upon which the description was based came from Guate-
mala or Mexico, where Captain Dow, for whom the species is named, also collected.
The specimens at hand were collected in Lake Guija and its outlet, Rio Lempa at
Suchitoto and San Marcos, Rio San Miguel at San Miguel, Lake Olomega, and
Cutuco. It was also seen in the Rio Sucio at Sitio del Nifio, but no specimens
were obtained there.

Order PERCOMORPHI
Family VII. ATHERINIDAZ
The Silversides

Body elongate, more or less compressed; premaxillaries protractile; jaws
with two or more series of conical teeth; lateral line usually absent, never complete;
two well-separated dorsal fins, the first formed of 3 to 6 slender spines, the second
with a short spine and 7 to 13 branched rays; caudal fin forked; anal fin with a
single spine and 12 or more branched rays; ventral fins abdominal, each with a

964 BULLETIN OF THE BUREAU OF FISHERIES

spine and 5 branched rays; vertebre more than 30; sides with a silvery longitudinal
band.

This is a large family of small fishes. Some of the members inhabit fresh
water, but the majority of them live in salt and brackish water.

10. Genus THYRINA Jordan and Culver

Thyrina Jordan and Culver, Proc., Cal. Ac. Sci., 2 ser., V, 1895, 419 (type Thyrina evermanni Jordan and Culver).
Melaniris Meek, Pub., Field Col. Mus., Zo6l. Ser., III, 1902, 117 (type Melaniris balsanus Meek).

Body elongate, compressed; trunk sharply compressed ventrally; profile in
advance of dorsal almost straight and nearly parallel with the posterior half of the
ventral contour; lower jaw included; teeth unequal, the outer ones more or less
enlarged in each jaw; scales often crenate, particularly on the back; origin of
first dorsal well behind the origin of anal; base of anal longer than head; pectoral
fins rather long, frequently longer than head.

A single species, which appears to be undescribed, was taken in fresh water.

11. Thyrina guija sp. nov.
Pepresca; Mansgupa; ALFILER; ROVALETE

Type No. 87273, U.S.N.M.; length 88 mm., Lake Guija, El Salvador.

Head 4 to 4.75; depth 5.5 to 6.65; D. III to V-8 to 10; A. I, 22 to 26; scales
43 to 47.

Body very elongate, compressed; abdomen rather sharply compressed, almost
but not quite forming a keel; caudal peduncle long and slender, at least twice as
long as deep, its least depth 2.55 to 3.8 in head; head rather short, flat above;
snout moderate, its length 2.9 to 3.2 in head; eye 2.9 to 3.5; interorbital 2.4 to 2.7;
mouth small; the small hidden maxillary reaching nearly to anterior margin of eye;
premaxillary fully protractile, the anterior margin strongly curved, posteriorly
greatly extended; lower jaw included, a little shorter than the upper; teeth in jaws
in bands, the outer series in each jaw enlarged, those of the upper jaw more so than
in the lower jaw and placed on the extreme outer edge of the jaw, curved inward
and exposed when the mouth is closed; gill rakers short, about 16 on lower limb of
first arch; scales mostly with straight edges, those on back with rather broad
indentations, small scales extending on base of dorsal but none on the other fins;
origin of spinous dorsal over the base of about the sixth ray of the anal, the spines
reaching less than half the way to origin of soft dorsal; origin of the soft dorsal over
or a little behind middle of anal base and about half as far from base of caudal as
the margin of opercle; caudal fin forked, the lower lobe larger and longer; anal fin
long, its base extending slightly beyond the end of the base of second dorsal, origin
of anal about equidistant from posterior half of eye and base of caudal; ventral fins
small, inserted about equidistant from margin of opercle and base of caudal; pectoral
fins longer than head, falcate, 3.45 to 4.25 in length of body; vertebre 20+ 20.

Color greenish above, silvery below; sides with a silvery lateral band, about
the width of the pupil, with a dark margin above; the scales on the back with dusky
punctulations; a dark vertebral line; upper surface of head dusky; fins mostly pale
green to translucent; a dark line along base of anal.

Buuu. U. 8S. B. F., 1925. (Doe. 985.)

Fic, 18.—Thyrina guija sp. nov. From the type. Length, 88 millimeters

FISHES OF EL SALVADOR 265

Many specimens, ranging from 20 to 105 millimeters in length, were preserved.
This fish was taken only in Lake Guija, its outlet, and in two localities on the Rio
Lempa. It appeared to be common in the Rio del Desague, the stream forming the
outlet of Lake Guija. In the other localities where the species was found it appeared
to be rather rare.

This fish appears to differ from T. guatemalensis (Giinther), recorded from the
Pacific slope of Guatemala, according to the very inadequate descriptions of that
species, in having a more slender body, more numerous scales in a lateral series, and
a slightly inferior mouth. According to Giinther (Proc., Zoél. Soc., London, 1864,
p. 151) T. guatemalensis has only 36 scales in a lateral series. Regan (Biol. Cent.
Amer., Pisces, 1907, p. 64) gives 36 to 42. Jordan and Hubbs (Leland Stanf. Jr.
Univ. Pub., Univ. Ser., 1919, pp. 58 and 60), however, are of the opinion that Regan
was considering three species under one name, and therefore the range given by
him may not be correct.

In the present species the range in the number of scales, based upon the enumer-
ation of 30 specimens, is 43 to 47. The mouth in guatemalensis is described as
terminal, while in the species at hand the lower jaw is included and is a little
shorter than the upper. The depth of the body in the length in guatemalensis is
given as 5 by Giinther and 4 to 5 by Regan. In the present species the depth is
contained from 5.5 to 6.65 times in the length.

Eight small specimens, taken in salt water in the estuary at Triunfo, agree
fairly well with this species, except that the scales above the median line of the side,
at least, have broad indentations, and the number in the lateral series appears to
be a little lower, 41 to 43. These fish are referred to the present species, although
further study and more specimens may show that they belong to a distinct species.

The contents of 4 stomachs examined consisted of insects and insect larvae,
entomostracans, and filaments of alge. The sexual organs were in a collapsed state,
showing that the spawning season was not near at hand when the specimens were
collected (January and February).

The specimens are from Lake Guija, Rio del Desague near Lake Guija, and
from the Rio Lempa at Suchitoto and San Marcos.

Family VIII. MUGILIDAZ
The Mullets

Body elongate, more or less compressed; mouth small, terminal or inferior;
teeth, if present, small, various in form; premaxillaries protractile; gill openings
wide, the membranes free from the isthmus; gills 4, a slit behind the fourth; lateral
line usually absent, never complete; scales large, extending forward on head; dorsal
fins 2, well separated; first dorsal composed of 4 rather strong spines; second dorsal
with I, 7 to 10 rays; anal fin with IT or III, 7 to 11 rays; caudal fin forked; ventral
fins abdominal, each with I, 5 rays. Most of the species of this family are marine
but several are strictly fresh water in their habits.

266 BULLETIN OF THE BUREAU OF FISHERIES

KEY TO THE GENERA

a. Stomach muscular, gizzard-like; teeth minute, slender; lower jaw angulate in front; anal with

SUSPIDES;.© XCD. Ue TI VeOTsy yiO UU Os sper ee eee ee Mugil, p. 266.
aa. Stomach not gizzard-like; teeth in villiform bands on jaws and vomer; lower jaw anteriorly
rounded; anal fin with 2 spines__-.=-~-_- 22222) 2222) eee” Agonostomus, p. 267.

11. Genus MUGIL Linnzus

Mugil Linneus, Syst. Nat., Ed. X, 1758, 316 (type Mugil cephalus Linneus).
Querimanna Jordan and Gilbert, Proc., U. S. Nat. Mus. V, 1882 (1883), 588 (type Myzus harengus Giinther).

Body rather robust, the back and belly rounded; head broad; mouth terminal;
jaws weak, the lower one with a median obtuse angle; teeth in the jaws minute,
flexible; eye large, with an adipose lid; scales large, extending forward on head;
anal spines 3 (2 in very young); stomach very muscular, gizzardlike. The species
of this genus are marine, but some of them enter fresh water and the one included
here appears to be a regular visitor to the fresh-water streams of El] Salvador, where

it is of some commercial importance and highly prized as a food fish.

12. Mugil cephalus (Linnzus)

“LIEBRE ANcHA;” “Liza”

Mugil cephalus Linneeus, Syst. Nat. Ed. X, 1758, 316 (Europe; based on Artedi); Jorden and Evermann, Bull.,.U.S. Nat. Mus.
XLVII, 1896, 811, Pl. CK XVI, fig. 348; Meek and Hildebrand, Pub., Field Mus. Nat. Hist., Zodl. Ser., XV, 1923, 275.
(For a fuller synonymy and additional references see one of the two last-mentioned works.)

Head 3.9 to 4.1; depth 3.9 to 3.96; D. IV-I, 8; A. III, 8; scales 42.

Body elongate, compressed; head low and rather broad; snout tapering, its
length 3.2 to 3.45 in head; eye 4.35 to 4.9; interorbital slightly convex, 2.2; mouth
rather broad, oblique; upper jaw projecting; maxillary scarcely reaching eye,
3.6 to 3.75 in head; teeth in the jaws minute but visible with the unaided eye;
gill rakers minute, close-set; scales moderate, 11 or 12 rows between origin of second
dorsal and base of anal, each scale with a finely serrate membranous border; origin
of spinous dorsal slightly nearer tip of snout than base of caudal, the longest spine
1.8 to 1.9 in head; origin of second dorsal about an eye’s diameter nearer origin of
first dorsal than base of caudal; the fin with a few small scales at the base posterior
to the anterior rays; caudal fin forked, the upper lobe longest, pointed, the fin with
small scales; anal fin similar to the second dorsal, its origin a little in advance of
second dorsal, with a few minute scales at base posterior to the anterior rays; ventral
fins inserted under the posterior fourth of the pectorals; pectoral fins a little longer
than the ventrals, failing to reach origin of first dorsal, 1.5 in head.

Color of a fresh specimen, 430 millimeters long, dark greenish brown above;
lower parts silvery; rows of scales on sides with dark stripes; fins all dusky, except
anal and ventrals, which are pale.

Two specimens, respectively 410 and 430 millimeters in length, were taken.
This fish, ‘although a salt-water species, occasionally runs upstream into fresh
water. The specimens at hand were taken in the Rio Lempa at Suchitoto, about
180 kilometers, following the course of the stream, from the sea and far above the
influence of tide. According to local fishermen this fish is a more or less permanent

FISHES OF EL SALVADOR 267

resident and it is not infrequently taken. It occurs in the market in the village,
and it was served on the table in a local hotel during our visit. Fish up to 500
millimeters in length are said to be taken locally. This fish was not seen elsewhere.

This widely distributed mullet is known from nearly all warm waters of both
hemispheres—on the American coasts from Monterey to Chile and from Cape Cod to
Brazil. The El Salvadorian specimens were taken in the Rio Lempa at Suchitoto.

12. Genus AGONOSTOMUS Bennett

Agonostomus Bennett, Proc., Zoél. Soc:, London, I, 1830 (1831), 166 (type Agonostomus telfairii Bennett).
Neomugil Vaillant, Bull., Soc. Philom., Paris, IV, 1894, 73 (type Neomugil digueti Vaillant).

Body elongate, compressed; mouth terminal in young, subinferior in adult;
cleft extending laterally to or past front of eye; lower lip never greatly thickened;
teeth in bands on jaws, vomer, and palatines; anal spines 2, the first one minute,
often hidden in the skin; stomach not gizzardlike. The species of this genus
inhabit mostly tropical rivers, some of them living in mountain torrents. A single
species was taken in E] Salvador.

13. Agonostomus monticola (Bancroft)

TEPEMECHIN; CHIMBERA; Liza

Mugil monticola Bancroft, in Griffith’s edition, Cuvier’s Animal Kingdom, Fishes, 1836, 367, pl. 36 (West Indies).

Agonostomus monticola Giinther, Cat. Fish. Brit. Mus., III, 1861, 464.

Agonostomus nasutum Giinther, Cat. Fish. Brit. Mus., III, 1861, 463 (Rio San Geronimo, Guatemala).

Neomugil digueti Vaillant, Bull., Soc. Philom., IV, 1894, 73 (Lower California).

Agonostomus salvini Regan, Ann. and Mag. Nat. Hist., 7 ser., XIX, 1907, 66, and Biol. Cent. Amer., Pisces, 1907, 68, Pl. XI,

fig. 2 (Nacasil, Guatemala).

Head 3.9; depth 3.8; D. IV-I, 8; A. II, 10; scales 42.

Body elongate, moderately compressed; upper profile gently convex; head
rather small; snout rather long, pointed, 3.4 to 3.55 in head; eye 3.4 to 4.1; inter-
orbital 2.55 to 3.1; mouth moderate, nearly horizontal; upper jaw projecting;
upper lip moderately thickened in a large specimen, rather thin in a smaller one;
maxillary reaching a little past anterior margin of eye but scarcely to pupil, 2.85
to 3.1 in head; teeth small in bands on jaws, vomer, and palatines; gill rakers
about half as long as eye, 19 on lower limb of first arch; scales rather large, strongly
ctenoid, extending forward on interorbital, present on cheeks, 11 or 12 longitudinal
rows between origin of second dorsal and base of anal; origin of spinous dorsal at
least an eye’s diameter nearer tip of snout than base of caudal, the spines strong,
the anterior one a little longer than eye and snout, 1.55 to 1.6 in head; origin of
second dorsal about an eye’s diameter nearer the origin of the first than the base
of the caudal, the outer margin of the fin concave; margin of caudal fin rather
deeply concave; anal fin similar to the second dorsal but somewhat larger, its
origin a little in advance of second dorsal; ventral fins moderate, inserted under
or slightly posterior to middle of pectorals; pectoral fins a little longer than the
ventrals, reaching to or a little beyond vertical from origin of spinous dorsal, 1.3 to
1.55 in head.

42885—25t+ 3

268 BULLETIN OF THE BUREAU OF FISHERIES

Color of a fresh specimen, 180 millimeter long, grayish black above; sides
silvery; under parts pale silvery; scales on sides with dark edges; an indistinct
dark blotch at base of caudal; first dorsal with dark spines and yellow interradial
membranes; second dorsal greenish, with a broad translucent margin; caudal
and pectorals plain translucent; the latter dark at base; anal and ventrals mostly
yellowish; iris golden. A smaller specimen, 82 millimeters long, lighter, the black
on margins of scales on sides not continuous, forming specks; caudal spot more
distinct.

Two specimens of this species, respectively 82 and 180 millimeters in length
were taken. It is improbable that this fish is as scarce as indicated by the few
specimens secured, for it appears to be well known to the native fishermen, according
to whom the species occurs in streams in several localities where it was not taken
by us. One of our specimens was taken in a deep rocky place in the Rio del Desague,
a short distance below Lake Guija, of which this river is the outlet. The smaller
specimen was taken in quiet shallow water on a sandy bottom in the Rio Lempa at
San Marcos.

According to the fishermen at Lake Guija the ‘“‘Tepemechin” spawns in June.
The eggs are much sought and considered a great delicacy. One fish is said to
produce a quantity of eggs, which in bulk is greater than that of the entire fish.
The species, according to native fishermen, reaches a length of about 250 millimeters.

This species, as here understood, ranges from southern Mexico to Panama,
occurring on both slopes of Central America and Panama. It is also known from
the West Indies. The specimens at hand are from the Rio del Desague, near
Lake Guija, and from the Rio Lempa at San Marcos.

Family IX. CENTROPOMID
The Robalos

Body moderately elongate, compressed; head long, somewhat depressed
above; mouth large, protractile; lower jaw projecting; teeth in villiform bands
on jaws, vomer, and palatines; preopercle and supraclavicle serrate; preopercle
with a ridge, usually bearing 2 spines; opercle without spines, produced as a flap;
lateral line more or less arched, continued to end of caudal fin; scales small or of
moderate size, ctenoid; two separate dorsal fins; the first consisting of 7 or 8
spines; second dorsal with 1 spine and 8 to 11 branched rays; caudal fin forked;
anal short, with 3 spines and 6 or 7 branched rays; ventral fins inserted behind
base of pectorals, with I, 5 rays; pectoral fins symmetrical; air bladder large,
with or without appendages.

All the species of this family are American and are included in one genus.

13. Genus CENTROPOMUS Lacépéde

Centropomus Lacépéde, Hist. Nat. Poiss., IV, 1803, 248 (type Scizna undecimalis Bloch).
Ozylabraz Bleeker, Arch. Neerl. Sci. Nat., XI, 1876, 264 (type Scizna undecimalis Bloch).
Macrocephalus Bleeker, Arch. Neerl. Sci. Nat., XI, 1876, 336 (type Scizna wndecimalis Bloch).
The characters of the genus are included in the family description. The
species of this genus are game fishes. Some of them reach a large size, and the

FISHES OF EL SALVADOR 269

quality of the flesh is excellent. They are chiefly shore fishes, frequenting brackish
water, and some of them ascend fresh-water streams. Three species were secured
in fresh water in El Salvador; others, no doubt, occur there from time to time.

KEY TO THE SPECIES

a. Seales small, 72 to 75 in a lateral series, about 8 rows between the middle of second dorsal
and lateral line; gill rakers few, about 8 on lower limb of first arch_____nigrescens, p. 269.

aa. Scales larger, not more than 65 in a lateral series, 5 or 6 rows between middle of second dorsal
and lateral line; gill rakers more numerous, not fewer than 12 on lower limb of first arch.

b. Scales moderate, 52 to 54 in a lateral series; gill rakers numerous, 16 or 17 on lower limb
of first arch; angle of preopercle with 2 abruptly enlarged serra; lateral line not in a

ark strea Ki gers PUENTE) Sale EN) ae a ele UN A 2 ere he robalito, p. 270.

bb. Seales smaller, 58 to 62 in a lateral series; gill rakers fewer, 13 or 14 on lower limb of first
arch; angle of preopercle with several gradually enlarged serrz; lateral line in a black

SUTORKE Spee eee PE! one ee be eat ee Be ee oe oe ey hy ee _pectinatus, p. 271.

14. Centropomus nigrescens (Giinther)

Rosato; RovaLo

Centropomus nigrescens Giinther, Proc., Z06]. Soc., London, 1864, 144, and Trans., Zod]. Soc., London, VI, 1868, 407 (‘‘Chiapam ’’,
Pacific coast of Guatemala); Jordan and Evermann, Bull., U. S. Nat. Mus., XLVII, 1896, 1119; Regan, Biol. Cent. Amer.,
Pisces, 1907, 50; Meek and Hildebrand, Pub., Field Mus. Nat. Hist., Zool. Ser., XV, Part II, 1925, 426, Pl. XLII.

Centropomus viridis Lockington, Proc., Cal. Ac. Sci., VII, 1876 (1877), 110 (Off Ascunsion Island).

Centropomus undecimalis (not of Bloch) Gilbert and Starks, Memoir., Cal. Ac. Aci., LV, 1904, 89.

Head 3 to 3.1; depth 3.7 to 3.85; D. VIII-I, 9; A. III, 6; scales 72 to 75.

Body elongate, compressed; profile concave over eyes; caudal peduncle long,
its least depth 2.75 in head; head long; snout broad, its length, 3.36 to 3.6 in head;
eye 7.55 to 9; interorbital 4.75; mouth large, oblique; lower jaw strongly projecting;
maxillary reaching nearly opposite middle of eye, 2.25 to 2.3 in head; preorbital
with a few small serre; preopercle strongly serrate, several serre at angle enlarged;
preopercular ridge without spines; gill rakers rather long and few, 8 on lower limb
of first arch; scales rather small, about 8 rows between middle of base of second
dorsal and lateral line, not greatly reduced in advance of dorsal, small scales extend-
ing on base of all fins, except spinous dorsal; origin of spinous dorsal nearly twice
the diameter of eye behind base of pectorals, the spines moderate, the third and
fourth of about equal length, the former not reaching the tip of the latter when
deflexed, the length 2.05 to 2.5 in head; origin of second dorsal notably nearer base
of caudal than preopercular margin; caudal fin forked, both lobes rather acute;
origin of anal under middle of base of second dorsal, the second spine somewhat
enlarged, not reaching the tip of the third when deflexed, 2.75 to 3.25 in head;
ventral fins inserted about an eye’s diameter behind base of pectorals, failing to
reach vent; pectoral fins reaching about to the beginning of the distal third of
ventrals, 1.9 to 1.95 in head.

Color of a fresh specimen, 715 millimeters long, bluish silvery above; lower
part of sides and abdomen silvery; upper surface of head and sides in advance of
pectorals yellowish; lateral line in a black streak; dorsals, caudal, and anal bluish
black; second dorsal, caudal, and anal with pale margins; pectorals and ventrals
yellowish green; ventrals with a broad white margin.

270 BULLETIN OF THE BUREAU OF FISHERIES

Several large individuals, ranging upward to 715 millimeters in length, were
taken with dynamite in the Rio Lempa at San Marcos. Only one specimen, 475
millimeters long, was preserved, which, together with notes and measurements made
in the field on still larger specimens, serves as the basis for the above description.

The ‘“ Rovalo” was reported by local inhabitants from several localities and
streams in El Salvador, but, as already stated, it was taken only in one locality.
Native fishermen at Suchitoto claimed that “ Rovalo” exceeding a meter in length
are occasionally taken there in the Rio Lempa. It is probable that this species and
possibly several others of this genus ascend the river to and beyond that locality,
as most of the representatives ascend fresh-water streams freely.

This species is known from Lower California to Ecuador. In El Salvador it
was taken in the Rio Lempa at San Marcos.

15. Centropomus robalito Jordan and Gilbert

Rosato; Rovaro

Centropomus armatus Giinther (not Gill) Trans., Zodl. Soc., London, 1868, 408.
Centropomus robalito Jordan and Gilbert, Proc., U. S. Nat. Mus., IV, 1881 (1882), 462 (Mazatlan; Acapulco); Jordan and Ever-
mann, Bull., U. S. Nat. Mus., XLVII, 1896, 1123.

Head 3 to 3.1; depth 3.25 to 3.4; D. VIII-I, 10; A. III, 6; scales 52 to 54.

Body moderately elongate, compressed; back elevated; head long, rather flat
above; snout long and broad, its length 2.75 to 3 in head; eye 4 to 4.15; interorbital
4.75 to 6.8; mouth large, a little oblique, lower jaw strongly projecting; maxillary
reaching nearly opposite middle of eye, 2.35 to 2.45 in head; preorbital with small
serrations; preopercle strongly serrate, 2 serre at angle notably enlarged; pre-
opercular ridge with 2 spines at angle; gill rakers slender, rather numerous, 16 or
17 on lower limb of first arch; scales moderate, ctenoid, 5 or 54% rows between middle
of second dorsal and lateral line, reduced in advance of dorsal, about 14 rows cross-
ing the back anterior to spinous dorsal, small scales extending on base of second
dorsal, caudal, and anal; origin of first dorsal scarcely an eye’s diameter behind
base of pectorals, the third spine about equal in length to the fourth but not reach-
ing beyond it when deflexed, 1.55 to 1.65 in head; origin of second dorsal about
equidistant from preopercular margin and base of caudal; caudal fin forked, both
lobes acute; origin of anal fin slightly posterior to middle of base of second dorsal,
the second spine much enlarged, reaching somewhat beyond base of caudal when
deflexed, its length 1.15 in head; ventral fins inserted shghtly behind base of pec-
torals, reaching to or a little beyond vent; pectorals not quite reaching tips of
ventrals, 1.35 to 1.6 in head.

Color bluish gray above, silvery below; lateral line not in a black streak; fins
all more or less dusky; spinous dorsal usually with more or less black on interradial
membranes; membrane between second and third anal spine, with black next to
the third spine.

Several small specimens, ranging in length from 120 to 140 millimeters, were
seined in the Rio Lempa at San Marcos in strictly fresh water and well beyond the
influence of tides. One small specimen of this species was taken in an estuary at
Triunfo.

FISHES OF EL SALVADOR O71

This fish ranges from Lower California to Panama. The specimens from El
Salvador are from the Rio Lempa at San Marcos, and from salt water at Triunfo.

16. Centropomus pectinatus Poey

Rosato; Rovato

Centropomus undecimalis Cuvier and Valenciennes (part), Hist. Nat. Poiss., II, 1828, 102.
Centropomus pectinatus Poey, Memorias, II, 1860, 121 (Cuba); Jordan and Evermann, Bull., U. S. Nat. Mus., XLVII, 1896.
1122; Regan, Biol. Cent. Amer., Pisces, 1907, 46; Meek and Hildebrand, Pub., Field Mus. Nat. Hist., Zod]. Ser., XV, Part

II, 1925, 421.
Centropomus medius Giinther, Proc., Zo6l. Soc., London, 1864, 144, and Trans., Zo6]. Soc., London, VI, 1868, 406 (‘‘Chiapam,’’

Pacific coast of Guatemala).
Centropomus grandoculatus Jenkins and Evermann, Proc., U. S. Nat. Mus. XI, 1888 (1889), 1389 (Quaymas).

Head 3 to 3.15; depth 3.5 to 3.7; D. VITI-I, 10; A. ILI, 7; scales 58 to 62.

Body elongate, compressed; profile concave over eyes; caudal peduncle long,
its least depth 2.55 to 2.8 in head; head long; snout rather broad, its length 2.8
to 2.9 in head; eye 4.15 to 5.1; interorbital 4.45 to 5.6; mouth large, oblique; lower
jaw strongly projecting; maxillary reaching nearly opposite middle of eye, 2.2 to
2.35 in head; preorbital with small serrations; preopercle strongly serrate, several
serre at angle enlarged; preopercular ridge with 2 spines at angle; gill rakers slender,
13 or 14 on lower limb of first arch; scales of moderate size, 544 or 6 rows between
middle of base of second dorsal and lateral line, somewhat reduced in advance of
dorsal, about 14 rows crossing the back anterior to spinous dorsal, small scales
extending on base of second dorsal, caudal, and anal; origin of spinous dorsal an
eye’s diameter behind base of pectorals, the spines long, the third the longest, reach-
ing to or a little beyond the tip of the fourth when deflexed, 1.5 to 1.7 in head;
origin of second dorsal a little nearer base of caudal than preopercular margin;
caudal fin forked, both lobes acute; origin of anal fin under the posterior third of
second dorsal, the second spine much enlarged but not reaching beyond the tip of
the third when deflexed, 1.3 to 1.6 in head; ventral fins inserted less than an eye’s
diameter behind base of pectorals, reaching to or a little beyond vent; pectoral
fins reaching somewhat past middle of ventrals, 1.7 to 1.75 in head.

Color bluish or grayish above; sides and abdomen silvery; tip of snout dusky;
lateral line in a black streak; dorsals, caudal, and anal all more or less dusky; the
membrane between the second and third anal spines darker than rest of fin but not
black; ventrals and pectorals yellowish green; tips of ventrals black in young, this
color disappearing with age.

Three specimens of this species, respectively 220, 280, and 350 millimeters in
length, were preserved. The largest of these was taken in strictly fresh water in
the Rio Lempa at San Marcos.

This species occurs on both coasts of tropical America where it enters fresh
water freely. On the Atlantic it ranges from Cuba to Panama and on the Pacifie
from Guaymas to Colombia.

The E] Salvador specimens are from the Rio Lempa at San Marcos and from the
salt water estuary at Triunfo,

QN2° BULLETIN OF THE BUREAU OF FISHERIES
Order CHROMIDES
Family X. CICHLIDA
The Mojarras

Body elongate, compressed; mouth large or small, terminal or subinferior;
teeth conical, incisorlike or lobate; vomer and palatines without teeth; premaxillaries
freely protractile; nostrils single on each side; lateral line interrupted under soft
dorsal, reappearing lower down on side; scales usually ctenoid; dorsal fin single,
the spinous portion usually longer than the soft part; anal fin with 3 or more spines;
ventral fins thoracic, with I, 5 rays; pseudobranchie wanting; branchiostegals
5 or 6; air bladder present. A single genus of this large family of tropical fresh-
water fishes is represented in the collection from El] Salvador.

14. Genus CICHLASOMA Swainson

Cichlasoma Swainson, Nat. Hist. Class., Fish., I, 1839, 230 (type Labrus punctatus).
Heros Heckel, Ann., Mus. Wien, II, 1840, 362 (type Heros severus Hecke)).
Hoplarchus Kaup. Arch. Naturg. XX VI, 1860, 128 (type Hoplarchus pentacanthus).

Body deep or elongate, compressed; mouth small or moderate; teeth on jaws
in bands, the outer ones more or less enlarged, sometimes forming canines; maxil-
lary exposed or not; premaxillary processes shorter than head; opercle entire; gill
rakers short and few (6 to 15 on lower limb of first arch); lateral line interrupted;
scales moderate or large, usually ctenoid, extending forward on head to the eyes
or beyond, also present on cheeks and opercles; dorsal fin single, not notched, with
XIV to XIX, 7 to 12 rays; anal IV to XII, 6 to 14; caudal round, truncate or emar-
ginate; ventrals inserted below or a little behind base of pectorals; pectoral fins
asymmetrical, with 12 to 18 rays.

This is a large genus with many representatives in Mexico, Central and South
America. Five species are represented in the collections from El Salvador.

KEY TO THE SPECIES

a. The outer teeth anteriorly in each jaw more or less regularly enlarged; premaxillary processes

not reaching middle of eye; mouth small, the maxillary failing to reach anterior margin of eye.

b. Dorsal XVII to XIX, 8 or 9; anal VIII to X, 5 to 8; body with black crossbars, the two

bands anterior to dorsal arched forward____.__-_--------------- nigrofasciatum, p. 273.

bb. Dorsal XIV to XVI, 12 to 14; anal V, 9 or 10; body with dark crossbars (at least in young),
none of them arched.

c. Body deep, the depth 1.8 to 1.95 in length (specimens 30 to 85 mm. long); profile
steep; eye large, 2.5 to 2.8 in head (specimens 30 to 85 mm. long); general color
bluish gray above, pale silvery below; sides with 8 moderately distinct cross-
DaTSen ode oi oper ee Au oe EO ee a aeaie ete Eee beh tia macracanthus, p. 274.

cc. Body more slender, the depth 2 to 2.25; profile less strongly elevated; eye smaller,
3.6 to 4.95 in head; general color darker; the crossbars (in young) more
distinct 222 Sisal a Lt Leen Sen 2 cele aie Sa eae ee eee a ee meeki, sp. nov., p. 275.

FISHES OF EL SALVADOR 278

aa. Anterior pair of teeth in upper jaw enlarged, canine-like; the anterior pair in lower jaw small,
followed by one or more enlarged teeth; premaxillary processes reaching to or beyond
middle of eye; mouth moderate, the maxillary reaching about to anterior margin of eye;

dorsal XVII to XIX, 10 to 12; anal VII or VIII, 8 to 10.

d. Body rather deep, the depth 2.12 to 2.47 in length (specimens 127 to 185 mm. long);
about 6 rows of scales on cheek; a prominent dark spot above origin of lateral line
(sometimes obscure in adults); no black sports or bars on opercle; no rusty spots on
bodyzandtverticalifims sss. lane se seeneee et ee eae Ee trimaculatum, p. 277.

dd. Body more elongate, the depth 2.54 to 2.74 in length (specimens 127 to 185 mm. long);

about 8 rows of scales on cheek; no dark spot above origin of lateral line; opercle
with one or two bars or blotches; adults usually with rusty spots on body and vertical
ELIT spon oe oa re ee aR eS se ee Eee motaguense, p. 279.

17. Cichlasoma nigrofasciatum Giinther

Burro; AcCHTBA; CHAMARRA; CHTNCOYO; CONGA; MOJARRA

Heros nigrofasciatus Giinther, Trans., Zodl., Soc., London, VI, 1868, 452, Pl. LX XIV, fig. 3 (Lakes Atitlan and Amatitlan).
Ciehlasoma nigrofasciatum Jordan and Evermann, Bull., U.S. Nat. Mus. XLVII, 1898, 1525; Regan, Ann. and Mag. Nat. Hist.,
7ser., XVI, 1905, 75; Regan, Biol. Cent. Amer., Pisces, 1906, 22; Meek, Pub., Field Col. Mus., Zo6l. Ser., VII, 1908, 189.

Head 2.55 to 3.35; depth 1.9 to 2.25; D. XVII to XIX, 8 or 9; A. VIII to X,
5 to 8; scales 27 to 30.

Body compressed, short and deep; dorsal profile rather steep, only slightly
convex over snout and eyes; caudal peduncle short and deep, 1.7 to 2.5 in head;
head rather small; snout moderate, 2.4 to 3.3 in head; eye 3.2 to 4.6; interorbital
2.05 to 3.4; mouth small, terminal; maxillary hidden, failing to reach anterior
margin of eye, 3.25 to 4.15 in head; fold of lower jaw interrupted at symphysis;
premaxillary processes extending to anterior margin of eye; teeth in the jaws in
bands, conical, the outer ones in anterior part of each jaw enlarged; gill rakers short,
7 to 9 on lower limb of first arch; scales rather large, somewhat reduced on nape
and chest, about 4 or 5 rows on the cheeks, 4 complete rows between origin of dorsal
and lateral line, a few small scales on the base of vertical fins; dorsal fin long, not
notched, the last spine longest, about equal to postorbital part of head, the soft
part elevated, the median rays filamentous in adults, sometimes reaching nearly
to end of caudal, origin of fin over margin of opercle; caudal fin broadly rounded;
anal fin with strong pungent spines, the soft part similar to that of the dorsal and
coterminal with it; ventral fins inserted slightly behind the base of pectorals, the
exterior rays more or less filiform, except in very young, and reaching beyond origin
of anal; pectoral fins moderate, reaching to or more usually past the origin of anal,
1.05 to 1.4 in head; vertebra 13 +15.

Color of a fresh specimen, 55 millimeters long, grayish green above; pale with
pinkish and silvery reflections on lower parts of sides; chest and abdomen nearly black
with a slight greenish tinge; snout black; sides with 9 black bars, usually broader
than interspaces, the first band arched, running across the nape and on opercle, a
second band concentric with the first and extending across nape to behind base of
pectorals, the third bar short, usually nearly and frequently completely connecting
with the second bar on median line of side, forming a V; the other bars nearly
vertical, the fourth to the seventh usually extending on dorsal fin, the eighth connect-
ing the ends of the dorsal and anal, the last one at base of caudal; ventral fins

274 BULLETIN OF THE BUREAU OF FISHERIES

black; all other fins dusky. Some individuals are much darker than others, but
the pattern is identical. The specimen described is one of the dark-colored ones.
These color varieties are recognized by the natives who, on Lake Chalchuapa, at
least, referred to the light variety as “plateada” and the dark ones as “negra.”

Many specimens of this common species, ranging in length from 25 to 120
millimeters, were preserved. This fish is the most common one of the family in the
fresh waters of El Salvador. However, it was not found in four localities visited,
viz, Rio Molino at Ahuachapan, Lake Ahuachapan, Rio Lempa at San Marcos,
and Lake Olomega. It is especially abundant in Lakes Coatepeque and Chanmico.

This species reaches a small size, probably rarely exceeding a length of 120
millimeters. It is used for food and also for crab bait. Its habitat does not appear
to be limited to any definite type of bottom, depth, or vegetation. In Lake Coate-
peque it was especially abundant among the rocks where, because of the very clear
water, it could be seen at a depth upward of 6 meters. In other localities it was
found in shallow water among vegetation and often in comparatively muddy places.
The species, according to the contents of 9 stomachs examined, feeds on small
animal and plant life of suitable size. Meek (Pub., Field Col. Mus., Zool. Ser., VII,
1908, p. 189) states that this fish deposits its eggs in April, May, and June in Lakes
Amatitlan and Atitlan, Guatemala. The sexual organs in the specimens examined
. (taken in January and February) were mostly in the early stages of development,
containing very minute ova that were not visible with the naked eye. A few
specimens, however, contained eggs upward of 3 millimeters in diameter and
probably nearing maturity.

This fish heretofore has been recorded only from Lakes Amatitlan and Atitlan,
Guatemala, and it apparently had not been taken in streams. The specimens from
El Salvador were collected in the following localities: Lake Guija, Lake Metapan,
Lake Chalchuapa, Rio Pampe near Chalchuapa, Lake Coatepeque, Lake Chanmico,
Lake Zapotitan, Rio Sucio at Sitio del Nino, ponds at El Angel, Lake [lopango, Rio
Lempa at Suchitoto, and Rio San Miguel at San Miguel.

18. Cichlasoma macracanthus (Giinther)

MogARRA

Heros macracanthus Giinther, Proc., Zodl. Soc., London, 1864, 153, and Trans., Zodl. Soc., London, 1868, 451 (Chiapam;

Huamuchal).
Cichlasoma macracanthum Jordan and Evermann, Bull., U.S. Nat. Mus., XLVII, 1898, 1518; Regan (part), Ann. and Mag. Nat.
Hist., 7 ser., XVI, 1905, 241; Biol. Cent. Amer., Pisces, 1906, 24, Pl. V, fig. 1.

Head, 2.5 to 2.8; depth, 1.8 to. 1.95; D. XV or XVI, 12 or 13; A. V, 9 or 10;
scales, 30.

Body very deep, compressed; the dorsal profile strongly elevated, straight from
snout to interorbital, then gently convex; caudal peduncle deep, its depth 2.2 to
2.55 in head; head moderate, deep; snout tapering, 2.7 to 2.8 in head; eye large,
2.5 to 2.8; interorbital 2.9 to 3.7; preorbital from two-thirds to three-fourths the
length of eye; mouth small, terminal; maxillary covered by the preorbital, not
reaching anterior margin of eye, its length 3.4 to 4.6 in head; premaxillary processes
reaching anterior fourth of eye; lower lip without a definitely free margin at the

WSheno,, Ol

. F., 1925. (Doc. 985.)

Fic. 19.—Cichlasoma macracanthus. From a specimen 78 millimeters long

SIOOUTT [IL CLT ‘YSuey ‘odA} oy} WO “AOU “ds 1yaIUW DULOSD]YIND — 0% "OLA

CEs6 90d) = “ez6T “A “A SA TAG

FISHES OF EL SALVADOR 275

symphysis; teeth in villiform bands in each jaw, the outer ones in each jaw an-
teriorly somewhat enlarged; gill rakers very short, about 6 on lower limb of first
arch; scales moderate, 5 or 6 rows between origin of dorsal and lateral line, some-
what reduced on nape and chest, small scales extending on base of vertical fins,
those above lateral line with smooth edges, those below lateral line with small
spinules on posterior margin and on a part of the upper surface of the exposed
portion; dorsal fin long, the spines graduated, the last one about half the length
of head, origin of fin over margin of opercle; caudal fin rounded; anal fin with
strong graduated spines, the last one a little longer and considerably stronger than
the last dorsal spine, origin of fin about equidistant from base of pectorals and base
of caudal; ventral fins inserted a little behind base of pectorals, reaching nearly or
quite to origin of anal; pectoral fins reaching to or a little beyond origin of anal,
1.1 to 1.2 in head.

Color of fresh specimen bluish gray with silvery reflections above; lower parts
pale silvery; sides with 8 dark crossbars, the fifth bar with an intensified black
blotch on median line of sides; a black spot at base of caudal above lateral line;
pectoral fins pinkish; all other fins dusky.

Nine small specimens, ranging in length from 30 to 85 millimeters, were pre-
served. These specimens agree fairly well with descriptions of C. macracanthus, a
species first described from the western slope of Guatemala. The specimens at
hand were taken in Lake Ahuachapan. This lake, although it has no visible outlet,
is situated within the basin of the Rio de Paz, a stream forming a portion of the
boundary between El Salvador and Guatemala.

This species has previously been recorded from the Pacific slope of Guatemala
and Tequesixtlan, southern Mexico. The El Salvador specimens are from Lake
Ahuachapan.

19. Cichlasoma meeki sp. nov.
MogarraA; MogarrA NEGRA; MoOJARRA PLATEADA

Type No. 87301, U.S.N.M.; length 175 mm.; Lake Guija, El Salvador.

Head 2.4 to 3; depth 2 to 2.25; D. XIV or XV, 12 to 14; A. V, 9 or 10; scales
28 to 30.

Body moderately deep, compressed; dorsal profile rather strongly elevated,
straight from snout to eye, then gently convex; adults of 200 millimeters and more
in length with nuchal hump; caudal peduncle short and deep, its depth 1.95 to 2.6
in head; head moderate; snout rather long and pointed, longer than postorbital
- part of head, except in very young, 2.15 to 3.2 in head; eye small, 3.6 to 4.95;
interorbital 2.5 to 3.6; preorbital as broad as eye in specimens about 160 millimeters
long, proportionately broader in larger specimens and narrower in the young;
mouth small, terminal; maxillary mostly covered by preorbital, failing to reach
anterior margin of eye, 2.5 to 3.8 in head; lower lip with an uninterrupted free
margin, the free part being very narrow at symphysis; premaxillary processes
' reaching anterior fourth of eye; teeth in the jaws in bands, mostly pointed, some
of them in the larger specimens frequently worn and blunt, the outer ones in anterior
part of jaws somewhat enlarged; gill rakers very short, 8 to 10 on lower limb of

276 ; BULLETIN OF THE BUREAU OF FISHERIES

first arch more or less developed; scales moderate, 5 or 6 rows between origin of
dorsal and lateral line, reduced in size on nape and chest, those above lateral line
with smooth edges, those below lateral line with small spinules on margin and on
portion of exposed part; dorsal fin long and continuous, the spines graduated, the
last one scarcely half the length of head, proportionately somewhat shorter in the
adult than in young; median rays of the soft portion produced in adults reaching
opposite about middle of caudal, origin of fin shghtly in advance of margin of
opercle; caudal fin rounded, sometimes more or less truncate; anal spines strong,
graduated, the last one much stronger and longer than the longest dorsal spine,
the soft portion similar to that of the dorsal, origin of fin usually about an eye’s
diameter nearer base of pectorals than base of caudal; ventral fins inserted just
behind base of pectorals, reaching to origin of anal in young, failing to reach this
point in large specimens; pectoral fins usually reaching to or a little beyond origin
of anal, 1.18 to 2.35 in head.

Color of a fresh specimen, 75 millimeters long, greenish silvery with bluish
reflections above; lower parts silvery; snout greenish; sides with 6 indistinct
dark crossbars, the third bar with a black blotch on median part of side; a black
spot on upper half of base of caudal; ventrals slightly yellowish; all other fins
olivaceous; iris yellow. In the young the black in each bar on the median part of
sides is intensified and somewhat broadened, suggesting a lateral band. In large
specimens the black bars on the sides, anteriorly at least, become very obscure,
the black blotch in the third bar disappears, and the caudal spot becomes obscure
and many of the scales on the sides bear greenish specks. Two color varieties,
‘““Mojarra negra” and “ Mojarra plateada,” which, however, show no structural
differences and which appear to intergrade, are recognized in some localities by the
natives.

This common food fish is represented by many specimens, ranging from 25 to
220 millimeters in length. The writer, however, has found it impossible to identify
the specimens with any known form and he believes them to represent a new species.
The species is related to C. macracanthus, of which small specimens were obtained in
Lake Ahuachapan. The present species, however, is more slender, the anterior
profile less steep and convex, the eye is notably smaller, and the color, while variable,
as shown in the description, is darker and the bands are more distinct. These
differences are most noticeable when specimens of like size are compared.

This mojarra is one of the most important food fishes of the fresh waters of
El Salvador, and it was taken, or reported, from nearly all waters visited. It is
said to reach a length of about 320 millimeters. In Lake Metapan it appeared to
be more common than elsewhere. A fisherman, who waded and used a cast net,
caught about 24 of these fish, ranging in length from 140 to 305 millimeters, during
4 hours’ fishing, and only a few of other species. The flesh of this fish is firm and
of good flavor.

According to information obtained from the natives, this species spawns along
shore in comparatively shallow water. A disagreement as to the time of spawning
(some saying that the species spawned in May and June, others that it spawned in
August) either shows that the spawning period is a protracted one or, more probably,
that little is known about it. The specimens dissected, which were taken during

FISHES OF EL SALVADOR DATEL

January and February, contained small ova, indicating that the spawning season
was not near at hand. The contents of four stomachs examined consisted of the
remains of fish, insects, insect larve, spicules of sponges, filaments of alge, and
fragments of higher plants.

The species is named in honor of the late Dr. Seth E. Meek, curator of fishes in
the Field Museum of Natural History, who contributed much to our knowledge of
the fishes of Mexico and Central America.

The specimens were collected in the following waters: Lake Guija, Lake
Metapan, Lake Chalchuapa, Rio Sucio at Sitio del Nino, Rio Lempa at Suchitoto
and San Marcos, Rio San Miguel at San Miguel; and Lake Olomega.

20. Cichlasoma trimaculatum (Giinther)

GuapPoTE; Mogarra; ISTATAGUA

Heros trimaculatus Giinther, Trans., Zod]. Soc., London, VI, 1868, 461, Pl. LX XVI (Chiapam and Huamuchal, Pacific slope,
Guatemala); Jordan and Evermann, Bull., U. S. Nat. Mus., XLVIT, 1898, 1529.
Cichlasoma trimaculatum Regan, Ann. and Mag. Nat. Hist., 7 ser., XVI, 1905, 333, and Biol. Cent. Amer., Pisces, 1906, 28.

Head 2.3 to 2.75; depth 2.1 to 2.45; D. XVII (rarely XVIII) 11 or 12; A. VII
(rarely VIII) 8 to 10; scales 29 to 31.

Body moderately elongate, becoming proportionately deeper with age; profile
nearly straight over the head in young, concave in adult; caudal peduncle much
deeper than long, its least depth 2.3 to 2.9 in head; head not much longer than deep;
snout rather pointed, 2.5 to 3.6 in head; eye 3.35 to 5; interorbital 2.3 to 3.8; pre-
orbital about half the diameter of eye in young (specimen 55 millimeters long),
nearly as broad as eye in adult (specimen 185 millimeters long); mouth moderate,
oblique; lower jaw projecting; maxillary partly exposed, reaching about to vertical
from anterior margin of eye, 2.65 to 3.85 in head; premaxillary process extending
about to middle of eye; lower lip with its lower margin free throughout; teeth in
the jaws pointed, the anterior pair in upper jaw enlarged, canine like, the anterior
pair in lower jaw small, the next two pairs enlarged similar to the anterior pair in
upper jaw; gill rakers short, 7 to 9 more or less developed on the lower limb of first
arch; scales rather large, somewhat reduced on nape and chest, extending on base
of vertical fins, 5 rows between origin of dorsal and lateral line, about 6 rows on
cheeks; dorsal fin long, the spines graduated, the last a little more than one-third
the length of head, the median soft rays produced, the filaments varying in length
among individuals, sometimes reaching the end of caudal fin but usually shorter,
the origin of fin over margin of opercle; caudal fin round; anal spines stronger than
the dorsal spines, the longest one longer than the last dorsal spine, about 2.6 in
head, the origin of fin about equidistant from base of pectoral and end of anal base;
ventral fins reaching past origin of anal to base of third or fourth spine in the young,
shorter in the adult, the outer ray usually more or less produced; pectoral fins reach-
ing scarcely as far back as the ventrals, 1.3 to 1.65 in head.

Color variable, a fresh specimen, 270 millimeters long, olivaceous, with a large
wine-colored area in pectoral region, extending on gill covers; sides with 5 blackish
blue blotches, the anterior one above origin of lateral line and the last on upper half
of base of caudal; fins all olivaceous, the pectorals paler than the other fins, the

278 BULLETIN OF THE BUREAU OF FISHERIES

dorsal with wine-colored spots, caudal with a broad light margin, anal with a dark
margin, ventrals with dark tips; iris brilliant red. Young, 40 millimeters long,
greenish; sides with 7 indistinct dark bars, broader than the interspaces; a black
blotch above origin of lateral line, another on median part of side above origin of
anal, a third on upper half of base of caudal; fins olivaceous, the anal a little paler
than the other fins, the median rays of ventrals dark. All of the smaller specimens
and some of the larger ones have dark bars, or indications thereof, on the sides.
Most of the larger specimens have 5 or 6 dark blotches on the sides, which in the
plainer colored specimens, having no dark bars, are reduced to 3, viz, one above
origin of lateral line, one on side above origin of anal, and the third on upper half
of base of caudal.

Many specimens, ranging from 25 to 185 millimeters in length, were preserved.
Only one specimen (185 millimeters long) has the exact color pattern described and
figured by Giinther in the original account of the species. Others, however, approach
this pattern. In structure the present species is related to the other ‘‘Guapote,”’
CO. motaguense, common in certain lakes in El Salvador. C. trimaculatum, however,
has a much deeper body, particularly in the adult. When a large series of various
sizes is measured, however, due to variation in depth with age, the extremes for
the two species, as shown in the description, overlap, but the average difference
in depth remains evident. The average depth in the length of the body in 13
specimens of C. trimaculatum, ranging in length from 40 to 185 millimeters, is 2.57.
In a similar series of C. motaquense it is 2.29. The difference in depth is not as
great among the young as it is among larger fish. For example, in 5 specimens of
C. trimaculatum, ranging in length from 127 to 185 millimeters, the range of the
depth in the length is 2.12 to 2.47 (average 2.22). In C. motaguense, in an identical
series, the range is 2.54 to 2.74 (average 2.65). In @. trimaculatum the head and
snout are somewhat narrower and more pointed and the checks are not as deep
and are provided with only about 6 rows of scales instead of about 8, as in @.
motaguense. The differences in color, as shown in the description, are pronounced.

This species, although taken in streams in a few instances, is principally a
lake fish. In Lakes Guija and Ilopango it is the most important food fish. The
fish is said to reach a length of 355 millimeters, but the largest individual seen by
us was only 270 millimeters long.

This ‘‘Guapote” is taken with hook and line, with trot lines, and with cast
nets. In Lake Ilopango, a deep clear lake, the Indians, when fishing in deep water,
dive and cast their nets under water. In the quality of its flesh this species ranks
with the other ‘‘Guapote” (C. motaguense) taken in most lakes in E] Salvador. It
undoubtedly is the most handsome of all the fresh-water fishes of El] Salvador,
having both a pretty shape and pleasing color.

According to a native fisherman at Lake Ilopango this fish spawns from August
to October in water varying from very shallow to 6 metersin depth. It builds nests,
and the eggs and young are guarded by the adults for some time after hatching.
The sexual organs in the specimens examined (taken during January) apparently
were in the early stages of development, showing that the spawning season was

FISHES OF EL SALVADOR 279

not near at hand when the collections were made. The contents of 5 stomachs
examined consisted of the remains of fish, snails, insect larve, and plant fragments.

This species heretofore has been recorded from western Guatemala. The
specimens in the present collection are from Lake Metapan, Lake Guija, Rio Lempa
at Suchitoto and San Marcos, Lake [lopango, Rio San Miguel at San Miguel, and
Lake Olomega.

21. Cichlasoma motaguense (Giinther)

GuapPotTE; Moro; Panpo

Heros motaguensis Giinther, Trans., Zod]. Soc., London, VI, 1868, 462, Pl, LX XVII, fig. 2 (Rio Motagua, Guatemala); Jordan
and Evermann, Bull., U. S. Nat. Mus., XLVII, 1898, 1534.
Cichlasoma motaguense Regan, Ann. and Mag. Nat. Hist., XVI, 1905, 336, and Biol. Cent. Amer., Pisces, 1906, 29.

Head 2.35 to 2.8; depth 2.2 to 3; D. XVII or XVIII (rarely XIX), 10 to 12;
A. VII (rarely VIID) 8 to 10; scales 30 to 32.

Body comparatively elongate; dorsal profile gently convex over the eyes in
young, a little concave in adults; caudal peduncle short, its depth 2.55 to 3 in head;
head longer than deep; snout long, its length 2.4 to 2.7 in head; eye 3.25 to 6.1;
interorbital 3 to 4; preorbital only slightly more than half the diameter of eye in
young (specimen 45 millimeters long), a little broader than eye in adult (specimen
200 millimeters long); mouth moderate, oblique; lower jaw projecting; maxillary
partly exposed, reaching about to anterior margin of eye, 2.2 to 3.4 in head; pre-
maxillary process extending to above posterior part of eye; lower lip with its lower
margin free throughout; teeth in the jaws all pointed, the anterior pair in upper
jaw enlarged, caninelike, the anterior pair in the lower jaw small, with several
enlarged teeth on each side; gill rakers short, 7 to 9 more or less developed on the
lower limb of the first arch; scales rather large, notably reduced on the chest,
extending on base of vertical fins, 5 or 6 rows between origin of dorsal and lateral
line, about 8 rows on cheeks; dorsal fin long, the spines graduated, the last one
scarcely a third the length of head, the median soft rays produced, reaching opposite
middle of caudal in large examples, shorter in young, origin of fin over opercular
margin; caudal fin broadly rounded; anal spines strong, the last one a little longer
than the last dorsal spine, the soft portion similar to that of the dorsal, origin of
fin usually about an eye’s diameter nearer base of pectorals than base of caudal;
ventral fins inserted a little behind base of pectorals, the exterior ray produced in
adult, usually extending to origin of anal; pectoral fins extending scarcely as far
back as the ventrals, 1.5 to 1.7 in head.

The color varies with age and also among individuals. A specimen, 260 milli-
meters in length, when removed from the water, possessed the following coloration:
Very dark green above; lower parts of sides lighter; underneath dusky with punc-
tulations; snout brassy; sides with black blotches forming a more or less continu-
ous lateral band; a large black blotch on upper part of opercle; a black caudal spot;
sides of head with dark brassy spots; dorsal and anal very dark green with dark
spots; caudal somewhat lighter green with dark spots; ventrals dark; pectorals
plain brassy. Color of a large specimen, 300 millimeters long, immediately after
being removed from the water, bluish silvery on back; lower part of sides pale
silvery; belly dingy white; sides with dark blotches forming an interrupted lateral

280 BULLETIN OF THE BUREAU OF FISHERIES

band; body and head, except snout, everywhere with rusty spots; pectorals plain
translucent; ventrals dusky; other fins slightly brownish and everywhere with rusty
spots, usually surrounded by blue. Young with alternating dark and light cross-
bars; the dark lateral band more pronounced than in large examples, except occa-
sionally in the very young (50 millimeters and less in length), in which it is often
quite indistinct; these young with a prominent black spot on sides below lateral
line at end of pectoral fin and a prominent caudal spot. Most specimens have a
dark bar extending from the upper posterior margin of the eye across opercle; not
quite connecting with a dark spot just above base of pectoral; another dark bar,
frequently consisting of two separate spots, extending from lower posterior margin
of eye to lower margin of gill opening.

This rather common lake fish is represented by many specimens, ranging from
25 to 300 millimeters in length. The specimens in hand agree fairly well with
published accounts of C. motaguense, the types of which are reported from the
Rio Motagua, Atlantic slope, Guatemala, and subsequently other specimens were
recorded from the ‘Pacific slope of Central America” from El Rancho, on the
Rio Motagua, and from Belize. The El Salvador specimens were taken in lakes
and ponds, not a single one having been secured in streams. The apparent differ-
ence in habitat between the type specimens of C. motaguense and the El Salvador
specimens suggests that they may not be identical. In the absence of material for
comparison it seems advisable to refer the specimens in the present collection,
tentatively at least, to C. motaguense.

This is the most common and most important food fish in Lakes Ahuachapan
and Coatepeque. In Lake Guija only one specimen was taken, and the species
was not seen among the catches of native fishermen. In Lake Chanmico a few
small individuals were taken. In three small, spring-fed reservoir ponds at El
Angel, situated on private property where the fish are protected, this was the most
common species, and the largest individual (330 millimeters in length) seen was
taken in one of the ponds.

In Lake Coatepeque this is the only food fish of importance taken. The only
other cichlid found there is the small ‘“‘burro,”’ C. nigrofasciatum. The fish in
Lake Coatepeque are taken mostly with hook and line and in rather deep water.
Several fishermen, during our visit, made fairly good catches fishing at a depth of
25 meters.

This ‘‘Guapote”’ is a good food fish, the flesh being fairly firm and of good
flavor. It reaches a larger size than any of the other cichlids, the maximum size
attained, according to a local fisherman on Lake Ahuachapan, being about 500
millimeters.

The spawning season and habits are very imperfectly or not at all known.
The specimens examined had the sexual organs undeveloped, showing that the
spawning season was not near at hand when the specimens were collected (Janu-
ary and February). The food of this fish, according to 6 stomachs examined, in
the young consists of entomostracans, insects and other small animal life, and alge.
Larger individuals feed on fish, crustaceans, and probably on plants.

FISHES OF EL SALVADOR 281

The species is recorded from the Atlantic slope of Guatemala, from British
Honduras, and from the ‘Pacific slope of Central America.” The specimens from
El Salvador were taken in Lakes Guija, Ahuachapan, Coatepeque, Chanmico, and
in small ponds at El Angel.

Order GOBIOIDEA
Family XI. ELEOTRIDA

Body elongate, slender or robust; vomerine teeth usually wanting (present in
Gobiomorus); premaxillaries protractile; opercle unarmed; orbital margin not
free, continuous with skin of head; lateral line wanting; dorsal fins 2; caudal fin
convex; ventral fins close together but separate, composed of I, 4 or I, 5 rays.

15. Genus GOBIOMORUS Lacépéde

Gobiomorus Lacépéde, Hist. Nat. Poiss., II, 1800, 583 (type Gobiomorus dormitor Lacépéde).
Philypnus Cuvier and Valenciennes, Hist. Nat. Poiss., XII, 1837, 235 (type Gobiomorus dormitor Lacépéde).
Lembus Giinther, Cat. Fish., Brit. Mus., I, 1859, 505 (type Lembus maculatus Gunther).

Body elongate, anteriorly subcylinderical, posteriorly compressed; mouth
large; lower jaw projecting; teeth small, in bands on jaws and on vomer; gill opening
large, extending forward to under eye; scales rather small, ctenoid; dorsal fins well
separated, with VI-I, 9 rays; anal fin with I, 9 to I, 11 rays. A single species was
taken in El Salvador.

22. Gobiomorus maculatus (Giinther)

GUVINA

Lembus maculatus Giinther, Cat. Fish., Brit. Mus., I, 1859, 505 (Andes of Ecuador).

Philypnus lateralis Gill, Proc., Ac. Nat. Sci., Phila., 1860 (1861), 123 (Cape San Lucas); Jordan and Evermann, Bull., U.S. Nat.
Mus., XLVII, 1898, 2195.

Eleotris lembus Giinther, Cat. Fish., Brit. Mus., III, 1861, 121 (Western Ecuador).

Gobiomorus lateralis Eigenmann and Fordice, Proc., Ac. Nat. Sci., Phil., 1885 (1886), 69.

Gobiomorus maculatus Eigenmann and Fordice, Proc., Ac. Nat. Sci., Phila., 1885 (1886), 70.

Philypnus maculatus Regan, Biol. Cent. Amer., Pisces, 1906, 5, Pl. I, fig. 2; Meek and Hildebrand, Pub., Field Mus. Nat. Hist.,
Zool. Ser. X, 1916, 352.

Head 3 to 3.4; depth 3.45 to 5.5; D. VI-I, 9; A. I, 10 or 11; scales 55 to 59.

Body elongate, not much deeper than broad anteriorly, compressed posteriorly;
caudal peduncle rather strongly compressed, its least depth 2.5 to 3 in head; head
long, somewhat depressed; snout long and broad, its length 3.05 to 3.55 in head;
eye 4 to 5.2; interorbital 3.6 to 5; mouth large, oblique; lower jaw strongly pro-
jecting; maxillary reaching middle of eye, 2.35 to 2.6 in head; teeth small, pointed,
in bands on jaws and on vomer; gill rakers minute; lateral line wanting; scales ctenoid,
extending forward on head to end of premaxillary processes; origin of spinous
dorsal about an eye’s diameter behind base of pectorals, the spines weak; origin of
soft dorsal about an eye’s diameter in advance of anal; caudal fin rounded; anal fin
similar to soft dorsal; ventral fins inserted slightly behind base of pectorals, reaching
a little more than half the distance to origin of anal; pectoral fins reaching to or a
little beyond tips of ventrals, 1.3 to 1.6 in head.

982 BULLETIN OF THE BUREAU OF FISHERIES

Color of a fresh specimen, 80 millimeters in length, olivaceous, with indefinite
and irregular dark markings above; pale underneath; sides with a black lateral
band; preopercle with 2 horizontal bands, one extending backward from upper
posterior margin of eye, the other a more definite one extending backward from
lower posterior margin of eye; a dark band underneath eye; ventral fins pale; other
fins greenish; spinous dorsal with black punctulations. In some specimens the
dorsal and caudal fins are spotted with dusky markings which are arranged in
irregular vertical bars on the caudal.

Several small specimens, ranging in length from 30 to 115 millimeters, were
taken in the Rio Lempa. One small specimen was seined at Suchitoto, and all
the others were taken at San Marcos, where the species appears to be common in
the quiet, shallow, or disconnected pools of the river. No large individuals were
seen.

This fish inhabits Pacific-slope streams from Lower California to Peru. The
specimens in the present collection are from the Rio Lempa at Suchitoto and San
Marcos.

Part II.—ANNOTATED LIST OF MARINE FISHES COLLECTED AT
THE PORTS OF TRIUNFO AND CUTUCO, EL SALVADOR

Only part of a day was devoted to collecting at each, Triunfo and Cutuco.
Triunfo is situated on a salt-water estuary lying wholly within the Departmento de
Usulutan. Cutuco is on Fonseca Bay and is one of the principal ports of El
Salvador.

Fish appeared to be rather scarce at both ports during our visit, particularly
as to variety. It is only to be expected, of course, that the abundance of fish life
varies somewhat with the season. It is probable, however, that fishing is not
very profitable, as little of it is beimg done. So far as known, no regular fisheries,
in which modern equipment is used, have been established anywhere along the
coast of El Salvador. All fishing observed by the writer was carried on with hook
and line and with the cast net, and only very small catches were made. Because
of the great scarcity of fresh-water fishes it would be a great help to the Republic
if the supply could be more abundantly augmented from the sea. It is entirely
possible that this could be done. The duty assigned to us was an investigation
of the fresh waters; the two visits to salt water were only incidental, and the time
devoted to salt-water collecting and the results obtained are too meager to form
the basis for any conclusions. The results obtained, as shown by the following
list of specics and notes, therefore, should not be interpreted to mean that profitable
marine fisheries could not be established. Furthermore, it is not known that
fishing with modern gear has been given a trial.

Family CLUPEID

1. Opisthonema libertatis (Giinther)

Meletta libertatis Giinther, Proc., Zoél. Soc., London, 1866, 603 (La Libertad).

Clupea libertatis Giinther, Cat. Fish. Brit. Mus., VII, 1868, 433.

Opisthonema libertate Jordan and Evermann, Bull., U.S. Nat. Mus., XLVII, 1896, 433; Meek and Hildebrand, Pub., Field Mus.
Nat. Hist., Zodl. Ser., XV, 1923, 188.

Several juveniles were seined at Triunfo. The species is known from Mexico
south to Panama Bay.

Family ENGRAULIDA

2. Stolephorus exiguus (Jordan and Gilbert)

Stolephorus eziguus Jordan and Gilbert, Proc., U. S. Nat. Mus., IV, 1881 (1882), 342 (Mazatlan); Jordan and Evermann, Bull.
U.S. Nat. Mus., XLVII, 1896, 442.
Anchovia erigua, Meek and Hildebrand, Pub., Field Mus. Nat. Hist., Zodl. Ser., XV, 1923, 200.

Many specimens of this species were seined both at Triunfo and Cutuco. The
species ranges from Mazatlan to Panama.
42885—25} 4 283

984 BULLETIN OF THE BUREAU OF FISHERIES

3. Stolephorus panamensis (Steindachner)

Engraulis panamensis Steindachner (Sitzb. k. Ak. Wiss. Wien., LX XII) Ichth. Beitr., IV, 1875, 39 (Panama).
Stolephorus panamensis Jordan and Evermann, Bull., U. 8. Nat. Mus., XLVII, 1896, 448.
Anchovia panamensis Meek and Hildebrand, Pub., Field Mus. Nat. Hist., Zodl. Ser., XV, 1923, 207, Pl. XV, fig. 1.

A few small specimens were seined at Triunfo.

4. Stolephorus rastralis (Gilbert and Pierson)

Stolephorus rastralis Gilbert and Pierson, in Jordan and Evermann, Bull., U. S. Nat. Mus., XLVII, 1898, 2811 (Panama).
Anchovia rastralis Meek and Hildebrand, Pub., Field Mus. Nat. Hist., Zodl. Ser., XV, 1923, 209.

Many specimens, which appear to belong to this species, were seined at Triunfo.
This fish previously was recorded only from Panama.

5. Stolephorus brevirostris (Meek and Hildebrand)

Anchovia brevirostra Meek and Hildebrand, Pub., Field Mus. Nat. Hist., Zodl. Ser., XV, 1923, 198, Pl. XII, fig. 1 (Balboa,
Panama).

Two small specimens, which appear to belong to this species, recently described
from the Pacific coast of Panama, were seined at Cutuco.

Family ATHERINIDA:

6. Thyrinops pachylepis (Gimther)

Atherinichthys pachylepis Giinther, Proc., Zo6l. Soc., London, 1864, 25, and Trans., Zoél. Soc., London, VI, 1868, 443 (Panama).

Menidia pachylepis Jordan and Evermann, Bull.,.U. S. Nat. Mus., XLVII, 1896, 801.

Thyrina pachylepis Jordan and Evermann, Bull., U.S. Nat. Mus., XLVII, 1898, 2840; Regan, Biol. Cent. Amer., Pisces, 1907, 64.

Kirtlandia pachylepis Gilbert and Starks, Memoir., Calif Ac. Sci., IV, 1904, 57.

Thyrinops pachylepis Hubbs, Proc., Ac. Nat. Sci., Phila., LXIX, 1917 (1918), 307; Jordan and Hubbs, Leland Stanford, Jr.,
Univ. Pub., Univ. Ser., 1919, 62.

Several specimens of this species were taken in Fonseca Bay at Cutuco. The
species has previously not been recorded north of the coast of Costa Rica. The
known range now extends from El Salvador to Ecuador.

Family MUGILIDA
7. Mugil curema (Cuvier and Valenciennes)

Liza; LIEBRA ANCHA

Mugil curema Cuvier and Valenciennes, Hist. Nat. Poiss., XI, 1836, 87 (Brazil; Martinique -Cuba); Jordan and Evermann.
Bull., U. S. Nat. Mus., XLVII, 1896, 813, Pl. CK XVI, fig. 344; Meek and Hildebrand, Pub., Field Mus, Nat. Hist., Zod.
Ser., XV, 1923, 279. (For a more complete synonoymy and other references see the two last-mentioned works.)

A few small specimens of this species were taken in salt water at Triunfo.
Family POLYNEMIDA:

8. Polynemus approximans (Lay and Bennett)

Polynemus approzimans Lay and Bennett, Zod]. Beechey’s Voyage, Fishes, 1849, 57 (Mazatlan); Meek and Hildebrand, Pub.,
Field Mus. Nat. Hist., Zo6l. Serv., XV, 1923, 290.
Polydactylus approzimans Jordan and Evermann, Bull., U. S. Nat. Mus., XLVII, 1896, 829.

Two small specimens were seined at Cutuco. In some localities this fish is of
considerable commercial importance. Its range extends from California to Peru.

FISHES OF EL SALVADOR 285

Family CARANGIDZE
9. Caranx hippos (Linnzus)
AUREL

Scomber hippos Linnzus, Syst. Nat. Ed., XII, 1766, 494 (Charleston, S. C.).
Caranz hippos Jordan and Evermann, Bull. U.S. Nat. Mus., XLVII, 1896, 920, Pl. CXLI, fig. 387.

Several small specimens of this common and widely distributed food fish were
seined at Triunfo and Cutuco. The species is known from the warmer waters of
both coasts of America and also from the East Indies.

10. Oligoplites mundus (Jordan and Starks)

Oligoplites mundus Jordan and Starks, in Jordan and Evermann, Rept., U. S. Fish Commission, 1896, 344, and Bull., U. S. Nat.
Mus., XLVII, 1898, 2844 (Mazatlan).

A single small specimen was seined at Triunfo. The species ranges from the
Gulf of California to Ecuador.
11. Oligoplites saurus (Bloch and Schneider)

Scomber saurus Bloch and Schneider, Syst. Ichth., 1801, 321 (Jamaica).
Oligoplites saurus Jordan and Evermann, Bull., U.S. Nat. Mus., XLVII, 1896, 898.

Three small specimens were seined at Triunfo. The species is of no commercial
value. It occurs on both coasts of tropical America.

Family CENTROPOMID/E

12. Centropomus pectinatus (Poey)
Rosato; RovaLo

Centropomus unidecimalis Cuvier and Valenciennes (part), Hist. Nat. Poiss., II, 1828, 102.
Centropomus pectinatus Poey, Memorias, II, 1860, 121 (Cuba); Jordan and Evermann, Bull., U.S. Nat. Mus., XLVII, 1896, 1122.

Two specimens, respectively 220 and 280 millimeters in length, were taken
at Triunfo. A large specimen, 350 millimeters long, was taken in fresh water in
the Rio Lempa at San Marcos.

13. Centropomus robalito (Jordan and Gilbert)
Rosato; RovaLo

Centropomus armatus Giinther (not Gill), Trans., Zod]. Soc., London, 1868, 408.
Centropomus robalito Jordan and Gilbert, Proc., U. S. Nat. Mus., IV, 1881 (1882), 462 (Mazatlan; Acapulco); Jordan and
Evermann, Bull., U.S. Nat. Mus., XLVII, 1896, 1123.

A single small specimen was secured at Triunfo and several others in fresh
water in the Rio Lempa at San Marcos.

Family EPINEPHELIDA

14. Epinephelus analogus (Gill)

PARGO TIGRE

Epinephelus analogus Gill, Proc., Ac. Nat. Sci., Phila., 1864, 163 (Panama); Jordan and Evermann, Bull., U. S. Nat. Mus.,

XLVII, 1898, 1152.
Serranus courtadii Bocourt, Ann. Sci. Nat., Paris, 1868, 222 (La Union, San Salvador).

A single specimen, 235 millimeters in length, was taken at Cutuco. Therange
of the species extends from Mexico to Panama and the Galapagos and Revillagigedo
Islands.

286 BULLETIN OF THE BUREAU OF FISHERIES

Family LUTIANID/E

15. Lutianus argentiventris (Peters)

Parco; Parvo

Mesoprion argentiventris Peters, Berlin. Monatsber., 1869, 704 (Mazatlan).
Neomenis argentiventris Jordan and Evermann, Bull., U.S. Nat. Mus., XLVII, 1898, 1260.

Several individuals of this species were killed with dynamite in the estuary at
Triunfo, where this snapper appears to be common. It was not seen in fresh
water. The species is known from Lower California to Ecuador.

16. Lutianus novemfasciatus (Gill)

Parco; Parvo

Lutianus novemfasciatus Gill, Proc., Ac. Nat. Sci., Phila., 1862, (1863), 251 (Cape San Lucas).
Neomaenis novemfasciatus Jordan and Evermann, Bull., U.S. Nat. Mus., XLVII, 1898, 1252.

Many individuals of this species were killed with dynamite in the estuary at
Triunfo. One large individual, 475 millimeters in length, which could not at the
time be preserved, was taken in the same way in strictly fresh water in the Rio
Lempa at San Marcos. Measurements, counts, and a color description were made
in the field, and it is believed that this fish belonged to the present species. This
snapper appears to be one of the most plentiful food fishes in the estuary at Triunfo.
The species ranges from Lower California to Colombia.

Family POMADASID
17. Orthopristis chalceus (Giinther)

Roncan

Pristipoma chalceum Giinther, Proc., Zo6l. Soc., London, 1864, 146, and Trans., Zodl. Soc., London, VI, 1868, 415 (Panama Bay).
Pristipoma kneri Steindachner (Sitzb. k. Ak. Wiss. Wien, LX) Ichth. Notiz., VIII, 1869, 3, Pl. II (Mazatlan).
Pristipoma chalceus Jordan and Evermann, Bull., U.S, Nat. Mus., XLVII, 1898, 1337.

A single specimen, 185 millimeters in length, was secured at Triunfo. This
is an important food fish in some sections on the Pacific coast of Central America.
Its range extends from Lower California to Panama and the Galapagos Islands.

18. Pomadasis panamensis (Steindachner)

Pristipoma panamense Steindachner (Sitzb. k. Ak. Wiss. Wien., LX XII) Ichth. Beit., III, 1875, 8, Pl. I, fig. 1 (Panama Bay).
Pomadasis panamensis Jordan and Evermann, Bull., U.S. Nat. Mus., XLVII, 1898, 1331.

A single specimen, 170 millimeters long, was secured at Triunfo. Most of the
representatives of this genus frequent fresh water, but it is not known that the
present species enters streams. The species ranges from Guaymas to Panama.

FISHES OF EL SALVADOR 287

19. Anisotremus dovii (Giinther)

BERNEGATE

Pristipoma dovii Giinther, Proc., Zo6l, Soc., London, 1864, 23, Pl. III, fig. 1,and Trans., Zo6l. Soc., London, VI, 1868, 414 (Panama),
Anisotremus dovii Jordan and Evermann, Bull., U. 8, Nat. Mus., XLVII, 1898, 1317.

A single specimen, 165 millimeters in length, was taken at Cutuco. It was the
only one among a dozen of its nearest relative, A. pacifici. The species ranges
from Mazatlan to Panama.

20. Anisotremus pacifici (Giinther)

BERNEGATE

Conodon pacifici Giinther, Proc., Zodl. Soc., London, 1864, 147, and Trans. Zool. Soc., London, VI, 1868, 417, Pl. LXIV, fig. 3
(“Chiapam,”’ Pacific coast of Guatemala).
Anisotremus pacifici Jordan and Evermann, Bull., U.S. Nat. Mus. XLVII, 1898, 1316.

Several specimens of this species were seen at Triunfo and Cutuco, and three
were preserved. It appears to be among the common food fishes in Fonseca Bay.
The range of this species extends from Guatemala to Guayaquil.

Family GERRIDA

21. Gerres peruvianus Cuvier and Valenciennes

MoJARRA

Gerres peruvianus Cuvier and Valenciennes, Hist. Nat. Poiss., VI, 1830, 467 (Payta, Northern Peru); Jordan and Evermann
Bull., U.S. Nat. Mus., XLVII, 1898, 1376.

Several small specimens were seined at Triunfo, where the species appears
to be common. Its range extends from Mazatlan to northern Peru.

Family EPHIPPIDAZ
22. Chzetodipterus zonatus (Gerard)

CHOPA

Ephippus zonatus Girard, Exp. Surv. R. R. Route, Miss. R. to Pac. Ocean, 1858, 110 (San Diego, Calif.).
Chetodipterus zonatus Jordan and Evermann, Bull., U.S. Nat. Mus., XLVII, 1898, 1668.

A single specimen, 270 millimeters in length, was taken at Triunfo. The
species ranges from California to Ecuador.

Family GOBIIDAE
23. Gobionellus sagittula (Giinther)

Euctenogobius sagittula Giinther, Proc., Zo6l. Soc., London, 1861, 3 (west coast of Central America).

Gobius longicaudus Jenkins and Evermann, Proc., U.S. Nat. Mus., XI, 1888, (1889), 146 (Guaymas).

Gobionellus sagittula Jordan and Evermann, Bull., U. S. Nat. Mus., XLVII, 1898, 2228; Gilbert and Starks, Memoir., Cal. Ac.
Sci., IV, 1904, 171.

Seven fine specimens, ranging in length from 32 to 78 millimeters, were seined
in the estuary at Triunfo. The species ranges from the Gulf of California south-
ward to Ecuador.

&

Tas | er Ses HOCAT1A8 Id 10 2aMergsc 24a) :
(sod tgthd). tivob aumaastouind OF
ee Cur C lM ee He bs he a

é a
"

sfurtaine DT) 21b Boe! TV ae ee ati léo8., Asan he, t sath Pea Ok (3 fi Sie ohanl 908 wee Shae
US PON I AA ee Ja? 0 IGT necator has aabso!

af} env jl onda) de nednd ase Lise a. neratoscilti im GOL flagtingde | ;
ejspant edissge afl toiseng ho avits abet Jhetasnt ii lo nasob # yoom.
arate if p . th of ne

\ he OF

igltaD) iofliosg aurnortoaic OS...

ail
sat

AEA pos nam

& gh VIED AD Sib eos! AV oboe) ,.scd look anwk? haa git tat pee 008 quo, 20o1t “sini.
; Agtemotan) i tesco Moet
JOSE BRE HVEE & mat tat 6.0 Mud wuareorey Hh bas aut ot i

eon bas ooude> bas otauitl In Mose otw 2aiseqe aiih)) esomioaq? Ler
iii wooed iri eoileit boot nomics oft unonin sd oF Pthogqa af .baytsas
limpet out ab roetiaaetc al pe: abodtes & 2

AIAAIO. wilicastA

ooeunsiogiot T bows tai) aunsivureg novi) . £ cone
) “6 TARY

aa aaLol ;

gsueqssve bis aabyot: tied wiadtsio vi ai yad) Tor net LY lot tavs Bread atuniousts ¥ tua ioteat :
OTe veh LIVIX 80M ta 2

ir
#1L9 6 esineq? adit o1scdw etaniyD ja boned etaw- eflominggs Hoc
wT madiion of meltessM. mort chagsx9 szaer ail

‘ BOM ylicsel

ree) ate rox ee so

A (isd oysiG neB) ATi 288) rene 0 vat os Ta! i008 ALB. ‘ec deaeaet

BBY AOBE ca VAX eh Ae Ue ue eee se Es ASOT AMINA
-9AT otavirt is aadas enn fienal, al etotomillica ore _ faniipege. alum
sebanoal ot moa) ei aogt

“RANGOD viet

us idles TO ES Ay BRIT

ed Godan) nlustiges walloasideD' eS

* gotiniy A Lavine!) lo teKoo teew) & Het, wobriod 505: ‘fio A er if) ‘ofwttiyes
y ean qauT) abt eeael) kee ote u.B) 4d cert pfaeariey Mt baadn
oA dad, aiownots etna baa frodlii jHORE ecu ay cult 186 i v; ding ~Anacgwe gt | be op

Mipetad

benies stow se aioaaiittin 4V of C8 mort digaal Mi Qcivent enantivoge owst
-divoe siatolila) to te) edi mort eavaey coinage od’ olan da

- FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS,
1897 TO 1922

5
By LEONHARD STEJNEGER
Head Curator of Biology, United States National Museum

&
CONTENTS

Page Page

Iminoduction=2: 22.2 See hoo 289 | The Russian fur-seal islands—Continued.

Investigation of the Commander Islands History of the Commander Islands,

MM MORQN Ms MAW MeO Vie boo 292 etc.—Continued.

Ttinemary. 281 G4ueosiiisuss ile a 292 Conditions at the end of 1897_-_--- 299

The Russian fur-seal islands_-_-------- 293 Conditions after 1897 until the
Condition of the Commander Islands treaty of, 1910. 220-425-268 301
male We ta ek ee a 293 Pelagic sealing. ~3--454555---- 301
berinopislang on 8st oS. e 293 Sealing industry on land_------- 304
Reef, North Rookery, in 1922____ 293 Condition of the rookeriesin1910. 308
Kishotchnoye Rookery-_-_.------- 296 The treaty of 1911___------------ 312

South Rookery or Poludionnoye The Commander Islands rookeries
Rookery sie chlec_ do ses 3 296 iN}dMOMESiS iL i ade 8 314
Copper Island. ._---------------- 296 Rookery raids in 1911____------ 316

Glinka rookeries_-------------- 296 Period from 1912 to 1917, the 5-year
Karabelnoye rookeries__-------- 297 Tapuskien esse cela ee, 318
New hauling grounds_--_---------- 297 Killing resumed in 1917_-_-------- 322
Summary of conditions in 1922___._ 298 The Commander Islands after 1917. 327

Explanation of conditions_~_-__-__-_-- 298 Number of seals killed on the Com-

History of the Commander Islands mander Islands between 1917 and
fur-seal rookeries and fur-seal in- MOQQHe ree sabe wit pow. @ 330
dustry since 1897__._.--_------- 299).| ,Conclusionss..,os2-ecebece jock sce best 331

INTRODUCTION

In the spring of 1882 Prof. Spencer F. Baird, secretary of the Smithsonian
Institution and United States Commissioner of Fish and Fisheries, sent the writer
to the Commander Islands, in the North Pacific Ocean off the coast of Kamchatka,
for the purpose of collecting specimens for the United States National Museum
and to investigate the natural history of the islands in general. Bering and Copper
Islands were visited from May, 1882, to September, 1883, and it was thus possible,
during two full seasons, to study the fur seals and the fur-seal industry of these
islands and their distribution and to make maps and sketches of the seal rookeries.
At that time pelagic sealing had not yet begun and the Commander Islands, which
belong to Russia, were at the zenith of their productivity, the take of sealskins during

289

990 BULLETIN OF THE BUREAU OF FISHERIES

the years 1880 to 1889 averaging 44,363 per year. The insight into the fur-seal
industry thus gained during its most prosperous period was of great value later.
With the year 1892 began the terrible onslaught of pelagic sealers on the seal herd
of the Commander Islands, which resulted in,an average loss of bout, “400,000
seals a year in excess of the normal mortality: of. the herd.

In 1895 the United States Fish Commission, wishing to obtain reliable informa-
tion as to the status of the fur-seal industry if the North Pacific, commissioned
Dr. F. W. True and the writer to proceed to the seal islands, Doctor True to report
on the Pribilof Islands while to this author was allotted the investigation of the
Commander Islands. The stay on the Commander Islands lasted from July 3 to
September 16, during which time all the rookeries were thoroughly investigated,
photographed, and mapped. As a result of this trip there was published in the
bulletin of the United States Fish Commission (Vol. XVI, 1896) the writer’s report
on The Russian Fur-Seal Islands * (148 pp, and 66 pls., including maps of, both
islands, and every rookery, showing the distribution.of the seals in 1883 and 1895).

As a member ofthe, Commission on) Fur-Seal Investigations appointed by
President Cleveland in, 1896, to).act in conjunction with commissioners appointed
by Great Britain and:Canada with special reference to the effect of the regulations
made by the Paris Arbitration Tribunal for the protection of thé seal herd, the writer
again visited the Pribilof Islands, spending 10 days there. The Commander Tslands
were next, visited, and the rookeries on both islands were inspected and photographed
between July 30 ‘and August 8, after which the writer proceeded.to the Kuril
Islands belonging, to Japan, On. August 29 and 21, the, then Russian seal rookery,
‘Robben Island, in_ the Okhotsk Sea off the east coast of Sakhalin, was«anspected,
‘photographed, and mapped. During the summer of 1896, therefore; practically all
of thé fur-seal rookeries of the North Pacific, Bering Sea; and Okhotsk Sea were
‘inspected.

In. 1897 the writer again inspected the. Russian islands. , The entire sealing
season, from July, 7 to September 2, was spent on Copper. Island and Bering Island
studying the various problems with ‘special reference to the papier of pelagiqn rein
‘on the greatly reduced Russian herd. 1 les

During five sealing seasons on ‘the Commander Islands it nig thus ‘been ‘ idssible
for the writer to gain an intimate knowledge of the fur-seal question, not only
during the period of greatest expansion but also during the period of greatest decline
due to the nefarious practice of pelagic, slaughter by foreign sealers.

“***The final report on this work was published in volume 4 ‘of the lage Report
of ‘the Fur-Seal Investigations, 1896-97, as Part IV of The Fur Seals and Fur-Seal
Tslands of the North Paciife by David Stott Fordan Bhd associates, under nese
‘The Asiatic Fur-Seal Islands and Fur-Seal Industry (1898, 384’ pp., 113 pis. inelad-
ing diagrams and maps). In ‘this report, which will be referred to’ hereafter Merdh
as The ‘Asiatic’ Fur-Seal Islands, a detailed description is given ‘of the islands)'th
flora and fauna; the inliabitants, the fur-seal industry and its history on the Russian
islands from theib first discovery in/1741 to and including 1897, ‘accompanied® by
photographs and maps‘of?all the ‘rookeries showing the distribution of ‘the Seals! in
1882483: and 1895-1897. The’report ‘also includes a full account ‘of the Japanese
fur-seal islands and pelagic sealing on the Japanese and Russian coasts.

1 This report was reprinted in Alaska Industries, House Document No. 92, part 4, 1898, pp. 613-754, pls. 1-67.

FUR-SEAL, INDUSTRY OF, THE COMMANDER, ISLANDS 291

The investigations, of the commission, indisputably demonstrated the. fact
that the disastrous. decline of the various seal herds was due to pelagic sealing alone,
It “was made equally evident that, the measures of protection devised by ane Paris
tribunal for the conservation of the seals were utterly inadequate and that if relief
were not afforded soon the seal herds would eventually be destroyed commercially.
The regulations made by the Paris tribunal legalized pelagic sealing during the migra-
tions of the seals in the Pacific Ocean apie of Bering Sea. The Amarioan side of
the latter was closed to all pelagic sealing until August 1, after which date it was
permitted up to within 60 miles of the Pe bilor Islands, it beg supposed that, this
time limit and zone sufficed for the protection of the female seals during the breeding
season. It should be noted that the regulations of the Paris tribunal related only
to the Pribilofs, the American islands, and not to the Russian.or Commander Islands,
over which the tribunal had no jurisdiction. Russia, in the meantime, negotiated
a separate treaty with Great Britain for the protection of the herd of the Com-
mazuder Islands: but illusory as was the American protection, that devised for the
Russian islands was even worse, for the protecting zone was fixed at 30 miles around
the. Commander Islands with no restrictive time limit during the breeding season.

Tn the meantime the annual yield from the killing of peonete seals on land and
breeding females at sea was gradually dwindling, and a flourishing industry, the
income He which had more pie reimbursed the United States for the entire
ments of His initod States, "Gea Britain, Russia, al Japan finally realized mia
something had to be done. The negotiations that rela resulted in the abolition
of legal pelagic sealing by the nationals of the four powers in the North Pacific
and adjacent seas, a convention to that effect being concluded in Washington on
July 7, 1911.
~The beneficial effect. of this treaty became apparent almost immediately on
the Pribilof Islands. Despite the cessation of commercial land killing for five years,
which retarded the rehabilitation of the herd, the number of seals ered from
about 216,000 in 1912 to approximately 605,000 in 1922. Returns from the Japan-
ese Gorcnamient of the 10 per cent of the Mie taken on Robben Island indicated
similarly improved conditions there. Only from the Russian islands no authentic
information was forthcoming. The question as to whether this lack of information
was due solely to the disttacted condition of the Far East after the Russian revolu-
tion or whether other circumstances were responsible naturally arose, in view of the
fact that the treaty of July 7, 1911, was to continue in force for a period of 15 years
from December 15, 1911, aa thereafter until terminated by 12 months’ written
notice, which notice aoe be given at the expiration of 14 years—consequently on
or after. December 15,1925. Wane’ in 1922, therefore, the Department of Commerce
decided to make an investigation of conditions existing there.

Because of the writer’s previous experience in areseall matters in Asiatic waters
he was detailed by the United States National Museum to conduct, the investigation
for the Department of Commerce, and had to aid him Capt. Carl E. Lindquist, of
Oakland, Calif., whose 14 years of service on various vessels in Asiatic waters and
Poattiarity oak sealing conditions made his assistance invaluable.

292 BULLETIN OF THE BUREAU OF FISHERIES

The writer wishes to express his thanks to Mr. Henry O’Malley, Commissioner
of Fisheries, who made possible this visit to the Commander Islands, and who
placed at our disposal every facility for visiting the various rookeries and for study-
ing the fur-seal question from all angles, as well as securing for our assistance the
services of Capt. C. E. Lindquist. ‘To my old friend Captain Lindquist I also express
my appreciation and thanks, and I gratefully acknowledge the help and courtesies
received from all members of the crews of the Coast Guard cutters Mojave and
Algonquin, on which transportation was had to and from the islands, and particularly
from Ward T. Bower, of the Bureau of Fisheries, who contributed so much to the
success of the investigation.

INVESTIGATION OF THE COMMANDER ISLANDS IN 1922
- ITINERARY

According to the original plan the investigation was to have continued for
about one month, but because of bad weather, delays in transportation, etc., only
16 days in all were spent on the islands, during most of which time conditions
were unfavorable for photographing and inspecting the rookeries.

The first place visited was Glinka, the southern rookery village of Copper
Island, which was reached on July 24, 1922, on the Coast Guard cutter Algonquin.
No officials were located at the village, only a rookery guard of 10 natives. These
we interrogated as to the condition of the Glinka rookeries, intimating that it was
desired to inspect them, but the guards were unwilling to allow this without a
permit from the fisheries official located at the main village. We therefore pro-
ceeded to that point and received permission to go ashore from the government
manager of the fur industry of the islands. There being no sealing, all the
inhabitants were assembled in the village and we were met with the greatest
hospitality and good will. The condition of the natives was better than we had
anticipated, though we found them lacking in good clothing, especially shoes, and
also in certain provisions.

The authority recognized by the officials and inhabitants was that of the
Merkulof government at Vladivostok and of the governor of the Kamchatka district,
and the old white-blue-red flag of Imperial Russia was displayed at the signal bluff
above the village. We were told that a number of Japanese seal poachers were
being held in custody, having been taken prisoner the day before, and their schooner,
which had been seized, was seen lying at anchor in the little cove of the village.

As the weather absolutely prevented any rookery work on Copper Island we
decided to proceed as soon as possible to Bering Island. The island manager
accompanied us in order to authorize our stay at that island, and toward evening
on July 26 we landed at Nikolski, the main village.

On July 28 North Rookery was inspected and, as far as the rain and gale would
allow, photographed. The weather continued rainy with heavy fog until August 1,
but the time was not entirely lost as the writer was given free access to the official
archives from the records of which valuable extracts were made relating to the
fur-seal industry of the islands during the 25 years since our last visit.

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 993

The fog continued to hang so low over the island that it was impossible to
make further observations on the rookeries, and on August 8 we departed on the
Coast Guard cutter Mojave. Robben Island, in the Okhotsk Sea, was visited on
August 11, and on August 20 we arrived at Hakodate, where the firm that had
entered into a contract with the Vladivostok government for the handling of the
Commander Islands furs for the next three years and the bap Petre of the
inhabitants was interviewed.

On August 30 a brief interview was had at Tokyo with the fur expert who
visited the Commander Islands:in 1915 and 1916 and inspected the seal rookeries:
there, and in Yokohama much valuable informatién was obtained from a former
administrator of the Commander Islands, who served from 1907 to 1917.

THE RUSSIAN FUR-SEAL ISLANDS

In 1898 the fur-seal islands still belonging to the Russian Crown consisted of
the Commander Islands, located off Kamchatka between the Pacific Ocean and
Bering Sea, and Robben Island in the Okhotsk Sea. The latter island was ceded to
Japan in the treaty of peace following the Russo-Japanese War of 1904-5. There
remain, therefore, in Russian possession. at the present date only the Commander

Islands.
CONDITION OF THE COMMANDER ISLANDS IN 1922

The Commander or Komandorski Islands, so named after Commander Bering,

who discovered the group in 1741, comprise two main islands—Bering and Copper
(Miedni)—situated off the east coast of Kamchatka between 54° 33’ and 55° 22’
north latitude and 165° 40’ and 168° 9’ east longitude, about 750 miles west of the
‘Pribilof Islands. Bering Island is about 50 miles long by an average of 10 miles
wide, and Copper Island 30 miles long by 2 miles wide on the average. Both
islands are very mountainous, the altitude of the highest peak on Bering Island,
Mount Steller, being given as 670 meters ? (2,198 feet), and that on Copper Island,
Mount Stejneger; as 637 meters ? (2,090 feet).

BERING ISLAND

During my former visits the seals on Bering Island were distributed between
two rookeries—great North Rookery, situated on the northernmost prolongation
of the island (Severni Mys or Cape Yushin), and South Rookery, situated about
midway on the western coast of the island—but the latter rookery is now extinct.

Reef, North Rookery, in 1922.—This rookery was formerly the largest and most
important on the Commander Islands and has, perhaps, received the greatest at-
tention. The inspection in 1922 took place on July 28.

Upon arrival at the rookery great changes from former days were noticed.
Several of the old buildings were missing, among them the small house of the sealer
near the salt house, as well as the house marked on the 1895 map as the “cossack’s
house.” Nearly every trace of the “abandoned village” of the same map had dis-

2 Fedtschenko, Boris: Flore des fles du Commandeur, 1906, p. 1. Cracovie.
* Morozewicz, J.: Mémoires du Comité Géologique (N. 8.), Pt. 72, 1912, p. 45.

294. “LM BULLETIN OF ‘THE BUREAU OF FISHERIES:

appeared, and many of the ‘sod huts of the “new village’’ were in ruins.! In ‘their
place'a large‘frame house had been ‘erected by the Government just back of the old:
village for’ occupancy by the overseer. and other officials, evidently: built:at a time’
when ‘the rookery was'in a’ more’ flourishing condition | but! now plainly showimg
evidence of decay.’ As no sealing was going on at this’ season, the native population,
was congregated: at thd) Shrenws village, nearly 7 miles farther east,: busy: with
catching and drying salmon for winter food, and only a guard of a dozen young men
was’ located at the rookery village. The salt house, which contained nothing but
a small quantity ‘of old:and discolored salt, was in a very dilapidated condition.
The floor was'covered with mold and slime and ‘bad rotted ai, in Sai sa roof;
was leaking badly) °° « tretaionbe
The decay of the fquildines was as aie imeem competed ai the aoe.
lation presented to view from the salt-house platform overlooking the rookery, from
which the writer’s, first sketch of the teeming masses of seals was made just 40
years before (July, 30, 1882: see plate 20, Asiatic Fur- Seal Islands). . At first, onl
a few straggling bulls Esieh be discovered on the main rookery and a handful ©)
seals on the, Sivutchi Rocks, Presently, as, the overseer pointed them out, a ‘thin
line of seals,was observed above the grass to the left. In the beginning some diffi-
culty was experienced i in becoming oriented, as the first things looked for 7 were the
white ‘‘sands”’ and ‘(parade grounds,”” ncherc the breeding harems used to be ‘lying
in thick masses with the characteristic black “band” of seals obliquely across the
sandy, gently rounded peninsula’ (fig.-1)..| To the writer’s astonishment it bécame
apparent that this entire space, except a very narrow fringe of whitish sand along
the northern edge, was overgrown witha tall, dense growth of /coarse grass (Ely-
tus), and’ that not only was this: grand rookery depleted almost. beyond. belief,
but‘it must have been so fora bonkiderfichla period for this cal sisnialed fo |have
become so ‘densely overgrown with: vegetation. in, obive
Wedescended on the:rookery. Tt, was low: tide; pee we ales out fat the) large
rocks directly north of the peninsula without seeing a cow or pup seal, over ground
where formerly thousands of seals bred and where it would have been. impossible to
have walked except in the spring before the seals had arrived or during the sealing
season when taking part in a drive:, Occasionally a roaming bull or half bull, of
which there seemed to be a great superfluity, would amble across our path appar-
ently i in search of COWS. tT BY this'time the wind had increased’ almost to a gale so
that it waslimposs sible! to'sét ‘upp’ a’tripod’ and eamera,’ and » few instantaneous
exposures With thesmall kodak were the only photographs that could be! secured,
The extent’of'the are covered ‘by the seals was sketched im on'the map (fig. 8) + and
an’ ‘attempt was then madé ‘to estimate the number of seals present on the rookery.
‘The first location of our observation: point’ bemg found very) unsatisfactory
(fig. 3) >, we proceeded cautiously to'the western ‘edge ‘of the reef, and, there gamed
a ‘somewhat better view (fig.:4)! -Am attempt was then made,to,count the bulls,
at least, but the count was quite iliusory.. It is certain that among the, females
there were many ‘bulls which! we)could, not and did not, see. At a rough esti,
mate, ‘based;on the» vatious.‘‘counts’’, by. the. writer, Captain Lindquist IR

4 The accuracy of this has since been verified from ‘the photographs:

5 This was chiefly because only a few seals’ could ‘be.seen, For comparison with conditions in 1895 at 1897. shire Ee
from nearly the same standpoint are added (figs. 5 and 6),

Buuu. U. S. B. F., 1925. (Doc. 986)

Fic. 1.—North Rookery, Bering Island, July 13, 1897, 1.30 p. m., from driveway. Sunshine

Fic. 2.—North Rookery, Bering Island, July 28, 1922, 1.30 p. m., from same standpoint as above Rain

Apno[O ‘1 eNSiyY Ul UMOYS [ ‘ON Uorze}s OJOYd MIOIA “Ud O¢'ZT ‘ZZ6T ‘8z AINE ‘puLIsy BuIIeg ‘AIOYOOY YWON—'e “Old

cae ee

ee a

aie nigel = ee RT

(986 (90) ‘Sc6T “A “A “S ‘A ‘TIng

uley "LOANS Y Ul UMOYsS Z “ON UOreys OJoYd WOIA “We'd T ‘ZZ6r ‘Rg AINE ‘puL[s] BuLog ‘A1ayooYy YWON—'F ‘OL

(986 90d) “S26 “A “A “S “A “TING

€ ONS Sv YUIOdpuLys ouIvs OY} ALIvou WOT ‘Ud { ‘/68T ‘EI Ale ‘pUR[S] SUIIeg ‘A19yOOY YON JO WeG—s ‘ong

(986 90) “S26T “A “A “8S “A “ITN

¢ DANS Sv ULOdpURAS OUILS MIOIY —GGRT ‘PURIST BuLlog ‘A1oyooY YON Jo wWweg—9 “oy

(986 90) “S26I “A “@ “S “A ‘TINg

AJOYOOI JO pUd AOU WOT PITY} UIOYQION “ul 'd Zz ‘2681 ‘EI Ale ‘puULIs] SuIIog ‘A1ayooy sAouyooysryy— ) “OI

(986 99d) “S@6T “A “A “8S “A “TING

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 295

Pyeshkof, and the native overseer, about 100 bulls, ‘‘idle’’ or otherwise, were actu-
ally seen, but it would not be surprising if there were as many more lying down
among the females or hidden behind rocks which escaped our observation. _ After
a reconsideration of all the various factors involved, we reaffirm our first im-
pression that the total number of seals, exclusive of pups, on North Rookery on
July 28, 1922, did not equal the rough count of seals on Kishotchnoye Rookery

Distribution of Seals 922

SSS., - +x. Sivutchi Kamen
ee i
Large field of Ke 5
Gf ee a ee fern
VAP,
4
4 Bug
y le, .

3
Mali Sivulehi Kamen ww

Pal. Sta. £71 tN 10
Reef Rookery i

Rf
Ned
2 . é

of
North Rookery Bering Island
Learihordl Steger

Scale
‘. MILE
(2s 0 SAZH-

‘ Fic. 8.—North Rookery, Bering Island, showing « distribution of seals July 28, 1922, according to Stejneger

on July 16, 1897, viz, 3,000 (Asiatic Fur-Seal Islands, p. 166). The seals were
lying in a comparatively thin belt along the western edge of the Reef proper, from
the extreme northern point of the “peninsula” halfway down toward the large
rock marked on the map as Babin. In addition, two harems were seen on the
outlying rock known as Little (Mali) Sivutchi Kamen, while a+ few others,
probably less than a dozen, were located on the outer or northern side of Sivutchi
Kamen itself (fig. 8).

42279—25 }|—_2

996 BULLETIN OF THE BUREAU OF FISHERIES |

Returning from the rookery a final attempt was made to take a’ photograph
from the identical station and with the same lens employed in 1897, but owing to
the fierce wind and the rain the plate shows nothing beyond the fact that the grass
Hae extends over the entire area formerly showing up ee ie “sands” ‘and

‘parade grounds” (fig. 2). Ae

Kishotehnoye Rookery.—The chapter relating to the fur seals on this and many
other Commander Islands rookeries must now necessarily read like the famous
chapter on reptiles in Iceland: There are none. Even as late as 1910 there Meet
still a few seals left on this beach, which in 1882 was continuous with the Reef
Rookery, but the seals have long since left it completely (fig. 8).

South Rookery or Poludionnoye Rookery.—This rookery, which was lorie on
the west coast of Bering Island about 20 miles south of Nikolski village, is long since
extinct, as was to be expected. In 1897, when the seals on fae) al rookery
were courted, there were only 2 able-bodied harem bulls and about 526 cows.
The rookery, as such, ceased to exist shortly thereafter.

COPPER ISLAND

A visit to the Copper Island rookeries was prevented by the unfavorable
weather conditions, the shortness of the time available, and the lack of trans-
portation. The account given the writer by the overseer at Glinka, however,
was sufficiently detailed to present a fairly reliable picture of the actual situ-
ation. What was said of the conditions there seemed to us, who had not visited
the rookeries in question since 1897, almost incredible in spite of our faith in ‘the
reliability of our informant, but his account received a startling confirmation
when we witnessed the destruction that had taken place on Bering Island.! It
was a shock to learn that the big complex of rookeries known as the Karabelnoye
Rookeries had been totally wiped out, not a single seal: remaining, and that ithe
Karabelnoye village had been abandined years ago.

The fur-seal rookeries on Copper Island were formerly. distributed in'two large
groups on the west coast of the island with the corresponding sealing villages and
salt houses on the opposite side, viz, a northern group (Karabelni) about 814 nautical
miles south of the main yillage (Proobeheaceanay and Glinka, about the same
distance: farthér. south, near the southeastern end of the island... The capacity of
these rookeries at fate time was in excess of the Bering Island fookeries, and
Karabelni’s was’ slightly less than/one-half that of Glinka. For.the sake of uni-
formity the scant data available are given under their separate headings.

Glinka rookeries.—The Glinkarookeries consist of a\string of more or less
isolated beaches, each with a distinct rookery name, spreading over a stretch of
coast approximately 5 statute miles long. * From nonth, to south the principal fea-
tures. are named as follows: Lebiazhe, Urili, Zapadnoye, Sabatcha Dira,, Palata,
Zapalata, Sikatchinskoye (Vodopad), (awmusherata and Babitche. In 1897, the
first and the last were inhabited by bachelors only, and the others ranked in impor-
tance in the following order: Palata, Urili, Zapalata, Zapadnoye, Sikatchinskoye,
Gavarushetche, and Sabatcha Dira. In 1910, according to Suvorof, the sequence

FUR-SEAL INDUSTRY; OF THE COMMANDER, ISLANDS 997

was: Zapalata, Urili, Lebiazhe, Zapadnoye, Sikatchinskoye, Palata, Babitehe,
and Gavarushetche, while Sabatcha’ Dira had disappeared as.a,rookery. .

In 1922 the seals had. practically disappeared south of Palata. Where 25
years ago they were in evidence at Sikatchinskaya and. other, places, as!shown! by
Figure,9, now there was none. | Palata itself, one of the grandest views on the island °
(Plate 72, Asiatic Fur-Seal Islands), had shrunk 'to,a handful of seals... Khabarof
mentioned 25 big bulls and about 35 cows. Zapadnoye also had vanished, ,but
Urili and Lebiazhe were still holding, their own; that is, in comparison tit the
last. 10 years or more. ;

Karabelnoye Rookeries.—The rookeries opposite Klaanbelaane alae were more
contiguous than those at Glinka (see Plate 99, Asiatic Fur-Seal Islands) and extended
not more than about 114 statute miles along the coast... Most of the coves under
the overhanging clifis were difficult of access from the land side and. individually
less significant than the Glinka rookeries, hence they have always been. treated
of as a whole and no estimate made of the seals occupying each cove or rock. | In
1896 I thought there were still 10,000, breeding seals present on those beaches. In
1910 there were still some left at Nerpitcha Bukhta (or Nerpitchi Kamen), for
Suvorof estimated the number present at 3 bulls and 150 cows. In 1911 the last
seal disappeared, and in 1922 Karabelnoye rookery was ancient history.

By sh NEW HAULING GROUNDS

There has always, been a tendency among the Commander Islands fur. seals,
at least among the bachelors, to haul out in new places at certain seasons,., This
was noticed in 1882 and 1883, but at that time it was attributed to the continued
expansion of the rookeries, which compelled the bachelors to occupy new ground as
they were! being crowded out of the breeding area.. Thus on Bering Island. the
beach at Kisikof (Tisikof) (Plate 94, Asiatic Fur-Seal Islands) was taken possession
of in, 1882; similarly on Copper Island bachelors at Glinka were hauling outin
Gorelaya Bukhta north of Lebiazhi Mys and at the other end at Kulomakh beyond
Babinskaya Bukhta. At the latter place in 1895 there were a few half bulls left
when on August 2 of that year Mr. Grebnitski and the writer camped on that very
beach. Likewise at Lebiazhi Mys only bachelors formerly hauled out. It seems, how-
ever, as if with the diminishing rookeries the cows, possibly attracted by the half bulls,
that had hauled out there and become, unduly numerous in proportion, must also
have come ashore at some of these points, thus.creating new breeding grounds; for
in 1910 Suvorof reports 700 cows and a number of bachelors at Lebiazhe and 200
- cows and 200 black pups at Babitche, though in neither place did he see any bulls.

-It should therefore, perhaps, not cause great surprise that in 1921 and 1922 seals
were reported to have hauled out on Bering Island not far from Northwest Cape,

and at the southeastern extremity of Copper Island as early as 1917. In fact, it
was at this recently formed hauling ground that the Japanese raiders, to be men-
tioned later on in this report, were caught killing seals. It was said by the natives
on Bering Island that the seals hauled out near Northwest Cape consisted of some

**Palata [in 1882 and 1883] to the looker-on coming over the mountains, was probably the most impressive rookery view in

the whole Commander Islands group. | 'The solid blackening masses of breeding seals, filling the gulley to overflowing and extend-
ing under the bluffs and along the beach on both sides, was a sight never to be forgotten.” (Asiatic Fur-Seal Islands, p. 145)... ,;

298 BULLETIN OF THE BUREAU OF FISHERIES

bulls and many females, as well as bachelors. However, it was expressly stated
that no black pups were seen, and consequently the gathering did not‘as yet! rep-
resent a new breeding ground. The ability of the natives to distinguish females
and bachelors on shore at some distance is not always to be trusted, as the writer
has had opportunity to observe on several occasions. We have even seen them
quarrel over individual seals in the mixed drives on Bering Island. On the other
hand, if cows returning from the feeding grounds to the northwest of the island
should happen to pass by a band of half bulls and bachelors hauled out ona beach
or reef there would be nothing surprising in their going ashore, either for'a rest or a
frolic. With the large number of superfluous bulls and half bulls on the: Bering
Island North Rookery it is quite likely that some of them hauled out by themselves
in their search for stray cows. | It is therefore improbable that we have here to deal
with the incipient formation of new rookeries, although it is possible that when bulls
are disproportionately abundant new preeunae grounds may become’ established in
just such a way.

SUMMARY OF CONDITIONS’ IN 1922 1 Lovet

In view of the above, the condition of the Commander Islands seal rookeries
in 1922 must be characterized as highly deplorable. Large and flourishing rooker-
ies, the breeding population of which as late as 1897 was still to be reckoned in
thousands, had entirely vanished, while those remaining had shrunk to about one-
tenth of what they were at that time.

In addition to this numerical decline there is another ‘ominous circumstance to
be recorded, which bodes ill for the future of the Commander Islands ‘herd, viz,
the great superabundance of old males on the breeding grounds. The evil effects
of the five years’ prohibition of killing after the conclusion of the treaty of 1911,
which in spite of the predictions of experts was established on the’ Pribilof Islands,
have proved even more disastrous’ on the Commander Islands. “With” the
number of seals fallen below the'18,;000 mentioned in Article XII of the ‘tréaty,
the continued suspension of land killing ‘of males except such as may be necessary
for the support of the natives of the islands is bound to inérease the handicap 6f the
herd. ‘The disorganization of the orderly rookery service, the -trampling’of /the
newborn seals, and, as mentionéd above, the possibility of the unmated: bulls ‘at-
tempting to start rival rookeries, aré logical consequences of the poliey of leav
ing the management of bidlogienl problems like goo involved in this business
to persons unversed in such matters. AB 9Ot09 9 0R

. EXPLANATION OF CONDITIONS AP a

The debacle of the Commander Islands fur-seal:herd is not of tis- hoy or ips Ms
day; it dates back to the beginning of pelagic sealing on the Asiatic ‘side of the
North’ Pacific Ocean. . It will not be necessary to go into detail with regard to the
origin and early phases of this destructive business, as: they have already: been
fully dealt with in a previous report (Asiatic Fur-Seal Islands,’ pp. 190~216), and
it will suffice merely to quote from the summary of that paper on page 203210) 1.

* * * the known pelagic “‘ Asiatic catch” from 1892-1897 was about 295,000 skins, . Allow-

ing 8,000 skins for the Kurils and Tiuleni, the known loss in that period :to the Commander Islands
herd was about 287,000 seals, apart from the loss of wounded ones, etc. The number’ of ‘seals

Buty. U. S. B. F., 1925. (Doc. 986)

Fic. 9.—Sikatchinskoye Rookery, Copper Island, August 21, 1897, looking toward Stolbi

tae

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 299

killed on the islands in the same period was 119,708. The pelagic catch was therefore considerably
more than,twice as large as that on the islands, while the loss to the herd from that cause was much
greater, due to starved pups and seals shot but not secured. It is certainly no exaggeration to
say that the actual loss to the herd in those six years has averaged 100,000 a year, more than one-half
of which were females, * * *,

At will thus be necessary only to pick up the historical thread of the destruction

since the close of the investigations of 1897.

“S aes Bivutchi Kamen
GOP barge field of Help Bas +
ti %.
- 4 if, a
Ld a
A : AN
Distribution of breeding Seals Sal
1895 ts mo ee .< x
Babin &/ y m 2 _— =r 10
i bad
oe te

- Blizhni Mys “7
MAP

of
North Roukery Bermg Island

Leonhard Steyneger

Fic. 10.—North Rookery, Bering Island, showing distribution of breeding seals in 1895, according to Stejneger

HISTORY OF THE COMMANDER ISLANDS FUR-SEAL ROOKERIES AND FUR-SEAL
INDUSTRY SINCE 1897

CONDITIONS AT THE END OF 1897

Two factors played an important réle in the decrease of seals on the Commander
Islands—the action of the Russian administration, which, in negotiations with the
British ‘pelagic sealing interests, accepted the 30-mile zone limit without time
restriction, and the optimism of the local officials, whose estimates of the number
of seals still on the rookeries were based wholly upon the so-called counts of

‘800 BULLETIN OF THE BUREAU OF FISHERIES °

the natives and their Cossack overseers, whose ideas of numerical figures ‘that
exceeded 10,000 were very fantastic indeed. On one occasion a Cossack, being
told by Governor Grebnitski to count the seals on Bering North Rookery, returned
within an hour and reported having counted considerably more than 100,000 seals.

In 1896 the writer reluctantly made an estimate of the number of
seals breeding on the rookeries of both islands, claiming for it only a remote
approximation, the chief merit of which was that it would serve to eliminate fanciful
estimates that had no foundation in any tangible facts. The figure presented was
65,000 breeding seals on both islands, which would presume a maximum presence of

MAP

of
Karabelnoye Rookery
Copper Island
by

Leonhard Stejneger:

Scale

1 to 6, Photographic Stations

Distribution of Seals Aug. 1-3, 7895

Fic. 11.—Karabelnoye Rookery, Copper Island, showing distribution of seals in 1895, according to Stejneger

about ‘36/000 females (including 2-year-olds) at any one time; but for the same! year
the assistant administrator estimated the number of cows on both islands at
about 135,000.7. No wonder they went on killing as they did.

The rookeries on the Commander Islands were at that time ‘“‘raked and scraped”’
forthe last bachelor'seal to such an extent that while it might have been safe to
have killed 7,000 seals that year, actually 14,472 were taken, or more than twice
as'many. After 1897, however, additional factors appear, which make it impor-
tant to review the fur-seal: industry ‘on the Commander Islands in some — in
order to fully understand what has happened tothe seal herd.

7 Suvorof: Komandorski Ostrova, 1912, p. 179.

VUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 301

CONDITIONS AFTER 1897 UNTIL THE TREATY OF 1911

Pelagic sealing.—The first circumstance to be noted is that: pelagic sealing, which
started so destructively to the Russian herd in 1892, continued without let or hin-
drance. The 30-mile protected zone around the islands, without time limit, was
; ™

Sy

LOD

\S

* Bk 1yzeiqeT

is

phordz729
DILYRT
ae)
S
———=———\

Wa. MAO_ NOR Ta

LUVYHOD XICNI

9681 ‘4-2 buy spas fo

1334 00001
L- Bt ‘BUpYequg

B + vhuysuiqag

uaumy iQ

I9Bsaleyg 07 BUIpI000B “CEQT Ui S[VOS JO UOINQII}s!Ip SurMoys ‘purvysy seddogy ‘sasayooy eyUO— ZI ‘S1y

& sh nna
EN

Pit

ah woysuraesg .y"
sagoulajg preyuoey
hy
pueysy teddoy

solloyooy exUILH
jo
dV

oe ‘yg

absolutely valueless, as a,protection for the seals. Moreover, pelagic sealing on the
Asiatic side was fully four times as disastrous, in proportion to the stock of seals,
on the Russian islands as was the destruction visited on the American seals. The
pelagic catch fell off so rapidly as to become unprofitable to the Canadian and Amer-

302 '' BULLETIN OF THE BUREAU OF FISHERIES

ican schooners, which were gradually driven out of the business by the Japanese,
who, because of their cheaper outfit. and labor, were able to make a big profit
where foreign schooners would starve. In addition, the Japanese Parliament passed
a law in March, 1897, by which, from April 1, 1898, under certain conditions, a money
subsidy was paid to the Japanese sealers,* a law that was not repealed until 1909
(Imperial Japanese Decree No. 173, June 26, 1909).

The award of the Paris tribunal, with its partial protection of a 60-mile zone
around the seal islands and the close of hunting in Bering Sea until August 1, as
well as. the elimination of firearms, applied only to the American and not to the
Russian islands. In addition, the 30-mile limit thrown around the Commander
Islands was only a separate treaty agreement between Great Britain and Russia
and bound no other nations. Thus when the Canadians céased pelagic sealing on
the Asiatic side and the Japanese took it up, even that slim protection was: done
away with. Although utterly valueless as a protection against regular pelagic seal-
ing, the 30-mile zone had afforded a fair defense against raids on the rookeries by
poaching schooners, against whom the 3 miles of the Territorial waters was of very
little protection because of the frequent dense fogs which made it is an easy task
for the marauders to get close up to the rookeries unobserved. The Japanese were
not restricted to the use of bow and arrows, but employed the shotgun withall its
terrible waste of life. As their catches steadily increased in spite of the dwindling of
the herd, it is evident that the Japanese were gaining efficiency from their experience,
and there is also reason for believing that the increase was due, at least partially,
to the taking of oversized skins, for which the earlier Canadian and American
sealers had no market.

The seriousness of the Japanese pelagic sealing and the terrible inroads made
upon the Commander Islands herd may be fully understood by a glance at the follow-
ing table, mainly derived from Suvorof (Komandorski Ostrova, 1912, p. 42).

Number of Japanese sealing vessels and their catch, from 1897 to 1910, inclusive :
Schoon- |’ Skins | Skins per Schoon- | Skins | Skins per

Year ers taken | schooner Year ners taken | schooner
14 5, 247 375 29} 10,035 346
16 4, 860 304 29} 10,176) .. 3651
12 6, 518 543 35 | 10, 420 298
15 7, 533 502 31 | 13,355 430
19 6, 945 366 35 |. 10,465 299
15 7, 462° 497 37 8, 309 225
21] 11,240 535
28] 15,698 BBY eee tale NN SU VAUA A a 128, 263

If it is taken into consideration that about 30 per cent more seals were shot
at than secured, and if allowance is made for one-third of these seals having
escaped with their lives, it will be seen that the total number killed by the Japanese
alone during those 14 years can scarcely have been less than 154,000, of which prob-
ably not less than 92,000 were cows, not to mention the loss to the herd due to the
consequent starvation of the pups on the rookeries and the unborn young of the
cows killed. i

8 Por more detailed provisions of the law see Asiatic Fur-Seal Islands, p. 325.

FUR-SEAL INDUSTRY: OF THE COMMANDER ISLANDS 303

‘The next calamity that befell the Commander Islands seal herd was incidental
‘to the Russo-Japanese war of 1904-5. Details of the wanton destruction are
lacking, but the rookery raids, which were not uncommon before and especially
after, assumed the character of willful extermination. Thus we were told that on
Copper Island in 1904 a large number of Japanese landed on the Karabelni Rookery,
and in addition to killing and taking the skins of a great many seals, practically

-all of which were cows, shot and butchered every seal they could reach for the sheer
sake of destruction. This once flourishing rookery was thus to all intents and pur-
‘poses exterminated during the war.

Nor did the raiding of the rookeries cease with the end of the war. The schoon-
ers came early and went late. In 1907 the first one was noticed at Copper Island
on April 24 (old style). In 1908 they were observed in April, and in 1909 the first

“one was seen as early as February 26. Again, in 1910 the first one was reported
on April 24. The piratical raiders did not confine themselves to the rookeries, but
looted the houses of the natives as well. The sealing village of Glinka, inhabited

-only during the sealing season, was visited by the marauders on April 16; 1908 (old
style), and nearly all the houses were broken into, their doors and windows smashed,
and the household goods, such as dishes, linen, tools, and even the stoves, stolen; salt,
sacks, and ropes were taken. The robbers even went so far as to carry away the
supply of coal—about 3 tons. The same thing happened the following year when
the crew of one of the schooners landed on March 6, broke the windows in 17 houses
on Glinka and stole the contents (Suvorof, Komandorski Ostrova, 1912, pp. 246-247).
Under conditions such as these it is a wonder that any seals were left. We shall
later learn what happened to the remnant of the herd on land.

It must not be inferred that the Russians were not taking any measures against

‘this state of affairs. The natives were organized and well armed for watch service,

and during the sealing season several men-of-war patrolled the 30-mile zone and
visited the islands. However, anyone familiar with the storms and fogs of the
North Pacific and Bering Sea, the mountainous character of the islands, and the
difficulties of transportation, must realize that effective protection against the small
‘schooners with their reckless crews was very difficult. Nevertheless, several seiz-
ures were made during these years.

The transport Shika, on August 6, 1906 (old style), arrested the schooner
Kompira Maru 15 miles from Bering Island North Rookery. On May 16, 1908,
‘the schooner Miye Maru was arrested 7144 miles from shore. Besides seals she had
‘on board 1 sea otter and 1 blue fox, sufficient evidence of illegal catch on shore.
Again, on July 25, 1909, the Shilka stopped the schooner Tokiva Maru, belonging
to the firm Yeno- gonb-kaishal, and confiscated 14 sea otters and 6 séalsking, On
the same day the rookery guards on Copper Island caught a boat with three raiders

from the schooner Hosio Maru, 214 miles from the Bobrovi Rocks off the northern
end of Copper Island. These few instances, however, and the punishments inflicted
{the three men caught in the raid on July 25, 1909, were given sentences of three
*months in the Vladivostok jail) were of no avail 3 in checking the =e of
the raiders.

42279-2513

804 BULLETIN OF THE BUREAU OF FISHERIES

Sealing industry on land.—On the Commander Islands the sealing industry was
in the hands of a trading and sealing company, practically the same arrangement
that. existed on the Pribilof Islands except that the killing of the seals was done
through the agency of the Government while the company’s responsibility began
with the receipt of the fresh skins at the salt-house door.

In 1897 the leasing company (the Russian Seal Skin Co.) was operating under
a 10-year contract from February 19, 1891, to February 19, 1901 (old style). This
company, however, continued the coulitie Heer during the season of 1901, ship-
ping from the islands in that year 11,489 skins (Bering Island 5,438, Copper
Island 5,527, and Robben Island 524), for which, at the rate of 12 rubles per skin,
they paid the Government 137,868 rubles, as shown by the documents in the Bering
Island archives.

On August 4, 1901 (old style), a contract for the lease of the sealing on the
islands for the next 10 years from September 1, 1901, was concluded between the
Russian department of agriculture and public desaa and the Kamchatka Com-
mercial-Industrial Co. much on the same conditions as previously, the company to
pay 10 rubles for each sealskin, 200 rubles for sea otters of the first quality, 100
rubles for sea otters of the second quality, 18 rubles for first-class blue foxes, 9 rubles
for second-class blue foxes, and 5 rubles for white foxes.° Headquarters of the
company were in St. Dotenduiee g, with an agent in Petropaulski.

This company, the lease to which expired on September 1, 1912, was succeeded
on that date by anew company, Ivan Yakovlevitch Tchurin & Co., also known as the
I. I. Choorin Co., which held the contract for the skins to September 1, 1916 (old
style), when the system of leasing was abandoned and the administration of the
fur industry of the islands was taken over by the fisheries bureau in Vladivostok.
As the five-year closed season for seals fell within their term, they handled principally
sea-otter and blue-fox skins.

Then followed the revolution, and the subsequent events, in so far as ee
have any bearing on the sealing ees will be related separately.

In the three years, 1895 to 1897, when the walter investigated the Carmneoaticies
Islands. rookeries the hauling Bounds were “‘raked and scraped’ for the last
obtainable bachelor. At the same time pelagic sealers took tremendous toll of the
herd at sea, which had not even the slight protection of the 60-mile zone and the
closed. season before August 1. .There can be no doubt that the killing..on land
was much too close, with pelagic sealing still going on unchecked. If the estimate
of the females present on the Commander Islands in 1896 were only moderately
accurate, the number of male seals killed that. year (14,472) would not, have been
justifiable even if the herd, had been in prime condition and no pelagic sealing going
on, As the killings’ are now regulated, on the Pribilofs, though possibly oyer-
conservative, there would not have been taken 7,000 skins; but the worst of it was,

of course, that in the same year probably 20,000 female seals belonging to, the
.Commander Islands herd were killed at sea. The next year this toll of females was
scarcely less than 12,000, yet by the same process of scraping the ili eit
islands the company Secured 13,620 bachelors. ishist odd

9 The text of the eoninuct: with tie title Kontrakt na Vidicite f ‘Geena Kamtchatskomu Torgovo- Promishlennomu Obshtchestvu
pushnik promishlof na Ostrovakh Komandorskikh i Tiuleniem, is to be found as Appendix I in Suvorof’s Komandorskie Ostrova,
1912, pp. 281-285

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 305

In 1898 and 1899 the number of skins taken did not much exceed 9,000, but
the catches of the Japanese pelagic sealers were slowly increasing in spite of the
dwindling of the herd. The lease of the Russian Seal Skin Co. was now nearing
its termination, the last season under the contract being that of 1900, but extended
over to the end of the 1901 season as noted aboye.. Quite naturally, then, and
especially as every seal spared on the islands meant one more possible skin for the
pelagic sealers, the lessees bent every effort to get by any means whatever as many
seals as possible.

It was thought by us who witnessed the “‘raking and scraping” of the rookeries
in 1895 to 1897, when the extreme methods applied yielded respectively 16,056,
14,946, and 11,335 skins, that the limit of what was possible had been reached;
but notwithstanding the facts that similar methods prevailed in 1898 and, 1899
and that pelagic sealing still continued to take its toll, the lessees succeeded in
taking 12,540 skins in 1900 and 10,965 in 1901.

The new lessees, the Roeienaien Commercial-Industrial Co., were not able
to keep up this pace, but considering the circumstances and the condition, of the
herd they did pretty well, taking 7,107 skins in 1902 and 7,806 in 1903. During
the disastrous years of the war (1904-5) they even took 8,319 and 8,990, respec-
tively. However, these years brought about the final collapse of the Commander
Islands rookeries. It has already been mentioned how the war was taken advan-
tage of by the Japanese raiders to partially destroy the rookeries. It may now
be told how the war, by causing enormous losses to the leasing company, also
became an active factor in the ultimate demoralization of the killing on land.

While the Russians at the outbreak of the war were endeavoring by negoti-
ations to obtain permission for the company to navigate its ships under a foreign
flag, the Japanese seized the company’s steamer Kotik in Yokohama harbor and
shortly afterwards its schooner Bobrik in the harbor of Hakodate, less than three
weeks after the opening of hostilities in January, 1904, thus crippling the company
at the beginning of the sealing season, though it was able afterwards to charter the
American steamer Redondo for the island service. This vessel, however, proved
worthless, and the Mineola, which replaced her, was wrecked in the Okhotsk Sea
near Tigil on the west coast of Kamchatka, involving a total loss of the cargo of
goods and furs. During 1905 two more ships were chartered but both were
captured by the Japanese in August, 1905. This capture involved a loss of 4,030
fur-seal skins. It should be added that the rookery at Robben Island, which also
was included in the company’s lease, was lost to the Russians from the beginning
of the war. Altogether the Kamchatka company’s losses were. enormous, and
quite naturally its managers tried to save as much out of the wreck, as possible.

From 1895 to 1897 there was still a superabundance of adult male seals on Cop-
per Island, though on Bering Island the bulls were relatively much scarcer, due to
the fact that the character and situation of the rookeries were much more favor-
able for a close killing than on Copper Island.,, At. that, time, however, the
proportion of the sexes was not such as to cause alarm if pelagic sealing could have
been stopped within a short time, a hope that was, perfectly legitimate in view of
the unanimous recommendations made by the seal experts from England, Japan,

\

306 BULLETIN OF THE BUREAU 'OF FISHERIES °

and America at their meeting in Washington in the fall of 1897. To stop the land
killing of males even for a single season, or to limit the number killed without
stopping pelagic sealing, would have been not only useless but absolutely sense-
less, as the only possible result would have been the sparing of that many males
for the benefit and encouragement of’ the pelagic sealers. Owing to the delay
in acting upon the recommendations of the experts who investigated conditions'on
the islands, the seals went to their destruction, which was terribly accelerated by
the circumstances detailed above.

The Russian authorities on the islands must have seen the approaching catas-
trophe, but their optimism apparently was not seriously shaken, since it was
possible to average nearly 9,400 skins during the years preceding the debacle of
the war. It then became plain not only that the breeding herd had greatly
decreased but that the number of breeding males in particular was becoming alarm-
ingly small. In 1904 the natives at the North Rookery on Bering Island reported
tothe administrator of the islands that only 25 old bulls were left and that
bachelors were exceedingly scarce. To counteract this disproportion of the sexes,
and also to make up for the loss of the bachelor skins, the company proposed
that they should be allowed to stop killing bachelors and to kill females instead.
This was objected to, but by a revival of the old fiction of virgin and superannuated
cows the company succeeded in obtaining from the Ministry of Agriculture and
Public Domain in St. Petersburg permission to kill 8,000 unimpregnated cows and
1,000 bachelors. As the permission arrived after the opening of the season, 1 608
bachdters but only 6,282 females were killed that year (1905).

Thus began a new chapter in the destruction of this herd, since, as was to be
foreseen, this unique method of restoring the equilibrium between the sexes, far
from accomplishing its purpose only hastened the destruction. Durmg the next
two years comparatively few females were recorded killed in the official documents,
but their skins were taken unofficially and without permission, the Peso ln
being that only nonpregnant or superannuated cows were to be killed.’ sal

At this juncture occurred the death of Nikolai Aleksandrovitch Grebnitski
(fig. 13), who had been the Government manager of the islands for 30 years. When
he first landed on Bering Island on' August 21, 1877, the sealing industry was still
on the upward grade, reaching its zenith during the following 10 years. He was a
capable administrator; who was looking out for the interests of the natives in their
relations with the leasing companies, and he established sensible and suitable rules
for the regulation of the sealing business and blue-fox and sea-otter hunting.
Grebnitski, who had studied at the University of Odessa and in Germany, had
considerable biological training and was thus well qualified for his position. It
may be safely said that in the days before the beginning of pelagic sealing ‘Greb-
nitski knew all that could be learned about the seals on land and acted intelli-
gently upon his knowledge. “Moreover, he was practically the only’ man in the
whole Russian Government who knew anything about the seals and ‘the’ sealing
business; but, of course, when ‘pelagic sealing started in earnest on the Asiatic
coast hé was practically helpless and can hardly be blamed for the ‘Russian
failure to secure any practical protection for the Commander Islands seal ‘herd.

oe

Buuu. U.S. B. F., 1925. (Doc. 986)

Fic, 13.—Nikolai Aleksandrovitch Grebnitski, Administrator of the Commander Islands from
1877 to 1907

Buu. U. S. B. F., 1926. (Doc. 986)

erent si a
aeons ef

- Copaacne acne mt Conmnbpa. 1381 A ae NG 12590.
ofan’ ‘a I 0A.

“WAHMEHOBAHIE, HUBOTHBIXD — separ | Most. Traenin

Horaia, WOpcKMAD XOAOeTAROND »
ae cheasen ‘
“MaTOKe <
FiCLOPOCTKOB
‘aGpast mopenix }fo copra
: de
anb copra

SaayGuixe oeenoes tre sopra

Fic. 14.—Copy of ‘‘Viedomost”’ or ‘‘ Report on the number of fur seals, sea otters, and arctic foxes taken on
the Commander Islands during the period from January 1 to December 31, 1910.’’ From original in the
Bering Island archives

a_i

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 307

No measure short of total stoppage of pelagic sealing could have prevented the
final destruction of the seal herd, though, of course, the excessive scraping of the
rookeries for, male seals and the, tragic killing of females toward the end of, his
administration and against his protest natuxally hastened the process.

Col. Nikolai Pavloyitch Sokolnikof, who for 10 years had been administrator of
the Anadyr district, was transferred to the Commander Islands after the death of
Grebnitski.. Upon his arriyal in i907, finding that the killing of cows was practiced
upon the initiative and responsibility of the assistant manager, he put a stop to it
that same fall and protested earnestly against it to the department in St. Petersburg
without avail. The department issued a permit for 1909 to kill 4,000 cows and
1,500 bachelors, and in order to make sure that this maximum should be reached
the leasing company agreed to accept up to 20 per cent of stagy skins. However,
not more than 3,155 cows were secured; though the total quota was exceeded by
311 skins.

When this document was discovered in the archives of Bering Island in 1922,
we knew already that cows had been killed quite extensively, but were so Sanaa
to see this unblushing Governmental acknowledgment that a photograph of the
“Viedomost”’ (herewith reproduced as _ fig./14) was taken as a memento and as
proof. The work begun in 1909 was continued in 1910, only there were fewer seals
to take... The result on both islands was as follows:

Sealskins taken on Commander Islands, season of 1910

Island I Bachelors Cows Total

Sikes Tab TS col SO mle pS. SN 2a Ok, SS 206 1, 105 1,311
Copper Island. - _-_ 12228282 Se28t8 2 itty ligt! Tae SE Eat SES DEER, Ses Se. «2 Na, ta ee ee 502 1, 492 1, 994

ER otalmerepe er nese II PS Sy Ne Pa ee Pe ees 708 2, 597 3, 305

In order to obtain this number it was necessary to extend the killing season
to August 31 on Copper Island and to October 2 on Bering Island. In fact, of the
3,305 skins taken only 863 were secured before August 1, on which date, in the past,
Killin usually stopped. Finally, in the years 1907 tb 1909 over 2,500 gray pups
were killed-m.the fall for food for the natives. Just exactly what was the propor-
tion of the sexes among these is not known, but certainly they were not all males.

The alarming reports about the condition of the rookeries, the approaching
expiration of the lease of the Kamchatka company, and the proposed negotiations
for a treaty to abolish pelagic sealing induced the Ministry of Agriculture and
Public Domain to make an investigation in order to obtain information at first
hand. Hvgenij K. Suvorof, who was sent from St. Petersburg in 1910, does not
seem to have had any previous experience with the fur-seal industry or the fur seals.
This, however, was not of much significance, for all he could do was to confirm
the utter demoralization of the whole business. It was only when he attempted to
explain certain features of the debacle and its causes, and also when he attempted
to’ make’an estimate or a “count’’ of the number of seals remaining and the classes
of seals composing the herd, a feature which will be referred to later, that his inexperi-

308 BULLETIN OF THE BUREAU OF FISHERIES

ence became a decided handicap. He undoubtedly did his best, and his report
contains a large amount of valuable information which has been drawn upon quite
extensively, but so far as the fate of the seal herd was concerned it did not and could
not possibly have any influence. It was too late:

Condition of the rookeries in 1910.—Arriving on the Commander Islands in 1910,
one of Suvorof’s first endeavors was to ascertain the actual number of seals on the
rookeries. Apparently all possible methods of ascertaining the number of seals on
the Pribilof Islands had been tried, such as calculation of the area occupied by the

LO en ihe of Bp

From Suverof,/9/2,pl 6

° IK (Gj
Distribution of Seals 1910 AN 0 | o
y S

MAP

of :
North Rookery Bering Island
Leonhard Stejneger

Kishotchnaye *,
en
Blighni Mys 27244

Fig. 15.—North Rookery, Bering'Island, showing distribution of seals in 1910, according to. Suvorof

herd divided by the theoretical space in square feet occupied by a seal; counting the
cows in a number of harems, averaging their number per harem, then counting
the harems (that is, old bulls) on the rookery; counting from a distance the number
of seals on a certain area of apparent normal density and then applying the figures
obtained to the whole rookery. Every one of these methods, however, had to be
abandoned and the conclusion reached that there is only one reliable method, viz,
to corral, drive off, and actually count every black pup on the rookery. Their num-
ber ascertained, the number of the breeding females, only a varying proportion
of which are present at the same time on the rookeries, is consequently ascertained

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 809

with exactness, and that is the only figure that is of prime importance. With
that given, the number of the other classes can be observed and computed with suffi-
cient accuracy for all practical purposes.

In 1897 the writer declared that a count of the seals such as was undertaken on
the Pribilofs could not be carried out on the Commander Islands owing to the
difference in character of the rookery beaches on the two groups of islands.

With the excessive shrinkage of the number of seals during the next 13 years,
however, it seems certain from what was seen on North Rookery, Bering Island, in
1922, that acount of the black pups would have been feasible there in 1910, though
it would still beimpracticable on most of the rookeries on CopperIsland. Moreover,
on account of lack of suitable means of transportation, one person would be utterly
unable to cover all the rookeries during the short period in which such a count must
be made.

As Mr. Suvorof had no previous practical experience and probably had no
detailed description at hand of the technique of counting the pups as developed on
the American islands, he was restricted to his own resources and the application
of some of the methods mentioned above, as definite statistics and not vague
estimates—which can be of value only when done by persons of wide experience and
tried judgment—-were evidently expected by the authorities in St. Petersburg, who
had been accustomed to receive most minute figures from the previous administration
of the island based on the fantastic ‘‘counts” which the native guards, stationed at
the various rookeries, made at certain intervals during each season. When con-
fronted with the Bering Island rookeries—Reef and Kishotchnoye—his first
attempt was directed at ascertaining the number of seals present by the area method.
Red and white marks and signals were painted all over the two rookeries, and
calculations of various sorts started. Unfortunately, or fortunately, this work had
to be left unfinished, as by the middle of July (old style) he had to leave for Copper
Island. Here that method was plainly impossible.

By that time he had also discovered, what we others had learned long before, that
even if one can get near enough to even a depleted rookery to start an autoptic count
of the seals in sight, ‘part of the seals always slip away from the counting, and the
figures obtained invariably prove lower than the actual ones. * * * Part of the
animals are in motion, some come out of the water, others wend their way to the sea;
many are concealed by the unevenness of the beach, by jutting rocks and cliffs.
Many new-born pups, in particular, are concealed in the crevices between the rocks.”
He also observed that the weather, the temperature, and the time of the day made
a big difference. Nevertheless, he decided that this was the only method possible
on Copper Island, and he believed ‘‘that the error of the per capita count of the
adults would not exceed 10 per cent and of the new-born pups 20 to 25 per cent,”
adding that “the method of area never gave such exactness’? (Komandorskie
Ostrova, 1912, pp. 181-2).

During the latter part of July and the beginning of August, 1910 (old style),
therefore, he ‘visited all the Copper Island rookeries and carefully counted all the
black pups lying on the beach and splashing in the near-shore zone.” To the
number thus obtained he “added 14) per cent for the females without pups, and

10 BULLETIN OF THE BUREAU OF FISHERIES

in this way arrived at the approximate number of cows on the given rookery.” It
turned out that his figures “approached sufficiently near the autoptic estimate of
the rookery guard” on Copper Island. With regard to the Bering Island rookeries,
where his ‘area method” had proved a failure, he had to devise a different method,
and as his figures on Copper Island had “approached sufficiently near’’ those of the
watchmen, he assumed that those of the Bering Island guard were equally reliable.
He consequently adopted the latter, applying to them, however, a “correction

MAP

of
Karabelnoye Rookery
Copper Island

by
Leonhard Stejneger’
Scale

a ab sp ont

1 to 5, Photographic Stations

Distribution of Seals 7910
From Suvorof

Fic. 16.—Karabelnoye Rookery, Copper Island, showing distribution of seals in 1910, according to Suvorof :

coefficient” deduced from a comparison of his own count on Copper Island with the
eye determination of the watch. In this way he obtained the figures of the following
table, which in his opinion approximately express the number of animals on the
rookeries, the number of bachelors present being practically negligible. He remarks
staticllily that the estimate relates to the number of seals left after 1,017 cows ne
500 bachelors had been killed.

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 311

Suvorof’s table of seals on the Commander Islands rookeries about the middle of August (new style), 1910

Half 4 Black
Rookery Bulls Bulls Cows pups
COPPER ISLAND
Glink rookeries: %
50 20
570 500
1, 350 1, 200
550
600 550
1, 250 1, 100
ADO os ea NT GPF 2 MEN AE EIN BP a 200:|"" 200
Karaheliove rookery: : >
PROCDIMEERNReATTION,: ) NSM AEEVUEE AN MOS lO ne onan ce ann [conn no hee 3 150=f> 124
MTN OtAleMasaee SN he see UN A IP oe £0 2 oa BL oe ; 48 5, 420 2 4, 200
~ BERING ISLAND
North Rookery: ' ;
1 gH ST) ED SSS 7a ARYA eee ap ean aie ee 1 7 1, 800 1, 400
Orlof Kamen_------_---- 2 150 130
Sivutchi Kamen-- 5 250 220
Kishotchnoye- -_----- : 2 200 160
In water near rookery —-- Bil (600 s|E sane seee
Tota luemret ee TAN eA ESN DRY ee 1 16 4, 000 2 2, 000
Potali(bothiistands) Sass -— 6-228 Po fo ee aE Ot on ae 1 64 9, 420 6, 200
- 1 Partly bachelors. seh .
2 These figures are those given by Suvorof; the actual summation gives 4,199 and 1, 910, respectively.
3 Approximate. :
The corresponding totals of the watchmen are given as follows:
4 Cows Black pups
Copper Islind. 2. 5. — Pusgesta ef-s ees fe + ie 7, 070 5, 755
Bennet island! migra lg a Ro ont AR 1205 74) 25550
4 Botiuslanges les 06 ALE ll a CN 11,190 8,305

& i

lt is not easy to see how he has arrived at these figures and how he applied
the ‘correction coefficient,” as it appears from his tables that his own estimate on
Copper Island differs much more from that of the Copper Island guards than the
accepted figures for Bering Island do from those of the Bering Island guards. How-
ever, much as this census Raeehable differs from the actual figures, it is a sufficiently
eloquent demonstration of the terrible straits to which the Rirargan tes Islands
rookeries had been reduced by 1910. If we add to this that afterwards during the
same year 1,580 more cows and 208 bachelors were killed © and imagine the number
of black pups that must have starved to death, as their mothers were killed both
on land and at sea, we get a vivid picture of the conditions that existed at the close
of the season of 1910. To complete the picture of destruction it should be realized
that*because of the-decreasing number of male seals taken, the natives were allowed
to kill gray pups in the fall for food. The number of gray pups thus killed in the
three years preceding 1910 averaged 860. How many of these were females it is
not possible to say.

Suvorof’s figures receive additional confirmation by ee maps. These are
only diagrammatic representations on small-scale copies of the rookery maps pub-

“10 Altogether in 1910 there were killed on the Commander Islands 2,597 cows and 708 bachelors; of these 1,017 cows and 500
bachelors had been killed before the census was taken.

BULLETIN OF THE BUREAU OF FISHERIES

312
lished by the writer in 1896 and 1898.

His data, however, have been transferred to

the latter, as shown in Figures 15, 16, and 17, and to better show the destruction

the distribution of the seals as the writer mapped it in 1895 has been reproduced

on the same scale (figs. 10, 11, and 12).

JO.0ANg 09 BU:ps099¥ ‘OTE Ul S[Bes Jo MONG Isip Furmogs ‘pueey Jaddog ‘soleHooy vAUl[H— AT “Ol

(Soreres west) 0164 97029 fo uonngisig

Omer “st
ust

S33) reed AUWHO X2QNI

°
¢,
7%?

shy tyeuxorg

POU OMILD
uewey tq’

sa§eulayg preyuoay

Ke
pups] Jeddoy
SILIOYOOY CYUILY

J jo
dVW

THE TREATY OF 1911

It had become obvious to everybody by this time, to those who had profited
by the pelagic sealing as well as to the powers owning the breeding grounds, that
some arrangement would have to be made if the seal herds were not to be extermi-

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 313

nated, to the permanent loss of all parties. Fourteen years previously sealing experts
of all the powers interested had agreed as to the essential facts and the best methods
for protecting the seals, but political and financial considerations had prevented action
upon their recommendations. Such considerations would obviously be of no impor-
tance when the seals themselves should have ceased to exist, hence the belated
willingness of the four governments to compromise their claims.

Under the auspices of the United States Department of State an international
conference was convened in Washington on May 5, 1911, for the purpose of conclud-
ing a treaty affecting the fur seals of the North Pacific Ocean. The powers attending
the conference were Great Britain, Russia, Japan, and the United States. The
principle that guided the negotiations seems to have been an acknowledgment of the
right of the pelagic sealers to compensation for giving up their preying on the seal
herds outside territorial waters. At any rate, the governments possessing seal
rookeries agreed to pay the others 15 and 10 per cent of the sealskins taken. A
treaty was signed on July 7, 1911, and after ratification became effective on Decem-
ber 15, 1911, and was to continue in force for a period of 15 years from that date
and thereafter until terminated by 12 months’ written notice given by one or more
of the parties to all of the others, which notice may be given at the expiration of 14
years or at any time afterwards. |

By this treaty pelagic sealing was forbidden in the waters of the North Pacific
Ocean north of 30° north latitude and including the Seas of Bering, Kamchatka,
Okhotsk, and Japan, with the exception that ‘Indians, Ainos, Aleuts, and other
aborigines”? may carry on pelagic sealing in canoes without firearms under cer-
tain conditions specified, among them that these aborigines must not be ‘under
contract to deliver the skins to any person.” There is nothing, however, forbidding
these aborigines to sell the skins. Otherwise no person or vessel shall be per-
mitted to use any part of the territory of any of the signatory powers for any
purposes whatsoever connected with the operations of pelagic sealing, nor shall
any sealskins not certified to have been taken legally be permitted to be brought
into the territory of any of these powers. The latter agree to enact and enforce
such legislation as may be necessary with appropriate penalties for violations,
and to cooperate with each other in taking such measures as may be appropriate
and available for the purpose. In addition, the United States, Japan, and Russia
agree to maintain a guard or patrol in the waters frequented by their respective
seal herds. This would entitle Russia to maintain a guard off the Japanese coasts
during the winter and spring migrations of the Commander Islands seals.

This apparently throws the burden of protecting the seals at sea against the
illegal pelagic sealing by nationals of the other powers upon the one in whose
territory the herd breeds, thus relieving the government of the offending
sealers of any obligation to interfere with them, except to prevent them from
bringing the illegally taken skins into its territory and to try the offenders and
impose upon them penalties of its own making, when such offenders shall
have been delivered to its authorized official by the naval or other duly com-
missioned officer of the government making the seizure. In other words, in
order to protect its seals against pelagic sealers the Russian Government must

314 BULLETIN OF THE,.BUREAU OF FISHERIES

have an effective guard not only in summer in the, waters surrounding) the Com-
mander. Islands within, a radius of at least 150 miles around, the. islands, as the
breeding seals are known to go that far to sea in search of food, but also. in winter
and spring along the coasts of Japan from the latitude of Yokohama, to and inelud-
ing the Kuril Islands. The Japanese Government is under no obligation to inter-
fere with its own subjects in all that big stretch of ocean except to, forbid them
the use of its ports, etc., to prevent them from bringing in the illegally taken skins,
and to try and punish them according to Japanese law when they have been. seized
by a Russian officer and duly handed over to an authorized Japanese official.

Article XII of the treaty is of particular interest to the Commander dslonds
sealing industry. It reads as follows: net

It is agreed on the part of Russia that of the total number of sealskins taken/annually upon
the:-Commander Islands, or any other island or shores of the waters defined in Article I subject
to the jurisdiction of Russia to which any seal herds hereafter resort, there shall be delivered
at the Commander Islands at the end of each season fifteen per cent (15%) gross in number and
value thereof to an authorized agent of the Canadian Government, and fifteen per cent (15%)
gross in number and value thereof to an authorized agent of the Japanese Government; pro-
vided, however, that nothing herein contained shall restrict the right of Russia at any time and
from time to time during the first five years of the term of this Convention to suspend altogether
the taking of sealskins on such islands or shores subject to its jurisdiction, and to impose during
the term of this Convention such restrictions and regulations upon the total number of skins to
be taken in any season, and the manner and times and places of taking them as may seem
necessary to- preserve and protect the Russian seal herd or to increase its number; ‘but it is
agreed, nevertheless, on the part of Russia that during the last ten years of the term of this
Convention not less than five per cent (5%) of the total number of seals on the Russian rookeries
and hauling grounds will be killed annually, provided that said five per cent (5%) does not
exceed eighty-five per cent (85%) of the three-year-old male seals hauling in such year.

Tf, however, the total number of seals frequenting the Russian islands in any year falls below
eighteen thousand (18,000) enumerated by official count, then the allowance of skins mentioned
above and all killing of seals except such as may be necessary for the support of the natives on
the islands may be suspended until the number of such seals again exceeds, eighteen sougand
(18,000) enumerated in like manner.

THE COMMANDER ISLANDS ROOKERIES IN 1911

While this treaty was being negotiated and signed in Washington and during
the months following the signing things went. from bad to worse on the Commander
Islands, where two important rookeries literally became extinct.

On April 18, 1911, a council was held in St. Petersburg at the Department of
Agriculture, at which the condition of the rookeries and the fur-seal industry as
ascertained by Mr. Suvorof in 1910 was considered. As a consequence of the
alarming report made by him it was decided to institute a closed season for the
land killing of seals, to begin in 1912 and to remain in force 5 years. The reason
for not starting the closed period at once was that, according to the contract, with
the Kamchatka company which expired in 1911, the latter was entitled to,a quantity
of sealskins, though the Government had eee to itself the right to determine
each year the number and kind of seals to be killed. The Anes of seals to be
killed for skins was therefore set at 200, and as this would not furnish sufficient
food for the natives, the latter were allowed to kill in addition 200 gray pups.
Mr. Suvorof was also ordered to continue his examination of the condition of the
seal herd during 1911.

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 815

Upon his arrival atthe islands it’ was found that: the rookeries had been still
further depleted during the intervening winter by at least 30 percent, according
to his estimates, and that there were scarcely more than 200 bachelors’ present at
any time on the hauling beaches. Moreover, even if that many could have been
killed, to do so would have endangered the future supply of male life; hence, after
consultation with the local administrator of the islands, he decided to allow the killing
‘of 200 cows, 60 on Bering Island and 140 on Copper Island, in place of 200 bachelors.

The number of seals on the rookeries was now so small that he thought an
approximate count could be easily made. It is probable that this is the most
nearly correct estimate made of the Commander Islands herd at any time, though
‘the number of pups is probably considerably (possibly 20 per cent) underestimated.

The tabulation of the count on all the rookeries follows:

Fur seals on Commander Islands rookertes in 1911

| , Half Black
Rookery bulls Bulls Cows pups
COPPER ISLAND
Glinka rookeries (July 22-25, new style):
ROH ENTISHOUCIO tee ee teres ene sean een nme ne edn nate cet ena e ene neem cee ot | mentees | ons s aaa 8 37
BINiRatoninsk@ye, fs 94-222 -- ph tee bp en--t-----b fi 7- 8. $e 3 5 1 2 311 457
ARVO: Ui cio Spb oe abe ROSS eS So SSeS ES BESS SSE Se es nell 1 8 650 962
FD pS = a Oe a ER TT RE ee eee BS reer oak San a Ra eee Depeae Aare 4 184 585
Jy OGUNOY GA ess SSe Se SS SOS SS Sa es a a ae ae 2 216
Wixi dnd Perdsheyék..3-) b. .-- $40! ee eet eel ee ELSE ek deed OS eee 7 15 1,101 1, 078
Karabelnoye rookery (July 20, new style):
TRIG DTC TET Gre eee a eR Se SEE Be ee eee Ee eer tere ae ea | ee eas Sey 20 15
PRAT RIUM abe ret ra A SE are ce eg eft 8 oka he eB ad 9 31 2, 347 3, 350
BERING ISLAND 7
aes rookery:
Dif! cease Sty aes OPES Sip MERI SPL GE RIEge SSUES Waray Veneta Un PSL Weal SRY DR 4 1, 150 891
Ont FSerT Oi ee ene nl ee ce DON SRE NRT PRET SRNR A SANE) MY ENE DIE AA AY eae 20 35
Siviliohi Ramone ton syeee cl es he he hehe et eel te A | oe kd on 2 159 338
PRAT SETOPCLLILON.© ere ete eee ans ce ine ana eeen nance te nae nan tesco sane ot lence nea eee eee else ee 47
| a Oe SE Ee EEE Oe eae ee | ER 6 1, 329 1,311
* ‘Total (both FOSTER SED ae ala atl ad By geen ll Re Pa g7]2 tAd 137? 11991731676 4, 661

..) Suvorof (Viestnik Rybopromyshlennosti, Vol. 31, Sept., 1916, p. 446) gives 4,988 cows. as the ial for both islands.

In addition, there were counted on the various beaches 128 dead black pups
‘and in the water off the rookeries about 415 cows. If, however, allowance is made
for the probable number of black pups overlooked, the cows absent at sea feeding,
‘the number of bachelors hauled out (220), and those at sea (including yearling males
‘and females), the probability is that the Commander Islands herd in 1911 did
not exceed 15,000 of both sexes and all ages.

By contrast it may here be‘noted that at the same time the Pribilof herd num-
bered about 125,000 seals all told, or eight times as many. During the’ prosperous
“days of the fur-seal ’ industty, dbout 1880, the Commander Islands herd ‘was
nearly one-half that of the Pribilof Islands. Now, at the lowest ebb, the decline
‘had been ‘so entirely’ out of proportion that it was only one-eighth the size of the
‘American herd. ‘Nothing can better illustrate the disaster which had befallen the
Russian herd. Even during this year of its extreme decline the factor that
“brought | on the disaster was more active and destructive than ever before. The
‘STupanese raiders and pelagic séalers, driven to desperation by the gradually dimin-
ishing catches, were growing more bind more aggressive and reckless. It is necessary,

316 BULLETIN OF THE BUREAU OF FISHERIES

in order to understand fully the causes of the continued decadence of the Commander
Islands rookeries, to go into the history of the 1911 raids in some detail. The
following account is therefore based on the story told by Mr, Suvorof, who, as
stated above, was present on the islands that year.

Rookery raids in 1911.—¥From the table of the count on Copper Island (p. 315)
it will be noted that no bulls or half bulls arrived in 1911 on Nerpitchi, Kamen,
the last remnant of the once big Karabelni rookeries. Immediately after the
rookery guards had been withdrawn from this rookery in the fall of 1910 the seal
pirates raided it, with the result that when Suvorof visited the place on July 20,
1911, he found only about 20 cows and 15 black pups; ‘‘of bulls there was not one.”’
Five days later he returned and found not a single seal—‘‘there was not even a
carcass of a newborn pup.”, The pirates had apparently been there in the interval
and literally cleaned out the rookery.

About thesame time a raid on the Zapadni and Urili rookeries of the Glinka group
was successfully carried out without attracting the attention of the rookery guards,
for on July 21 they found there 80 cows killed and the carcasses of 121 black pups,
and at Peresheyek the carcasses of 2 cows and 6 pups were picked up the next
day. Here were also found, among other things, a Japanese oar, a double-barreled
shotgun, a pair of binocles, a boat compass, skinning knives, cartridges, and ropes
ready with slipknots for towing the seals killed. Evidently one of the pirate boats
had been upset in the breakers. The raid must have taken place some time before,
for the skins of some of the seals killed had already spoiled, and the birds had picked
out the eyes of some.

A much more serious affair took place on July 21. During the evening of that
day the guards noticed some Japanese schooners signaling to each other with flags.
A concerted raid was therefore anticipated and reinforced guards were placed at
different points along the coast. About midnight about 25 boats landed near the
waterfall at Sikatchinskaya. Alarmed by the large number of men disembarked, the
rookery guards fired several shots. Some of the raiding party then attempted,
aided by the darkness, to sneak along the beach and to land near Palata but were
met here by the fire of another squad of the rookery guards, so that they hurriedly
went off to sea. One empty boat, however, was afterwards taken by the Aleut
guards near the same beach. In the meantime another party hastened in the
opposite direction but soon disappeared in the fog. At daybreak, however, a boat

-belonging to this party, in which were the corpses of two Japanese, was found at
‘Babinskaya. Knowing that in the raiders’ whaleboats there usually are three or
four men, a search was made along the coast for the missing ones, and soon a live
Japanese was taken on the beach near the rookery. During the whole of the fol-
lowing day a thick fog prevailed, out of which the firing of guns was heard, pre-
sumably the schooner summoning its boats. The captive Japanese was ‘afterwards
taken. to Vladivostok for trial in order that an example might be established which
would put a stop to the ever-increasing boldness of the raiders. The Russians
believed that at this raid they killed about 14 raiders in addition to capturing the
two boats.

Whether this repulse discouraged the pete is not known, but fe usual raids
on the Bering rookery did not take place. However, the marauders for the first
time landed on the southern shore of Bering Island. , During the first half of July

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 317

three natives of that island, who made a trip to Tolstoi Mys, noticed. tracks of
people who had piled up dry grass around the fox-hunting shelter hut at that
place and set it on fire. The fire had gone out and the hut was spared.

While these events took place on shore the naval guard was kept even busier.
At the request of the Department of Agriculture in St. Petersburg the Russian
Navy Department detailed the gunboat Manzhur to do guard duty around the
Commander Islands. In spite of adverse circumstances she probably averted
further attacks on the rookeries and achieved considerable success in protecting
the remnants of the seal herd by compelling the pirates to leave the islands earlier
than usual. During the summer months the Manzhur made six trips to the islands,
seized one schooner and three boats, examined a long line of schooners, and con-
fiscated a quantity of skins. Nothing will better illustrate the methods of the
pirates as well as their recklessness and boldness even in the face of the naval guard
vessel than a detailed account of some of these occurrences.

_ During the second cruise to Copper Island the Manzhur, on June 26, seized the
Japanese schooner Kofudzi Maru about 214 miles from Sikatchinskaya Bukhta.
A boat belonging to this schooner was captured at the same time about 2 miles from
shore. On board the schooner there were found and confiscated 3 sea-otter skins
(one perfectly fresh and with flesh still partly adhering) 6 salted fur-seal skins, 2
quite fresh, and 1 hair seal. The vessel was supplied with small 37-millimeter guns
and military ammunition.

On its fourth cruise the commander of the Manzhur was informed of the raid
and shooting that had taken place at the Glinka rookeries on July 21. On the 31st
of that month he therefore proceeded to the strait between Copper and Bering
Islands’in search of Japanese schooners, hoping by an examination of their muster
rolls to discover which ones had taken part in the nocturnal affair. The first to
be examined was the Chitose Maru No. 2 (Dai-itchitose Maru). Unfortunately a
confrontation of the crew with the muster roll was impossible, since 8 of her 10
boats were hunting away out of sight at sea. However, one of the boats remaining
on board as well as one arrested near by showed traces of having been hit by bullets,

‘mute evidence of their having taken part in the shooting affair. An examination
of the hold of the schooner revealed 59 skins of adult seals and 15 skins of black
pups, some of them just turning gray, proof positive that they had been killed on
the rookery, since the pups can not. swim at that age. The skins were consequently
confiscated, but the schooner itself was released, as the search was made outside
the 3-mile Territorial zone.

The next schooner examined was the Toer Maru. All its boats were away, and
in the hold were found only 3 adult skins and 2 black pups. The skins were seized
but the schooner was released. The Hashiman Maru was next. searched, but no
skins: found. All her boats were away. While the search was going on shots
from, the direction of the coast were heard on the Manzhur.. Leaving the schooners
at once and steaming toward Copper Island, three boats were met and. seized
within the 3-mile limit, not far from the Bobrovi Kamni at the northwest extremity
of the island. The first of them, which belonged to the Chitose No. es contained no
skins, but. in the last two, belonging to the Toei Maru, 3 freshly killed seals with
the skins still on were found. It was at once decided to seize both schooners, but

318 - “BULLETIN ‘OF THE BUREAU OF FISHERIES

these, observing the arrest of the boats, put on all’sail in a hurry and were: soon
lost to sight. Nothing can better illustrate the hardihood of these raiders than the .
fact that. they continued their depredations inside the Territorial limit: with ‘the
Russian naval guard ship plainly on the horizon.

The Manzhvur then turned toward Bering Island again and caught up will two
more schooners: On the lvaki Maru skins of 13 adults and 1 black pup’ were
found, and on the Kinkai Maru 63 adult skins and 4 black pups. The skins were
seized and the vessels released. On the Konzen Maru only 2 adult sealskins' were
found, and these were restored to the owner. On this schooner the log book proved
to be in perfect order. However, on all the others the logs were found to be very
carelessly kept, and the last two days not even written up.

On this trip the Manzhur searched 6 schooners, seized 3 boats with 10 cpasanebe
sailors, and confiscated 3 sea-otter skins, 138 skins of adult seals, and 22 of black
pups. Asa result not more than 5 soherlers were afterwards observed from Copper
Island, and that a long distance off, while no more raids took place on shore that
year. The schooners did not leave the neighborhood, however, for on August 24
the Manzhur boarded 3 vessels in the sea between the two islands, among them
the same Hashiman Maru that had been searched three weeks before. As no'black
pups were found, and as it was impossible to prove that the adult skins had been
taken inside the Territorial zone, no confiscation could be made.

Taking into consideration the vastness of the sea, the precipitous nature of the
islands, the constant fogs and incessant storms, the impudent recklessness of the
pirates, and the difficulty of chasing a fleet of staal schooners in such dangerous,
uncharted seas with a large naval vessel hundreds of miles from any harbor, it is
small wonder that it has been almost impossible to protect the islands effectively.

PERIOD FROM 1912 TO 1917, THE, 5-YEAR ZAPUSKA

With the treaty of 1911 going into effect and a 5-year Zapuska or closed season
decreed, the Russian authorities began to hope for better times for the Commander
TIslandsseals. At theend of the Zapuska in 1917 it was estimated, though without
any particular pretense whatever at exactness, that the Commander Islands herd
would number 40,000 seals, and that by 1926 it would be possible to take at least
18,000 to 20,000 bachelor skins. Unfortunately these expectations were selpacey
too roseate and could only lead to bitter disappointment.

The available material for a history of the Commander Islands fur-seal industry
from 1912 to 1917 is rather scanty. The archives on Bering Island gave scarcely
any information of value. Colonel Sokolnikof continued in charge of the adminis-
tration during this period, but as he was generally at odds with the officials in the
Lesa of Agriculture at St. Petersburg the latter relied chiefly on their own

“specialists.” Finally, in 1915, the fisheries and the fur industry were organizéd as
a separate administration with headquarters at Copper Island, independent of the
district official. In 1917 the department in St. Petersburg again sent’ Mr. Suvorof
‘to the islands for the purpose of taking the necessary steps for the resumption “of
seal killing after the expiration of the five-year closed season. m

‘ These five years were on the whole quiet. The seals apparently were becoming
‘more numerous. The increase in the number of bulls and males in general was

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 319

particularly noticeable. Besides, raiders and pelagic sealers were not troubling the
rookeries to any extent.

Unfortunately there is available no reliable or even approximately rational
census of the increase. The treaty of 1911, it is true, assumes that an official count
shall be made as a basis for all calculations regarding the permissible number of
skins to be taken.. Accordingly, instructions were given the local authorities to
count the various classes of seals present on the rookeries at regular intervals through-
out the season.

In all previous reports the writer has emphasized the physical difficulties in the
way of a reliable count of the seals on the Commander Islands. The difficulties en-
countered on the Pribilof rookeries were multiplied and amplified enormously on the
Russian islands, and, besides, there were no persons on the islands competent to make
even an intelligent estimate. The count was placed in the hands of the native over-
seers of the rookery guards and the results are positively grotesque. There are
available the certified official counts for Copper Island from 1915 to 1921, inclusive
(except that of 1917, which was missing), and the counts for Bering Island during
1914 and 1915, which were extracted from the original documents in the archives on
Bering Island. A glance at these figures demonstrates beyond question the above
assertion. It would serve no useful purpose to set them forth in detail, but, con-
sidering the importance of establishing the facts once and for all, there are given
below the tables for the year 1914 for Bering Island, 1915 for both islands, and
1916 for Copper Island.

Count of fur seals on Bering Island by rookery guards in 1914

Date Bulls Hulls Bachelors} Cows Pups siete Total

September 15
September 20
September 27
October 1
October 4
October 11
October 16.
October 18
October 25

320 BULLETIN OF THE BUREAU OF FISHERIES

Count of fur seals on Bering Island by rookery guards in 1916

[No distinction made between harem bulls and idle bulls; yearlings not counted separately]

Date

‘ANI SUISt 220) eee i
Septemiber 4.) fifyi lestsio sais
September 94-22... - 02, se
September 18 __-
September 26. --_-_.-.---=:-

October 3245. £2 s2dLey Le

OctoberdO es oe heen see

Oetober VA ss io Lh slad ier

October2a"2- - oc. oe eee

Octobers0k: Ssh ee a eee

November'6 =... 252 325.2. JE es oe ees

Bachelors

Seen in

Cows Pups thé'sea

Count of fur seals on Copper Island by rookery guards in 1915

| Total’

(Schedule probably misunderstood; the figures under ‘‘Idle bulls” are probably intended for both harem and idle bulls]

Harem Idle
Date bulls | bulls

Half
bulls

Bache-
lors

August-12-204 __- 42822
August-27 22% 26s 42588.
August-31 20% -.__.4_22f.
September 3_ ___-.-------
September 14_____.__-_--_2_-.2.---|----

September 18's. ___!. 220522222 2 ]eL

September-23_)2___- 4.220s5.20 2 4] 3

September. 28_. _.-- 4.01222.) Re }ez-- s
October 7 At wt. .. | Ue Lee 5
October 2s! bak La Shed he ee Ss u
October 22% i. pea eS ee z
October. 28740 a eek eT! =

Cows

Year- | Seen in
lings the sea

Pups

Total

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 821

Count of fur seals on Copper Island by rookery guards in 1916

Harem Idle Half Bache- Year- Seen in

eon) Date bulls bulls bulls lors Cows Pups lings the sea Total
Pe ea EC aD Fe Ce CS a SL en) rere eee) A ae ere ee 2
1 1 Gilet 95 c0 selene | deve otot [egos senses 8
5 ese eae 421 Pea Se alone See Roe eee ace fate toe 60
36 1 IS rie emcee | inten wre th ere ere tae ls 60
63 6 OA SeIe es st ete at oe 121
99 4 72 0 pee ee 2 a 185
103 82 383 257 130 7 12 1,174
358 91 322 672 376 13 58 1, 890
162 50 276 978 854 3 180 2, 503
164 33 188 595 657 16 147 1, 800
267 34 289 2, 755 PG tal bee ete ee 1, 710 7, 733
275 39 3, 750 4, 400 2, 004 147 431 11, 046
214 163 1, 136 2, 600 2, 324 398 846 , 681
September) 7_-.-.- 222 wese2ee-|-e--4-1-2- 230 145 1, 290 2, 360 1, 970 376 4, 484 10, 955
eprember 13.2. 62 ce 271 75 174 1, 300 2, 178 1, 605 335 1, 675 7, 613
September 26-_-___ 149 81 113 2, 100 2, 326 1, 438 573 1, 980 8, 760
RICO TR sree at eA Soe ee a 190 135 1, 002 2, 500 2, 358 191 1,715 8, 091
October dOt+ ofp feese eg celelb nse 82 26 468 900 1, 150 60 2, 863 5, 558
CAA STR A aa rae 140 118 1, 316 1, 900 1, 220 67 1, 348 6, 110
NWowember’ 6.2? | is0 fice tl ncot-ccu 31 6 240 270 560 | ys 2. - oe he 719 1, 826
INOMOM etre. oS eeaec anes OC aE Tes 13 2 100 139 204 FOS SS ose 133 681
Teper PIM Bere see eee Re IL aS 4 aes pape! 15 19 AS j}t ett igee 90 154
Wovemlberm eens tere OS ieee 5 2 5 6 S| A ae a i te 17 38
Nonmember 2rsicsev sd. picts. fA faye eked logs i lec eh dos lees gase sg feeb eee | ep le eet ee 4 4

Certain features of these extraordinary tables which otherwise might seem
incomprehensible are explained hereafter.

In the Copper Island count it will be noted that the column designated ‘‘ harem
bulls” has been left blank. Evidently the official who made the count did not under-
stand the difference between “harem bulls” and ‘‘idle bulls,” hence the latter may
be considered as including both categories. No such distinction was made in the
Bering Island count. However, in 1916 the Copper Island census taker actually
counted harem bulls on two dates. Thus we find enumerated on September 7
no harem bulls but 230 idle bulls; on September 13, 271 harem bulls and 75 idle
bulls; on September 26, 149 harem bulls and 81 idle bulls; and on October 1, again,
no harem bulls and 190 idle bulls. In 1918 he was evidently better instructed, for
now we find harem bulls consistently enumerated from June 28 to September 2.
‘In 1919, however, according to his figures, the harems lasted from July 2 to October
11, in 1920 from June 15 to August 380, and in 1921 from July 27 to August 30.

Another peculiarity of the Copper Island census is the detailed figures repre-
‘senting the seals counted!in the sea off the rookeries. Thus the pretense of having
counted 7,150 seals in the water on August 23, 1921, will be appreciated by anyone
who has ever attempted a similar task. The Bering Island enumerator, on the
other hand, shows better judgment in this respect, since his figures, when he deals
with more than 30, are round number estimates up to 2,000, except on September 4,
1915, when he counted 341 in the sea.

Utterly hopeless as these enumerations are, at least one general conclusion may
be reached, namely, that the bulls have been steadily increasing since 1911. The
bulls form the only conspicuous class of seals that may be counted with some approxi-
mation to the actual number present. It is true that even in this regard the tables
‘leave much to be desired; nevertheless, the fact of the increase can not well be
doubted. From the Copper Island tables I have compiled the following significant
figures:

322 BULLETIN OF THE BUREAU OF FISHERIES

Bulls present on Copper Island, 1915 to,1921 }

Gount. between June;12:and Wea 222 i ee a ey ae ID ee ae eee 104 | 103 | 287) 254} 404 271

Maximum count (on various dates before Aug. 15)__._____-..___-._-_.------------------ 507 | 449 | 682 | 572] 643 822
Average of the counts in July (not including half bulls)..-__----.-.-------.-------------- 230 | 197 | 469 | 343) 355) 477

1 Count for 1917 missing.

Remembering what has been said about the Copper Island census taker not
understanding the difference between the categories of bulls, and without attempting
to account for the glaring inconsistencies and the lack of an even approximately
orderly increase, the significant fact remains that in 1911 a maximum of 40 bulls
was counted on Copper Island. This figure, from its very smallness, can not have
been far from the actual number. We may therefore safely conclude that the
number of males engaged in reproduction increased from less than 50 in 1911 to not
less than 250 in 1918 and 300 in 1921.

It must be admitted that the census of the Bering Island North Rookery is less
fantastic than that of Copper Island. The maximum of bulls for 1914 and 1915-
namely, 106 and 145, respectively—may not be much out of the way, and taken in
connection with the fact that in 1911 there were said to be only 6 bulls on that rook-
ery it is fully in agreement with the indications for Copper Island.

While a certain degree of credibility attaches to the estimates of bulls present
on the rookeries, the same can not be claimed for the so-called count of cows and pups.
The figures presented are nothing but wild guesses, and are only less fantastic than
the figures presented for the seals counted at sea in front of the rookeries. The
acme of absurdity is reached by the Copper Island census taker who, for instance,
on August 13, 16, and 23, 1921, reports, respectively, 6,090, 7,162, and! 7,150. seals
sporting in the water off the Glinka rookeries.

KILLING RESUMED IN 1917

In 1917 the five-year closed season expired and the question of resuming’ killing
and determining the quota of young bachelors to be killed was being taken up by
the authorities in St. Petersburg. Mr. Suvorof, who made the last examination of
the rookeries in 1911, was again sent to the Commander Islands for the purpose of
looking into conditions on the rookeries and deciding these hci: barr
accordingly.

It will be remembered that at the conclusion of the treaty in 1911, Mis Sirvotbe,
on the basis of a combined count and estimate of about 4,800 pups, made by him-
self, indulged in the hope that by 1917 the total herd might have grown to 40,000
seals of all ages. After a combined count and estimate, presumably made upon the
same principles and according to the same methods employed by him in 1911 (see
page 315), he came to the conclusion that in 1917 the total number of seals was about
13,500. Naturally he was disappointed. However, before concluding this discus-
sion we believe it will be shown that on the one hand his expectation of 40,000 was too
sanguine and on the other his estimate of the size of the herd, both in 1911, a in
1917, was too low.

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 323

-\As'shown above, the great calamity of the Commander Islands during the years
preceding 1911 was supposed to have been the disproportionate falling off of the bull
element, which had reached such a low ebb that the Russians thought it necessary
to slaughter cows by the thousands for the purpose of establishing the desired
proportion between the sexes. It was with some satisfaction, therefore, that upon
his return in 1917 Mr. Suvorof found this condition changed. It may be assumed
that the count of the bulls is approximately correct—at least the number is probably
not too low. He says that 572 bulls and 172 half bulls were counted, but that a
number of bulls could not be counted on Lebiazhe and Babitche rookeries and that
consequently the total was considerably higher. The number of harems given is
265, and the number of cows per harem about 18.8, but inasmuch as the uncounted
bulls on the two rookeries were not included in the calculation the average harem
was even smaller (though it is not explained how the pups could have been counted
on rookeries where the bulls could not be counted). Bearing in mind that the average
harem in 1911 was supposed to be over 96, this reduction to 18 or less in 1917 is
certainly astonishing, especially if we compare it with what took place on the
Pribilofs during the same’ period and under the same conditions, where, with a not
unduly depleted bull stock in 1911, the average harem in 1917 had not been lowered
below 26 cows.

The conclusion is therefore inescapable that if the number of bulls was even
approximately correct that of the breeding cows was estimated too low. This is
the more plausible because of the methods used in taking the census, as remarked
before. Inevitably the number of pups was underestimated both in 1911 and in
1917. In the latter year for the first time Mr. Suvorof proceeded on the theory
that the number of black pups on the rookeries is equivalent to the number of breed-
ing’ cows.’ His figure for the two classes of seals for 1917 was 4,982. It would
not be surprising if this were an undervaluation of 20 per cent.

» . While it is admitted that the estimates made by Mr. Suvorof in 1911 and 1917
are comparable with each other because they were made according to the same
principles and methods, no value whatever can be attached to the figures given
for each of the intermediate years based upon the enumerations made by the local
authorities. The figures received in St. Petersburg during the years 1912 to 1916
(except 1915, for which none were submitted) are given by Suvorof in ‘an article
published in 1916 (Viestnik Rybopromyshlennosti, vol. 31, Sept., 1916, p. 446),
as follows:

Comparison of figures received in St. Petersburg for Commander Islands seal herd

Class of seals 1912 1913 1914 1916

Susie ps Si aie Se RO Re NO gn 37 64 249 533
Half bulls) 22-22-22. RIM A ain ar ne eee ea Ware at NR nae ee We A 19 74 255 252
RS PROS ee SL ee Ra ayn ok oh gel oe eed ee 5, 329 5, 240 4, 983 3, 326
BEES ORT PLNDIS een ee cere NC Ee en ES EO Be ee 5, 212 5,000 4, 647 4, 671
BachelorgeeG § ous oo sted sid oe esec ctr sense sltoisete ee 375 783 671 1, 365
PISO DIB LENSE LOD a 2 en See aya Seren ou eo ee SEE ate SEE oN ene eae ne pote een eee mtoos noctalue re roan 4, 000
TRIN gO I IE EST Pa LP eee 10,972 |. 11,161 |} 110,775} 114,169

1 These figures are those given by Suvorof; the actual summation gives 10,805 and 14,147, respectively.

324 BULLETIN OF THE BUREAU OF FISHERIES

Just what relation these figures bear to the official counts, the absurdity of
which has been demonstrated above, can not be stated, but no system of calculation
or interpretation can possibly produce credible figures for those years. Mr. Suvorof,
after his visit in 1917, submitted the following figures 4 for breeding females: |

EW shad a salt laser en A S30) 1914. oo oT oe
bi pape ican AR rh i nl DB, DOA TOU G3 22 trun cnn aie ena 4,769
191g! wedoiuds oh deiol deo So a4'geo/rorpiicosiii ai All ud silt to di ‘4,982

He thus came to the conclusion that while the male, life had progressed satis-
factorily, the number of females practically remained stationary, and) he tried, to
explain this alleged fact by the curious assumption that pelagic sealing had caused
a great surplus of superannuated cows, which, after pelagic sealing ceased, died. off
as fast_ as new ones were born; but he failed to explain how old cows would have a
better chance to escape the pelagic sealers than young ones. The true explanation
seems to be that the Russians reckoned with a lower mortality than experience has
shown to occur. The whole history of the fur-seal industry of the, Commander
Islands seems to indicate that the normal mortality of the Russian herd during the
migration season is even higher than that of the American herd; but, if that is true,
the expectation for 1917, based on Suvorof’s own estimate of the living black pups in
1911, should not greatly exceed the number he reported having found in that year.”

The total number of seals present on the Commander Islands rookeries in
1917 according to his calculation was 13,267. It has already been said that his
figures for 1911 probably were too low, and also that those given for 1917 are cor-
respondingly low. This belief is based, first, on the conviction that the combination
count-estimate of black pups, from the way it was undertaken, must have’ fallen
considerably short of the actual number, and, second, on the fact)that his caleula-
tions, at least those of 1911, did not take into consideration the number of seals
that remain at sea during the whole season. With these probabilities in iatind, we
have indulged in the following speculation:

Assuming for 1911 an undervaluation of 10, per cent in, the ied of ‘black
pups alleged to have been on the Commander Islands rookeries at that time,, viz,
4,661, we start with a round number of 5,100, and accepting the count, of 46 bulls
at its face value we would be justified in estimating the total number of seals of all
the other classes to have been, in 1911, including those| absent/at sea, as follows:

C020) (ee PEN EO EME AN OSD Wess aM, PSL mE TS BS RN RLS a 5, 100

Black Pups ee Se oe aig a eg ae ea 2g ea 5, 100

Bulli. ise oa oe ae fe a a Er le 46

Bachelomse coer sa ae ee a eee ta ak te oes or oa ee 1, 350

2-ybar-oldfétmales eee) ec see ae rer se eee ree el, oe a a eRe 1, 200

"Vearlinge femialege See Sere ee ee eee oe a ep OO 1, 275

WWearling tales )) U6) iu fotos tat yebee nd ate ley a alpiien maaan ania data neta ENR 1-27 ben
ho) 7) [eR pe sey epee es Nau nt Vegi EN YS 15, 346

11 These figures differ somewhat from his tabulation given above.
12 His own estimate in 1916 as to how many might be expected in 1917 was 5,830 cows (Viestnik Ry baoeagiyan eerie
vol. 31, 1916, p. 448). ! ;

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 825

Assuming again a mortality equal to that believed to occur in the Pribilof
herd,'the Commander Islands herd in 1917 should have been composed somewhat
as follows: ifiyt ¢

Wows). 22 222-2218) AY eustyss! LSS WCMVAHE DGS ty Ave ado 6, 640
VB ENGR FOO) OS 2 ee te eg leo pe an 6, 640
Bullsa(haremsisurplus, and idle). 202.222.2242 eal ee 800
BACHE OVS Sse wes o rd Soler aI mememememeee Se ashe ek op EE oe Ls a 5, 708
D=VCAr-OlG stem ales sulla 20 yaene Meemiemre ria Te ier ees es a ee Me 1, 609
eearling females se wc enaiws wiikey hh MACE NT eee he aren in tng cea oe ear eee eS 2, 066
Wearling males). iss ee ek Oe ee ee seen ee 2, 052

PERG He ENGST a a de ee eg a Le RP - 25, 515

COUTTS a a es a a Pg ea a oe Oe ae 4, 982
TECH BIS 2) fa Ea ail die th pee a fp re lp agile 4, 982
Bulls (harem, surplus, and idle) over______-____-_-+----_L------ _ + -- 744
aphelorsutey Gar lto doilpiienones bap biile suns odoet ine) gay 2, 559

It will be noted that he left out the yearlings, which, according to my figures,
should have amounted to 4,118, as well as the 2-year virgins, 1,609, a total of 5,727
seals not accounted for. Presumably the 2,559 bachelors enumerated are those
supposed to be seen on or about the rookeries without accounting for those absent
on the feeding grounds. Suvorof expressly stated that the number of seals on all
the rookeries in 1917 was about 13,267, and that the total number in the Commander
Islands herd was 16,957. With the above explanation it will be seen that there is
less discrepancy between the figures than is apparent at first glance. At any rate,
it seems certain that the number of seals of all classes present on the rookeries,
including pups, yearlings, and 2-year-old females, did not exceed 18,000.

The question now presented itself as to the number of seals to be killed in
1917. As already noted, Article XII of the fur-seal treaty stipulates that during
the last 10 years of that convention not less than 5 per cent of the total number of
seals on the Russian rookeries and hauling grounds will be killed annually, pro-
vided that said 5 per cent does not exceed 85 per cent of the 3-year-old male seals
hauling in such year. 0

Suvorof’s figures indicate that at least 2,559 bachelors hauled out during that
year, but apparently he did not ascertain the relative number of the various year
classes, which would, in fact, have been an impossible task. It is thought, however,
that no matter what figures we accept for the total the number of 3-year-old males
hauled out did not exceed 700. The Russians would consequently not have been
required to kill more than 600 seals all told; but that article of the treaty provides
further that if the total number of seals frequenting the Russian islands in any
year falls below 18,000, as enumerated by official count, killing may be suspended.
It is taken for granted that by ‘frequenting’ the treaty means hauling ashore,
for surely nobody, not even an official count, can enumerate the thousands of
seals remaining hundreds of miles away at sea without going ashore. As stated
above, it is almost certain that even under the highest estimate the number

826 BULLETIN OF THE BUREAU OF FISHERIES

of seals hauling out on the Commander Islands in 1917 did not exceed 18,000.
Nevertheless, the Russians decided to proceed with the killing. Mr.., Suvorof
gives the following result:

Seals killed on the Commander Islands in 1917

es Bering | Copper
Classes Island! | Island | Total
3-year-old males 296 238 534
4-year-old males_ 4 211 215
Half bulls..____- 35 35
Bulls Lew ls feue ) 17 17
COWS 2 eee eee eeee oe eee eee ee nL ei eee Les NEN SO mee 3 9
TOta ee roe ree oe ro eee eee ee ek rn ie Wena A Ae 306 504 810

1 From the ‘‘akts” in the Bering Island archives the seals on that island were chiefly taken in three drives: July 18, 60; July 25,
152; and Aug. 3, 88 seals.
and

The few females were killed accidentally or because they were ‘‘sickly,”
the bulls and half bulls to reduce the dangerous excess of these classes.

The latter fact is the most eloquent demonstration of the value of the! treaty
in protecting the Commander Islands seals against pelagic sealing. The amazing
reduction of bulls before 1911, which even served the Russians as an excuse for the
desperate expedient of the wholesale killing of breeding females, was not only over-
come, but it was even found necessary to kill off some of the superfluous adult
males to protect the cows and pups against the dangers resulting from the crowding
and fighting of the bulls.

In order to complete as far as possible the history of the sealing in 1917, the
following details relating to the drives on North Rookery are given:

Weight of skins taken in drive on North Rookery, Bering I séiad: July 18,1917.

[ Weight | Number Weight | Number
Class in pounds * of skins Class in pounds) of skins
534, 1 84 dl
6 ; r 8i4 6
64 1 834 2
6% 4 9 3
6% 6 9% 2
7 ii gy) ° 2
A 9 934 1
74) 3 104) 1

734 7 :
8 + is | eae O's 12° talaga iigagaeelmeetpey st asl alas er 6

Count of seals on North Rookery, Bering Island, July 18, 1917
, nb ;
In water} ) |‘
Mali Bolshoi
., Class of seals Reef Hang | \| Slyutchi | Sivutchi | , styin Total
8 Kamen | Kamen By pes
rocks
i me

HDS coe sect ee roe. ae a Sa ee oe ck ree ae 96 AB 4 _£89_ 25282 y) bi see eee ee 138
EET bell 0) CPi EO ae eee Sales Smee eeen ee 20 CY Fe ee See Die a ae 64
5 £5 (1015) (0) y ae le fal BP a Lt) 4 hte 15 379 5? Sire. sees 402
"OFA ff Sea SEY Sai ae ier Oa ae ee oe i CRBD eee ? BT) epee 1, 222
MBL ACK DUES es ee re ee A ETE ARE eS NES Pret PSTGT | 2A8. FAD 7 pl ee 6B | 1, 159
(Dead blacks mups so. ee ee ee eke ee 65 gs ee ee Cl erase = _ 68
‘Badchielors Arid cows.~ 2 conceal ee Se Le oo Weed. SID RR SES AEP SR AE ee 87 87
Mngdn. OO ONG OTe TAD Sid NOU 2,/384 434 5 230'| ©. '87 |) (13,140
UNE a ag

1 Of these, 59 bachelors and 1 cow were killed, as shown in preceding table, and 25 3-year-old bachelors were branded on the

shoulder with the figure 7.

FUR-SEAL INDUSTRY OF |THE COMMANDER ISLANDS 32:7

On July, 25, 1917,) there was another drivye,on, North Rookery, Bering Island,
in. which 152, bachelors! were killed... On, the succeeding day the following count

was made: oteo7ibel¥ :
F Count of seals on North Rookery, Bering Island, July 26, 1917

Wien Hh In water
Sivutchi or on Total

Class'of seals’ | big Reef Kamen | outlying

..y, On) the same rookery a drive, on, August 3, 1917, yielded 87 bachelor skins,
minimum weight 6 pounds, maximum weight 914 pounds, and 1 cow skin weighing
6 pounds, . rt
rid THE COMMANDER ISLANDS) AFTER 1917.

"While the somewhat exaggerated expectations wére not realized, nevertheless
there had been noticeable progress, and the outlook for the future was undoubtedly
promising. Commercial killing had been resumed and with proper management
from then on should have Buedndé a steadily increasing source of revenue. The
total of 810 skins taken ‘was obtained from those age categories of seals that were
practically’ nonexistent five’ years before. ' In 1911 in order to obtain 200 skins it
Was necessary to kill 200 cows; in 1917 four times as many young bachelors were
killed without difficulty, ‘and not only with no danger to the future welfare of the
herd but to its decided advantage. Evidently the restoration of the Commander
Islands fur-seal herd was only a question of time. Unforeseen complications, how-
ever, intervened to the great detriment of the herd.

j ‘With the Russian revolution of March 15, 1917, a new chapter in the history
of the seal herd was inaugurated. The immediate Hs on the islands was the
retirement of, the old tite In 1920, after the Bolshevik Soviet Republic had
taken. over the Russian Government, Pietr Aleksandrovitch Khramof, as the head
agent for. the fisheries and sealing industry, was sent to the islands, He arrived at
Bering Island on September 6 and afterward made his headquarters on Copper
island. He was young, enthusiastic, and energetic, but without experience.
On October 24, 1920, two Bolshevik representatives of the Communist Go vern-
ment of Kamchatka i in, Eatpopectee <i arrived. They stayed on Bering Island long
ae to organize a, Communist Government and then departed.

In the spring of 1921 two representatives of the Soviet of the Eastern Republic
were landed at the islands. To them the natives of Bering Island delivered the
furs taken during the winter of 1920-21, to be converted into supplies of various
kinds. According to the official receipt issued to the fisheries agent on May 27,
1921, the following furs were taken: 614 good blue fox skins, 2 damaged blue and
6 white fox skins, besides 4 whole skins and 1 damaged skin of gray fur-seal pups
and 1 skin from a full-grown seal found dead.

8328 BULLETIN OF THE BUREAU OF FISHERIES

In the meantime the Soviet Government in Vladivostok had been superseded
by a “white” one and plans were laid for the supression of the Kamchatkan Federal
Republic in Petropaulski.. The ‘‘white” government in Vladivostok was in turn
overthrown by the forces of the “‘ Chita Government,” or the so-called Far Kastern
Republic, during October, 1922. Shortly afterwards the Petropaulski Government
was also deposed and the whole Coast Province, including Kamchatka and the
Commander Islands, was again in the hands of the ‘‘red”’ or Soviet Republic.

The details as to what took place on the Commander Islands since 1917 have
been set forth in order that the conditions on the Russian fur-seal islands during
this highly critical period, when the seal herd needed the most careful nursing, may
be fully understood. As will be seen, it was mainly a period of instability and lack
of authority and discipline. The natives had been taught not to pay any attention
to the orders of the officials, and naturally all law and regulation practically ceased.

That under such circumstances complete chaos did not ensue speaks well for the
common sense of the natives, yet it was but natural that the seal herd should suffer
much as it did during the ‘‘interregnum”’ from 1868 to 1871 (Asiatic Fur Seal
Islands, p. 117). Alcohol now, as then, played an important role. The natives,
unrestrained by the strong arm of the Russian officials, would sacrifice anything
for the chance to get drunk. With the many vessels of varying allegiance—
Japanese and ‘‘white”’ and ‘‘red’’ Russians—touching at the islands, there was
no lack of supply, and the means of obtaining the much desired alcohol were the
furs of seals, foxes, and sea otters surreptitiously killed and secretly stored away.
The number of fur-seal skins disposed of in this way by the natives may not have
been very great because of their bulk and the difficulties connected with curing
them, but there is reason to believe that in 1919, 1920, and early 1921 there were
shipped fur-seal skins in considerable quantity of which no official records were
kept.
Of course, all these transactions being illegal, it was next to impossible to
obtain definite and reliable data. Even the crews of the Japanese men-of-war
that visited the islands several times each year were reported to trade extensively
with the natives, obtaining furs in exchange for liquor and old clothes. Copies
of official reports to that effect were furnished the writer.

With the practical abandonment of the naval seal patrol by the Russians after
the revolution, the seal pirates began their destructive work again, but while before
the treaty of 1911 it was lawful for them to sell their skins in the markets of at
least three’of the contracting powers, they were now compelled to dispose of their
illegal catches surreptitiously. Consequently, while formerly it was possible to
demonstrate by figures the pelagic losses of the seal herd, it is now only a matter
of conjecture how many skins were taken at sea and in raids on the rookeries.
If it be recalled with what boldness the seal pirates operated about the islands in
1911 (see pages 316 to 318) almost in the presence of the Russian men-of-
war, it may easily be understood that during the critical years of the revolution
and after, when no naval authority was there to check them, their activities and
recklessness increased from year to year.

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 329

Schooners were hovering around the islands and were sighted from time to
time by the natives when the fog lifted long enough to afford a clear view of the
ocean, but as to their catches nothing definite is known. It was only when the
crews attempted raiding the seals on shore and were discovered in the act by the
rookery guards that tangible facts became available. Thus in 1921 North Rookery,
on Bering Island, was raided. The schooner escaped, but five Japanese sealers
were caught on shore, and were afterwards taken off by a Japanese transport
and presumably taken to Japan and given a trial according to the treaty.

In 1922 no less than four schooners were observed hanging around the rookeries.
On July 11 a Japanese schooner anchored off the seal rookery on Copper Island
and sent a boat ashore with five armed men who started killing seals. The native
rookery guards, who had been watching the performance from behind the rocks,
suddenly descended on the beach and seized the boat, at the same time firing at
the landing party, who ran away. The natives also shot at the schooner, which
was compelled to weigh anchor and disappear, leaving the captain and the owner, as
well as the senior machinist, behind. They, with two sailors, surrendered to the
guards the next day and were taken to the main village of the island, where they
were locked up by the fisheries agent.

Less than two weeks after this affair another schooner anchored off the
rookery near the southeastern end of the island less than 300 feet from shore and
was caught, together with the captain, the owner, and the entire crew, 14 men
altogether. On board the schooner were found 96 fresh sealskins and 5 live blue
foxes, which it was stated had been caught on Bering Island where the schooner’s
men had also been ashore.

It is not even necessary to give credence to the assertion that the natives had
on occasion accepted whisky from the raiders in return for permission to kill seals
in order to understand that with the lawlessness and the loosening of all obligations
which followed the revolution, as set forth above, the slow but sure increase of the
seal herd up to 1917 met with a decided check, so great in fact that orders were given
in 1921 immediately to stop all official killing of seals on land. A report at that time
represented the exact number of seals on the Bering Island rookery as being 4,339,
including the newborn. On Copper Island the day of inspection was very warm
and besides it was very late in the season after the harems had broken up, hence
only an approximate estimate could be made, the total number of seals there being
estimated at 6,000. As the total for both islands consequently fell far below the
18,000 stipulated in the treaty prohibiting further killing seemed justified.

The writer was assured the Bering Island census of 1921 had been made
‘‘head by head,” and that it was not a mere estimate. It was taken on August 16,
1921, and the detailed figures follow:

LCs to uns ee eraete oe een et Una LE) MLM LMU SIG, oes UMN Me a ge 48
MOU OU ben Meare Teese Aen Mi rece es Hon me eee ees 60
Tet AVAL oy OT BESARe SD Nh) TY SB ee ee 56
Bachelors. -J4ob 240) Beit eign Sospllfoh sory cod dbo egle | ard 418
Cowisste Sa ee yy hie aly ee ae Teh ie or ba Se Ye eee teers 2, 221
roa EE ete SOAR a} s4 eta the Obs gg DOB He Say 2g Ce ty Oy tN OR wR OE 1, 488
Pea Thin gsm c pana rence vente a. SN Reis! tee ua Dy lee oe ae el 98

830 BULLETIN OF THE BUREAU OF FISHERIES .

Experts will know how to value this “count,” and the writer calls attention
merely to the curious proportion of cows and black pups, there being nearly half
more of the former than of the latter, a feature that characterizes so many of the
Russian estimates. .

How the estimate of 6,000 seals for Copper Island was arrived at is not known,
but it was certainly not based upon the periodical count of the rookery. guards,
which for August 16, the same date as the above Bering Island count, gave the
following figures: ;

Maret bullsi!OP8 Bie Olh DAT TSR IO Ta Shea OT 8e NE SATS Bea Od 118
Buills:without haremse Jao. sclt Bo rested sce ore aocule: weap ecrplo ps 264
EPA TE RWIS, Bas apt og Et Ba as On a i ce El) lk: a eg 220.
Ba @hedors iy sage a ye a cp a CU eg 2 0 1, 381
COWS! 2 Ue See a TE hee Ra re ee te RO we 2,680 |
ij F-Xed egg 0) 0a ea ae Nh lg a ey Cpl oa ~ 2, 200
Yearliipede OO) UR) JOT LOR ae TRS En SE AR TE SOW: 2 NCR 80"
Inthe! water_dinbseoedi sorte sb ago gett been saysle ele glass eyez oe by 6 7,62)
gl 0 2) bl eh hed ing gf Gh Sa diene one cl ie a At Ss 14, 105

Counting the 7,162 seals in the water must be regarded. as an astonishing per
formance, but does not inspire great confidence in the accuracy of the rest of the
count. However, disregarding fice fantastic figures, in view. of the status of the
rookeries as the writer devel them in,1922 we Ane no hesitation in accepting the
statement that in 1921 there were less than 18,000 seals frequenting the rookeries
and that there was ample justification for stopping the killing of seals for skins.

NUMBER OF SEALS KILLED ON THE COMMANDER [ISLANDS BETWEEN 1917 AND 1922

Unfortunately complete official returns as to the number ‘al skins taker 1. since
land killing was resumed in 1917 until it was again stopped in 1921 haye not been
available. The writer has seen only Mr. Suvyorof’s figures for 1917, as, given above,
embracing the killings on both islands, amounting to 810 skins, , In, the Bering Island
archives were found reports of killings on that island only for 1918 and: 1919, and. with
regard to these it is not certain that they are complete. for 1920 we know of no
record except a statement said to have been taken from ‘‘A report published, by. ;the
special delegation of the Far Hastern Siberian Republic to the Washington Conference,
1922,” to the effect that the production of fur-seal skins on the Commander Islands
during 1920 was 1,000, probably only an approximate figure, which may or, may not
include 813 skins shipped by A. S. Yakum in the Admiral, Zavoiko in May, 1921, as
the Magnit took 400, skins away from Bering Island on September 9, 1920. On
Bering Island in 1920 the natives, were given permission to kill 100 seals, but
they killed 400. These were the skins shipped on the Magnit. ; The Yana, in October,
1920, took the skins of 73 half bulls and 6 pan, and in Petropaulski we were
informed that the sailors of the Yana brought “a lot of dried sealskins.”” It was
also said that 52 skins were landed there is 1921 from the Kamchatka Manus,

On Bering Island the following killings were made in 1918: July 16, 115
bachelors, and July 24, 187 bachelors and 1 half bull, a total of 303 skins; and in
1919: July 3, 107 ee ee and July 7, 68 bachelors, a corel of 175. T1921, 66 seals
were killed July 7 on Bering Island for food and leather for the natives, and 32
seals were killed for the same purposes on Copper Island.

FUR-SEAL INDUSTRY OF THE COMMANDER ISLANDS 331

~ When the writer left the Commander Islands on August 8, 1922, no seals had been
killed Officially, nor was there any evidence in the Bering Tibnd salt house that
any skins had been salted, and the natives complained bitterly over the lack of
fresh meat. We were informed shortly afterwards that a killing of 500 seals on
both islands for food would be permitted later. The killing of over 2,700 seals is
thus ‘accounted for. ‘To this number about 1,600 robably should be added for
Copper Island during 1918 and 1919.

CONCLUSIONS

Taking all the above facts into consideration, there should be no surprise over
the fact that the Commander Islands seal herd has not recuperated as rapidly as has
the Pribilof herd. The writer’s dismay at thesight of North Rookery on Beri ing Island
on July 28, 1922, was primarily due to the fact that it had not been seen since 1897
and no information giving an adequate idea of the conditions as they existed in 1911,
the year of the seal treaty, had been received. It was, of course, known to me in a
general way that the herd was then in a serious state, and even that females had been
killed, but nothing had reached Washington to make one realize the full extent of
the. disaster. Furthermore, the complete story of the depredations that have
taken place since 1917, the Humaber of seals killed on land and at sea during those
years of lawlessness, and the lack of authoritative control on all sides probably
never will be known; but allowing for all this, there seems to be a further circum-
stance which may have had a baneful influence in retarding or even checking the
rehabilitation of the Commander Islands rookeries.

The northward movement of the two seal herds in the spring along the American
and the Japanese coasts, respectively, is approximately alike on both sides. .How-
ever, as has been pointed out heretofore," the Commander Islands seals off the Japanese
coast in the spring congregate in relatively greater masses and for a longer time in
certain limited areas, a circumstance that makes them much more susceptible to
close offshore attack.

At the beginning of May the seals commence to crowd the Gulf of Mororan,
between Hondo and Yezo, and for nearly three weeks they are assembled here,
during which time the heaviest damage is done to the herd. Toward the end of
May the seals move northeastward along the coast of Yezo, comparatively near
land, making a last stand to the east of Iturup about the middle of June.

The islands of Yezo and Iturup are the home of the Ainos, the aborigines of
this region. Now, Article IV of the treaty of July 7, 1911, provides as follows:

It is further agreed that the provisions of this Convention shall not apply to Indians, Ainos,
Aleuts, or other aborigines dwelling on the coast of the waters mentioned in Article I, who carry
on pelagic sealing in canoes not transported by or used in connection with other vessels, and
propelled entirely by oars, paddles, or sails, and manned by not more than five persons each, in
the way hitherto practiced and without the use of firearms; provided that such aborigines are
not in the employment of other persons, or under contract to deliver the skins to any person.

In practice this article is probably the easiest to evade in the whole treaty.
It has been found so on the American side. Effective control is very difficult, and
there is a strong suspicion that every provision of the article is being violated.

13 Asiatic Fur-Seal Islands, p. 263 and PI. 113.

Pt el egaahetp sy

‘Z

832 BULLETIN OF THE BUREAU OF FISHERIES

Dr. G. Dallas Hanna says," referring to this question, that dispatches in the daily
press indicate that the figures as to the number of fur seals killed at sea under the
treaty provisions are assuming proportions little short of alarming. Of course,
official information as to the illegal killing and disposal of the skins can not be
obtained, but the loss to the herd, especially as such a large percentage of the seals
thus killed undoubtedly consists of gravid females, can readily be appreciated.
If conditions are thus unsatisfactory on the American side there is no reason to
believe that they are better on the Asiatic side, since in view of the peculiar bunching
of the seals there, as stated above, the losses are probably proportionately greater.
If there is ground for believing that the mainstay class of the Pribilof herd has been
reduced in a single season by over 1 per cent, the losses to the Commander Islands
herd presumably exceed that figure considerably. The future prospects for the
Russian fur-seal herd are thus less bright than for the American herd. The annual
mortality is almost certainly greater, hence the recuperative power of the herd is
correspondingly less.

This condition can be remedied only by a more effective control at sea according
to Article VII of the treaty, both during the spring migration and the breeding season
of the seals, as well as by a strict enforcement of Article IV. With regard to the
latter, the only rational remedy would be the revocation of the privilege of the
aborigines to take seals at sea, as suggested by Dr. Hanna.

14 Report of U, S. Commissioner of Fisheries, 1921 (1922), Appendix VI, p. 110.
18 Hanna, op. cit.
54

GENERAL INDEX

Page

PMoantinnpifilosass 22208 oe 108, 142, 144, 145, 146
CUE GU TER Ga ke eae a a Dd a 142, 144, 145, 146
MeO PROTITUL Se: see ees ere hs at ee 142, 144, 145, 146
occurrence; .1922)and 19232222 2 ee 142

_ 141, 142, 145, 146, 151, 162, 170

278

162

123

Actinoptychus undulatus: --2----------.--- 2-1 113-115, 120
adriaticum, Rhabdonema..- 2-22-22. 2 eek 113-115
adspersus, Tautogolabrus____-_...-..-_------- 105, 166, 170, 172

eenus, Myoxocephalus

-. 165, 166, 170, 172

MPeTAvONADTOMUS eee: 2 Sot ee 244
MBauUalism Mh ysanopodazce—.._.. 2S A nee ees 153
mstiva, lwabidocerass =... 222 141, 145, 146, 151, 162
Qfinis; Wpogeblas~—s. +... 22 eek 156, 158, 159
age at maturity, Pacific razor clam, Siliqua patula

(DWE ay 8)) A a le ER a 201-236
FATAL ADEs ING XUICOD 0) 00: cla ea ai ea a 136
Aglantha conica__________ 125, 130

UPItaliss Hee wn oe 124, 130
Agonostomus._______- esa a a Ge Waa 266, 267

ie aGyay PLO) re A Lat oc plage cll a la es 241, 267

SGFINS) T(E UR as © ls er ele el i a a Ee 267

SHENAE es oe a Se gaa a 267

LEVIN GSO 3s a sleds ay pe nee eg 267
ALLER OLN aMeruetaente ee se Sy oes kOe 129, 130
alata genuina, Rhizosolenia_____________ Pe LZ LLG lly, 120)
alata gracillima, Rhizosolenia.________.___________ 112, 116, 120
alatawivnizosolenia:- 922 = 22 2 | 108, 104/110; 119
MPAs sOrmoOpleurac- 2020 oo 163
PMOenamarvargiensise) 2-026 2100 set LM a a 137

131, 133

264

153

SIERO CE ae ene ee eee ia HAG
Dterawsesiadulphign ss ost 2) ee 113-115, 120
atennagum Bittium: .. 622. ee 137
Alteutha depressa....-..._.....-..-2----2--4---22. 141, 145, 146
Ameiurusmebulosus: _- +... 22-22 pseu dgectueslecs 9
americana, Ammodytes......__.-_-_-2-s2-22sel222 166, 170, 172

Neomysis 152

SATS LO Late ns len. enaPe en Sie S.C tae mpl IES Rey Oh eg ame 141
emenicanus,, BlOmarus: <2 0.24 205002 22 ee oe ee ee 156, 157, 159

Pseudopleuronectes 133, 166, 168, 172
PAMETUTINOGHYILES x reas hs oe ke ET We 164

americanus .. 166, 170, 172

occurrence, 1893-1907.
mmMmpelises Compressa.....--._---.------------see2_-une
PTIAGHOCO Nala saernes NL ee ge 150, 151
SDURUDES Mme mtad remem Tato Nok a ee 150, 151

amphibians, enzymes

Aum phineura 22 ees

ADD pod aes ee anime same Ling 800) eigenen 149

amphipods, occurrence, 1922 and 1923 ________________ 149, 150
80709—26——2

we

Page
Amphithoéionewmang se ssse sees eek ee 150, 151
TULL ICAL eee ee et ata ae Su mend fe eee ERMENO) By, 150, 151
PATTI PICT Ge ONAL Ae tee ee ee en et Nee RU 131, 133
amylase 192-195
Anablepid____
MADICDS* oste wee se eae is ae eee A Ee eae Se
CO Vili Bete Neer weer ee een aa stern ime k
CO WLieewaie scape eyes Mpc Cro, Pte mee a es
ANADIEDS © ODILIS teeter ee ee ie aes sre beer 261
PATI CHS OVALS eee ee eee ree ne amare ie ye Parsee ee nn 137
ANACYRUUS PUAVCUIICDSIS sos ete oo eee ee ee 246
analogus sw pinephalusues ses ses bee eee eee 285
Anarhichas lupus_______.____ 168, 172
AM CHOVIaNDICVALOSLLSee ee ce oe ee ane ee ee eee 284
DLO Wil eeee ee emer ene: emer Ae eon Meaty S 167, 172
CXISUA Meee eee peas B Lata Soe pie Rea bos 283
pangimensisi sf) foe eesti Srpeage 284
RAs trellis Heese = ies a ae ares alta prec enn tate 284
Anguilla rostrata_________- 167, 172
anpiineus WProteuS) 2) oon Se se aby 184
ATIISOUTCMUSIAOVilemer eats ee ee Va ey ee 287
ACN CITE tee eee aie cs Seba pe a) Vella aobrrnatien 287
annelids;‘occurrence; 1922) 27.) etd ee 131
MASINI AG a cepa ae eee ay, SR eth San bel al el | 130-136
annulatussGammartisecs= se soe le ree 149-151
Snnulipes ehacurusmeesee ste: ee eee eee 158
Anomalocera patersoni___.___._____________- _ 102, 143, 145, 146
“Amoplodactylus lentusie fs see ee ec a 161
tADeLLCS eee te emcees ese ae eee eo Oe ues ae 64.
quadracus-._-.__-- See Oe la Ee le ee 167, 172
apicatasStOMmotoca sae ae wanes de 123, 125, 129, 130
apiculata,iChsetopleurawee =) see. 2. ena, pee ee 138
SANTO OTICICtILaAT ge umen Manns Sie CL anos Wes ALS el a Lo 162
OTL 1 Caled Cl RUB 3h SR et EF SI she vl ind laden a # 163
lorcunTencey 1893! tOnlO0/ sae nasee cn ee ones Meiers oe rhe 163
19221 an Cyl O23 tee sues eee ee ee st eR BD
appendiculatus, Stephanopyxis.__________________ 113, 115, 120
approximans,, Polydactylus..._......-_.--___---!.-tu-. 284
IBOLYMeMUS ON sees es eee ek Sas Mag tert. ween Ae ae ay Od
MArabellatopalinaty- asses eee oe ee ne ee 131, 133
Anenariay Viva 3 ee wees Ne) a Oe ul 137, 213
argentatus, Astyanax________ 244
argentiventris, Lutianus 286
INPeSODrIOT WAU ane E fr na is ile pee yo 286
TNGomseris Minima 2) ile EE ht eae 286
PANIC op Baste semen ENCORE LN IU ia alleen ee eben Sie. 243, 248
PAT IUS Ds eIN EON ee od Cg 2 SAN Dee UO he de wee 250
248
250
251
euatemalensisa- meen ee Seu Leg 249
(Pay LOT eenaes eewey easier tee hn SEF ice a8 age 241, 242, 250
farius;¢him cloduse.. 24s ila a ula Sk eos pga gd 250
armata, Lysiosquilla___..________ el OU sane ey sayy SAU a 154, 155
armatus, Centropomus.--_.__-.._..-----L--222-2-----: 271, 285
PAW GA TOSUTACA icra etree Le, ee ORI Pa 149-152

304

‘Asellopsis sp 2-222 U0. S55 ee Lo ke EP ee epee 146
Aspergillum
Asterionella japonica
Astyanax

argentatus--___-
fasciatus seneus--_
AStyrislunatalc 20s! S250 se ee eee
Atherinichthys pachylepis
Atherinids
Thyrina- -__-
guija.
'Thyrinops!pachylepis- 2.22. 2eses = eee ee ae 284
surantiacus, Balanoglossus:: 222. --- 2222-5 ee ese eee 162
We 2iid Soci Ln eee eB een en ee ee 285
a ee eer a oy eee ee 124-127, 149
103, 125, 126, 129, 130
OCCUILENCE, (1005100 LOU ha = ees = ere 126
Autolytus alexandri 131, 133
COMNUtUS: 222 oe Ee Se ee avant dale 131, 133
QIN OLCOW 2 ees Ny ae Ee ea pee aa = een 131, 133
longisetosis 131, 133
OMMACUS oie eee eee eee see 131, 133
varians..-_ 131, 133
AV ALAA MACHIS. 22 Se eae ee ee ee eee es es 137
ayaniutjanus- os 02-6. oiler eae eee ee ete 182
bacciferum, Sargassum : 22-2202 = 220222 <2 2 eae 121
bachei.-Nemopsis222¢2 32s 2 eee = re _--- 124, 125, 130
Bacteriastrum delicatulum 118, 114
Varlansieets oop lenses encca oe 115, 120
DIC es a er re 240, 250
Balanoglossus aurantiacus ----_--------- DE Te eee rns 2} 162
balanoides, Balanus 147, 148
Balanus: balanoid esac ses sss eee anes 147, 148
CFenatus. 26 a sae Se ee ree ee 147
QbUMNCUS = =~ 52552 oe ee a ee ee 147
palticd; Widotheas.2 22-2 ss. no ae ae eee ee 151
barnacle, larve, occurrence, 1922 and 1923__s___-_--__._- 147
BSS LOC Rea re a eee pre aa ee ee 5
Srnallmouthiblackwys sass oe ee ae ee eee 3-6
Win te thes ra Se ae sa ee 182
pepticidigestion ss sce = Sas==- saan eee ere 185
Batea secunda 149-151
beli,|@lnysem ys 29s sen seen a ae 182
Bellerochea malleus_-__--_------------------------- 113, 115, 120
belsanussMielanirisa= oss ses seen == === een een een 264
bergonii, Cerataulina 118, 114, 120
Bering sland eee soak eee ae es oes eee ere ae eee 289-332
Kaishotchoye rookeryes) 222522 S82 2-2 SS eee 296
‘Poludionnoye rookery-- - ==2- 22-2. 222. eee 296
Reef, North Rookery JIZZ
South rookeryese- oe ose ee ee ae ee 296
Dernegat@sess 2-2 sane een ase = ae eee 287
BLOGS aren ae ae eee ee ee eee ee Nat 124
CuCcUM ISS Ss ae ei a5 Seen eae pease _-. 129, 130
picophora,| Olytiasssss=se= asa a saa ae nn een ene 125, 130
Biddulphia‘alterans: 22-2: 22-2222 2=- 72S eeaeee = 113-115, 120
biddulphianasss a=. eewe see eee arte ee ee 113-115, 120
fAVUS See eee eee ee eee a eee _-. 118-115, 120
PTAnUlatases ee asses eee eee eee een 114, 115, 120
rhom busts se 2 ote ae eee ee ee aaa eed 113, 115, 120
VOSICULOSAs2e+ = ek ne sec se nent ence eee eees 114, 115, 120
biflosay, Acantiave.2a=2 sea ee ener er ees 103, 142, 144-146
bilinearis, Merluccius.—--+--_--_----_-- -2-fivedeue 165, 166, 170
bispinosa, Euthemisto 149-151
Bittium alternatum—-—2-2---2-2=-="--_-- =" eee 137

GENERAL INDEX

Page
black bass) smallmouthse 2000 ee a eae ae 3-6
black crappie 182
Black tumor, catfishos2) 295520) See ea 9-13
description: 25. te) ee 10
microscopicistudiye= = 20222 aoe eee 11
studies'and experiments=— 222522202) ceaneeee pene il
bluegill
digestion tests with invertase
food
peptic digestion
IB olin aa be eee oe a
alata, 129, 130
Bopyrids. 22-2. 22 2 ie ee 151
boreale; Chtsetoceros.- — -.-. 9 2 ee ae 116, 118-120
borealis;'Cancer. 2. s...4. 2 See 159, 160
borreri, Melosira-< 2. =... -_ 2.2 ee 114, 115, 120
bostoniensis; Macelina..2......_-_.c.. 9) eee 137
boucardi,:Poecilia. <2... ee 255
bouchellei, Roeboides:_.¢2___ 2 aa 248
Bougainvillia carolinensis_- 123-125, 129, 180
superciliaris 123-125, 129, 130
Brachyura_) 22s oe ees bs 1S 159, 160
occurrence, 1922 and! 192322229). - ee. 5. 5 ees 159
brevicornis, Oithona_____- Rg 141, 146
Dachidius_Y2--*:_ 20. 2 eee eC ee 144-146
brevirostris,, Anchovia-_--+-- 225 Season ame ae 284
Pallene: ..-.--=-----...-...---2 See
Stolephorus__--___
Brevoortia tyrannus
brightwelli, Ditylium__----___--.___- 103, 105, 113-115, 118-120
Bros.nius brosme
brownil; Anchovia.- 02-2022 -._. 2 5 Se 167, 172
Buceinum undatum: 232 22_) = 2-25 eee 136
bucera, Nephthys _- 131, 133
bull snake______-_- -- 182, 184
digestion tests -- 192-196
with amylase_ 2.220 -25 2 ee 192-195
with invertase. _---22-204 52 ee 195, 196
ereptic digestion- --- 191
peptic digestion=-- 2-- 225-22 see 184, 185, 187
tryptic digestion... ---..-2) 2." -- ees 188, 189
bullhéad!=... 222 --- 2.2.2.8. cet eee 2, 3, 6, 13
9
273
Busycon canaliculatum)---- 2-25) 2252 522 136
Byblis serrata..-2---2-22 22422222255) See 150, 151
esxea, Chiridotea...--/4--52242..--2:---5 eee 151
Calanidi:...24. 25-2 4222222--- 4-205. eee 146
Calanus'\finmarchicus. 22222225 22225 eee 143-146, 175
calear avis, Rhizosolenia --- 112,115, 120
@aligidee..2. 222-5 2525-223 22222724532 4he i ee 146
Caligus schistonyx--=*-+— => 252222 lass. sae eee 144, 146
callarius;Gaduss= 22. ===222=2 22225

Callianassa stimpsoni
Callinectes sapidus____------------
Calliopilises2222 222-22 en ee 2 oe oe ee ee
Teoviuiscuilus.2 cee 2a Ce ED eo eee ae
campanula, Melicertium
canaliculatum, Busycon
@ancer borealis: 2 i22225.s2.4s4e 22442 oes Se
ATFOFAbUS= O25 42-5222
capillata, Cyanea
Waprellan 22 te eet
geometricas.--------5 =
Jimearis: 2. 2.22 22.-22454 Sein CO gu ge ea 150, 151

GENERAL INDEX

Oy arenrypai cl ee Meee eine a cere pin es a rape pad fr) ea
(Caranxiippos estes oS ee et lal ye
Oligoplites mundus

SEvtLIsTIS aeraes ale bees ag

Caranx hippos-_---__-

Carcinides meenas-__-_

POCA ba a

__. 155, 158

Carinogammarus mucronatus___._.__.--___-.--------- 150, 151
Carneay Podocoryne....----.--.-_------=225412£22. 123, 125, 129
carolinensis, Bougainvillia____..____.__.___- _ 123-125, 129, 130
carolinus, Prionotus...-...------- 165, 166, 170, 172
CENTOS 2 eS Um 52 2 SOLD 182, 184, 186
TIRE SBIOM ee Ve on a oe eae a oa eee eee wattle 184
kitgestionitests2.. 22.) RRL Ob Se aes 193-196
wtbjamiylasee Ss. <. --=---. alk Lees: ab alg 193-195

ETE LRITVOLL ASG satin seen eee ee te ULL 195, 196

erepuic, digestion=—---22=-- 1222522. eae sl 190, 191
Peppichdipestion== oss ee eet et AVES 185, 187
tyne GIPCStIONE Aesth osteo eee UL 188, 189
carpio, Cyprinus = IS H82
Catena sPhvillodecet=ss%e2-—- 2. 2 ewe ce SOS 131, 133
CULIS enna enema tee ss 2 Ph ee 12, 240
lacksimmMoraeen tos! 2 a otis! 9

Repencertete se i cee ee UDO eer 248
@atostomus commersonii-_--__._.._-_.--.-0 2-2 L212. lli.. 182
catulawiilysiellast 24) 222.22... 22 el bi Ll a eAOREAE ES 137
cavolinii, Tanais 151
Centropages hematus_----_-___- Liahicuisibes eal 103, 1438-146, 162, 170
occurrence, 1922 and 1923____ 143
AVAIL UTS eee een ne ane ste a GN a lpn 141-146, 151, 170

er Pere ee Stee eer eel LAPSE HL LRilE Mell nla PRD 146

Lia cc incre a eh Rapa 243, 268, 285

ie Soe abe ts eames ceo foe AL IE 268

Le ey cis a Ue ae a 240, 241, 269

240, 241, 269, 271, 285

240, 241, 269, 270, 285

Le Sh ye fete a gn ald 268

ep SAS NE Coan an 27 , 285

Lee parc ace eR AE = dee tesa i ny RPTAL

eS eee See SSR so pee epee ls ays Lita le 271

Lee ates nh rte cee Sea Seaeratags| etal satget 240, 241, 269

240, 241, 269, 271, 285

240, 241, 269, 270, 285

Eee faeen eet ai Siren 9 - Urs s Sr 269, 271, 285

eee sae EOE

re aati nereit. < eeraw enw AOANLT SEL OOP 240, 241, 266
Ts eee 113, 114, 120

Src SSS bets SES Hae aE mee ener eas See 121-123

i Ie a Re es are eel ee 123

TRAC LOCOLOSS 2a eee sere Soe ee 2 net re tg 121-123
tripos___--- 121-123
Cheetoceros 105, 117
abundance, 1922/and 192320) 2202 se 117
atlanticum. .. 116, 119, 120

{01} GEN (Ey A Se ee See eee ee 116, 118-120
cinctum-___- pe ae aoe oes eas 115, 119, 120

116, 118-120
116, 118-120
116, 118-120

115, 118, 120

-. 115, 118, 120

salOUe 116-120

densum lala et a ee N -- 116, 118-120
Giadem assy sge ee 2 O8 eee eile 6 pauls 115, 118-120

didymum! ses 4) aise ies aie ee Nl BONY 115, 117-120

335

Page.
Cheetoceros—Continued.
distribution, 1922 and 1923__-_.--_-_-. asst eeee 118, 119
RACINIOS A. eRe aaa n okt eee aOR eae 115, 117-120
lorenzianum 115, 118-120
boo) is eran ee 115, 118, 120
peruvianum 116, 118-120
SCHULLER soeens ae east Oe eee ei SoU es 115, 117-120
SOCla] Owe sent Ry memes ae Mee cee ee Sew Sie 115, 117-120
CORES Sasi ee tts So ok eee 115, 119, 120
Willoieeeneeceae tenet) Bie Wes 116, 118-120
Cheetodipterus zonatus___ 287
Cheetopleura apiculata___ 138
cheetopterana, Pinnixia___ 159, 160
Chestop tents wees ss eee eee eee aa) FE erased 160
chalceum, Pristipoma-_--_---- ele oo eae SU Sy 286
chalceusOrthopristis#:- 2. 20-22-22 = see 286
IPTISEL PO Be ee are ee ae eee ee ree ED 286
chamarras sees een
Characinide
Astyanax! eo ce tok oe ca bh see ee 244
Roeboides_--_------- 246
Characins ee ee 240
chelifer, Harpacticus__ . 145, 146
Ghreloniate. Sa idiv Ue Riles AM as eared oye ae 185
@helydraiserpentina een eh aed eae 182
Hevh ga} of cy ys seh eS a pe ret A 267
Chrimibo layne eee APRN Nal Ce a iy 240, 253, 255, 258, 260
ETIT] COSY Orme teste arc es an ete Hw 2h et oT ee 273
chinook salmon—
age: determinations]. sesso. o- avetn Siena See 18
Columbia/River itakenec. 2.02.0. 0 een ena, 28
age groups, abundance_-_.._.:....--.----------- 43
fish more than 1 year from maturity, percentage_ 42
fish taken in ocean, relative maturity 29
FA WON (UN 0 eel eg Uke te es st a 48
immature fish, age 39
percentage_____._.._--_-- We tee a ear io 40
correlation between size of eggs and size of fish__-__- 27
MrakesiBay takenee see te er ecn tote ce act ee
Bort Bragg shiirome sess ue on So
growth and degree of maturity in the ocean
maturity, determination. ---2)---- 2) ee ene

Monterey Bay, taken__-_____
age groups, percentage

TelagivennAaAlurity. =e econ see ene el oe 66

SIZOSS eR acre eerie ens tad eo eer 63
@hiridoteascesCa- tees Cre a eee ee eee teas S 151
@hloridellaes eas a ae MeN ere On ees cee ete 153, 154
(2) acl) 0) SE ke ae rae oe tes debe ek ek Sel 153, 154
occurrence, 1893 to 1907_--- 154
chopa..-. 287
Chordata___ . 162-164
@Hromid css ee eee ee eee eee) aoe 272
Cichlid'se yee ens waren eee er mene a Meets 243, 272
Cichlasom ates eae ee oe oe ee 272
macracanthussesssse0=-s-- eases 241, 272, 274, 276
Pemeckiviee? een aera an 241, 242, 272, 275
motaguense _- 239, 241, 273, 278, 279
nigrofasciatum 241, 272, 273, 280
Crimaculatumie = ee eee eee 241, 273, 277

@hrysem ys) Delis a ten weeen Rees aa rs ee aoe 182
CINCT AR Siete sone ale eh erent horse No eee 182
CDEYSODS #@ROCCUS seen enemies sep oee Le Shs Se cee 182
STENOCOMUS ss mew oy een unt iop tes a2 165, 166, 170, 172

. 147, 148

GWichlasom asin meee cea Oe nei etree UE Neer Pepa ra els 272

336

Page

Cichlasoma—Continued.
miacrocanthus. 2.2.0 o-oo SS BEE 241, 272, 274, 276
mieekil! os 2258 le AAD IONE E275
THOLAPUCNSC Los ooo __. 239, 241, 278, 278, 279

241, 272, 273, 280
Beene etre |! 241, 273, 277

@ighlideett.ts oie ee ee Be sae 2 SEIZE 272
@idhldsomia fesse eee ey ee aera ae 272
MBCTOCANTNUSE: sas sosse ss Sere eae 241, 272, 274, 276

TOOK ces fs og os oe pcs ee eNO DADRA ORO TO
motaguense___-__._.___.-..---.-__ 289, 241, 278, 278, 279

nigrofasciatume_s..2- s-_--.--=- 536208 241, 272, 273, 280
trimaculatum 222.22 eae 241, 273, 277
Gilintasseseddoso tail ts ne ee eee 155
cimbrius, Rhinonemus.---__-----.------_B200£ 165, 168, 170, 172
cmctum, .Cheetoceros_-______-_.__ ________ ss inva 115, 119, 120
Ginera; \Chrysemys- oe eee a 182
Circolana concharum-~--__..----- anna. SST
Cirripedia
cladophora, Cemmaria--.----2-22-25s2222<2522252/50e 123, 129
clam, razor (Pacific), growth and age at maturity, see
PAZOMClAM 24 soles oe sul aoe Dee 201-236
SOlG. 4 e.2shccscccccesseesceeseceecs =e 213
claparedii, Thaumaleus-_-____-_-_--------------=--- 143, 145, 146
clausii,-Acartia__.....--.------ _. 142, 144-146

Gltone limacina:- = +2 222222222 2nesseel2see-s2sere552228% 137
closterium, Nitzschia_--_____- _-. 118, 114, 120
@lupea harengus: +: 2-2 22sctas=2 5252 aaso sea ste es Aoi eel 164, 172
libertatisys 222. ses 2 Se ee ee oe 283
occurrence, 1893-to 1907-2 == --+:=222.- 0S LeS ee 171
@lpeidsees $2: sys. 2222222222422: A Oe Bee 283
Opisthonema libertatis_.....-.--...---.------------- 283

_---- 125, 130
116, 118-120

Clytia bicophoras.22. 2207 Se bie
coarctatum, Chetoceros

Coarctabtus;sEVasiasot sa =a-2= shake sass AEHOTA 189.
@obitis anableps:::-22=s2s2222=22=2-2s0 ey Oe 261
CODEC Cant airs * Taka fhe ogee Case fe ee e166; 168,
Gcplonterathea= sass = tee ee ee Se LOS ytz3a—130)
CONOSSITON Shorter eno t eS eee eee Sse ee ae 15, 43
@ollianassastimpsoniae = a0 n ss sen enee ee eee 155
Collinectes sapidus 161
Commander Islands, 1897 to 1922, [ur-seal industry_..- 289-332
Bering IS and ewe tee ee eae ee eee eee ee eee 293
CONGIbION O22 Mee: eae ee aha eee een a res 293
Bering islands ease! Sis oe Ree eee eae eee 293
Kishotchnoye rookery ---_------------------- 296
Poludionnoye rookery=----------2----- ase 296

Reef, North rookery 293

South rookery-- 296
Copperuslands esses esse 296
Glinkawookeniess. 22: 2oe es eee eee 296
Karabelnoye rookeries_-_-------------------- 297
explanationess nesses nena ne ee eens Naps 298

SULIT Typ es ee ee a eet 298

Sy Airey heh ete ee Bees A A 301

he a hy ge es a 2 BS

after 1917_---.--- 327
hauling grounds 297
SCOT ee ee eee eee 299
TIVES bie abLON syil O22 yee ore moet eerie ene ee eee 292
pelagic sealing 801
rookeries, 1911 314
FOOKerysneidsa ses es eae 316
sealing, land 304
YSN SU Yeo LA ST I 8 i Se Om i a Se Se 330

Castel ney SHOP ITU Os 5 2 eee eee ena 312

GENERAL INDEX

Page

commersonii, Catostomus__.-. 22: -2_ 12222 ee 182
common) bullhead__.----20 222. .-22 2-245.) eee 9
compressa, Ampelisca-..--..---.-..-- Spay souks 150, 151
@lenodinium 2.2. -.22.- 20d. 22.222 123
compta; 'Grubiasins”. ... 4. .2..242 10) eee 150, 151
concharum, Circolana_________ Pik e200
Dodecacera=:------ 2-2. -.-.5 4-302 2 soa uaeee ee eee One
Conga! _. 0.2222 JL ee 7)
| conger,; Leptocephalus-. -_..__.._.21.1__ LJ aapeee 168, 172
conica, Aglantha_.-___.........-.--- NU aot elannepierge
Conodon pacifici._.-....-.+.. 22... ..-4. ASO ey
contortumy Cheetoceros__-___-. 2-2) eee -. 116, 118-120
@opepoda:.-.- 2... 222225). ee 141-146
occurrence, 1922 and 1923.__-___________ _tdeat ine diyen 145
copepods, food for larval fish___-_________s2ulsiuu dil 170
Copper Island! _..-..._--------.-_2 3. Sassari 296
@linka rookeries_..............--._._ Hota 296
Karabelnoye rookeries___.__________-__---u Lucu inil2 297
Corethron valdiviee___.___.__ -- 105, 113, 114, 116, 117, 120, 121
eomutus, Autolytusi22i22..--- 2-2... aaa 131, 183
cononata, Doto...----222502--1--=2--- 42. Sees 137
Coronatusongipedia ste es ees ae 1/2 0c Seema 145, 146
Pseudodiaptomus!.!-_uef2 103, 141, 142, 145, 146
Gorophium, cylindricum =o Soo sass eee nee oe 150, 151
Corymorpha pendula 125, 130
costatum, Skeletonema_---_--_-___- 108, 105, 108, 113-115, 118, 120
courtadii, Serranus._..______-___-/ <= -2 peed 285
@rago,septemspinosus__--__-__-.--+-_-. Sap 155, 156, 157, 159
erangonoides, Naushonia_-__--___---.2-2.2--.+---.. 155-157, 159
: -n- -eoesy il 186

Beemer am Sf es, 182

digestion tests 192-196
Withiamtylase =e i920 192-195

with invertase__.__........--2._ apes 195, 196
pepticdigestion!=-. 25. 22 eee SH dkeadase 185
tryptic digestion_-__- Se Sees Pe eee 189
cxenatus, Balanus. se. .- _. 5 147
esiophilum; \Cheetoceros. 2-3-2228) hae ees 116, 118-120
Crocodilia.s1222s2.-5---_<..-+-+.-3. a eee
crani, Halithalestris:.-..-20 210.2 12 1) ee 146
crotenensis, Fragilaria___- 114, 115, 120
@rustacea—- 325 so = 2 52 2 es 139-162
Cryptacanthodes maculatus. -_________.__-..-__--2.4. 168, 172
Ctenophora. 103, 128
maximum seasonal distribution, 1893-19G7__________. 124

occurrence, 1922 and 1923
cuatro-ojo---
cuatro-ojos----
ClHICUMIS; "Beroee Yee aa ee eee

Cyanea- Be ee ae ee LL ee 124-127, 149
Capillata ste oo ose ere ie eee ae 127, 129, 130
occurretive, 1893-1907 127

Cyclaspis variens
Cyclometopa, occurrence of common larva, 1922________ 161
@yclophora tenuis. -.+-.----4. <2 22 eee
@yelopidte:. 52-2022 eee
Gyclopterus lumpus----+- 5-2. 2-5-8 a
eylindricum, Corophium
@ylindroleberis marise______---_--_-___-__- 2 ee
zostericola. 2... 2 ese Ss) ee ee
@yprinodontevs225 2-2 ee
Gyprinodontes.-..-~--.=2 <2 5-45
Anablepidz

GENERAL INDEX

Page

Cy prinodontes— Continued.
Cyprinodontide 243, 253
PTOLUMG UUs ee oe eet ee eT ea 253
punctatus____ 241, 253
PGS CLIT Gas eerie re een UN ose co eateee aeons 243, 254
WTO MIENeSIAe nee oe ce eu le ane cee See er ee 255
SDUGHODS=sessee so) aoe! eo eee ees 240, 241, 255
IpTIapICht Dy sees cetes 2h 2 ee She ARN Reale pen 258
fOSLERI Reet see ret Ss aa aa 241, 242, 257, 260
letonai 241, 242, 256-258
Cyprinodontide. 248, 253
PEO TUMOUIUIS hates eel CM I LD els a 253

Cyprinus carpio 182
RUmnISM MLO OUUOS 32.8 = oe a oe So ee 149-151
(Crphereisiemarginatas 2.22250) oe ey ee 141
PACH VLOIICULA Ss Soe e ae eee ee ata oe eli 124, 125

quinguecirras: see ee ee 127, 129, 130
Dactylopusia vulgaris-2-..-.--_...--_-- eee lt ts 141, 145, 146
@amicium,.©hetoceros.. 2 115, 118, 120
danicus, Leptocylindrus__.______________. 105, 113, 114, 118, 120
davenporti, Microstomum_-_______.._..___-__.----_---_- 135
davidomi, Tintinnopsis--.-----_. 2-22. el 123

debile, Cheetoceros . 115, 118, 120
decipiens; Chetoceros_--+-.-2...-. 222 116-120
BIAGSIOS Ra Sees eset eee ee ae a ire 114, 115, 120
delicatula, Rhizosolenia...............-.---------- 112, 115, 120
delicatulum, Bacteriastrum_____....___._._.______-_-- 113, 114
democratica-mucronata, Salpa___-_______________-.__- 127, 164
d@ensum,) Cheetoceros..-- 22 116, 118-120
depressa, Alteutha Pea ba 141, 145, 14
MeDLOSSHIM wROriOd nit: | 121-128
depressus, Eurypanopeus_________.--_ 22/2222 Li llle 159, 160
diadema; @heetoceros_.:-.._. 22-2 115, 118-120
diademata, Tiaropsis-.---....-.._ eee 123-125, 129, 130
WiASivlistpolitac=- 4-222 eske ek lela SI 98 152
MUAGriSpInosa==Ss2-2keesusslccclkeewbeu the Shee 152
SIOULE OS A a at ee te 152
@intomiss:- =. -.=..- 2... 103-175
distribution __ . 118,114, 175
TIL PAS es NE ge te De 113

TASER | espe pete Ra ln el ees ee ALE 114

eastern Atlantic waters..._..........-._-_-__--- 175

YE) S) SRD ay Se a en a cp aca Na A a 115
oceanic_-.---__ 116
tychopelagic____ 115
Dictyocha fibula__.______.____- 118, 114, 122, 123
@idiyamtm,Chretoceros!.-- 2222-2. SE 115, 117-120

Digestive enzymes in poikilother:nal vertebrates. In-
vestigation of enzymes in fishes, with comparative
studies on those of amphibians, reptiles, and mam-

TaN ISHE Meee EUR We tela bh te veut Sy li bs NLL 181-200
GugipalismAplanthas oe Soi ee es oleh 124, 130
Gare timNe OMe iets es ene ce ena eG pea 267
dinoflagellates. _____ 104
dioica, Oikopleura 163
iipunena strangulatas.s-) 2229. 2 se 123, 125, 129
discaudata, Tortanus-_- _--- 108, 141, 142, 144-146
MispanuPedophylaxsetesw ey. ey eee ee 131, 133
Iistephanus speculume= 222-00. ess ee 113, 114, 122, 123
distribution, plankton, Woods Hole region__-_______.__ 91-179
Ditylium brightwelli___..-.--._-____ 103, 105, 118-115, 118-120
Dodecacera concharum 131, 133
(Oe) E i ht SMI ct eae ea DE ieee Se pa ar ee ay ticle 182

ereptic digestion 190, 191
DDH CICIFestioneessewne Re LOO e! Rate Aa ele 185, 187

tEY DUC hivestiOn= = ee aanennee sa ens cea enuenes 188, 189

Doto.coronatas 22-22-2222 lithe ee eat
ovoid aoph se aie wy Wee Gael ee Lee ea ESE
Douglas Lake, Mich., description
dovii; Anablepsii2 222222222222 eres dis
Amisotremus:= 2-222 22222 2202512222. + Jeehrferen = Soe
Pristipomas==2-~ 25 22 22. A eet paint!
dowiis:Anableps_--....2-.4_- cise fesscOeess ehh ee
dubia, Libinia
Dysmorphosa fulgurans.....-.-222-220.-2-224ecl224-_
eDUIMCUS HB Alan UISmeee ees neni ke write aaa se 147
IB Chinod Crmia banc tet eben ectv iy Tear en lls as oes 138, 139
Ectopleura ochracea _ 123, 125, 129, 130
Edoteartrilobatesscn ss cee seer oenc sar nee sane re 151
OC eTL SAVE y GL] Us eeeeem eatin asl are I Ne nee tn eee 137, 214
CO WALUSIL NIONOCIOGESS. cocoa oe -- e eeee 149-151
BOLUS TAGS mere eee Me er ces mt ae Le ee ar 125
i lpSalvadon sishessssemee ts sec ea ee een apes eek Fak 237-287
FAT DIOP 1G be ee seer = er ale en Sie ne ye 243, 261
Anablepsidovilies seos. sao eee 240, 241, 262
Meh Ee ashi elec hla lay eae SD et) Di et UO eee 243, 248
Ariusutavloniessnes se nee 241, 242, 250
Galeichthys guatemalensis 240, 241, 249, 251
PAP Henini ee ees esas cee a> Aree nara ANe eres 243, 263, 284
AM ab spabolsyfqb Oh oes ees eta 241, 242, 264
Thyrinops pachylepis 284
ISREAT CO Sob eee eer en eine ee 285
Garanx ui QOSete see oe oe ee eee eee 285
Olizgophitestmundus=se2=s ssn. sane Ses ey NEY Ons 285
SAUITUS Sedans = See need Sams a= ON oe 285
entropomid sew eeee meso. tose nee emir nh teens 243, 268, 285
Centropomus nigrescens_-._...2272 0. el) 240, 241, 269
DECUDAULIS Seer | aeen eae _. 240, 241, 269, 271, 285
robalito 240, 241, 269, 270, 285
Characini dees: sam nue Nate ere RY ee ee ee rey 243, 244
Astyanax fasciatus seneus_____.-_._----_-L_-.- 241, 244
RHOIGeSisaAlvad OLiSta==sseen. ese eee 241, 242, 246
Cichlid somesessk= se ewee sa ase = a knees Ait wel ee 272
Cichlasoma macracanthus _.--__-___-- 241, 272, 274, 276
TTT RL ee eee Renn one SA IM PET A 241, 242, 272, 275
motaguenses fhe ssssi sls 239, 241, 273, 278, 279
nigrofasciatum_- 241, 272, 273, 280
trimacnlatumesa-ss sae ne eels ee: 241, 273, 277
TTD eld ce a ee eas ote eee we er nee renee) ernie eae 283
Opisthonemailibertatism soso. 2 seston sta a cen oe 283
Cyprinodontidcoseeeseestew anaes wees a oe 248, 253
Profundulusipunctaluse=sses as see sae 241, 253
BESS ay hg Ke =i a oe ae Lec nln ei 243, 281
Gobiomorusimactllatus2_ ee ee 241, 281
Engraulide i 283
Stolephorus;brevirostris ss: o.. 22s esses ee 284
OXISTUS Mand sient lie uae aU tect earn ioe Pree 283
DATIATN CN SIS see eete a ee eee ieee eee eo lea 284
TASLL CIS see ete he See es ae en reas a 284
HE pHippideet ces cseaecccen 287
Cheetodipterus zonatus 287
Epinephelide 285
ypinephelus;analoguss = ss-se ao as pea ee os 285
Gorrid se bene teem. Clue fee omen’ sel Lo 287
Georresiperuvianuse see se eee eee oe 287
Glob iid ss fee ene i Lee NUS a ae ce ae ae 287
Gobionellus sagittula_ 287
Mniitignid weement es ete 8 eee eee ee ee eae ae 286
Lutianus argentiventris___.________--- needa sa 287
MOVeMiasciatUuses--< soos see nee ee eee woe 286

338

Page
E] Salvador, fishes— Continued.

Mupilidee: too sys ie sue ae . 248, 265, 284
Agonostomus monticola____.________-_-_-__-.-- 266, 267
IMugilicuremads. 2252255 4/235524J5 07 ee

cephalus.2223 42. ss 3Jens 2202 ee

Pimelodidsesie yo eas ve SU Ss We
Rhamdia guatemalensis

eeciliidss sss oes sate Wa a OR 243, 254
Mollienesia sphenops--____.-_.-_-_-_____-- 240, 241, 255
Priapichthys fosteri 241, 242, 257, 260

letonai 241, 242, 256, 258

Polymemid cavern e ieee esa uO oI aaa Vive ule ve Cs 284
Polynemus approximans 284

Bomadasid asi aay oo ot ey Nae an len 286
Anisotremus dovii---_._._____._--.------ 287

PAC CiSse eS aie Be a an 287
Orthopristis;chalceus-222 22 ee ee 286
Pomadasis panamensisses: see wen sepa ea eee 286

IASMOPUS IS ViS secs as a tare te Se Nee ee tas 150, 151
elegans, ISagittas: fe Bic.0 oe 8 te TERE en So pare 133, 134
Pleo trides- 23. £2. c dias eR ene. veers 243, 281

Goblomoris 25 As see et Oe ere pee, Le 281

maculatus 241, 281
Eleotris lembus________- 281
elongata. Ostrea=2 0 ee ON eters eee ee Na 213
elongatus, Pseudocalanus_____________________ 103, 143-146, 170
Elops, leptocephalus
Hilysiellaicatula ts. 2225222 soo Se ee
emarginata, Cythereis

ibinig® Of ae oe Pee eee eee ee 160, 161

merita talpoid a. = 52 = ee ets 156, 158, 159
emertoni, Autolytus __ 131, 133
empusa, Chloridella 153, 154
Mngrawlideas.c.25. 22 oot Le ee ps 283

Stolephorus brevirostris
Oxiguuses.-2 ee -- 283
panamensis_ ~~ 284
rastralis-22 ere ee ae ee as eee 284

Hineravlisspanamensisa:--o-o.-senee ee eee eee 284
enzymes, digestive, poikilothermal vertebrates, etc... 181-200

carbohydrate-splitting 192-195
digestive! tests2-22: = 2-2-2 | ea ee ~ 192
INVObIN Ge. Slee Ee aie k Eee es he nIne 195
Epenthesisifolleatas<-2.- =. .- = <2 s22ue ee es 124, 125, 130
Riphippidseseie 2. sired oles is ee eek a a 287
Cheetodipteruszonatus= 22s - =e eee 287

IE PIP PuUS!ZONALUS ase ae ee ee as eh ain NE 287
Epinephalus analogus 285
Hpinephelid'sa.: 222 cos ge tok ok ened sled ale od ee APRN 285
Epinephalusianalogus2: 222222 ea eee 285
Erichsonella filiformis______- See alisyh
nicht huss=-sesseee es - 153-155
Ericthonius rubricornis_ oa) S151
Enythropsese esis. coe a we eee ee een 152
FUSOXSIUCIUS sess SUE Le oe Uy Saree cea 182
Euctenogobius sagittula----....---.---.---..----------- 287
(SJE RECY.c) oop WSS Se en akan DAS RN SUED ROE Su beReyCiS fs 34
WUPALUTUSSD eee ee ee Oo enna eae eee 156
HUphausiakrohmiie!* ows Pe Oe See se eeu 153
AILS 0 =) 72 Nena ey pa ea Lo ape ga LN Bee 153
Eurypanopeus depressus---------_---_---1------------ 159, 160
Eurytemora herdmani 143-146
Ihirundoidessss at eer eee oe teen ae 143-146
HMuthemistosbispinosa:---2-- 25.222 sess 149-151
rubricornis 150

DOR LAD OGY: Ve 00bb Yay 1c bap ene Maes et MR a Sep ele ec aos) 125, 130

GENERAL INDEX

Page

Eivadne nordmanni ee see ee ee 138, 139, 140

Spinifera. ioc osee ete we es ae ee es

tergestina
evermanni) 'Thyring ee a eae sr P43):
exigua, Anchovias tater: 2 RE ee a aaa a 0125
OXISUUS, SLOLEPROTUS nse se eee ee 283
Hacelina bostomiensis! 22222-22220) aa eee 137
feerceensis, Rhizosolenia______._.__...._....-_-___- 112, 115, 120
fasciatus seneus, Astyanax____________.___________--__ 241, 244
favs; SBid@ulphigc ees ee ee eee ee 113-115, 120
feliceps,'Galeichthys 22-0) 22 ei Saar 248
ferruginea, Limanda 168, 172
bulla, Dictyocha e252 2) ae ee eae 118, 114, 122, 1238

filiformis, Erichsonella__ 151
Alin eee Ae aM 252
filipendula, Sargassum__-_-.--.__ 121
finmarchichus, 'Calanus!--9__+_2-_ 2.) eee 143-146, 175
Podom. 2-020) 2 6 yok ea 139
Fish, Charles J.:; Seasonal distribution of plankton of
Woods Hole region 91-179

peptic digestion seaep edge:
bernegate._- 2-222) 2-8 ere 287
black bass, smallmouth_______________ peer ys scree 1 3-6
black crappie 182
bluegill: 2.03 2e 2 a 182, 186
bullhead 2.222222)... ee 2, 3, 6, 13

characins

chimbera } 267
chimbola 240, 253, 255, 258, 260
ehincoyol2 4) is pe a ee 273

chinook salmon, see chinook salmon.

digestion tests___
ePticidigestionee apes suas ae eee 185

iY DlICidigestione = seeker eg 189
CUATLO-OJ Oi ae Ey le 262
CUALLO-O] OS! dap Sa Nee ws A a oe 261

GENERAL INDEX

fishes—Continued. Page
determination of ave. ..2..-.:.2.---2-2225-222 22 Le 18
eS alvadOrwtrs os of lel walk ST 237-287
descriptive catalogue...._.........-...-.-------- 243
Taa2NG) oy te ee a MeO UE Ae ae aad 240
enzymes..________
eulachon_________-
10 Ub ole
flounder, winter
HOUIBROV Oban HUE A one ys ad Bee 262
PtP OLS me Lea le os om Pk iy a2 ad te si ET 277, 279
(pg sata >) SU tg ey pe Aa a POT Coats en ee
hormpout.-—..=_-__-
immature, canning.___-_.__.__.-
destruction... +. .2:-2-2-__.-
results of taking
ASURUGPUTE Sf semi aace nara Raed a i ol OI
Meilllifishes2. -c 2-2-2
larval, relation to food supply_.._.---__---.---------
Mebrajanchawe etn ul eel la OEY 284
Hip rere Chia seme mene yet TO Nk a 266
TIE, 34 hp ANSTO 0 PEALE ED 266, 267, 284
ROSH CHCHPMia eee i be et ee at 4
inaye' otf UCU 3) Tp a ODT 264

marine, collected at Triunfo and Cutuco, El Sal-

mud puppy
mullet

red snapper--___- Vaan OR BRO Oe a I 182

BONGHIG Sethe
rovalete

eggs, Measurement

silver
Sarndinges— 2 soe.
sea catfish 248
silver salmon 15
silversides 263
smallmouth black bass 3-6
Snapper, red 182
SUG ere 2 eee 166
starfish 138
sucker 2, 3, 5, 182

digestionktostsus me wale Ly ON aur ues 193-195
EOD STAULTIT OWE: ses ae ee eee RS es SA 239, 240, 254
SWIC D ASS eee eee ete oe nT Been teen aL 182

peptic digestion 185
WINTER LOUNGE ee eee eee ee Ue deers 133

339

Page

Fishes, Republic of El] Salvador, Central America__._ 237-287
flaccida, Guinardigecs sss sce eso ee on ae eae 113-115, 120
flavellata, Liemophora_...._.--_---------- --- 113,114, 120
flavescensPerca es tae sa selmi h We GRRE GCS ahr EEA 182
MavidtlawAureliacs.2 nse. 2e. sels deers Ble 108, 125, 126, 129, 130
flounderwintersssss 25222 cas week aes BA NA oy UE 133
folleatarsE pent esis sass es sae eee UIRCLEAS 124, 125, 130
formosa, Heteromysis é 152
AVI 2 RUE Ae Re RIN a es PN See lly 2s Ne 124, 125, 130
fosteri, Priapichthys_.--2 2 022220220 2s. _... 241, 242, 257, 260
LOUT-C YC rare eee ren ders Pee Le SON oe Sls A Lae 262
Hragilaniaicrotenesista. ster ee. ee ean re 114, 115, 120
frauenfeldii, Thalassiothrix_____ ROP LPS Ete eS cates. 20S 114, 116, 120
fulgurans, Dysmorophosa 123
Podocoryne- 252. ----- --- 123-125, 129, 130
Hana iSeeebeaeee ak eH nee aS HO Te Mie AONE heh 164
ZuAtemaAlensisnmeswe 23 ees tes ue eae eran Me ete ae 253
DUNCLATUS meats eae rereme yak Ls ORE ere ie 253

Fur-seal industry, Commander Islands, 1897 to 1922, see
Commander Islands, fur-seal industry.

LUTFS Cals Pee ae ee es ye eh eel Sele ee ee Pele ee
hauling grounds, Commander Islands__-____________
history, Commander Islands, 1897 to 1922
number killed, Commander Islands, 1917 to 1922____ 330

pelagic sealing, Commander Islands_. _-____________ 301
rookery raids, Commander Islands _.-__-_-___._____ 316
sealing industry, land, Commander Islands_________ 304
fuincata diy spree ee eames week 2 UE Rasa Ig Le 144-146
furthilsyATus eae nen ee lees Sele a) 251
Miscus ys yn enalhuss esse lees ole see eee e se 165, 166, 170, 172

fisus;}@erativme 2202) 225 ee ey ee oe A a}

Gadusicallarius# 2222 2222222-_22-_-0 == = 164-166; 167, 1170; 172

morhua; erepticdigestion==__- 2-22 4) le 196
Galeichthys 248

felicepse= eels s tee uae 248

guatemalensis)sasssas sr nee ete _-. 240, 241, 249, 251
Pallionisisynedras escent ee wee eee eee Rnee 113-115, 120
Gammarusjanntlatusss-s2 see seaens ee anne ons 149-151

HOCUS Cates eee ees errs Reeve ais Sapte METER Ling Sy 150, 151
garter snake, digestion 184
PaSbropodilanycomeumeme tins SU ee eemmen 3. wee Gee a eee 136
Gemmaria cladophora_- 123, 129
Gerres peruvianus--_____ 287
Gerridsos2 eee se s8 eas 287

Gerres peruvianus 287
gibbosusisepomissscaentetes eee eae bee a eee 255
SLi ex HO DNOLUSS Maton mney We aNyoem nua alin) ye tnapteas 255
GTauCcoth oe Sees Sees is coin ee leon eaNY pF RON 158
Glenodiniumicompressasaees ee eee ee ane ene 123

glometrica, Caprella
Gobiidss sare Sane
Gobionellus sagittula

Gobloided sania sews vrese se is 281
lO Our G a3 aes nme eC et Teer ale LR eat NNN 281
Gobi orm or tis eee ie tesa AEE PA as 281
TMACULAGUS sme ee wae eaten AR eee ite 241, 281
Goblomortis hese seem ame mats ee) SPAN IE NNO ho 281
LALCr eis Pewee cine eM fan Pree Mle Je oe CE deo 281
maculatus =sese.

Gobionellus sagittula-_
Gobius longicaudus

Gonyaulaxitricantha sss ee ee ee 123
gracile, Gymnodinium___.__________ I es Maa os epee eats] 23
pracilisMSctellammeemer cr snes! il Teen ies aie 143, 145, 146
@Grammatophora;marina= 2229-22-62. ee 113-115, 120

Ser Dentin deer vee meeawe ses eee. See ve eee Ss 11520

340

Page
grandoculatus, Centropomus.-_-_____._._-._.-----_------ 271
granulata, Bid dulphigws-s 2222) ss esas eee 114, 115, 120
HMeterocrypta 2222252 Lee eee eeaeee 159, 160
grapsoid larvee, occurrence, 1922_____.__._----_-.-.--.-._- 160
grata; Lizzie ston eo ea .- 123-125, 129, 130
groenlandica, Zygodactylaz_..-_-..)= 2-2 124, 130
exonlandica, (Phyllodocesa sense eee 131, 183
Growth and age at maturity, Pacific razor clam, Siliqua
patula: (Dixon). =~ see eee ek ae Oe a eae ae 201-236

Growth and degree of maturity, chinook salmoninocean_ 15-90
Grubia compta 150, 151

ZUaADOLCZs sen eee eee Ae ee 277, 279
guatemalensis, Anacyrtus__22 2222-2 23en 2 el eps 246
HATAUS Perens. Were ae Sens eS ee OE 249
gaa as ce Fo Sree UAE 2s Nee Sime eee 253
Galeichthys2 vs renees: hla weee oe seneurie ne 240, 241, 249, 251
Pime]OGUS I: St eres eae Re leh gre ee 252
AVS Mia si SOLE Le Ae Se ee Oe Cee Ge ee 241
ROCDOLGCS yo ee a ee ee oe oa 247, 248
Tachisurus=* 2-22 20 eo feo eens 2 oes pee eee 249
Ab yrina 2s 222 Ao Oe se ee ee ee ee 265
fuilat Thyring ress i ahi" Vote gr eee merge aay Goes es 241, 242, 264
Guaimardias) accida sess eee eee 118-115, 120
ao sboV stoi eyo Me ee ee 165, 166, 170, 172
PUWINAs sooo ose oe ee eee ee eee ee ee oe 281
Gymnodinium gracile 123
Gymmnopleds 220 52h Mie ice gees ae eg ete 141
Halitholestrisicromie ie = 5. cee nese oe ma uamennn tenia eta 146
harengus, |Cluped: -- 222525 22 CR es eens ee et 164, 172
IVEY RUS HS SOe tte et eee aa ee 266
Harmothdenmbricatave= 2050. seen eee eee 131, 133

141, 144, 146
___. 145, 146
__.. 145, 146

Harpacticus chelifer-
UniremMis ee = eee eee eee

hanvardiensis WAlderia= =. c- = o- e eee e eee 137
hebetata (semispina), Rhizosolenia___._____.____.____- 116, 120
helgolandica, Tomopterus...-...._.____-_---_..:_- 130, 131, 133
hematus, Centropages--.--._._.-_.-_-_--_- 103, 143-146, 162, 170
EHenricia‘sanguinolentate. = so sse 2 seen eee sees 138
herdmani,Hurytemorae soe eee ee 143-146
Hermanellatspeaaeess sates eae! Sd 46
Tier OS stk oat 18 Ber 2 SSS IC Ae LoS GOTT A Norn ele 272
macracanthus! = txt es ene ee eee. ae ee 274
motaguense= 2.222" Ss {see ~ ee ee ee eee 279
MIGTOLASCLATUS Sa cee ee ee ere eee Ne eee 273
severus 272
trimaculatus 277
Hesionide 131
Heterobranchus sextentaculatus_--__-__.----.----------- 251
Heterocryptaigrantataa si: eo ae 159, 160
Heterofusus retroversustesscss see ee eas eee area 137
Heterognathinee re 22 ues s0 ii Wee aoe ceo ate 244
Characini dense cece 0 ee eae ens ceed 243, 244
A'S ty an axe 22s ee eae On ae ae Sere eee 244
fasciatus aenus

TOS D OL ES 2 ne yer eee ears Si ne ee
Salvadorisias 00122022 see yee ee 241, 242, 246
Ee teromysisyfOnmos ase ao aoe ae ne 152
Heteronereis_-_------------ ee neers See aye ie 131
EXeterophrys/sOlie sk vite! 2 eee be SN ss ee 121-123

Hildebrand, Samuel F.: Fishes of Republic of El Salva-
dor, CentralvAtmenl cases mes ae ae eure ee ne nee mene 237-287
Hippoglossoides platessoides 165, 170, 172
gecurrence, 1893 ito 19075 222 bose soe oes ee ee 173

Eppolyteizostericolass2= seas ae ae ae a ee ee 155-159

GENERAL INDEX

hippos, Caranx__
Scomberssee ene sees
hirundoides, Eurytemora
Holmes, H. B., F. W. Weymouth, and H. C. MeMillin:
Growth and age at maturity, Pacific razor clam, Siliqua

patula (Dixon) 222250 2 25 2 ec a 201-236
holothurians.... 22: =. 222. 22220) 138
iomarus americanus: sos-- 9-8) eee 156, 157, 159

occurrence, 1893 to 1907__ 157
Bioplarchus: 2!) 203 sie. a2) 272

mectacanthus!/\— 2... 62: Pe 272
HOMPOUts ee sea seas wana tens 9, 13
hyalina, “Thalassiosira.- ...__- J...) ee 114, 115, 120
Mryalodiscus'stelliger-_.--.....-._+_ ae 114, 115, 120
yas coarctatus—"--.-- 2-222. 2i22 2) eee 159
Hybocodon prolifer_---_______ 123-125, 129, 130
Hiydromeduss- 272.2... --_-_2_2_) 2 103

occurrence, 1893 to 1907_ 124

1922 and 1923-2. 2! 2225 jt ee 123
Pyperiidse. 2222502 ese os) 149
Ichthyobdella rapax_-_:_____ 131, 133
Idothea==- = 2-26 5 ee 22 ee a ee ee 151

Dbaltica.- = 3.82 6-2 eee ee ee 151

meétailican — 272 hen, hee: ee gee 151

phosphorea.. -. 2-2 5-22-4224 es 151
Idya furcata_- 202... 2. Joos ee. ee 144-146
Hyopsyllus'sarsi= £2222. 0 228 2.002 0 So ae 141, 145, 146
imbricata, Marmothoes 2.2... 2222.22 3c oe 131, 133
Impressa, Loxoconcha=-=--- 22: --- 242) plea ae 141
incisor, Lepomis----____- Epps: 7)
Inenmis, ontopemian. =. sss=s5 eae u - 150, 151

Thysandessat 2225 sii 23. = es eee eee

irgidans, Pecten= =. --2- 5-2. -.22222-2 1.020 eee
irrorata, “Unclolassse 5. eae

irroratus, Cancer
Tsopeda-- = 22 22 ie oN ee
istatagua....22. 2021-264. sb iS ea

japonica, Asterionella-—.----____________ oats 113-115, 120
Jassaymarmorata-— 32252 2202 8 22 2 ee 150, 151

Kenyon, Walter A.: Digestive enzymesin poikilothermal

vertebrates. Investigation of enzymes in fishes, with

comparative studies on those of amphibians, reptiles,

and mammals: 2025. Uso 2 o.oo ee eee 181-200
ASHES soe see ne ee IE eae wien AER)
Kirtlandia pachylepis -- 284
Knerl,, Pristipomes ) 8. yo a ea 286
krohnii; aphasia. a2 oP eae 153
Labidocera sestiva_~_---_._--_--_____ _.-- 141, 145, 146, 151, 162
Labrus punctatus: — 2225.02 ee Pe ees 272
laciniata, Staurostoma---_______ - 125, 130
Jaciniosum), Cheetoceros2522--- === eee 115, 117-120
Mactophry suri s MUS soe eee = ae ee

WUACUN A VINCtAS = soos oe ee ees ET AU
lsevis, SulasmoOpUs i 2 eee ee e e ea
leviusculus, Calliopius
languida, Oceania-_--__
Laophonte sp----__-
lateralis, Gobiomerus

Philypnus = oes oe LC Ss ee

GENERAL INDEX

. Page
latipinna, Mollienesia- _----...---------------------4:+5 255
Jeidiyijeinemiopsis-_..._-.--....--------------- _ 129, 130
ieembpus macwlatus._.-......-.-..-----.--4------ss845-- 281
lexinispheotris=— 222802 tee ee 281
lentus; Anoplodactylus_ -_-_.-.------___-_4s-4s-besiet2- 161
OHSS SDR Aas ee ee eo oe as Oo 147, 148
Lepidonatus squamatus----.-.--.-------------------- 131-133
menomis'ei bbOSUS: =. ------=------.-- seks sods psamee tied
MNICISOM are aes cee eS
Leptocephalus conger--. -_----
WOpSGe sane e nce sese oe
eptochelia savignyi---=---------- 22-2 e eo tog eee
NS HOC AM NOLS see see | oe ee aera oe ee 152
Leptocylindrus danicus--~.----__- Lek SLERES 105, 113, 114, 118, 120
Leptosynapta inherens----_--------------- Ee els 139
letonai, Priapichthys
leuckarti, Podon__-.---_-------
libertatis, Clupea------------
TOTES a ae
Wpisthonemassstere le -- eee 8 soe sok k lee ERE 283
ADIT ONC DI Ae ane scene oe rg 160, 161
GMI ALPIN ALAS meee ate Se ee 160, 161
Pe eae ee a a 159
Licmophora flavellata --- 118, 114, 120
lynebyei-.-.---.------- --- 113, 114, 120
ESTES Dine) NC ORO Se en ape eee ee eee ee BE 284
UN AGN) OVE) I ee a eo a 266
limagina, @lione ..+2-82-.5--.---.-- 2-222-252-2262 -4e8 137
Limanda ferruginea.__-_---.-------2-4--------2s2--4e 168, 172
Mmmbata, ANCLeIS.2=- == -----=----<--- tees 130-133
Limulus polyphemus-......--------=------.-. =524-4+4-- 161
linearis, Caprella______.___- 150, 151
Miniope'seutigera.._-..-....----=-------- NS oh yn ee 125, 130
DUG ONC WALLONIA ts: 6 ene ate 134, 136, 137, 170
IT POMMMAVNLOLea wees eee eee 134, 136, 137, 170
Malliata ween oueeene he ee ee ee 136
UERGIS MERE 5 ty aS ae ee Sk os le 136
SY ss a) | gg ee 266, 267, 284
Lizzia grata _-- 123-125, 129, 130
JocistaGammartis: 2.2) 02 ee eee 150, 151
NGpeperehiseneee resus sok Coe ee ee 4
LONE OCA a Ree ace ema seer a 138, 166
larval forms, occurrence, 1922 and 1923 138
Nonpiceanplis bagurus--22.0202 220 88 158
longicauda, Appendicularia_____-__---_------------ 163
OIRODICUT AME ee eee ce ena nk wea celts Se ed, 163
longicaudata, Thysan6essa 152, 153
Foneicaiidtsn GODIUSssaeseee cect s cole ose oe son ul 287
omercormis, Memorges.. 1-22 228222. lie 103, 143-146, 170
Voneimana Agmphithoe:22 2-2 9222) eee 150, 151
Longipedia coronatus_. 145, 146
NGneMMeS @CratillM: secs see eS ec oeean ee ace 123
MOUPINEIIUS) ACATta:-4--c2ss ees eee Nee ee 142-146
lomgisebosis, Altolytus.- 220222 2 2ole eof 131, 133
iqupisctma i Nitzschiaws- 22 le. 2 esos Lee 114, 120
Thalassiothrix____- 113, 114, 116, 120
AOMMIMSHDISCALOTLUS( 2-2. so 2 2522S 164, 168, 172
Lophopsetta maculata ___ 165, 166, 170, 172
Torenzianum, @heetoceros --_-_-------_--------_---- 115, 118-120
epxOconena IMIPYeCSSAs5----2-- 6 \ a once ten eee
TUTE CCREIDU Ea oly aco RS ee peep oie Lee
NICOUSMIVUCLTUGI as semen. ee ae ee as
lucida, Siliqua__-----_-
lucifera, Odontosyllis__
lucius, Esox -__.-------
PMO TrIMeris | LENWise =a. s eee eau eee Oe 131, 133
ltumpus) Cycloptepuse soe eee eee 168, 172

lunata, Astyris_..___-
lupus, Anarhichas-

DEAD A ESS obi(6 [2 2 gees a re eae ae eee

iLutionusrargentiventrise/] 22 seus ee eo resents 286
MOVOMIasclatusaewe: soe ea je a eS eh 286

Eutianus argentiventris__-- 2. - 2-22 hee beee eae? Louw esG
Moy emfasciatus sas watee ns Fe NS ee - 286

Mag amlisia ya: wok te wees ee ee eye ae hy 182

lyngbyei, Licmophora

Mysiosguillatarm atawecwee se eee meee See

PIC OSs eee eect eae eee eee at a eb ea
macrocephala,-Ampelisca 22-222 2u vse el ue ee eee 150, 151
Midcrocep lalussszsasew ssa eees Uae ee rem Salve Soe ee 268
Mescroceros; Geral eee eee ee eee 121-123
macrocheless eh OlyOnyxXeres a2 see esa eee ees 159, 160
VES OCU Ta eerie ce terse pence ier mea, ee See he cles 155-159

larval, occurrence, 1922 and 1923__.___-_-....-.--.. 155, 156
maculata, LO pNOPSeLta.-2=.- o2- sce oe ee 165, 166, 170, 172
maculatus; Cryptacanthodes: 2-2-2 = se se 168, 172

Goblomorus aries Meee) Sees ce sees a lace ae 241, 281

J AA(SS 0a) OY UTS ga a SS ace pe 281

iPInlypNUSss eee NI Got 281

Rinnotheresn snes te a even Sie Sh Sa te a 159, 160

Spheroides____....---- -- 165, 166, 170, 172
maculosus;wNeCuUnIS: i 52--ees seb ese bow ee i 182, 184
meonas, Carcinidess oss. Ssuke = a ess Sesese nan see 159, 160
Wiagelona rosea sess Silene fae a2 2k On al ons So 131, 133
mialloustmoellerocheaar= steam e ea ewe eee ee 118, 115, 120
MIANUNITAIS FenZ VAMOS Me sect en ali Ne eee ase 181-200
Man jud ase Pee eee cs ke eee. sere ee eek a 26d
Marie ve VUNGrOleDeNis yest se ee Oe ee ee 141
marina, Grammatophora_______-.-_-__-----_-- 113-115, 120

IIN(ST AS Sapa See teers AEN re Se AUS TAY Sets Sh oy ee as 239

iOS CT EN mr Werle SiR FUn Hain Naas WS aa See kat ett 125
MIBTIMOL Atay) ASSAM versa Se Se nse cb de oe 150, 151
maturity, chinook salmon in ocean, see chinook salmon.
maturity, Pacific razor clam, see razor clam.

McMillin, H. C., F. W. Weymouth, and H. B. Holmes:
Growth and age at maturity, Pacific razor clam,

SHIGE DALLA GIDIXON) pee enees eee eee Gees nen 201-236
mead, rontellasessses- 2-2-2 - Se ee ee 102, 148-146
mediusmC@entropomuses tress ewe eeno sie Nene ees 271
meekin Wichlasomasuwueweene sce de cee eee 241, 242, 272, 275
megalopsrLlatynerelsee wa oe eee el ee ee ee 131, 133
Meganyctiphanes norvegica._____------_---------------- 153
Melaniris_---------- 264

balsanus2 22.2! -- 264
Meletta libertatis 283
Melicerthumycampanulace. = 22222225 oes ee 123, 129
Weelosina Orrerite.ce pasos se ooh ether see oe CS 114, 115, 120
ANDO aes es Se eee a nek SP oll pera aided 164

Meni Giamovatars sues oes wae se eee ae 165, 166, 170, 172

OCCUTTENCE, WISOShtOn 90 fanaa he ee ee ee 173

pachylepis.s=scesns sees 2 5- a See fee Woe Tee 284
menidia notata, Menidia:. 2) _2--- fn 165, 166, 170, 172
INONLSAUS RIA eyDOCCILUS sme eee SU eee Nec 255
Mercenaries VCRUS Sera ere eee mc ek 137, 170, 214
Mierluccins bilimearisacssiesssess seats nie. ou 165, 166, 172
Mesoprian argentiventris____.___- 53 A ES 286
anetallicawMldogheamrnnuee wale uel LMM En a guygu Wt 151
IVECTASTOSISS wae cere en cee oe SL eS Ue AE 10

WMietridialiucenssentos a ease wre oe ees 4 3— 146
IN ChOgaAdUS bOMCOdNee ees cee e ese ee eae 165-167, 170, 172
Microsetella norvegica__.-_.___..-___-- ~-.--s-, 144-146

MOS EB yee yet eee arate aren Le a ea 142-146

342

Page

Microstomum davenporti- 2): 2222-22 122. se Seas 135
minor} Lepfocuma_222- 222 asec es ae top?
MinnowSs = 22522 29.24 Lue. ee ew ees east}
TOUGS e222 eee cles eee Seen ea dy 4

OPS 2s) SSeS Be ee ee ee 2 e SS 239, 254
minuta, Parawestwoodia___________________-________- 145, 146
minutus, Planes --- 159, 160
mira; |Hutima-2oo. ete ee ene ee 125, 130
mirabilis \Syncorynes=2- eee eee ee 123-125, 129, 130
mitra? Cheetoceros=e a= eoias ae ee eee sees 115, 118, 120
Winemeopsis ses eases = Sete ee eet pea 98, 124, 125
Nobis Liat ter: Sie Oe ene Od 129, 130
occurrence; 1893'to 1908" 202 ease ee, Lo
mojarra 272-275, 277, 287
Mojarra negra-- -_- 275

Dlateada #22 == 22S) aa. 0 sae ek a
Mollienesiat ©2222 28 = ee a
latipinna- - ----
sphenops- -_-----
sphenops tropica-
WOW USCa Sot eee ee eee eee ee a ae

eae ea a ee ea as 149-151
--- 241-267
Ure see eee Le ee Oe hee eate cat mame ar 267
Oe ae ee eee 168, 172
De eee eee ee a an ee on ee eee 279
motaguense, Cichlasoma-_----_-------_-_- 239, 241, 273, 278, 279
SERGr OS See eecce ean nat See eee eee mer a 279
mucronatus, CarinogammaruS___--_--.._--._________- 150, 151
BIG" MINNOW see. Se ee pen apr 4
mud puppy
NMiupilsa 2 e iae
cephalus
curema-----
monticola-___-
Mugilide
PA'PONIOSLOMUS! Sceewer oe oom nen ene oe ne eee 267
MONUCOlG Desc cans tn ooa ee eee ee ee 266, 267

mundus, Oligoplites 285
IMUSSe] M22 ee ee 214
SULT GL OS AE CLI eRe ee ne nee ee tc ee Teele URLS ETS 159-161
iy ayarenarias sree ot encore ane ne eee eee eee 187, 213
Myodocopa ale ph pl Re MN Lente cet st dl naan AB V1)
Myoxocephalus eeneus-__------------- RRO Ve .. 165, 166, 170, 172
occurrence; AS9s to 1O0h2 t= = wasn een e sane ee 169

152

160

OULigse a es a eats = ea eek s ie GE 137, 214
Mbyxus harengus-:-=-------<220 02-0 Lo diel ke lates REA 266
IN ayasimarin ys e2 teehee Rel R es ee cop eer tee lee gee 239
nasutum, Agonostomus 267
Naushonia crangonoides__-_-_-_-_-----_-----_--____-_- 155-159
MeDuUlosus Amer TUS a eae oe ae ee ee ee ee 9
INectonemaragilet. .a- eye ee ST Ei SUSE 136
Necturus, digestion tests_._...__--.-_--__--.--_------- 193-196
ereptic digestion 191
MACUIOSUS See e es ee ee edema 182, 184
pepticidigéstion eee eee kes 8 ae ea 185, 187
tryptic digestion__ 188, 189

Nemathelminthes._-.-.---.-------- Ssh eh a ae ER 135

GENERAL INDEX

tayloris-22 25222525) S272 aes
Galeichthys.2:-2--2.5222 22. 2S Se eee
guatemalensis:---+2-=-- 2s. 2eeune 240, 241, 249, 251
Nemopsis bachei 124, 125, 130
Neomenis argentiventris____._.._..__....-__-_-__l_--- 286
novemfasciatuso. 2._ + 22.5.2 2) 25 22 ee
Neomugil: 4 seus hes sees seas Rea
digneti
Neomysis americana
Neopanope texana sayi

INephthys' butera'!\S.<.. 222525220 ee ee ee
Nereidee
Nereidiformia
Nereis limbata:is+ss2:-2222225222:5/ eee 130-133
pelagicasc:24:2222222222222227202) ss eee 131, 133
Vibens os cec A mess ccs ee cnt CSS ee en 130, 133
Meriticidiatomssse ss eee 115
nigrescens, Centropomus_-______._____-__-_-_-__-- 240, 241, 269
Migrocinctus, Driphoriseos ase css e es nae eee 137
nigrofasciatum, Cichlasoma-_--______- --- 241, 272, 273, 280
nigrofasciatus, Herosese: 1-22" 252 ses Sacer eeeees 273
Nitzschia‘closterium= "22252522 5222522 Ua ee ee 118, 114, 120
longissima: #2222582 222222 224 23s 2 114, 120
paradoxa:s: 22222222212 525067) eee 113-115, 120
Setlataysts: ccs arerte hurt nies 104, 105, 113, 114, 116, 119, 120
nitzschioides, Thalassiothrix____. __--.--_--_-- 113-115, 118, 120
nordenskioldii, Thalassiosira__.-.._.--.-..-_-.---- 114, 115, 120
Hordmanni, Hivadne- i= --2-- 22 2- een SUN tS 138180
norvegica, Meganyctiphanes_____- Eira teritedaties michal. i 51}
Miicrosetella- 222 b2- 2222 -------.--- 144-146
novemfasciatus, Tutianus--= 222 - + 200s See ee ees 286
Neomisenis: 2228222222508 0s = See eee eee 286
nudibranchs
nutricula,Purrtopsis==== 22" = 5 seo es aie eee 123, 124, 129, 13
nuttallisSitliquar(Solon) 22222222522 322 see 202
Obelia'spy cinco 5 S02 222 2222 oe eee eee 128, 124, 129, 130
obscura, Podarke 131-133
opscurus; Amphiascus. 922 =s=22 25s sss Sense 141, 145, 146
occidentalis; Roeboides-- — =" 22.—_ 2.8. eee 247, 248
Oceania languida
Oceanic: diatoms: +2325 0055525 25 Ser re
éceanicum, Peridinium)) 2232225 232 ee 121
oceanicum var. oblongum, Peridinium-__--_--__-_--__- 121-123
ocellatus; ‘Ovalipes®. 2) 22228202222 22 oe ee 159, 160
ochracea, Ectopleura 123, 125, 129, 130
Odontodactylus 2%. oo SSPE See ee 154, 155
Odontosyllis lucifera 131, 133
SP SERA MM ACI ae Ne aes
Oikopleura
albans
dioica
Jongicsiida: -- 75 2 2522 a2 3 eae ee ee 163
vanhoffeni 163
OitTHonAreviCOLMISe= 22 se oe le eg nee 141, 146
GUTTA Se SE ee ma 141, 145, 146
Oligoplitesimundus= 27-2): 2 ee 285
we Oe aes EO Pe oT DENT Eee oe 285
164-166, 170, 172
SI EL Se week hs ha Lp CR a 131, 133
CQ)y0)s' 0c bt: pene ere aN RNa Le RROMENAL HAN INN vena Dey ig os 185
283
164

GENERAL INDEX

Page
onpiculare,. Nanystylam = -.see2.2 0 OO 161
ornata, Amphitrite_____ 131, 133
ornatus, Autolytus__.___ 131, 133
@xrthopristis;chalceus2+-=2.2 22222222 2b ee es 286
Osburn, Raymond C.: Black tumor of catfish_.______.. 9-13
@Smerusrmordax:cos: = s=-=.2 sere 2225 Se 168, 172
SINAC Oda ements seas = re Aen Lol Peles miele tena ea MBSE 140, 141
Ostrea elongata__________..--___-._--------- a an 213
lostreum,- Pinnotheres: - ---.. 22.22.22 .22 2022222 159, 160
Ovalipes ocellatus 159, 160
ORVIR DTA Kehoe 2 eae ae Ve sss vel eee wien ole So OE NS 268
Oxyrhyncha, common larve, occurrence, 1922__-_______- 161
ROnnPNOSt ylIS:Smithios¢c6 2s ak ee we see MRR ES OS 152
ON SHCA Mem a Ite AUN Ue ie ee eb ea RM alle IMM AMA Mala R iis fEE Aa 213
pachylepis, Atherinichthys_-______..-_______.._--.---_- 284
BERG eA curd GU Spe scream nape ts Ee ae 2 2 epee 284
IRR EA NE Ck eh eens, Ah dean a reed, 284
SIR WAN Gees ere te ee US Sips ayes 284
MM by Tin OpSo ace a eee Hepes PR Fahy YY Ahn NE Late ey rie 284
Pacific razor clam, Siligua patula (Dixon), growth and
age at maturity, see razor clam_____._.__._--.---iLuJ 201-236
pacifici,, Anisotremus.................-.--.-2--22--- 2228 287
{OVSTRUANG LG) a) spel SRN ee SR en SED 1 287
pachheusMhaleichthys._._..2/..--.. WeDo eee 34
edopnviaxGispar-e: 222000 sooo seb keen 131, 133
JERS eg i CONES I A Pe De ey ee Sev ees be 158
155
158
158
159
PAUCEdsbUTblel see AL | ae ele tie ey 182
epuicidigestionesai_ Leer oll Chee a! eevee 185
ifspuleidigestion-. 22) 02 2 eeu yy os beset) 188, 189
Palesmonetes\vulgarise! oso Ui ee st ee eles 155-159
Pallene brevirostris_______-__- 161
palliata, Littorina 136
panamensis, Anchovia 284
SEM PT ANUIS SA eee ne a soe os eae 284
Serlapichthyswe abe eet Jeet Be 8 esol see 261
SLOG DNOGUS Ste sae sien 2 SS Le a Stet see pete 284
FATIGUE ee oe ramet 2 BUN oe eee 279
paracalanusipanyls==s9 225 o20 22262 lavish eel se 145, 146
(patadoxa,,Nitzschia_...-_..--.-.------------- _. 113-115, 120
paraliasilcatass 02-2222 40222692. ce ee ele ~ 113-115, 120
iParaphoxus Spinosus_..._.-..--...,-22---.------_-2423 150, 151
Parategastes spheericus-----.__...-.-------------. 141, 145, 146
Rarayestwoodia minwtas 0.22 ee ei 145, 146
286
285
286
ManwUs pearacalanuse.. oa wee Ae ew se ke 145, 146
watersoni,, Anomalocera_.-._.--_-__.-___--_._4_-._ 148, 145, 146
pattersoni, Anomalocera._-_.._>.---_----.-------_--=-- 102
patula, Siliqua, growth and age at maturity___________ 201-236
DEAMIMIEOUEO seem oe ee So eee oe ee 138, 166
mectacanthus, Hoplarchus—_—----______--_---2 422 -_ 222. 272
RectenniTrnidanssaatenae Heimer stents Le MsAme ee Seee Tee! 214
pectinatus, Centropomus- --__..._._------ 240, 241, 269, 271, 285
pelagic sealing, Commander Islands___--_-__--_.-------- 301
treaty forbidding in North Pacific Ocean__._________ 313
pelazicawNereisumss tet oles eek ee 131, 183
Beleco pod ana arene eemecceme seas oe ye eee 137
Peliagmtltichs es ere eee race een Lau Ne ae Kee Keb 159-161
pendula, Corymorpha slo eels tee 125, 130

pennata, Pontella_.-...-.----------------------------- 143, 146

Percomorp lise ss. nee le ee oe 263
Atherinide_ 243, 263, 284

MAB Ge UO pal OD BONE a ee AI SNE CREE Le 264

BUT aac ee Ne eer yey 241, 242, 264
Gentropomid coseense sae eee een ee, 248, 268, 285
Centropom US Samus eos eae ce eee id 268
MIGTESCENS Syste. ed hag Th 240, 241, 269

_.-. 240, 241, 269, 271, 285

240, 241, 269, 270, 285

Miupilid'sss 2 clea ee, ee a ue 243, 265, 284
Agonostomussik Misco ei er eal hy cee 266, 267
MONtICOlA eee a BL a sd 241, 267

INGA AE eee hl She a Nev nd a 266
CODDANISE foresee eee pen vel tay Pyles oe 240, 241, 266
Peridiniumdepressum..- 2-9-2 20-2 eee 121-123
Oceanicummtee ee eee eo eye 191
121-123

peruvianum, Cheetoceros_______________________-. 116, 118-120
pertivianus, Gerrestes a A ag, 287
Rhilypnusilateralissesseee see oe oe ey ee aay 281
maculatusve ss eg. 2S bet lesser toArUW Ry (Waban 281

eb OLISBie ee aye eee ee en eral! 164
gunriellus tse ne e lel ea ite 165, 166, 170, 172
eceurrence, 1893;to 1907- -. - 8. eeu ork 167
miHOsphores; pid otheawe sss. sims seu eee eet ely Veter 151
Phyllodoce catenula__.____._____ 131, 133
gronlandicalectscowres Vue ON Sees i toa sua 131, 133
Phyllopod ayes ks Paras fe ore leer _.. 139, 140
oeccurrence;,;Wioods Hole, 1922). 6) a 138
In y Salli a teen a enna SS sl ee ie ytany bea 101
phytoplankton sy 227 te as 95-175
pickereli..- 2.22). 13, 182, 184
digestion tests____ --- 192-196
Guisyouple blastp a _-- 190,191
eptic;digestionswa9- 2 sees) fee eae _-- 184-187
brvplic digestion 0. oe eat 188, 189
pileus@Pleurobrachiass-- 9 se lse6so ieee 128-130
IPAM CIENOUS See eee eee yk eS age 251
ValSODT EME s. ce veeee teem te oes els be | aaa 251
Pimelodide---_-__- . _ 248, 251
Ram dia ple see ee Sat eT RR adie dete t 251
guatemalensise see. ei aera ee 241, 252
PimelOodus/arius sa epeven 2 ewe wihled eA, Paden nf. yee es 250
guatomalensiss sees 51a WU ws eet oe ha as elute 252
Quelentie. ee Nee aie ae & 251
iRinnixiaicheetopteranae 225-2 eo se eee ee 159, 160
SHV ATO Sewn sane wee en gets ome ele eee oes ee 159; 11160
PINNOUBCLeSHMACUlAl IS eee eee eee Seg 159, 160
OSURCUM SSeS. ee moni es 1 ei Tk 159, 160
esnlo0, 151

ee os eae ere SGN Pane tae Reng EN) trey 150, 151

_ 164, 168, 172

Se Ep se eee ee as 5 ee eer ee 182

pe a ee eee) ear nee ty 159, 160

Sees es Se eee ae te ee eM Ue eas Shee 91-175

Bp EA Nn Sil ser st Se WS 101

a ett ae) eee © LN ce Pee ae a 130

ATODTOSUT ACA Teme n aise LUN sect cA Neh. Gi cone 149
Brachyura 159

OY AY 0) TOUTE a NRT, ee lL 162
Cirripediaie upee epee NT pay Wiser mel 147
C@oplenterata 2. Selah wy create eran 123

344

Page
plankton—Continued.
general discussion—Continued.
Copepoda re as Se ee en 141
Crustaceans ec one PUES Ti ar 139
CumMacéa == Sore Rare at Uae ame mate 152
Diatoms and other plants. eu LOL
Mchinodermatawe. less a eee eae ae 138
164
155
136
140
139
121
Pycnogonida 161
PCHIZOPOG a eee ee ey oe 152
Stomatopoda 152
Vermes 130
Miphosuraj eh eae 161
IMMUeHCE; Salinity = as ee A ie es 175
temperature. -- se alfa
Neritic. we a a eee 101
Oeeanici ss F286 SN oe ee Oe 101
planorbis} Skenia.2 20. a ES 137
DIGLCRd Ayo yoo ae os ae no ce tosee eae ee 240, 244
plateadon 23205 oo ee a a ee 246
platessoides, Hippoglossoides.-.-___.__..________- 165, 170, 172
Platyhelminthes -W ops eos eos ee soe ee ee 135
Platynereis:meégalops_-_..__. 22st _ 131,133
Platypoecilus mentalis...____. 32 4 052 See - 255
TTOPICUS =) Wo Sse ce ST 255
Pleurobrachia= | 2 a 8 124,125
URC UIS ore ee a TO),
occurrence, Woods Hole, 1893-1907_____-_______- 128
Podarke obscura_._.__..___. SSeS) BO ORE SEEST 133
IRodocoryne|carmeiess ans Steer ncc ss LISEIZOSTZ5
PUT UTANS | oc = eh hae 123-125, 129, 130
Podon finmarchichus 139
intermedius _ 138-140
leuckarti 139
polyphemoides- 2+ = 222225 222 140
Peecilia boucardi Way th 255
salvatoris______ _. 240, 255
sphenops- - ---- - 255
PONS See 22 Ee SL ee EY Mine oe a 255
Posciliid se: 227 ae sie ee ee ee
Mollienesia_______._-
Priapichthys
poikilothermal vertebrates, digestive enzymes-_-__--___ 181-200
animals used in investigation.__________.___________ 182
carbohydrate-splitting enzymes_____________________ 192
conclusions 198
190
184
188
184
197
INVETLINZIENZVIMCS! ane bare see pee enneres Mejen Seve eae 195
methods, description ==: ssa a ie 183
results ssigieakse4
polite, -Diasty lis Ue ee eke se eae ONT es SR NE ee 152
PoVaAchrussviLens aes ee Ne Ne ee ee ee ee ee 172
occurrence, 1893 to 1907, Woods Hole 169
Polydactylus approximans 284
Polynemidess eee 284
Polynemus approximans 284
Polynemusiapproximans lasses ee eee 284

Polyonyx macrochelestese oes soe eee cee eee ae 159, 160

GENERAL INDEX

Page
polyphemoides, Podon2. 2-212.) 1 - u eeeee 140
molyphemus,, Limulus! 22 -) 22) 23) ea 161
(Pomadasidees 25 ool N 8 oe 13)
Anisotremus dovii 287
pacifici 287
Orthopristis chaleeus: —— - .. ee 286
Pomadasis)pansmensise - 2-2 2225s ee eee 286
Pomolobus pseudoharengus.____________ _ 168,172
Romoxis sparoides. a ee 182
(Rontella mead. 0) le ee . 102, 143-146
penmata. a. 2s a 8 ee ae Tea ee 148, 146
Pontellid te 22. seer e. 26 ac etsece Hoe Boe ee 146
Pontogenia inermise 2-2 eo ee ee aa 150, 151
Poronotus triacanthus--_._..2.-.--2_ _- 165, 166, 168, 170, 172
Priapichthyse-) 0 2.020) ee ee 258
fosteriet}s.! = - 8 oul. 2 a ee 241,242, 257, 260
letomaleeee Eire Ss Pe is Tee 241, 242, 256-258
PAN AMOENSISH Se Foo ee Le 261
Prionotus carolinus 165, 166, 170, 172, 173
occurrence, Woods Hole, 1898 to 1907______._.__.___. 173
Pristipoma\chalceume 2) __ ses2e ae. a eS
chalceus

kneris 24. ieee. oe ee
producta, ‘SSyncoryme._.__._.______- 1 aaa ehpaie Pees
Protimaulys eee ee ee
DUNCbAp US oe SLs) Ue ee
prolifer, Hybocodon----_-___
Proteus. anguineuss 20 oe
Protozoa. 22 ey eee | Le
occurrence, Woods Hole, 1922 and 1923____________ 121, 122
Pseudocalanus elongatus__._-._.-.._.___------- 103, 148-146, 170
occurrence, Woods Hole, 1922 and 1923.____-..--_-_- 144
Pseudodiaptomus coronatus--.--_-___--_- 103, 141, 14:2, 145, 146
occurrence, Woods Hole, 1922 and 1923______--_____ 142
pseudoharengus, Pomolobus--____-.--___-=_LL2-22-2-_ 168, 172
Pseudopleuronectes americanus-________.-__-_- 133, 166, 168, 172
Pseudosquillajciliatas.:-__..___-1-.-_. iv ee 155
Pieronotus 2.2 2252 eke es ee 251
pteropods_..--2-. =. 2.1.2 -s2e-4n50- cece eS 137
Ptilocheirus pinquis 150, 151
piers sk. 2. Uo ee eee ee Se 170
pugilator, Wea. 22222. 2 2- + 222242222 Se a 159, 160
OVE (2201; 5. Cl OCC: We ae 159, 160
punctatus! Fundulus..-._...22--222---2..L. 2s 253
Wabrus.\ js. kin 2 se sbshi 2-25 See 272
Profundulus- +.) uv... 2) 2.2.22 Se ee 241, 253
Pycnogonidal._-i2..2-2.--5..--2..-. hie ee 161, 162
Quadracus; ‘Apeltess 2 Vit sisi soon = ee 167,172
quadrispinosa, Diastylis 152
QUAHOR TE? S MVE nh es Aru eee 214
quelen, Pimelodus_______________- 251
‘Querimanna tices ble esa eee ee Ne Se Dea ee eee oe 266

rapax, Ichthyobdella__--____.____-
rastralis, Anchovia_2-..-- 22. 5.
SUOlepNOnUS ao. sau eee een
Tazorelaim,, PAacihies 20.) 2-. oes
adit; erow bases nee eee
age, determination
ANALOTIUY oe ye ER tae De Cea hie siete ne Ae
digestive system 22022 22a lie eee eee 207
mervous' system oo Ne eee ome 207

GENERAL INDEX

Page

razor clam, Pacific—Continued.
commercial catch: --..=-.----=---2--==2-S222252. 2202 231
description.__-....-..--. 202
larval development 213
setting, time._____-- 214
IpaomiOwON sei e ole Meee eee elec cree weer eewenen 207
PETE te Bed ORI 224
224
229
227
occurrences sh eo bee a ee 202
PENIS IONPTOW bese oe lw Le cee ode SU) 203
SOWING O Woolies b ee ee SU 208
relation of water temperature________----------- 209

Co kt i aes gente 137
113-115
251
guatemalensis 241, 252
TUTE UGS ae A Sage a a nC an AS YEE LF 252
Rhegmatodes tennuis______________________----+------ 125, 130
Rhinonemus cimbrius_--_-.___-.__.____------- 165, 168, 170, 172
IRMIZGSOlMIaAS Yk eee apcnielon 112
elatig site is CMe 103, 104, 110, 119
alata genuina-____ 112, 116, 117, 120
BNET CouevesH oaths = AA a ee LE ee 112, 116, 120
TSE GDIE CATS eS Se a a 112, 115, 120
LENT ey tre ey aoe aes oe SOT SUNN 112, 115, 120
PEPREBO NSIS ee a i ana 112, 115, 120
hebetata (Semispina)__.-________________-L Lacing 116, 120
BEMIS Dinas se ee ee a ee 103, 109, 112, 116, 117, 121
BEI BOTA ue eee ees 112, 115, 116, 118-120
SHIT SOLG eater clans oe Ae ee 112, 115, 116, 118, 120
RIReTORENIS es oh eS EGU 112, 120
THombus, Biddulphian <2 es 113, 115, 120
Rich, Willis H.: Growth and degree of maturity of
chinook salmon in ocean 15-90
robalito, Centropomus.-_____.______ _. 240, 241, 269, 270, 285
MODAL Bs nae neee eS 268-271, 285
Moceusiehrysops.......-..---..-- 182
MOC KAD ASSH pee ee 2 So Oe eet 5
IVOROIMES ea ee eu bdo testi 246
pouehellei- 222.2 latyoenen Bindinies 248
Piratiom alerisiss2s. va eth e sel eee teed 247, 248
Breinpntalise Ne a ee int eli 247, 248
SHEVACORIS etanu dna! De It yeah RS 241, 242, 246
OT CATs snes an unON ne elie Saree We nay he Me 286
rookery raids, Commander Islands 316
MOSCA ViaPelON As eases oeen Hak hoe tou eae 131, 133
MViicrosetellaweaeenn nse eee o abide Snake Sie 142, 145, 146
OSLLAL AP AIR LLL ayaa ee eee ete ents SUES eS ess 167, 172
TPR ATA E Shas a VE Le ne Loe ea 264
ee oe ea ULE ee _. 269-271, 285
MipHLCaua, Amp OIGHOeseaas ole es Seek Leek 150, 151
Mepricormis, Wricthenivs-< 22 kets sl nese AEE ere 151
MSCITNSTIStOMee sean nae Oe) xs NEE ea chimes ane
rudis,-Littorinal:¢2 2-222 ys 2

Russian fur-seal islands

LSEEYE EIT BEY CLUSTER | JGNN 2S SA OS oO

occurrence, 1893 to 1907, Woods Hole-___ 135
1922 and 1928, Woods Hole 184
Sermbdentatas .:s525 5245226 voice oases ges BS 133, 184

Page

Sagittula, Euctenogobius__.___._._.____-.__-_1-_ 2-2 tLe 287
Gobionelluss 2222 Leese es sok oe ee ee lata ae SUES NOR 7,
SebUera Ord ALE es Ve A ot SEA SARIS Be ae 15-90
Chim ookesee el ra a Ms fall Oe ek ee a el 15-90
age, determination! 222) 2 {he a eS thas 18
Columbia! River, taker! aro y ohne Ie! 28

age groups, abundance___--_-_-____-__-_.____ 43

Growth. Sheil. Soe belt lel 48

immature fish, age and percentage_.______ . 89,40

TS GUL ye ee ag SARE f 29
correlation between size of eggs and size of fish__ 27
Drakes Bay, takkems Js. OHsii) an) Se NO 67
Horty brace vtakenas- ee sorta alike kT Ate 67
growth and degree of maturity in ocean.________ 15-90
maturity, determination-~ 2222-222 S22 20
Monterey, Bay, taken. 22.2 -22t 4 bedi 63

age groups, percentage-___

SIZ Ie
COL BERLE) SETH IEY HIER LEAS ED ALO i DEE Ps BALLS
eggs, measurement
ISI Vers a oes ee ees eine oN x
Salpa democratica-mucronata______.__-__________---__ 127, 164
salvadoris, Roeboides_____--___- octet pete 241, 242, 246
salvatoris, Pcecilia_._..._.-____- hoe en I 240, 255
salivini, A‘gonostomus__-=-- -__- ---__- = OE 267
sanguina lenta, Henricia______.._.___ Lise 138
sapidus, Callinectes__......2-.=.-..2-22e eee 159-161
Saprolegnianos 9 ee eo oye ee a OEE 13
ret CANES 1 BES iil
240, 244, 246
BAN SASS URMILA 101, 102
pacciferumis sie eT en ee OTH ED ATS 121
filipenGula: =~. So COT 121
Barsiliyopsyllusie2. 2 _ 141, 145, 146
Sarsiellavamericana- 2. 22-222 2< 141
SUT ASE Loses ase en ae sete s son ee 185
saurus, Oligoplites. -.. 285
Scombersss2.2 --Sseseec sso LES, 285
savignyi, Leptochelia ; 151
Sayanaahinnixige 220. 2. a eS BO 159, 160
Savi; Pituophisss=-22.55ses22sis2 Se 182
schistonya, Caligus: 222.222 sos SU ee) 144, 146
Schizopodans ie sesscs USO Gash OST ETO FT 2 152, 153, 155
schuttil»@hsetoceros=2222-sis2 so sss EE ee 115-120
Gl senavunaeCimMaligs=as 4 == seen asl Nee ren MiRtrRae tee 268
Scomber hippos-_--_----------- Be SOU,
SAULUS es ewe Nis Se ae ne ae a een ee ons 2 OV ES 285
Sil pta, Diastylisesa2 =o) sons) arse Pon eh nt at faleigh Saco 152
Scutigora;HLirlopesnsasns 2 eter meek mel a rine 125, 130
SCY DHOMCOUS seaman sem arene raion ne neene ae ALSZeE 103
maximum seasonal occurrence, Woods Hole, 1893 to
LOD SSL sane eee nes Soe toes oes t ees 124
occurrence, Woods Hole, 1922 and 1923______________ 125
SCAN CH ELIS Hers om same nein Meee set nro mee ETP S ARES YR 248
sealing industry, Commander Islands________________- 289-332
seals, fur, Commander Islands__-_.._.._.___-_-__-_-__- 289-332
Rilled;; Commander Islands ois 2222 2222s 2 5ll ee re 330
Seasonal distribution, plankton, Woods Hole region. _- 91-179
SCCUNGASBAleaL erase eres les Sie Teno Be oe oe ae 149-151
semispina, Rhizosolenia-_-__- --- 103, 109, 112, 116, 117, 121
Seplemspinosus, Cragozs:2s22.0 > fes 2 ee 155-159
seriata, Nitzschia__._...___- _ 104, 105, 113, 114, 116, 119, 120
Derlolayzonataeessnes sa smeewsserd hie eR Glen shee el eh 168, 172
serpentina, Chelydra- ==. ---+222--2------22. -. 113-115, 120, 182
DELranus:COUrLAG Sas enas ne tae bes Das os None Seaton ners 285

SBITALA ADV DLISH a= sesh umenine ae Revie a hin Sina rise een 150, 15 1

346

Page
Serrodentata, Sagitta_2- 9-2-2 2 sees _-- 133, 134
Setellaigracilis=--- 9s i eens posne eect 143-146
setigera, Rhizosolenia__-_--._---_____- 112, 115, 116, 118, 119, 120
Setpsa,iSpiose tt. as 28 Bert eee ee Ne ee ee eee 131, 133
SEVCRUS WEL CLOS Ses aoe eee 272
sextentaculatus, Heterobranchus 251
shrubsolei, Rhizosolenia-------.-.-------- 112, 115, 116, 118, 120
Silicoflagellata, distribution, 1922 and 1923, Woods Hole. 113,
114
Siliqua-lucidas 2. = Soe a a ort e c 204
patula a. ce betes ops ee a eee pe hel aly 201-236
growth and age at maturity____________=_____- 201-236
(Solen) nuttallii_.-.. 222 22-22. ae 202
Silver 'Salmons jecss soe ence Shee eee ert 2 15
silversides= 2.2. -2. 2-22. -- 1. ete ee epg 263
similis; \Oithonas-- 222. -- 2). = eee ae eet ereeege 141, 145, 146
Skeletonema costatum-___-_---__--- 103, 105, 108, 113-115, 118, 120
Skenia'planorbisss5 222522 2: 2 3a ee ae ee 137
smallmouth) blackibass=-25-- = 2.22222) ae a eee 3-6
Smith, Frank: Variation in maximum depth at which
fish can live during summer in moderately deep lake
‘withithermocline:)- -- 252-2 22 See enone ee sane 1-7
smithi, Oxyurostylis
snake; bull: .-.-. 2222.2 22-- 25s 2s eee ey eee ae
ereptic digestion. _-----.----°-..--.--2. pie Ae. ee 191
peptic digestion_-_-.--.------
snake, garter, digestion
smapper, Teds 22.5252 s2.--- =~ os
snapping sturtle..-2:=:2-.-2- 55225 22 eee ets 182
digestion tests. 5-22.22 Soke eee eee 192-196
ereptic digestion 190, 191
Depticidizestions: sas a= eee ee 184, 185, 187
tryptic'digestion---.-4- 2522 2h eee 8 = are reg t 188, 189
sociale;sChetocer0sses- 22 - = eee eee .- 115, 117-120
soft Clamisc. 2252226252225) 5- 22-225 sch 222 oes
sol, Heterophrys
Sparoides; (ROMOXS: ase a se nee ee ee 182
speculum, Distephanus:--_---_----=-2--_- = 233 113, 114, 122, 123
spenops tropica, Mollienesia-__---_--_--___-_ __ 2 sae 255
spheericus, Parategastes-..-----=---- 2. = shee 141, 145, 146
sphenops, Mollienesia=== +0 222-2 -- 2-2 eee 2A0, 241, 255
Poocilian ss 2a eae ae ee eee Oe 255
Spheroides maculatus.__.--------------------- 165, 166, 170, 172
occurrence, Woods Hole, 1893 to 1907__..--.--------- 173
Spinifera,Iovadne. 05 e= i ie ee a ae peo abl 139
spinipes, Ampelisca._.. -- 222-2 - 22222 22 tee eee 150, 151
‘SDIM0suUs, Paraphox uses esses. see anes Se eens ene 150, 151
Sploisetosase ns a. bes Ra ee ces A a ee ee re 131, 133
squamatus, Lepidonatus__--_=-- 42-2 =_ 2 eee 131-133
SOUT Ce ee ae en eae a 166
Squillidse: occ 2s) Fats eS ok en ee ee 153
starfish 138
Staurostomajlaciniata cosas sass asa ese ae eee a ee 125, 130
Stejneger, Leonhard: Fur-seal industry of Commander
Islands 189 7;torlO 220s seu ee eee 289-332
stellatus,,Chthamalus:. 22 2-4) ee pee eee 147, 148
stelliger, Hyalodiscus___-.----------+--+-+-------- 114, 115, 120
Stenothoticy pris sos 2-2 - ieee ee ees 149-151
Stenotomus chrysops---_---------------------- 165, 166, 170, 172
occurrence, Woods Hole, 1893 to 1907___------------- 171
Stephanopyxis appendiculatus____---_--.---------- 118, 115, 120
stimpsoni;, Callianassac 22-2) 222-2 see en 155-159
Aoritomoiusus in. sates ess AL ee 137
Stolephorusibrewirostrisesss 2a sae a ee ee eee 284
exiguus ~ , 283
panamensis: s22h 2.2 ean Oe) ON ane UO gay Lee Ue 284
rastralis: 2.3. 2.22 oe 2 See eee 284

GENERAL INDEX

Page.
Stomatopoda. 2. - 2-2-2 4 22 bape ea ane ee 152-155
Stomotoca apicata 123, 125, 129, 130
strangulata, Dipurenas_ 22205022 22 Ss 123, 125, 129
Striatella unipunctata_-2 222 - <2") eae 113-115, 120
styliformis, Rhizosolenia______._.------_--25- 5. 22sLike 112, 120
suckers 200s VS Ns Seine oe Moe er 2, 3, 5, 182
digestion tests... 22.5202. 122 0. 0) eae 193-195
sulcata, Paraligue. 22 oon. oe eae 113-115, 120
superciliaris, Bougainvillia....__._. _________. 123-125, 129, 180
Sylligters so ee ee eee 131
Syncheta triopthalma } 135
Synchelidium'sp__.2---0) 2202.02 ee 150, 151
Symcoryne mirabilis. _/2_-2.°-- see ee 123-125, 129, 130
producta__.__....---.oSuieieceeal a Ba 123, 129
Symedra‘gallionii. 222228 he eee see 113-115, 120
aNd wlat a ee ee aN 113-115, 120
Syngnathusifusciss=-) 0-22 -2 ee 165, 166, 170, 172
occurrence, Woods Hole, 1893 to 1907__---_.----.---- 171
Tachidius brevicornis___..-._-__--_-2ussctee APE 144-146
Tachisurus guatemalensis___-___.__-_- iL ss2l ell ills 249
talpoida, Emerital 200202 8202 0 cere ee saris 156-159
Tanais cavolinii 151
Tanystylum orbiculare_. 0-28.25" s = ial eee 161
tad, ‘Opsanus 2 2722S FA Ee een ae 164
Tantoga onitis. ~~... a ee 164-166, 170, 172
occurrence, Woods Hole, 1893 to 1907__.-----.------- 169
Tautogolabrus adspersus____---..-.-------- 165, 166, 170, 172
occurrence, Woods Hole, 1893 to 1907_____----------- 167
faylonifAriustt (222225 noes he eee 241, 242, 250
Telepsavus 132
larveele ae oe ee ae 181, 183
telfaini, “Agonostomus=-=2_0 22-2222 eee 267
Temora longicornis___..__.-..----_---- - 108, 143-146, 170
tenera, Euphausia._._..-_ 2+. == +_s odie Seite 153
tenuis, Cyclophora.____.--.-.-___-_- ---- 114, 120
Dumibrineris 2822.5. eee 131, 133
Poeciliagiut ses 2 ae ee ee 255
Rhegmatodes_
tepemechin....222/...2:_-- 22222222422 eee
teres, @heetoceross._ 2 - Ses ysus ses ae eT
tergestina,. E-vadne___.---.-_ 22 eee eee 138-140
Tetragonopterus seneus--_--_-__-----_-- 222 ses usec 244
texana sayi, Neopanope 159-161
Thalassiosira decipiens____..-..-.----.---------22. 114, 115, 120
Tyan. feo ae ee ee ela 114, 115, 120
nordenskioldii__._-...--.--.--- ---- 114, 115, 120
Thalassiothrix frauenfeldii
Jongissima so 6222 eee lee cee
mitzsehioides!: sls. eka, Sees eee

Thysanoéssa inermis

longicaudata 23-2 c2 2: ess Se ee
occurrence, Woods Hole, 1898 and 1899__-.------ 153
Mbysanopoda..._...-. 5 ane ee ee 153
PBC Ualig: 8 aes oe 2 Oe 153
SD ee oer NE nS I 2 a 153
Tiaropsis diademata __.......-..-.------------ 123-125, 129, 130

Dimarformosa (cso os ae 124, 125, 130

GENERAL INDEX

Page
ERIC URE ODSIS ee sae ee eee en eee a eae 121-123
Gavi G Offle a ae mene NU Ru NEE oe) ncn eae e a 123
tomcod, Microgadus.....--------------------- 165-167, 170, 172
Momopterus helgolandica: _---_--..----.---.---------- 130-133
tonsa, Acartia. 222.2)... _ 141, 142, 145, 146, 151, 162, 170
PE DEINE OW ie eee ae ee cee ee econ aoe 239, 240, 254
Mortantis’ discaudata:_.-..--.-.-------=----------- 103, 141-146
occurrence, Woods Hole, 1922 and 1923__---.-------- 142
triacanthus, Poronotus___---.-------------
tricantha, Gonyaulax_____-----.----------
trigonus, Lactophrys--------
triloba, Edotea.....-----...
trimaculatum, Cichlasoma__
brim AcUlatuss Hl CrOSeee == 2.5222 22=-2 eae ee anno 277
triopthnalma; Synchseta. -_ 205505 23a eee eae
Triphoris nigrocinctus- -_-
tripos, Ceratium-___-_---
Tritonocfusus stimpsoni-
PROpIcus,Platypoacllussssseesss-22-----~-------=--=----
Binvp hose pind Wiss. semen soe a 252-22 2 elie
tuberosa, Acineta__-__-
Turritopsis nutricula--
turtle, digestion-------
AIM TCC Meee oe serene ones con esnenccns
MED CiGIgestlon: <2 2 aoe een aera een enone ea 185
iy UCiIGeStlONe == soe see s= se een eae 188, 189
SHIA oli) Gaeta Ne oe So oa Se a 182
BLODELCIGIZES LON eee oeeon can at sone 190, 191
peptic digestion_.-____------- 184, 185, 187
tiyplicidigestione..sseess ones sae Male eee 188, 189
tvchopelagicidiatomse ess) a. ses lo poce eee tee nee 115
typicus, Centropages__.-.--.------------------ 141-146, 151, 170
itynannus;)Brevoortia--2---2-2-2 ~~ -- - ee 165-167, 170, 172
Weaspusilatons: eset aera oo oe ee cee sees 159, 160
DUE siakepe = eee an sane eee ote ot Sa osdese 159, 160
SD Bes ees Lok Se ec seen lobe eeecese 159
iWiltminpaomee tee Oho So. eases eee eseel sled 246
Unciola irrorata_____-- 150, 151
unaattim, Buccinume--.2--.2-----2 2-2. =. 4 22-2 o enna 186
undecimalis, Centropomus 269, 271, 285
SiatraG Ba eo es 5 ee eee a 268
landulatar Synedrass=..=ss2---- os 2 se 2252 shes 113-115, 120
undulatus, Actinoptychus-_-.........-..----------- 113-115, 120
unipunctata, Striatella.—-.-----._-2.-..-----2----- 113-115, 120
uniremis, Harpacticus----.._--.--- 145, 146
Upogebia affinis---- 156-159
WOOLY CISIS Pe ane eee eee teeth seb belenel sas 165, 170, 172
occurrence, Woods Hole, 1893 to 1907_--.------------ 171

valdivisee Corethron_--- 105, 113-117, 120, 121
warthnofitents Oikopleursaceoas--ceee ones seseac ceed anscas 163

varians, Autolytus_-
IBACteriastrum sess e werceae oh etuneae ac oe av eee
Variation in maximum depth at which fish can live
during summer in moderately deep lake with ther-
mocline
peneraliplan: ofioperations..sesse- 2-0 -e-. se eee ee 2;
tests arestiltsmeeee ee neee ee ae
variens, Cyclaspis--
Venus mercenaria---

AVIGLINCS een ne ene tear A amen eS EO ee Se et eee 130-136
vertebrates, poikilothermal -----__.__._._____.-______- 181-200
Gigestive:cuzVmes tees se ass eee cao eee 181-200
carbohydrate-splitting enzymes_...._..._______- 192
INVeruneenZyIMesaseno ee eee sees ee 195
digestivestractreactlonsesssse=2-- see sess eee aes 184
erepsin, occurrence, mucosa of intestine ___ 197
erepticidigestionss-sees ses. coat one eee SUE. ee 190
pepticidigestionss 222 sot ssa sos eos sel eces ecu eee ee 184
esophagus’: te sace ses scene Sr og en 184
Stomach] S=smon ie eee eae Ve OR oe 185

cry pticidigestions: ssseessa-2 2a eee ene eee 188

vesiculosa, Biddulphia.

vilsoni, Pimelenotus 251
Vincta, Lacuna. --22---..-..- cee Lebplor
WITENIS WNELCIS seo saesoe ees ena 130-133
Pollachinseaesssss soe. cae See Oe Ae 172
VIDIGIS © CNGLOMOMIUS seen aes ae einen nae eee ere ee ed 269
WUlgaris wD actylopusidessssa=. tase ea eee neta ee 141, 145, 146
(Paleomonetes as scoeees poses te ae 155-159
WANE WRN AMCs Seerel se heed ee me ee ee eee 252
Weymouth, F. W., H. C. McMillin, and H. B. Holmes:
Growth and age at maturity of Pacific razor clam,
IStUGuapatulasGDixXON) ees seen eee eee ee elas 201-236
WiDILO}DESS Sees ne et oe eh ES ce ee 182
pepticidicestionsavew= == Somes fie ee 185
willeiK@ heetoceroSeas. soe o os Bees en ene eee 116, 118-120
wintersflounder sass = este ee id oie eee 133
Woods Hole, seasonal distribution, plankton_._________ 91-179
Xa phophoxusieil lie a hae ee eee ee 255
XG HOSUTa Meet sek eens ee eet Oe ee ae 161, 162
ZONALAN Ser Ol asec see tee seen aeons Soo ae! 168, 172
ZONaLUS iC hetodipteruss.22- ee eta see eee ese eee 287
IE DDL DDUS See eee oh ra eee eed, SER Ee ose eer 28;
zooplankton, influence of pelagic diatoms_______________ 174
Zosteranmaring= 2-225 5.22. see ee 125
zostericola, @ yilmdroleberis=. ssessssss 2-2 eee eee te 141
HI pDpolyteh ae cose ce ser ne es ste ee te 155-159
Zygodactylaigroonlandica-____2-22 222-2 22-2 124-130

So

ata Hi iad o
Fels cuert, Te Ristarin¥
elisera barr ett mount Sairahy
{Lh pitiioons

itu pedee

‘ i sal deun Meegrr ee iQ Aisa
rarneaniy
th s oleae eid ps STA
Suto Lae 2 a pa Hputniohiaiine, woah dirt
7 «ih LAI pa = Javea wv aah
ne S Rt -2}3 ViPZ Od pay

rai Boe yal
moose dunn with

OTE GUEST OFC RE REE OH 8 i,

SOTA EO
ook doth tolaeedanten

a Mary ¢ PY} Se tt} 6 eae ea J
Brisa lls

Gina

> SAA AMR a Pda pa a ae

BAL OR Riise Dyan NaN sty
eioibs ete (ate
ROM Py ste

BE ONE renee

wa emw eOlENue Oii58 a SAT Tyo ATs
whe Asser a wintpeeerel ian ee Ato eats ole
ec te wale Nha eal asi daa Ale a er WESCpaCe
i adie : : STN WINKCLORD
. pee i ris gm rnd Uehiie ORNL,
ns a oeigs hem eink (alent ad aS ALD AAR RE SI Gu ale daa Ae daha ed
Seay Sect i leh lobe ve wolualeye | VR MY ECe ae
i ae : -elidonointth jatokliy | hd aha
wT a 5 epee ty Atom akitiv } 5 Ae a ORD reat
tee Fide Wasi age ita Talon a ice Shh owe ey EOE aoaly ide Serer cheap
, Hs ost ss werey a= MOREA D Oy f | i's ee : *
py pee ~ Hs EE ey woth UOC TORUS ye UMe REN! EWR Vali cal nee ce ait rel oree
fae EL ie phe pbolvinntl shea toy | EL) ADRS yg Ate RB)
Het oye es Giake nie ye eee on, 4 TAL oot RR, Fe aes ibi
EB ore See a RIGO, hearsay ‘ Mie er uaek
‘ TEE bias UE DAO) Ne A eT AY ce pe eet Aone oto a rer
ish jortr ofiivel to whiten Ja oa Rate feet) ' bie erage PROM at ca ete
re ii dees te Cee eae _. -{toxdQ)) alisenehrs phe aE LF eo ar fae
Svc tes id li pagel oR RO Fromueae Ug Ne nahn
gente . rolinnwily sieqiag b wa ngehtss pees arlene
ital wonmaolneds) delitecl Me ToL Loe Oe At ict aly ac en
: os each witrige FSO Getta Nas
Ei) : ri a staal g loge Likely Laie gope eh ein CN y oi ves Oona Pea eat a
Gia sawn PN Ne +
-Hltg aekonpaug x (PER YAS wus thoy
ii. F Sasi pe aijeod gid, lems th ol oa ees
NA eae Et ee Ee ae
i. hath ea catireoe | bieioiinie at tru
y 4 ¥ iQue zittatae | bi raibeet Wethereanns aa
oer : eee tt eae enudlia ; aes ae ae! ii
verte tails shuabuap My aaa Tiere tie HAlgeiqons } ie RoR iat sabes hid
hai 1 UR ee sexta botiaediaee walkin Sao, fi rip) ben Yume Sat “bene ‘a
2 Ae ; milo) Alba Modes if
a Pes, Ibi Vege tH ‘tbadieard ween {St Wet Ahh ETAT CALS Hie eine ae
#4 ; fae sas pea lc daa ae

vielen

Piha,
Lent
me Te

Pa
R as Ha
Pia

ote LN
v

be
Hn

vy Th,
Wn Tyre

i
Ate
ed

Bt). OA

oN ae
toa

OL eiiagl
opt
uN fiver)

i
(

wey
, MP

Wehs
AY
as