(AO- 1*15

17 -A CONTRIBUTION TO THE LIFE HISTORY OF DIBOTHRIUM CORDICEPS
LEIDY, A PARASITE INFESTING THE TROUT OF
YELLOWSTONE LAKE.

BY EDWIN LINTON, PH. D.
[With Plates cxvii-cxix.]

I. — INTRODUCTION.

In July and August, 1890, I was requested by the Hon. Marshall McDonald,
U. S. Commissioner of Fish and Fisheries, to investigate this interesting parasite.
My instructions were to join with Prof. S. A. Forbes, of Champaign, Illinois, in the
field work of an expedition sent out by the U. S. Fish Commission for the purpose
of investigating the life of the principal lakes aud streams of the Yellowstone National
Park, Wyoming, the special work assigned to me being to determine, if possible, the
cause of the prevalence of parasites amoug the trout of Yellowstone Lake.

I desire in this connection, to express my deep gratitude to Professor Forbes for
the unabated interest which he took in my branch of the work and for the valuable
hints and suggestions which I received from him. I should also be remiss if 1 did
not express my personal obligations to Capt. F. A. Boutelle, D. S. Army, Superin-
tendent of the Park, whose enthusiastic interest in the expedition aud prompt assist-
ance in many ways contributed very much to the successful conduct of the investi-
gation. Mr. Elwood Hofer, our esteemed guide and friend, and Mr. F. D. Booth did
much volunteer work which receives but inadequate compensation in this mention of
their names.

II. — BIBLIOGRAPHY.

The literature relating to this parasite is confined for the most part to the reports
of the U. S. Geological Survey, under Dr. F. Y. Hayden.

The first mention of the worm that I find is in the Report on Montana and Adja-
cent Territory for 1871, p. 97, where Dr. Hayden remarks as follows respecting the trout
of Yellowstone Lake :

We were able to discover but one species of fish in the lake, and that was trout, weighing from 2 to
4 pounds each. Most of them are infested with a peculiar intestinal worm, which has been described by
Dr. Leidy in a subsequent portion of this report as a new species, under the name Dibothrium cordiceps.

337

Bull. U. S. F. C. 89 22

338 BULLETIN OF THE UNITED STATES FISH COMMISSION.

I make the following extract from a report prepared at Dr. Hayden's request by
Mr. Campbell Carrington and published in the report cited above, pp. 97-98.

A curious fact * * * is connected with these fish, namely, that among their intestines, and
even interlaced in their solid flesh, are found intestinal worms varying in size, length, and thickness,
the largest measuring about 6 inches in length. On cutting one of these Trout open, the first thing that
attracts your attention are small oleaginous looking spots clinging to the intestines, which, on being
pressed between the fingers, break and change into one of these worms, small, it is true, but never-
theless perfect in its formation. From five or six up to forty or fifty will be found in a trout, varying,
as I said before, in size, the larger ones being found in the solid flesh, through which they work their
way, and which in a very short while becomes almost putrid. Their number can generally be esti-
mated from the appearance of the fish itself; if many, the trout is extremely poor in flesh, the color
changes from the healthy gray to a dull pale, it swims lazily near the top of the water, losing all its
shyness and fear of man ; it becomes almost savage in its appetite, biting voraciously at anything
thrown into the water, and its flesh becomes soft and yielding. If, on the other hand, there are few or
none, the flesh of the fish is plump and solid, and he is quick and sprightly in all his motions. I
noticed that it was almost invariably the case when a trout had several scars on the outside of the body
that it was free from these worms, and therefore took it for granted that the worms finally worked
their way through the body, and the flesh on healing up leaves the scars on the outside ; the trout in a
short time becomes plump and healthy again.

Mr. Carrington, after proposing a theory to account for these worms, further
states that "while all the fish above the upper falls are more or less affected by
them, below and even between the upper and lower falls such a thing as a wormy
trout is never heard of."

Allusion to the wormy trout of Yellowstone Lake is made by F. H. Bradly in
Hayden's Keport on Montana, Idaho, Wyoming, and Utah, 1872, p. 234.

Capt. William A. Jones, in his Eeport on Northwestern "W yoming, including Yel-
lowstone National Park, 1873, p. 22, speaks of this parasite as follows :

We find, as others before us, that the trout of the lake (Yellowstone) are perfectly splendid in size
and condition, but are full of parasitic intestinal worms, which leave the intestines and enter the flesh.

The specimens submitted to Dr. Leidy for examination were in a bad state of
preservation. It was therefore difficult to make out details of structure. Upon com-
paring my specimens with Dr. Leidy's figures I was at first led to think that there
might be two distinct species of the genus Dibothrium represented by the parasites of
these trout, but I now think that the difference must be due to the macerated condi-
tion of the material upon which Dr. Leidy based the original description.

I make the following extracts from Dr. Leidy's notice of this worm, which was
published in Hayden's Eeport on Montana and Adjacent Territory, 1871, pp. 381-382 :

Among the specimens submitted to me were several of the worms inclosed in oval sacs imbedded
in fragments of flesh. The sacs having remained unopened preserved the contained parasite from the
general decomposition of the others, so as to enable me to ascertain its character. It belongs to the
genus Bothriocephalu8, or rather to that section of it now named Dibothrium. Two species have long
been known as parasites of the salmon and other members of the same genus of fishes in Europe, but
the tape- worm of the Yellowstone trout appears to be a different one.

Two of the best-preserved specimeus of the tape-worm measure 5 inches in length by a line in
width at the broadest part. The head, almost a fourth of a line in diameter, is obcordate, as repre-
sented in the magnified figures subjoined. The two bothria or suckers are thick and discoidal, placed
back to back, obcordate in outlino, and directed with their broad and slightly depressed surface toward
the margin or narrower diameter of the body. The body is flat, thick, with rounded margins, and is
narrowly annulated. The annulations appear to be due to muscular bands, and number about ten to
the line. If other segments exist, independent of these annulations, as a character of the worm, the

UPB

LIFE HISTORY OF A TROUT PARASITE.

339

conditiou of the specimens does not allow of their distinction from transverse fractures at irregular
distances. No genital apertures could he detected at the sides or at the margins. Internal organs of
any kind could not he seen, hut the soft interior tissue of the hody is filled with round corpuscles
resembling in appearance starch-granules. These proved to he composed of carbonate of lime, as they
were completely dissolved by acetic acid with the evolution of carbonic acid. From the sliape of the
head this tape-worm might appropriately ho named Dibothrium cordiceps.

My attention was first called to this parasite in December, 1889, when specimens
were sent me for identification by the U. S. Commissioner of Fish and Fisheries.
These had been collected the previous autumn by Dr. Jordan, who published some
accouut of them in his report on the fishes of the Yellowstone National Park.* These
specimens, although in a good state of preservation, were in poor condition to show
superficial characters owing to their crumpled and contorted condition. The bothria
on the larger specimens were indistinguishable. The larva have the power of retract-
ing the head, which character indeed seems to be retained to a certain degree in the
larger specimens, and specimens on this account often do not exhibit the specific
characters plainly. The report in which I published an accouut of this parasite, and
which contains notes on the anatomy of this and another Dibothrium from the sucker
(Catostomus ardens), gives, therefore, a rather meager description of the superficial
anatomy of this worm. In my first report on this worm I suggested that some of the
fish-eating birds inhabiting the lake would probably be discovered to be the final host
of this parasite of the trout. When, therefore, the opportunity was afforded me of
investigating the matter, I made an examination of such piscivorous birds as I could
secure, with the result given below.

III.- LARVAL STAGE.

In the larval stage, Dibothrium cordiceps occurs either in cysts among or on the vis-
cera of its host, the trout (Salmo mykiss); free, on or among the viscera; beneath the
peritoneal lining of the abdominal cavity ; or burrowing in the muscular tissue of the
body-wall.

The cysts, which are in reality blastocysts or nurses, within which the larvae develop,
are of various sizes. Some were found less than 1 millimeter in diameter, others were
12 millimeters or more in the longer diameter. The smaller cysts are globular or sub-
oval, the larger ones are oval. The smaller larva? up to a centimeter or more in length
are lanceolate, flattish, tapering rather abruptly and uniformly to each end. The larger
specimens are linear, increasing slightly in breadth posteriorly, with a blunt truncate
or emarginate termination; the body is crossed by fine transverse wrinkles, closely
crowded together, but in the louger specimens presenting the phenomenon of distinct
segments one-half millimeter or more in length. In the living specimens the vascular
system is quite evident. It consists of four longitudinal vessels, two near each margin,
the smaller vessel of each pair being near the margin. Branches from each of the
marginal vessels extend to the other and from each of the larger vessels to the other
These brauches also anastomose with each other by means of secondary branches
(Fig. 11).* The aquiferous vessels unite with two oval vessels at the posterior end, which
is often emarginate, with a terminal pore at the emargination (Fig. 13).

The head in the living worm is small and extremely variable, sometimes stretching
out until it is linear and not more than one-fourth 'he ordinary diameter (Fig. 6) ;

* Bull U. S. F. C, 1889, pp. 41-63.

340

BULLETIN OF THE UNITED STATES FISH COMMISSION.

sometimes abruptly linear from the body, which then appears to be shouldered ante-
riorly (Fig. 4) ; at other times the whole anterior part of the body becomes exceedingly
attenuate : sometimes it is swollen or truncate at the apex (Fig. 5) ; sometimes rounded
liuear or spatulate ; sometimes triangular, wedge-shape, and even retracted until it
is no longer distinguishable. The bothria, or suctorial pits, are lateral, and the head
is flattened so that its broadest aspect is seen when the worm is lying on its edge or
shorter diameter (Fig. 9). The edges of the bothria are rather thin and lip-like and
are often crumpled or thrown into sinuous folds owing to longitudinal contraction of
the head (Figs. 1, 2, 3, 7).

The color is usually ivory white or translucent white with an occasional tinge of
yellow.

There is no indication of genital organs other than that mentioned in my previous
paper of clusters of nuclei, shown in thin sections, lying along the median line near
one of the lateral faces, thus indicating that in the adult the genitalia are median and
the apertures lateral.

DISTRIBUTION.

This parasite occurs, so far as known, only in the Eocky Mountain trout (Salmo
myMss). I have found it in the trout of Yellowstone Lake, Yellowstone River above
the lower falls, and in Heart Lake. I did not succeed in getting any fish from below
the lower fails, for examination. I am told, however, that wormy trout are never found
in the river below the lower falls. It is very probable, however, if careful search were
made for them, that an occasional trout in the river and its tributaries below the lower
falls would be found with cysts of this parasite. At the Grand Canon Hotel I exam-
ined some trout which were said to have been caught below the upper falls and found
one with cysts in the abdominal cavity and a large larva among the abdominal muscles.
In Heart Lake I found the trout not infrequently infested with this parasite, occurring
in cysts and free on and among the viscera, but not among the muscles.

Dr. Jordan reports that the trout of Riddle Lake, which drains through Solution
Creek into Yellowstone Lake, are apparently free from parasites. It may be that this
conclusion would have to be abandoned if an examination were made of several of the
large trout of that lake.

Following is an abstract of notes made at the time of collecting, inserted here for
the purpose of showing the actual occurrence of this parasite in the fish examined :

HEART LAKE.

All of these trout were taken in front of our camp near the Rustic Geyser Basin,
with trammel-net and hook and line.

(1) July 26. — Found one trout near shore with wounds on its sides, from the effects
of which it was dying ; had evidently been struck by a fish hawk ; no dibothria.

(2) Ten trout examined ; a few cysts found on serous coat of intestine and pyloric
coeca; only two of the number had many cysts, about a dozen each. The serous coat
of the intestine of these two was somewhat congested and the swim-bladder was more
deeply colored than in the others.

(3) July 28. — Eight trout examined, most of them with a few cysts, as in No. 2.
In one there was a larval Dibothrium in the abdominal cavity on the outside of the

LIFE HISTORY OF A TROUT PARASITE.

341

serous membrane. It was translucent white, bluish with transmitted light, 20 milli-
meters in length and 1.75 millimeters in breadth.

(4) Four trout examined ; a few cysts in each, and a free larva 31 millimeters in
length, yellowish white ; one cyst, 9 millimeters by 4.5 millimeters in the two principal
diameters ; when opened, liberated an active larva 15 millimeters in length and 1 mil-
limeter in breadth. Along with larva in the cyst was a considerable amount of milky-
white nutrient material. Larvae were seen to make their escape from cysts that had
been left lying for an hour or two in water.

Mem. — The Dibothrium cysts are found in the majority of the trout, but usually
not in great numbers. Often the development appears to have been arrested, and the
cyst degenerates into a calcareous or waxy calculus.

(5) July 29. — Six trout examined ; only one found with cysts in abdominal cavity.

(6) July 30. — One trout examined ; a few cysts among pyloric cceca.

YELLOWSTONE LAKE.

(7) August 1. — Five trout examined ; taken in trammel-net near camp on west
arm of lake near Lake-shore Geyser Basin ; all with cysts, two with many, one with
worms in flesh, three with worms escaped from cysts, and one with worm under peri-
toneum, not yet in flesh. Of these five trout the first was a male about 35 centi-
meters in length, few cysts ; the second a female, same size, about thirty small cysts,
the largest 3 by 6 millimeters, on pyloric cceca, some of the cysts with the larvse
escaping, some yellowish and waxy, two cysts on ovary, the largest 12.5 millimeters
in length, cyst about size of a small pea in the liver; the third, a male, same size,
several cysts on pyloric cceca, one larva about 5 centimeters in length nearly free from
its cyst ; the fourth, a male, same size, two or three cysts on pyloric coeca, one cyst
under peritoneum, not in flesh ; the fifth, a male, a little larger than the others, about
thirty-six cysts on the pyloric cceca, one larva 38 millimeters long outside of peri-
cardial cavity, two cysts in testes, one larva in flesh a short distance back of right
ventral fin 5 centimeters in length, coiled irregularly, flesh surrounding it inflamed
and sore, pit made from peritoneal side, two others under peritoneum of air-bladder,
surrounded by inflamed tissue, with apparently a good deal of lymph exuded. One
of the larvae taken from the muscles of the latter specimen, after lying in water for
two hours, measured 24 centimeters in length, aud an hour or two later 34 centimeters
in length ; greatest breadth 2.5 millimeters.

(8) August 2. — One trout caught with hook and line near warm water of geyser
basin, a spent female; two hundred and twenty cysts of various sizes on pyloric caeca;
one larva in flesh, beneath peritoneum, on left side back of pectoral fin; one on left side
back of gill slit; two on right side back of gill slit; one in liver; a few cysts scattered
along intestine and ovary; fish in poor condition. Five trout examined, caught with
hook and line a short distance from entrance of warm water; few cysts, no flesh worms.
Five others from near warm water, caught in trammel net, more or less infested with
cysts; one large larva in intestinal wall of one of the fish, intestine adherent to body
wall and larva beginning to penetrate the latter; one larva in flesh above the lateral
line, near the dorsal fin ; four others from same lot ; several cysts, none in flesh. Nine
others, caught with hook some distance from warm water; no worms in flesh; a few
cysts on intestines and in ovaries of a female.

342

BULLETIN OF THE UNITED STATES FISH COMMISSION.

(9) August 3. — Twenty-two trout examined ; seventeen of these with from few to
many cysts; six of the latter with larvae escaping from cysts or under peritoneum ; one
larva measured 54 centimeters in length after lying for eighteen hours in water; five
without any parasites.

(10) August 4. — Three trout examined; two males with few cysts, one female with
several cysts, and one larva burrowing under the ribs.

(11) August 6. — Four trout caught with hook, north end of lake, not far from outlet ;
one or two cysts in each of three males; one female had a few cysts on pyloric caeca ;
no flesh worms.

(12) August 7. — Thirty-one trout caught with hook, north end of lake; five with
worms in the flesh, twenty with cysts only, for the most part with only two or three;
six without parasites.

(13) August 9. — Nine trout caught with hook in southeast arm of lake, not far
from inlet; one with several cysts in abdominal cavity and larvae under peritoneum.
Six taken from near mouth of cold stream from mountain side ; four of the latter with
cysts on pyloric caeca, and one with a larva about 30 centimeters long in flesh near
back-bone; five others, no flesh worms, two with cysts.

(14) August 13. — Examined a number of small trout from 12 to 20 centimeters in
length, caught in the head waters of Alum Creek, South Fork, above hot springs. No
cysts or other evidence of parasites discovered after very careful search.

(15) August 25. — Examined twenty trout caught in Yellowstone River near outlet
of lake ; only one was found to be much infested with parasites. It was a female and
in poor condition. The parasites were on and among the pyloric caeca ; no flesh worms
noticed. On same date examined another lot of about a dozen trout superficially.
Only one of them was in bad condition. Upon opening it a large number of larval
dibothria were found in the abdominal cavity. The soldiers who were fishing there
said that they found the fish of the river less commonly parasitized than those of the
lake.

(16) August 26. — At Grand Canon Hotel examined three small trout caught just
below the upper falls. There was no indication of dibothria in these fish. A large
trout, 36 centimeters in length, also said to have been caught below the upper falls,
was in poor condition, and had several cysts and migrating dibothria on the pyloric
caeca. There was, in addition, a large larva under the peritoneum and burrowing
through the kidneys into the muscular tissue. Another specimen, said to have been
taken above the upper falls, was also in poor condition. There were many cysts in
the abdominal cavity, but no larvae in the flesh. Another from the lake was in good
condition, but had several cysts on the pyloric caeca.

In the above extracts from my notes the only parasite of which account is taken
is Dibothrium cordiceps. As a matter of fact the trout are infested by a number of
parasites. A small Bistomum is common in the lower intestine, and a slender, white
nematod in the intestine in the vicinity of the pyloric caeca. Nematods are not uncom-
monly found encysted among the viscera, and some very peculiar soft globular cysts,
filled with a granular fluid, looking like tumors, covered with a layer of peritoneum,
which is richly supplied with capillaries, when opened liberate a small nematod.
The latter have not yet been studied. A lerneau parasite is also common, usually on
the fins or at the base of the fins, and not rarely in the mouth.

LIFE HISTORY OP A TROUT PARASITE.

343

When the size of the trout is not mentioned it may be understood to be about 35
centimeters long.

Spent females seemed to be more commonly parasitized to a serious degree than
others, although my examinations have not been extensive enough for safe general-
izations in this particular.

From this investigation one might conclude that the trout of Yellowstone Lake
are not so badly infested with the flesh parasites as they have been in previous years.
This may be due to the fact that the observations were made earlier in the season
than those of Dr. Jordan for example, who found the worms more abundant than they
appear to be now. It is to be noted, however, that larva; in various stages of de-
velopment are obtained from the same individual. In one case cysts no larger than
a grain of wheat were associated with others larger than a pea, and others from which
the larva; had escaped and begun to burrow into the flesh of their host, attaining a
length of 16 centimeters or more and a breadth of 3 or 4 millimeters.

It may properly be inferred from this that the source of infection continues through
several months of the year, thus showing marked contrast with the large ligula of the
Witch Creek sucker, Catostomus ardens, in which case the nearly uniform size of the
parasite points to a source of infection supplied, perhaps, by the short sojourn of
some migrating piscivorous birds.

With regard to the escaping of these parasites through the skin of the host, men-
tioned by Mr. Carriugton, I was unable to find any confirmatory proof. In a few
instances I found that the parasites had penetrated the muscular tissues and were
lying immediately beneath the skin ; more commonly they were on the peritoneal side
of the body wall or burrowing in the muscles. I saw no evidence in the fish which I
examined that parasites had escaped. If it can be demonstrated that these parasites
really do leave the intermediate host and take to the uncongenial medium of the
water, it would furnish an instance of self-destruction unusual among cestod parasites.

I was frequently asked, while in the Park, if these worms are in any way injurious
to man. I think it can be safely answered that they are not, except as their presence
might make the fish less acceptable to the palate. Fish, perhaps more than any other
animals, are required by nature to harbor parasites. There is probably not a food
fish in the world that does not furnish a home for one or more species of parasites in
some stage of the latter's existence. Fortunately fish parasites, as a rule, do not
live in man ; at any rate, the various processes of preparation for food to which fish
flesh is subjected effectually destroy the vitality of the parasites. It may be, if not
pleasant reading to the fastidious, at least consoling to the timorous, to know that
forms closely related to the subject of this sketch are in some places actually eaten as
food and esteemed as delicacies by those who eat them, who, it may be inferred, ask
no questions either for conscience' sake or for the sake of knowledge. Ligulre, para-
sites of the European tench and of other related fish, are used as food in Italy, where
they are sold in the markets uuder the name maccaroni piatti, and eaten usually under
the mistaken notion that it is the roe of the fish. It is also eaten in Lyons by many,
where it goes by the appropriate and truthful name of the ver blanc*

* Donnadieu, Contribution a l'histoire de la Ligule. Extrait Journal de Anatomie et de la Physio-
logic, P. 1.

344

BULLETIN OF THE UNITED STATES FISH COMMISSION.

IV.— ADULT STAGE.

I have found a large Dibothriuui in the white pelican (Pelecanus erythrorhynchus),
which is evidently the adult form of B. cordiceps, of which the trout (Salmo mykiss) is
the intermediate host.

Of the four birds examined, two contained this parasite. These were situated
about the middle of the intestine. They were in several fragments in each case, but
there was no tendency shown to separate into individual proglottides. Some of the
fragments were slender and attenuated, and when studied subsequently proved to be
the older parts of the strobiles degenerated into slender ribbons of connective tissue
and still containing eggs. In one of the birds, the length of the fragments, which
evidently belonged to a single strobile, was about two meters. There was in addition
to this an approximately equal amount of attenuated fragments of degenerated por-
tions of the strobile.

The fine impression made by the stately movements of the pelican while on the
water or in the air is not sustained on closer acquaintance. It has an abominably
rank and fishy smell.. It is grievously tormented with parasites. The alimentary
canal of an adult bird is from 2 to 2£ meters in length, and throughout its whole
extent it is liable to be infested with various parasites. Least numerous and least
painful are the dibothria in the intestine.

The mouths of each of the four birds examined contained hundreds of some mal-
lophagous parasite. These wer<>. attached to the mucous membrane by the head, and
required a sharp pull with the forceps in order to detach them. They were attached
in clusters, so that a dozen or more could be removed at one time with small forceps.
They were on the inside of the pouch, near the larynx, in the larynx itself, in the
beginning of the oesophagus, and in the buccal cavity generally. The oesophagus con-
tained an immense number of a rather slender nematod, 10 to 15 millimeters in length ;
these were usually attached to the mucous membrane, and left a small round hole
when removed. In the lower part of the oesophagus and in the stomach there were
also large numbers of nematods. These were larger than those in the oesophagus,
with thickish, usually dark-colored, bodies; they were not attached to the mucous
membrane, but in the stomach were in the midst of the food.

The stomachs of each of the birds contained practically nothing but partly
digested fish.

Much of the stomach contents of each of these birds had suffered so little from the
processes of digestion that the size of the fish could be easily estimated. In each case
the fish which had constituted the last meal of these birds were from 30 to 36 centi-
meters in length, or, in other words, the average size of the trout of the lake. There
was a little gravel or coarse sand at the bottom of the stomach and occasionally a little
vegetable d6bris, the feather of a small bird, etc. These apparently had been swallowed
incidentally to the main business of eating. These birds were found breeding on
Molly Islands, in the southeast arm of Yellowstone Lake, July 10, 1890.

SUPERFICIAL CHARACTERS.

The following description of the superficial characters is based on an alcoholic
fragment with scolex attached, and measuring about 75 centimeters.

LIFE HISTORY OF A TROUT PARASITE.

345

Head, in lateral view of body, elongated, wedge-shaped (resembling exactly some
heads of D. cordiceps from cysts in Sahno myMss, which I have collected), largest at
base, tapering slightly towards apex, narrower than neck, except immediately at base,
where it is of same breadth as neck (Fig. 17). In marginal view of body the head is
broader than the neck and snbsagittate (Fig. 18). Bothria lateral, lips rather thin and
flexible.

Millimeters.

Length of head 2. 00

Diameter, corresponding to marginal diameter of body .80

Diameter, corresponding to lateral diameter of body 55

Diameter of neck, lateral 80

Diameter of neck, marginal 60

Neck, immediately behind head, crossed by fine lines which, less than 2 milli-
meters back of the head, divide the body into short but evident segments. The body
increases in breadth slowly and uniformly, and at a distance of 30 millimeters from the
head is 2.5 millimeters broad; at 60 millimeters, 4.5 millimeters broad, at which point
the segments are about 0.5 millimeter in length. At about 40 millimeters from the
head a faint, darkish, median line is discernible, which becomes darker and more evi-
dent farther back. With a lens this median line is seen, in maturing segments, to be
occupied by the reproductive apertures. The dark color is due to clusters of ova.
The reproductive apertures are lateral, near the median line of the strobile, not far
from the anterior edge of each segment, and on but one of the lateral faces of the
strobile. From some of the apertures a short, blunt cirrus was seen protruding.

For about 15 centimeters the strobile was thickish with entire margins, then for
about 10 centimeters it was characterized by remarkably ruffled margins. This was
in the widest part of the strobile. The segments were here much crowded. The
strobile in this ruffled portion is also thickish, at least when it is compared with the
succeeding portions. At this point the breadth is 7.5 millimeters, the broadest part of
the strobile; the length of the segments, 0.34 millimeter ; the thickness about 1 milli-
meter (Fig. 21).

Beyond the ruffled portion, the strobile becomes flatter, much thinner, and decreases
in breadth, while the segments become squarish. A characteristic scalloped margin
is produced (Fig. 23) by the frequently concave margins of segments and their promi-
nent posterior edges. Near the posterior end of this fragment the segments had the
following dimensions : Length, 2 millimeters; breadth, 4.5 millimeters ; thickness, 0.5
millimeter or less. The ruffling of the margin of the median portion of the strobile
may be due in part to the action of the alcohol.

Another fragment, smaller in most of its dimensions than the above, presents some
characters which should be noted. The entire length is 22 centimeters. The head is
somewhat contracted longitudinally and is bluntly rounded in front.

Millimeters.

Length of head 1. 00

Diameter of head, corresponding to marginal diameter of neck 0. 75

Diameter of head, corresponding to lateral diameter of neck 0.75

Marginal diameter of neck 0. 50

Lateral diameter of neck 0. 75

Neck almost immediately crossed by fine transverse stria?, which give rise to
distinct segments 5 or G millimeters back of the head, where they are one eighth of a

346 BULLETIN OF THE UNITED STATES FISH COMMISSION.

millimeter in leugth and 1 millimeter in breadth. The terminal segments in this frag-
ment are 2.75 millimeters broad and 1.25 millimeters in leugth.

This fragment, for the most part, resembles the larger specimen, being flattish,
white, roughened by transverse wrinkles, and having its margins at one part of its
course ruffled. In one place, however, 7 centimeters from the head, and for a distance
of 3 centimeters, it is shrunken, attenuated, and yellowish. A segment in this part
measured 1.5 millimeters in length and barely 1 millimeter in breadth. Immediately
beyond this the normal segments were 5.75 millimeters in length and 2.5 millimeters
in breadth.

Some slender, attenuate fragments, previously alluded to, looking when first col-
lected like tangles of narrow braid, were at first supposed to be a distinct parasite.
Subsequent examination demonstrated them to be fragments of mature portions of the
large Dibothrium. Following is a more detailed account of these fragments than has
yet been given : One fragment, measuring 34 centimeters in length, was very thin, and
about 1.5 millimeters broad. In some places segments could be made out. These
were much attenuated. Some of them were measured and found to be 3 millimeters
in length. In some shorter fragments, which were quite irregular, being flat in places
and in others cylindrical, segments were noticed which were 4 millimeters in length
and 1.25 millimeters broad. Another fragment, 33 centimeters in length, was reduced
throughout the greater part of its length to a mere filament one-fourth of a millimeter
or less in diameter. Another fragment, about 30 centimeters in length, was reduced
to less than one-half of a millimeter in diameter throughout most of its extent, but
widened in each direction, so that each extremity was a little over 1 millimeter in
breadth.

A thin fragment was placed in glycerine, and was then seen to contain ova along
its entire extent. The ova lay along the median line in clusters, which were somewhat
continuous. In this the joiuting was indistinct. In some of these fragments the joints
are distinct, although there is no tendency in the joints to separate. A fragment,
35 centimeters in length, showed distinct joints in part of its course ; in other places
the joints were indistinct. This fragment was flat and thin and about 1.5 millimeters
wide where it was flattened ; in other places it was not quite so wide, owing to a tend-
ency to roll up at the margins. Both the flattened portions aud the attenuated,
cord-like portions of these fragments were found to contain ova, which were seen to
be abundant in specimens made transparent in glycerine. These slender fragments
are evidently portions of the larger strobiles, in which the muscular tissue has degen-
erated, and the whole structure is reduced to a narrow band of connective tissue con-
taining ova.

ANATOMY.

In order to make a comparison of the structure of the adult with that of the
larva, portions were stained with borax carmine and cut into thin sections. These
were made in three directions, transverse, longitudinal parallel to the margin, and
longitudinal parallel to the lateral face.

Sections of the head show that while that organ is somewhat smaller than it is in
the larger larvae, the muscular fibers are much more abundant and more strongly
developed (Fig. 34). In outline and arrangement of parts they are the same. There
appear to be three principal systems of muscles or contractile fibers in the head,

LIFE HISTORY OF A TROUT PARASITE. 347

longitudinal, circular, and radial. Some of the radial fibers, however, are continuous
with longitudinal fibers of the body. The first of these, the longitudinal, is fairly well
developed in the larval stage, the latter two, the circular and radial, but feebly
developed. Each is represented by many strong fibers in the adult. In both larva
and adult there is a layer of well-developed longitudinal fibers immediately beneath
the cuticle of the bothria. (Fig 34, s c.)

The layers of the body have the following disposition (Fig. 27) :

(1) The cuticle, which appears, when highly magnified, to contain minute circular
fibers.

(2) A thick granulo-fibrous layer, the outer portion of which consists mainly of
longitudinal fibers, which, in transverse sections, are seen to be continuous, with
radial sheets of conuective tissue. The inner portion of this layer is less distinctly
fibrous and more granular. In transverse sections the connective tissue of this
layer appears as radial fibers, which become thicker near the cuticle, where they
are more or less parallel, somewhat branched, the branches anastomosing. Toward
the inner part of the layer they lose their parallelism and form a me^h of fine fibers.
This layer contains the peripheral system of vessels, which, however, is not so promi-
nent as it is in the larval stage. Towards the inner edge of this layer in mature and
maturing segments the granular parenchyma is collected into roundish masses, form-
ing the vitellaria.

(3) A layer of coarse longitudinal muscle fibers. In the interstices between
these fibers there are fine connective fibers which extend from the outer granulo-
fibrous layer, and are continuous with connective fibers of the central core.

(4) A thin layer of fine circular fibers surrounding a central space.

(5) The central core of the body. This space, where not occupied by the geni-
talia and the vessels of the water- vascular system, is filled by a net- work of connect-
ive fibers in mature segments, but in the anterior immature part of the strobile it
contains much granular protoplasm, from which later the genitalia develop, the tes-
ticular lobules developing towards the margins, while the cirrus bulb, vas deferens,
vagina, ovary, shell-gland, and uterus develop in the vicinity of the median line.

The marginal canals and the aquiferous vessels, which lie in the central core, have
practically the same characters, proportions, and disposition as they have in the larval
form, as elucidated in my former paper.

A few nuclear clusters, which appear in some of the transverse sections made
near the anterior end of the head, are probably nerve ganglia (Fig. 34, n). Calca-
reous bodies, so abundant in the larva, appear to be entirely wanting in the adult.

GENITALIA.

The genital apertures are close together, but distinct, lateral, about on the me-
dian line; the male orifice much larger than the female and situated uear the anterior
edge of the segment; the female orifice is situated behind the male orifice and a little
to one side. Sections show that the male orifice (Fig. 25) communicates directly with
the relatively large, oval, cirrus bulb, which contains muscular fibers in a loose, open
tissue, among which the cirrus is usually retracted. The bulb is embedded deeply in
the segment, lying for the most part in the central core between the longitudinal
muscle layers. At its base it is in connection with a sub-globular organ (Fig. 25, p) y
which appears to be a short vas deferens, or, since its walls are rather thick and dense

348

BULLETIN OF THE UNITED STATES FISH COMMISSION.

and composed of fine fibers, presumably contractile tissue, it may also function as an
ejaculator\ r duct. The vas deferens joins this from the posterior side. The cirrus is
apparently iuvaginated by means of numerous retractor fibers, which are inserted on
the inner w.dl of the bulb and to the cirrus (Fig. 25, a). The testes are oval granular
bodies lying in the central core rather towards the margins (Fig. 32, t). They com-
municate with the median vas deferens by means of minute ducts.

The vagina opens near the cirrus, posteriorly to and a little to one side of it. It
pursues a somewhat tortuous course posteriorly, apparently reaching the ovary on its
dorsal side; near its beginning it expands into a seminal receptacle (Fig. 26.) The
ovary (Figs. 27, 33) appears to be a single mass of nucleated cells lying near the
posterior edge of the segment and towards the ventral side, near the ventral layer of
longitudinal muscles. Its longest diameter is transverse to the longitudinal axis of
the segment. The shell gland, as I have made it out, lies near the ovary, somewhat
posteriorly on its dorsal side (Fig. 33, s g). The uterus is a very voluminous organ be-
ginning on the dorso-posterior side of the ovary and lying in broad folds rather toward
the dorsal side of the segment, the folds extending some distance on either side of the
median line and occupying nearly the entire length of the segment ( Figs. 27, 28, 33).
Its caliber is much larger than the diameter of a single ovum. In the older segments
the walls of the uterus apparently give way, since the ova in them are seen to be in
masses in the central part of the strobile. (Fig. 5i4.)

There is no indication that the segments separate in any other way than in long
chains when mature.

Sections made parallel with the lateral faces oi the strobile show a small pore lying
a short distauce back ot the genital apertures and on the same side, i. e., the ventral
side of the strobile. This is apparently the external orifice of the uterus from which
ova may be discharged.

The ova are rather large, about 0.07 millimeter in length and 0.035 millimeter in
breadth. The walls are thin and usually collapsed in alcoholic specimens on one side,
so that, in mounted specimens, the ova appear to be bowl-shape. They are of an
amber color, do not stain readily, and their contents are granular. They lie in some-
what transversely parallel clusters along the median line, the masses of ova occupying
a space approximately equal to one-third the breadth of the strobile.

By counting the ova in a series of sections carried through a mass of ova in a
maturing segment, it was estimated that the segment contained 2,300 ova. The frag-
ments from the intestines of one of these pelicans contained approximately over 2,000
segments. This would make 4,600,000 eggs from this ntrobile. This estimate does
not include the ova in the attenuated fragments, which would probably yield as many
more. These numbers are only approximations, but they probably are far below the
actual numbers. It would be much within the bounds of probability to say that for
each pelican on Yellowstone Lake in any season there are 5,000,000 eggs of Dibothrium
cordiceps discharged into the waters of the lake, under such conditions that make it
probable that a small percentage of them eventually obtain lodgment in their proper
intermediate host, the trout.

V.— REMARKS ON CESTODS.

During my sojourn at the hotel on Yellowstone Lake, at which time I met several
tourists and others from various parts of the country, from England, and from France,

LIFE HISTORY OF A TROUT PARASITE.

349

all of whom were more or less interested in the so-called " wormy trout" of the lake,
and were especially anxious to know something of the cause of the malady, I found
that with an occasional exception it was necessary for me to preface any remarks I
had to make in answer to queries by giving a short disquisition on tapeworms in
general before I could make it clear that I was not a raving theorist when I stated
that the cause of the wormy trout was probably the wormy pelican.

Since this report is likely to fall into the hands of some who may be interested
enough to read it, but whose previous reading on the subject of tapeworms has never
extended beyond the obtrusive headlines of some quack advertisement, I have deemed
it expedient to give a very brief account of the life history of a typical and familiar cestod
worm, following this with an equally brief statement of what has been found out with
respect to the life history of a near relative of Dibothrium cordiceps, before giving an
account of what is the probable round of life of the latter worm.

The cestods, then, are a peculiar natural order of worms, all the members of which
are parasitic during all or at least the greater part of their existence. As a rule, two
animals of different kinds, and related to each other as eater and eaten, are required
to enable the cestod to complete its life history. One of these is called the final and
the other the intermediate host, the cestod being in each case a more or less unwelcome
guest.

Probably the best-known member of this order is the common pork tapeworm
(Tainia solium), whose intermediate host is usually the pig, in whose flesh it passes the
larval or encysted stage of its life, constituting the so-called " bladder worm" or
" measles ' : of measly pork. In order to attain the adult stage it is absolutely neces-
sary that the bladder worm be swallowed by the proper animal. In this case the
proper animal is man, in whose intestine the bladder worm becomes the adult tape-
worm. The pig is therefore an intermediate and man a final host of the common tape-
worm. Tne life history of the tapeworm proceeds, therefore, in this wise : Man, par-
taking of improperly cooked, pork containing larval tapeworms, swallows one of the
latter, which, being liberated by the action of the digestive fluids from the cyst of con-
nective tissue in which it is inclosed, soon finds a lodgment on the walls of the duode-
num or other part of the small intestines, where it clings by means of an exceedingly
small head provided with minute hooks and four small sucking disks. The body gives
rise to a chain of joints or segments, which rapidly mature, and are voided with their
contained eggs in the natural way, and, under certain conditions not necessary to detail,
find their way into the stomach of the pig. A minute embryo is there developed from
each egg, which penetrates the walls of the stomach or intestine, burrows through
the tissues, and finally comes to rest, usually in the muscular tissue, where it becomes
encysted and develops into the bladder worm. Thus the humble round of its passive,
though somewhat eventful, life is complete.

The worm which infests the trout belongs to the genus Dibothrium, or as it is fre-
quently written, Bothriocephalus. It has been known for some years that certain forms
related to this genus, which as larvre, known by the name Ligulcc, iufest many of the
European fresh-water fishes, more especially the Cyprinidce, reach their final or mature
stage in a variety of aquatic birds. This has been demonstrated by the experiments
of Duchamp and Dounadieu, who succeeded in raising mature Dibothria in the intes-
tines of ducks, which had been fed Liguke from the Tench. The migrations in this
case, as made out by Donnadieu, are as follows : The eggs develop in the water,

350

BULLETIN OF THE UNITED STATES FISH COMMISSION.

where they give rise to ciliated embryos, which bear a close resemblance to ciliate
infusoria. These pass into fishes, particularly the Cyprinoids, where they become
established in the peritoneal cavity. The round of life is completed in the intestines
of aquatic birds, where the eggs are rapidly formed.

VI.— REASONS FOR REGARDING THE PELICAN AS THE FINAL HOST OF

D. CORDICEPS.

While there seems to be no reason to doubt that the tapeworm found in the
pelican is the adult of the trout parasite, it may be well to sum up the evidence which
has led to that conclusion.

In the first place, it is to be noted that the parasite of the trout is a true larva and
shows no signs of assuming the adult condition in the trout. Even the largest speci-
mens, which have left their blastocysts and migrated into the tissues of their host,
show but the faintest beginnings of the reproductive organs. Moreover, no cestod
is known to attain the adult condition elsewhere than in the alimentary canal of its
proper host. Again, out of the large number of trout examined there was not a single
case of a mature worm of this genus in the alimentary canal. A few of the large trout
of Yellowstone Lake were convicted of cannibalism, since their stomachs contained
remains of trout, the only species of fish in the lake. There is no doubt that a large
number of trout of the lake are eaten by larger fish of the same species. If the flesh
worm of the trout ever matures in the intestine of the trout, or, in other words, if
the trout is both intermediate and final host of this parasite, I should have found
some evidence of it. Failing to find the adult in the trout, search had to be made
among the animals which feed on the trout. A very brief consideration of the fauna
of the Yellowstone region was sufficient to make it clear that the adult form of this
worm must live in the intestine of some of the fish-eating birds that inhabit the lake.
Moreover, since the cause of infection must extend through several months of the year,
as shown by the variety of sizes of parasites occurring in a single fish, the final host
is seen to be more probably a bird that stays through the summer than one which is
only a visitor. Pelicans abound on the lake, one or more of them being usually in
sight, on any part of the lake, at any time of day during the summer months. They
have numerous roosting-places and at least one breeding place on the lake. They are
known to be notorious fish-eaters. It is clear, therefore, that collateral evidence alone
points strongly to the pelican as a, if not the, final host of the flesh worm of the trout.

Evidence of a more direct nature, however, was obtained by the capture and exam-
ination of four pelicans. Their stomachs contained partly digested remains of large
trout, and practically nothing else, thus demonstrating their ability to capture the large
trout of the lake, and showing that they live exclusively dn a fish diet ; in the intes-
tines of two of them were tapeworms, which there is no reason to doubt are the adult
stage of the trout parasite. When one of these worms from the pelican is compared
with a parasite from the flesh of the trout, the head with its characteristic bothria,
or pits, is found to be practically unchanged. The strobile, or jointed body of the
worm, is, as might be expected, much longer and larger in every way. There do not
appear, however, auy characters in one that are contradicted in the other.

On August 9, on the shore of the southeast arm of Yellowstone Lake, I picked
up at the edge of the water three fragments of a Dibothrium strobile. The largest of
these fragments was 70 millimeters in length and 6 millimeters in breadth. It was

LIFE HISTORY OF A TROUT PARASITE.

351

mature, contained ova, and was evidently identical with those found in the intestine
of the pelican.

This fact confirms what seemed probable from a study of the worms from the
pelican, viz : that the strobiles are passed in chains from the birds. The excessively
parasitized condition of some of the trout may thus be accounted for.

Again, some of the contents of the rectum of a pelican, in whose intestine speci-
mens of D. cordiceps had been found, was preserved in alcohol for examination. Upon
examination of a small portion of this material several ova were found which were
easily recognized to be ova of D. cordiceps. A small fragment of the strobile of the
same parasite was also found, which was much frayed evidently by the digestive
processes of the host.

It may therefore be taken as demonstrated that both ova and fragments of the
strobile of D. cordiceps find their way into the water where they may be swallowed by
the intermediate host, the trout.

Of course this argument does not quite amount to a demonstration. If some one
who has the time, opportunity, and inclination would conduct a series of experiments
of feeding specimens of trout parasites to ducks, in which they might develop, or,
better, to pelicans, in which I think they will certainly develop, it would serve to raise
a part of the history of this worm entirely out of the regions of conjecture.

There remains also to be ascertained the fate of the eggs of the pelican tapeworm
after they have been consigned to the water.

VII.— CONSIDERATIONS RELATING TO THE PARASITISM AMONG THE TROUT

OF YELLOWSTONE LAKE.

In these considerations two problems will be discussed: First, to account for
the abuudance of parasitized trout in Yellowstone Lake ; and, second, to account for
the migration of the parasite into the muscular tissue of its host.

In order to reach a proper understanding of the matter, a brief review of some of
the physical features of Yellowstone Lake and the surrounding region is necessary.
Yellowstone Lake is a large body of water of very irregular outline, containing about
150 square miles of surface and having a coast line of approximately 100 miles. It
lies near the great continental divide and empties through the Yellowstone River into
the Missouri-Mississippi River system. About J. 8 miles below the lake there are two
falls in the river. The upper fall is 109 feet in height, the lower, which is one-half
mile farther down stream, is 308 feet in height. On the western side of the continen-
tal divide there are three lakes, much smaller than the Yellowstone, but still quite
considerable bodies of water. These are called Heart, Lewis, and Shoshone, respect-
ively, and each is not more than 8 miles in an air line from the nearest point on
Yellowstone Lake. Lewis and Shoshone Lakes empty through Lewis River into Snake
River, a tributary of the Columbia. Between them and Snake River there are falls of
some 60 feet in height. Heart Lake empties into Snake River through Heart River,
but there are no falls on this latter stream.

The natural distribution of fish in these lakes presents some peculiarities which
should be mentioned here. In Lewis and Shoshone Lakes there are no fish. This is
not due to an absence of food nor to the presence of conditions unfavorable to life,
since these lakes were found to be swarming with amphipods, entomostracans, and

352

BULLETIN OF THE UNITED STATES FISH COMMISSION.

insect larvae. Plainly, therefore, the presence of falls on Lewis Kiver, insurmountable
by fish, must be regarded as the real reason why there are no fish in these lakes.

In Heart Lake, on the other hand, there are three or four species of fish, viz, the
trout (Salmo myTeiss), the chub (Leuciscus atrarius), the sucker (Catostomus ardens),
and probably the blob (Coitus bairdi), all common species in the Rocky Mountain
waters. This fish fauna is therefore what we might expect to find in the lake.

In Yellowstone Lake there is but one species of fish, viz, the trout (Salmo mykiss),
identical with the trout of the waters on the western side of the continental divide.
The lake was found teeming with this species of fish and no other by the first explorers
of the region. It is not at all probable that the lake was stocked by the aborigines,
and the explanation given by Dr. Jordan is doubtless the correct one, namely, that
the trout gained access to the waters of the Upper Yellowstone through Two-Ocean
Pass, where the waters of the Yellowstone and the Snake arise from the same swampy
meadow on the great continental divide. Prof. F. V. Haydeu has shown (Bull. U. S.
Geological Survey, vol. v, No. 2, "The so-called Two-Ocean Pass") that during times
of high water caused by the melting snows, there is actual connection between Atlantic
Creek, a tributary of the Yellowstone, and Pacific Creek, a tributary of the Snake.
It seems reasonable, therefore, that the trout, an active and somewhat gamy fish and
fond of the colder streams, should make its way over the divide, while it would be
exceedingly unlikely that the more logy sucker and chub should, under the circum-
stances, attain a like distribution.

Since there are no fish in Lewis and Shoshone Lnkes it is obviously necessary to
make a comparison here only between Heart and Yellowstone Lakes.

The invertebrate life of these lakes, while affording one or two interesting con-
trasts, presents no differences that would have any bearing on the presence or absence,
abundance or scarcity of parasites. There is but little difference in elevation, Heart
Lake being 7,409 and Yellowstone 7,741 feet above ocean level. The depth, so far as
known, is much the same, although the Yellowstone, being much the larger body of
water, will probably be found to have the greater depth. A depth of 146 feet was
obtained at a distance of about 800 feet from the west shore of Heart Lake, where the
bottom temperature was 40° P. A depth of 159 feet was obtained on the west arm of
Yellowstone Lake at a distance of 2,000 feet from the shore, the temperature at bottom
being 42° F. A depth of 195 feet was found at a distance of about a mile from shore at
the north end of the lake. The temperature of the surface water varied with the
time of day, but was practically the same in the two lakes. Near shore, at our camp
on the west arm of the Yellowstone, on August 2, at 9 a. in., the surface temperature
was 54° F. So far as the temperature of the lake water is concerned and the inverte-
brate life of the two lakes, the fish in Heart Lake and those in Yellowstone Lake are
living under substantially similar conditions. Whatever influence the presence of
warm water from hot springs and geysers exerts, the conditions are practically the
same, since each has geyser and hot spring regions on its borders and each receives
warm tributaries from such regions.

I think it likely that after all the only difference between the two lakes, that
touches this question, lies in the fact that, while in Heart Lake the trout are associ-
ated with the chub and the sucker, and consequently suffer or profit as the case may
be by the mutual reaction which this association implies, in the Yellowstone they are
alone and neither profit by the presence of another species, which they might use for

LIFE HISTORY OF A TROUT PARASITE.

353

food, nor suffer from having a part of their food supply appropriated by another
species, nor receive partial immunity from sudden death by having their co-species
furnish a part of the food of their common enemies.

Indeed I am not sure that the proportion of parasitized trout in Yellowstone Lake
is so overwhelmingly greater than it is in Heart Lake. While I did not find any flesh
parasites among the Heart Lake trout, I found that a great many of them had the
parasites in the peritoneal cavity, both encysted and free among the pyloric cceca.

Pelicans were seen frequently during our stay on Heart Lake and their breeding
place at the south end of Yellowstone Lake is ouly 10 or 12 miles away. It would
be very strange, therefore, if* the trout of Heart Lake were found to be free from
these parasites.

Neither am I at all certain that the parasites of the Heart Lake trout never pene-
trate the flesh of their hosts. However, after examining a great many trout from this
lake without finding any penetrating the flesh, it may be worth while to consider this
question: Why should the parasites of the Yellowstone trout have the habit of bur-
rowing into the flesh of their hosts, while those of Heart Lake seldom or never do %

I am not sure that I can give an altogether satisfactory answer to this question.
It appears to me, however, that the reason for this difference is to be found in the
peculiarly isolated and circumscribed situation of the trout in Yellowstone Lake.
In Heart Lake the trout have at least two species of fish besides their own to feed
upon ; in Yellowstone Lake the trout, if they are to eat fish at all, are obliged to resort
to cannibalism. This is certainly done to some extent. In a few cases I have found
evidence of it in the stomachs of the Yellowstone trout. In each such case the canni-
bal was a trout above the average size. Mr. Elwood Hoferhad also observed this fact
and stated as the result of his observation that "occasionally a cannibal is met with,
and when it is, it is sure to be a big fish."

The Yellowstone Lake trout are confined to the lake and the river above the
upper falls. It is true that this species is found in the river between the falls and be
low the lower falls, being quite abundant in the Yellowstone Eiver and its tributaries.
It is not likely that large fish could be carried over the lowerfalls and live ; small fish
and ova, however, might be so transported uninjured. Whatever may be the truth
with regard to the passage of fish over the falls, it is certain that no fish could return
to the lake after having once made the descent. The trout of the lake, therefore, are
compelled to pass their whole life within the limits of the lake and its tributaries, or
if they do leave, the door is shut behind them with no hope of its ever opening for
their return. In Heart Lake the case is different ; not only can the trout leave the
lake at will, either by tributary streams or by the outlet, but having left they can
return again.

To what extent these fish migrate with the changing seasons and diminishing
food supply,. I have no exact knowledge. Whatever may be their habits in this
regard, the trout of Yellowstone Lake are forced to limit their migrations to the lake
and its tributaries, and to the less than 20 miles of river between the outlet and
the upper falls. The trout, being thus circumscribed in their range, if from any reason
their food supply should fail, must suffer the consequences. Tbe3 r can not seek new
feeding-grounds. If their food should be of such a kind as to produce peculiarities of
flavor, or if it should contain the germs of disease or parasitism, they will be continu-
Bull. U. S. F. C. 89 23

354 BULLETIN OF THE UNITED STATES FISH COMMISSION.

ally exposed to the source of contagion or parasitism with no respite, with no seasons
of relief except what may be incident to the nature of the germs. It follows, therefore,
from the peculiar conditions surrounding the trout of Yellowstone Lake, that if there
is a cause of parasitism present in successive years the trout are more liable to become
infested than they would be in waters where they had a more varied range. Trout
would become infested earlier and in greater relative numbers, and the life of the
parasites themselves, that is, their residence as encysted worms, must be of longer
duration than would be the rule where the natural conditions are less exceptional.

Again, in such cases of parasitism as that under consideration, where the parasite
is a larval cestod and the intermediate host the only animal of its kind in this region,
the final host is more likely to partake of parasitized food than he would be if part of
his food consisted of other species of fish not harboring this parasite. For example,
it may be supposed that pelicans, when in the vicinity of Heart Lake, feed indiffer-
ently on trout, chubs, and suckers. Even if the trout stood an equal chance with the
others of being eaten, which is hardly to be supposed since it is a more active fish than
they, it would then constitute but one- third of the food of the pelican. The chances
of the pelican's becoming parasitized would therefore be diminished to one-third what
they would be from its diet on Yellowstone Lake. But with this diminished parasit-
ism in the final host would go a smaller amount of eggs from the adult parasites to be
disseminated where the fish are likely to get them.

As a matter of fact there can not be this difference between the parasitism of the
trout of Yellowstone and of Heart Lake, as these bodies of water are so near together
that the pelicans found ou the one are the same individuals which visit the other.

While in camp on Heart Lake, and also on the west arm of the Yellowstone, we
saw several pelicans, but secured no specimens. After we reached the hotel on YelloV-
stone Lake I made an excursion to the southeast arm of the lake for the purpose of
securing specimens of these birds. The southern end of the lake is as yet almost never
visited, and we discovered that the pelicans not only roost, as was known, but also
breed on the small islands there. Four pelicans were obtained on Molly Island in
a deep bay on the west side of the southeast arm of the lake. Our party were in
two rowboats, Mr. F. D. Booth, with our only shot-gun, in company with Messrs.
Thompson and Coughlin, visited the island and shot the birds for me. As we had found
the pelicans thus far rather shy, I and my companion, Mr. Curl, did not approach the
island, but rowed across to the northern shore, keeping well outside of the island for
fear of frightening any birds which might be there. I thus, much to my subsequent
regret, missed seeing the breeding-place of these interesting birds. Mr. Booth reported
that they could easily have secured a boat load of the birds if it had been deemed
necessary. He estimated that there were at least five hundred pelicans, young and old,
ou the island, less than one hundred of which were young. The young were large
enough to take care of themselves by running and swimming, but could not fly much.
The old birds refused to leave the neighborhood of their young and so could readily be
shot. The specimens secured were all adults and measured 92, 96, 97, and 98 inches,
respectively, from tip to tip of the wings. They had an exceedingly rank, fishy smell,
and the gentlemen who visited the breeding- place bore testimony in a variety of select
epithets that the smell of the place was horribly bad. The result of an examination
of these birds has already been given. Many gulls were also found breeding on this
island.

LIFE HISTORY OF A TROUT PARASITE.

355

It is probable that other small islands in the southern end of the lake, for exam-
ple, Peale's Island in tbe south arm, may also be breeding-places of the pelican. We
saw them on our return, roosting on the south end of Frank's Island, but we did not
pass near enough to ascertain whether it was a breeding-place or not.

It is easy to see from the foregoing why the fish of Yellowstone Lake should be
greatly infested by these parasites. There are probably not less than one thousand
pelicans on the lake the greater part of the time throughout the summer, of which, at
any time, not less than 50 per cent, are infested with the adult form of the parasite,
and since they spend the greater part of their time on or over the water disseminate
millions of tape-worm eggs each in the waters of the lake. It is known that eggs of
other Dibothria hatch out in the water, where they swim about for sometime, looking
much like ciliate infusoria. Donnadieu found in his experiments on the adult Dibothria
of ducks, that the eggs hatched out readily in warm water and very slowly in cold.
If warm water, at least water that is warmer than the prevailing temperature in the
lake, is needed for the proper development of these ova, the conditions are supplied
iu such places as the shore system of geysers and hot springs on the west arm of the
lake, where for a distance of nearly 3 miles the shore is skirted by a hot spring
and geyser formation with numerous streams of hot water emptying into the lake, and
large springs of hot water opening in the floor of the lake near shore. Trout abound
in the vicinity of these warm streams, presumably on account of the abundance of food
there. They do not love the warm water, but carefully avoid it. Several persons with
whom I talked on the subject while in the park assert that diseased fish, that is to say,
those which are thin aud affected with flesh-worms, are more commonly found near
the warm water, that they take the bait readily, but are logy. I frequently saw peli-
cans swimming near shore in the vicinity of the warm springs on the west arm of the
lake. It would appear that the badly infested or diseased fish, being less active and
gamy than the healthy fish, would be more easily taken by their natural enemies, who
would learn to look for them in places where they most abound. But any circum-
stances which cause the pelican and the trout to occupy the same neighborhood will
multiply the chances of the parasites developing in both the intermediate and final
host. The causes that make for the abundance of the trout parasite conspire to
increase the number of adults. The two hosts react on each other and the parasite
profits by the reaction. About the only enemies the trout had, before tourists, ambi-
tious to catch big strings of trout and photograph them with a kodak, began to frequent
this region, were the fish-eating birds, and chief among these in numbers and voracity
was the pelican. It is no wonder, therefore, that the trout should become seriously
parasitized.

It may be inferred, from the foregoing statements, that the reason why the para-
site of the trout of Yellowstone Lake migrates into the muscular tissues of its host
must be found in the fact that the life of the parasite within the fish is much more
prolonged than is the case where the conditions of life are less exceptional.

VIII.— REMEDY.

A natural inquiry following the discovery of the true nature of this parasite will
be : What remedies, if any, are proposed ?

A very effective remedy, and one which suggested itself immediately to Captain
Boutelle, he being an enthusiastic lover of the gentle sport of fishing, is to externa-

358

BULLETIN OF THE UNITED STATES FISH COMMISSION.

nate the pelican. It would, indeed, be a lamentable result of my investigation if that
dire calamity should bsfall the unwitting cause of this peculiar malady. One of the
most charming minor effects of the singularly beautiful scenery of Yellowstone Lake to
my mind is produced by the presence of these noble birds. I do not think such
heroic measures are either called for or advisable. The trout of the lake can never
figure as the food supply of a large number of people. Their abundance or scarcity
neither raises nor lowers the price exacted of tourists by the hotel association. They
are destined to contribute more to sport than utility. A speedy remedy is therefore
not necessary.

With the increase in numbers of visitors to the lake will go greater destruction
of trout by enthusiastic fishermen. This will probably reduce the number of diseased
fish at a more rapid rate than it will that of the healthy ones, and if the precaution
be taken not to leave dead fish on the shore, and not to throw them in the water, where
in either case they would probably be eaten by the pelican, the chances of the hitter's
becoming infested with the parasites will be correspondingly lessened. And particu-
larly if the lake be stocked with some other species of fish (of which I think the chub
of Heart Lake is most suitable, since it is an omnivorous feeder, and therefore not
likely to interfere seriously with the food of the trout, while furnishing the latter
with much-needed animal food, and at the same time lessening the chances of the
trout's being eaten by the pelican, and since the parasite does not develop in the chub)
a lessened parasitism of the pelican would result ; and with fewer parasites in the
pelican would go a diminution in the number of ova disseminated in the water, and
consequently a lessening of parasitism in the trout.

It is probable, also, that the presence in the lake of some fish which would form a
part of the food of the trout would result in imparting a more vigorous constitution
to the latter and make it better able to withstand the strain of excessive parasitism.

At any rate, before a war of extermination is waged against the pelican it would
be well to ascertain whether or not Dibothrium cordiceps develops in other fish-eating
birds. My own investigations have not been extensive enough to enable me to decide
this question. Beside the pelican, the only birds of the lake that I ha 1 an opportunity
to examine were three species of duck, none of them piscivorous, one hawk, one heron,
and three gulls. In one of the latter I found a Dibothrium bearing some resemblance to
the trout parasite, but evidently a distinct species.

Washington and Jefferson College, January 1, 1891.

LIFE HISTORY OF A TROUT PARASITE.

357

EXPLANATION OF FIGURES.
The following letters have the same significance iu relation to all the figures where they are used.

c. cuticle. p. v. peripheral vessels. ' ov. ova.

b. bothrium. m. v. marginal vessel. P. cirrus.

1. to. longitudinal muscles. a. v. aquiferous vessel. P<. cirrus bulb.

c. m. circular muscles. in*, vitellaria. d. vas deferens.
/. v. fibro- vascular layer. o. ovary. t. testes.

Drawings by the author.

Plate CXVII.

Figs. 1 to 14. Larval stage of Dibothrium cordiceps Leidy, from peritoneal cavity and muscular tissues

of the trout (Salmo mykiss).
Fig. 1. Sketch of head from life, marginal view, lateral of body, from muscular tissue of host ; after

lying in water three or four hours this specimen measured 34 centimeters in length and

2.5 millimeters iu breadth, nearly linear, X about 12.
Fig. 2. Head of specimen from flesh of host, marginal view of head, lateral of body, alcoholic, X

about 12.

Fig. 3. Same view of another specimen, alcoholic, X about 12. Lateral view of same specimen
shown in Fig. 8.

Fig. 4, 5, and 6. Sketches from life showing some of the diverse shapes assumed by the head and neck

of a living worm, each magnified about 12 liuear diameters.
Fig. 7. Head and neck of an alcoholic specimen, X about 12.

Fig. 8. Lateral view of head, marginal of neck, same specimen as sketched in Fig. 3, X 12.

Fig. 9. Lateral view of head, marginal of neck ; large specimen, 35 centimeters in length, from flesh

of host, alcoholic, X 12.
Fig. 10. Median segment of same specimen, X 12.

Fig. 11. Anastomosing vessels of water-vascular system, middle of body, compressed, sketch from life.
Fig. 12. Water- vascular system, posterior end of large specimen, compressed, sketch from life.
Fig. 13. Posterior end of another smaller specimen, compressed, sketched from life; a, pulsating ves-
sel; t.p., terminal pore.

Fig. 14. Posterior end of large specimen, alcoholic, X 12. The emargination is sometimes deeper

and sometimes not so deep as this.
Figs. 15 to 34. Adult stage of Dibothrium cordiceps Leidy, from intestine of the white pelican ( Pele-

canus erythrorhynchus).

Figs. 15, 16. Marginal view of head, lateral of neck of two different alcoholic specimens, X about
15 and 24, respectively.

Plate CXVIII.

(Adult stage of Dibothrium cordiceps Leidy, intestine of white pelican.)

Fig. 17. Marginal view of head, lateral of neck, alcoholic, X about 24. Same specimen as shown in
next figure.

Fig. 18. Lateral view of head of specimen figured in No. 17.

Fig. 19. Lateral view of head of another specimen, X 15, alcoholic.

Fig. 20. Segments of body towards anterior end, alcoholic, ventral surface, x

Fig. 21. Antero-median segments, ruffled margin, dorsal surface, alcoholic, X 4.

Fig. 22. Median segments of small fragments, alcoholic, X 12.

Fig. 23. Mature segments from large specimen, alcoholic, X about 5.

358

BULLETIN OF THE UNITED STATES FISH COMMISSION.

Fig. 24. Longitudinal section, near median line, parallel to margin of strobile ; the uterine walls have
been absorbed and the ova lie in masses ; the segmented appearance of the body is here
shown to be only superficial.

Fig. 25. Transverse section through cirrus bulb, highly magnified ; a, retractor muscles of cirrus.

Fig. 26. Transverse section through beginning of vagina, showing the enlargement of that organ near
the vaginal orifice into a seminal receptacle, highly magnified.

Fig. 27. Transverse section of mature segment showing the different layers of the body.

Plate CXIX.

(Adult stage of Diboihrium cordiceps Leidy, intestine of white pelican.)

Fig. 28. Diagrammatic sketch of mature segment, showing relative position of genitalia, etc.

Fig. 29. Ovum as seen in a thin section of a mature segment ; the knife has cut away a part of the

shell exposing the granular contents collected into spherical masses, highly magnified.
Fig. 30. Diagram of female generative organs, from ventral side ; v. d., vitelline duct.
Fig. 31. Section of vagina, highly magnified, showing ciliated interior.

Fig. 32. Portion of transverse section of segment, highly magnified ; Par., parenchyma of inner core
of body.

Fig. 33. Longitudinal median section, parallel to the margin, through cirrus bulb, uterus, vagina, etc.

Fig. 34. Transverse section of head ; n, nerve ganglia, a row of which extends across the section from
the inner margin of one marginal vessel to the other. The heavy stippling represents the
cut ends of longitudinal muscles ; «. c, subcuticular longitudinal fibers particularly well
developed in the bothria.

Bull. U. S F.C. 1889._(To face page 358-2.) Linton. Trout Paras„e

Plate CXVIII.

t