IN THE EVENING.

MISSOURI

BOTANICAL GARDEN.

SEVENTEENTH ANNUAL REPORT.

ST. LOUIS, MO.:
PUBLISHED BY THE BOARD OF TRUSTEES.
1906.

BOARD OF TRUSTEES OF THE MISSOURI
BOTANICAL GARDEN.

EpwarpD OC, ELIOT.
JOHN GREEN, M. D.
GEORGE C. HITCHCOCK.
LEONARD MATTHEWS.
WILLIAM H. H. PETTUS.
JOHN F. SHEPLEY.
Davin S. H. SMITH.

EDWARDS WHITAKER.

«

President,
RUFUS J. LACKLAND.

Vice-President,
DAVID F. KAIME.

ADOLF ALT, M.D.,
President of The Academy of Science
of St. Louis.*

WINFIELD S. CHAPLIN,
Chancellor of Washington Univer-
sity.*
GEORGE R. LocKwoop,! 3
President of the Board of Public
Schools of St. Louis.*

DANIEL S. TUTTLE,
Bishop of the Diocese of Missouri.*

ROLLA WELLS,
Mayor of the City of St. Louis.*

A. D. CUNNINGHAM, Secretary.

*Ex- Officio.

1 Elected President of the Board of Public Schools of St. Louis, October 9, 1906,
to succeed Robert Moore, who had held that office for one year. :

(2)

PREFACE.

Under direction of the Board of Trustees, the seven-
teenth annual report of the Missouri Botanical Garden is
presented to the public.

The sixteenth volume was issued on May 31st, 1905,
which is to be regarded as the date of publication of the
scientific papers it contains, except that by Professor
Hitchcock, separates of which were distributed on March
29th, 1905.

These reports are sent to scientific institutions and
journals in exchange for publications or specimens desir-
able for the Garden, and, when possible, reprints of the
botanical articles they contain are presented to botanists
occupied with a study of the subjects they refer to. Any
of the Garden publications not out of print may be pur-
chased at approximately the cost of publication from Mess.
R. Friedlinder & Sohn, Berlin, Germany; W. Wesley &
Son, London, England, or the undersigned.

Witiram TRELEASE.

St. Louis, Mo., October 10, 1906.
(3)

CONTENTS.

PaGr.
1. REPORTS FOR THE YEAR 1905:—
a. Report of the Officersofthe Board . ....... #7
b. Seventeenth Annual Report of the Director . . . . . 28
2. ScrentTIFIc Papers :—
a. Studies on the lignin and cellulose of wood.—
ee eae ee Oa 41
b. Studies upon some chromogenic fungi which Mickhen
wood.—
By George Grant Hedgcock . .. . a 59
-¢. Zonation in artificial cultures of dics had otheottien and
other fungi.—
By George Grant fo Oe ee an 115
d. Some new Texas plants.—
By B.D ae 119
e. Ascidia in Gasteria and Agave.—
By J. Arthur Harri = Sekt : 126
J. Prolification of the fruit in Capsicum ap Passiflora.—
By J. drthus Fes Ce Ort. ee oe ee
_g. Fasciation in Oxalis crenata and experimental produc-
tion of fasciations.—
: By Henri Hus Se ek es oe A Pee eb boste Bene 2 4
h. Constriction of twigs by the bag worm and ichiitans evi-
dence of growth pressure.—
By Hermann ven Schrenk . . 2... 1: , 153

AS

(5)

LIST OF ILLUSTRATIONS..

Frontispiece: In the evening.

Miepg the DrOOK =~. - 6s et the
Without the walls .

Improvement plans . +--+ +s * * * * * ° ;
A tangle of Ee Or a ee oe ed
Se a ee eae ae Sean hn | A ea

In the North American garden . ++ +
Plates 1-2 Si a RE Tree ee 5 |g a aS
Wee Sale... oy 5. eae et
Piates 10-16"... 0 eH

Plates 17-19 7 . . . . . . . . . . . ® . .
Plates 20-26, Diagrams 1, 2 ee ee eT a

(6)

. Facing p.

“

ec

cs

‘ec

x.

BROOI

ALONG THE

REPORTS FOR THE YEAR 1905.

REPORT OF THE OFFICERS OF THE BOARD.
SUBMITTED TO THE TRUSTEES JANUARY 10, 1906.

To the Board of Trustees of the Missouri Botanical Garden:

This report marks the close of the sixteenth calendar
year of the Board’s administration of the Missouri Bo-
tanical Garden and of the revenue property devised for

its support.
PURPOSES.

The General Assembly, by an act approved March 14,
1859, authorized and empowered Mr. Henry Shaw to con-
vey certain real estate and personal property to Trustees,
for the establishment of a perpetual fund for the support
of the Missouri Botanical Garden, to be maintained for
all time for the use and enjoyment of the public, and as a
means of education and research in botany, horticulture
and allied sciences.* Fuller details of Mr. Shaw’s pur-
poses are contained in his will.f

TRUSTEES.

Mr. Shaw died on the 25th of August, 1889, and his
will was admitted to probate on the second of September
of thesame year. As Trustees, he designated «* M. Dwight
Collier, Henry Hitchcock, Wm. H. H. Pettus, Dr. John B.
Johnson, Adolphus Meier, Wm. G. Eliot who is now Chan-
cellor of the Washington University, and his successor in
office, Charles F. Robertson, who is now Bishop of the

* Rept. Mo. Bot. Gard. 1: 26.
+ Rept. Mo. Bot. Gard. 1329.

(7)

ee eee ee aN Se
oem

8 MISSOURI BOTANICAL GARDEN.

Episcopal Church of the diocese of Missouri and his suc-
cessors, David F. Kaime, James Yeatman, Judge Saml.
Treat, Joseph W. Branch, Gerard B. Allen, Rufus Lack-
land, Judge G, A. Madill, the President for the time being
of the Public Schools, and -his successors in office, the
President for the time being of the Academy of Science of
St. Louis, and his successors, the Mayor of the City of St.
Louis and his successors in office, Dr. Asa Gray of Cam-
bridge, Massts., and Profr. Spencer F. Baird, Secretary of
the Smithsonian Institution, Washington City. The two
last named as honorary Trustees are added to the trust in
recognition of their scientific eminence and ability.”’

Both of the honorary members, as well as the Chancel-
lor of the University and the Bishop of Missouri, had died
before the composition of the Board was made known by
the publication of Mr. Shaw’s will, as had Mr. Allen and
Mr. Meier of the personally designated members. By
direction of the courts the membership of the Board was
therefore fixed at fifteen, comprising the ten then living
designated members and five ex officio. Organization was
promptly effected by a majority of the members thus con-
stituting the Board, Mr. Lackland being chosen President —
to which office he has been re-elected each year, and Mr.
Pettus acting as Secretary for a short time until provision
was made for the regular employment of a Secretary,
when Mr. A. D. Cunningham was appointed to that office —
which he has since held.

The Board were promptly put in control of the Garden
and revenue property, and assumed their responsibility as
Trustees from the nominal date of September 1, 1889.
One of their first acts was to secure care for the Garden
by the appointment of a managing Director for it, Mr.
Shaw’s selection of Professor William Trelease of Wash-
ington University being confirmed and the duties of the
Director being defined as ‘the duties prescribed for that
office in the last will of Henry Shaw, deceased, and such

REPORT OF THE OFFICERS OF THE BOARD. 9

other duties as may from time to time be prescribed by
this Board in pursuance of the trusts declared in said
will.”’

Of the original members of the Board appointed as
individuals, only three remain in office, Mr. Lackland,
Mr. Kaime and Mr. Pettus. Mr. Collier resigned in 1889
because of removal from the city. Judge Treat, because
of his advanced age, retired in 1890. Mr. Yeatman and
Judge Madill died in 1901. Mr. Hitchcock, who from
the first had been Vice-President of the Board and Chair-
man of its Garden Committee, died in 1902, and Dr. John-
son and Mr. Branch died in 1903. Vacancies were filled
for atime by Dr. George J. Engelmann, — elected in
1889 to succeed Mr. Collier, and resigned because of
removal from the city in 1895, and Mr. George 8. Drake, —
elected in 1890 to succeed Judge Treat, and resigned in
1895. Membership in the Board is now held further by
Mr. Leonard Matthews, — elected in 1895 to succeed Mr.
Drake, Dr. John Green, — elected in 1896 to succeed Dr.
Engelmann, Mr. John F. Shepley, — elected in 1901 to
succeed Mr. Yeatman, Mr. Edwards Whitaker, — elected
in 1902 to succeed Judge Madill, Dr. D. S. H. Smith, —
elected in 1902 to succeed Mr. Hitchcock, Mr. George C.
Hitchcock, — elected in 1903 to succeed Mr. Branch, and
Mr. Edward C. Eliot, — elected in 1903 to succeed Dr.
Johnson.

Bishop Tuttle, who had succeeded the late Bishop
Robertson before the organization of the Board, is the
only ex officio member who has met with it continuously
from the first. Mr. Chaplin, who became Chancellor of
Washington University in 1891, has served since that time
as a member of the Board. A number of well known
gentlemen have further found place on it, from time to
time, while serving as Mayor of the City, President of
the Academy of Science or President of the School Board.

~ mu ie
aes. a ‘=
ie cd. eae te
ee Ot er

a ae, eee gee

26 K Bi hE ie far
Cn” ee RG BEY 5, tO

ee OA te a aaa ne pe) 0 RRL ak SR aT TER aan a ee en
4 % ‘fe 3 ROSS

10 MISSOURI BOTANICAL GARDEN.

INCOME, TAXES.

When the Trustees were charged with their duties the
revenue property was appraised at $1,241,274.00,—
much of it being unimproved real estate, yielding little if
any revenue and burdened by taxes which year by year
have become more oppressive. By gradual increments the
gross rents have grown 324%, from $91,258.32 in 1890
(when the general taxes were $22,470.53) to $120,742.94
in 1905, the general taxes meantime having increased to
$36,397.57, or 62%,—nearly double the percentage in-
crease in rents.

Special street and sewer taxes average $10,075.75 per
year for the entire sixteen years, in addition to unusual
large expenses for sewer and street improvements, averag-
ing $47,360.05 for each of the last two years. The average
annual cost of insuring the revenue property and keeping
it in suitable repair has amounted to $12,940.94. Com-
missions and necessary legal and other occasional expenses
have averaged $2,336.75, and the office expenses of the
Board average $5,000.00, per year, — leaving an average
net income, nominally available for the purposes of the
Garden, of $52,261.85. Out of this, however, no less
than $137,384.22, an average of $8,586.10 per year, has of
necessity been reserved and has been ultimately entirely
spent for street and other purposes apart from the Garden
itself, to which the remainder, except for special bequests,
has been devoted for improvement or maintenance.

TESTAMENTARY CHARGES.

The testamentary provisions of the founder of the
Garden constituting fixed charges on the revenue are: (1)
Maintaining a net income of $3,500.00 from property
deeded during his lifetime to Washington University for
the support of the School of Botany; this has resulted in

WITHOUT THE WALLS.

=.

Bee
ss
“ ,
*

REPORT OF THE OFFICERS OF THE BOARD. 11

an average payment of $1,272.60 per year. (2) The sum
of $1,000.00 per year for a banquet to the Trustees of the
Garden and invited guests, the average expenditure for the
past sixteen years being $948.56. (3) For a banquet to
the gardeners of the institution and invited florists, nur-
serymen and market gardeners, $400.00 per year, of which
an average of $368.15 has been used. (4) For premiums
or prizes to be offered at a flower show held in St. Louis,
$500.00 per year, the expenditure averaging $360.75. (95)
For a sermon to be preached each year on the wisdom and
goodness of God as shown in the growth of flowers, fruits
and other products of the vegetable kingdom, $200.00
per year, which has been paid in full.

REAL ESTATE IMPROVEMENTS.

Among the real estate improvements made by the Board,
in addition to sharing in general street and sewer develop-
ment, is to be noted particularly the improvement of Flora
Avenue, by which the main gate of the Garden is ap-
proached from Grand Avenue. This street, nearly a mile
long, with the co-operation of other persons owning prop-
erty along it, has been converted from a narrow dirt road
into a wide macadamized boulevard with granitoid walks,
shade trees, and a well planted park strip down the cen-
ter, the expenditure of the Board on it amounting to

$44,840.78.
UNPRODUCTIVE PROPERTY.

It has been found impracticable to carry out Mr. Shaw’s
plan of long term residence leases for the unimproved real
estate in the vicinity of the Garden, and the courts have
granted to the Board power to sell this property, on con-
dition that the proceeds shall be reinvested so as to yield an
income for the maintenance of the Garden. Under this
permission some sales of building sites have been made and

as 7 aed ae 2 ae Be a’ i ee ae ke - thie Sid | +
Be TT ee MRE ES OIE CA eo te OMAR at UO eee Re eS AK SN: aR Ra
- ’ 7 ss hie

12 MISSOURI BOTANICAL GARDEN.

the Board hope gradually to sell all of the better class of
this property at a fair valuation; but much of the land is so
situated with reference to railroads and the factories, etc.
that have clustered along them as to have little prospective
value for residences, while, as has been sdid, the taxes
on it are becoming heavier each year. The Board are
therefore being forced to the conclusion that this part of
the endowment real estate must sooner or later be parted
with as a whole, and the proceeds reinvested in property |
of a better description, capable of yielding revenue. Upon
this conversion of the entire endowment into revenue-pro-
ducing property obviously rests the sole hope of the Board

, to meet the Director’s wishes— which they share—for a
much needed addition to the buildings used for the Garden
office, library, herbarium and laboratories, for additional
and larger plant houses to relieve the overcrowding of the
present collections so that they may be better displayed

ie and suitably enlarged, for further extension of the grounds
according to plans already prepared, and for the development
of the institution into the great and productive research —

pet center planned by the Director in accordance with the wish
of Mr. Shaw, — the realization of all of which, however,
is believed to be only a question of time.

THE PARK STRIP,

When Mr. Shaw presented land to the City for the

4 establishment of Tower Grove Park, he deeded with it a
strip 200 feet in width entirely surrounding this park, ex-

cept for the necessary entrance ways, with the provision,

agreed to by the City, ‘*that the Board of Commissioners

of Tower Grove Park shall from time to time cause to be

leased the said strip of land of 200 feet in width surround-

ing said park, in convenient lots not to exceed 200 feet in

front, nor less than 100 feet in front, to any one person

for periods of thirty years before renewal, for the purpose

REPORT OF THE OFFICERS OF THE BOARD. 13

of erecting villa residences thereon only; and all the gross
rents received from said leases, without deduction, shall be
forever paid over to said Henry Shaw and to his heirs,
executors, administrators and assigns, so that he and they
shall forever enjoy said rents; and said city shall execute
the proper leases therefor, which shall contain a clause
that there shall be only one residence on each tract so
leased.’’ The purpose was distinctly stated to be to make
of the 200 foot strip about the park not only a source of
ornament to the park but a source of revenue for the
maintenance of the Garden. A small portion of this strip,
adjoining the north gate of the park, was subsequently
released by Mr. Shaw to the City asa building site for a
residence erected for the superintendent of the park. When
the Board assumed charge of the Garden property it was
found that a few houses for park employees had been
erected, at Mr. Shaw’s expense, on the strip, andthe rental
from these was turned over to the Board; but these un-
sightly buildings have all been removed by direction of the
Park Commissioners and, with the exception of the small
temporary rents referred to, no revenue has yet been se-
cured from this portion of the revenue endowment of the
Garden. The Trustees should asearly as possible take steps
either to derive such income as may be secured by having
the park strip used in the manner described by Mr. Shaw
in his deed of gift, or to induce the City of St. Louis to
lease or purchase by condemnation the entire strip sur-
rounding Tower Grove Park, and thus enable the Board
to realize from it an income for the maintenance of the
Garden.

GARDEN IMPROVEMENTS.

At the Garden itself, the Board are pleased to note con-
stant and great though necessarily expensive change for
the better, and its valuation, with permanent improve-
ments, has nearly doubled, having increased from the

14 MISSOURI BOTANICAL GARDEN.

original appraisal of $125,160.00 to $227,185.94. When
the property came under the charge of the Trustees it was
a country home on which for some years inadequate
expenditure had been made for maintenance. The streets
about it were without sidewalks, and the walks within were
of such construction as to be impassable in wet frosty
weather. The surrounding walls were crumbling, the
plant houses were limited in capacity, of antiquated and
inadequate construction, and very badly out of repair, while
the residence assigned to the Director was found by a
committee of physicians to be in a most unsanitary condi-
_tion and a little museum that had been maintained for
many years had so deteriorated under the care of house-
hold servants that the Director was advised to close it.
The provision of room for administration, library, her-
barium and research purposes also confronted the Board as
an immediate necessity, as did provision for a supply of
city water, improvement of existing temporary sewers, and
some arrangement other than open grates for heating the
residence and museum building.

On these needed improvements the Board have heen
compelled to spend large sums of money. The full length
of the Garden front on Tower Grove Avenue has been ter-
raced, provided with a granitoid sidewalk and planted to
an avenue of maples,—already becoming beautiful and
yielding in summer a pleasing shade,—while cinder paths
have been maintained along the Shaw Avenue and Magnolia
Avenue sides; the larger of these improvements costing
$3,087.56. The principal walks of the Garden have been
suitably remade; the expenditure for this and the replace-
ment of much of the brick edging amounting to about
$3,000.00. The gate-house, walls, and fences were put in
repair at a cost of $5,056.03. The area covered by plant
houses has been more than doubled, at a cost of $29,608.45 —
the new houses being of modern construction, except for
two small temporary structures, and permitting the growth

REPORT OF THE OFFICERS OF THE BOARD. 15

of plants that could not otherwise have been cultivated.
Granitoid ponds, steps, etc., needed for the replacement of
decaying old wooden structures, or tofurnish opportunities
for growing such plants as the Amazon lily, have cost
$1,931.26.

At the Director’s residence the temporarily built, un-
sanitary and discordant east wing was replaced in 1890
by a conveniently planned wing, at a cost of $19,000.00.
A -water supply, including necessary piping along Tower
Grove Avenue, supplemented by a provisional sewerage
system, necessitated the expenditure of $4,589.10.

In compliance with an express direction of Mr. Shaw’s
will for the provision of a cottage to be occupied by @
person charged with caring for the mausoleum and its sur-
roundings, a neat building has been erected which also
serves as a gate-house for the private gate at Cleveland
Avenue by which the Director’s residence is reached from
the street. The cost of this improvement was $4,538.05.

Under another provision of Mr. Shaw’s will, requiring
its ultimate reconstruction on Tower Grove Avenue in some
convenient situation in contiguity to the Garden, his
former city residence was removed in 1892 from the cor-
ner of Seventh and Locust Streets and rebuilt, mainly in
fire-proof construction, in the Garden, on Tower Grove
Avenue between Shenandoah and Botanical Avenues, the
large expenditure of $33,478.89 for this purpose being
made by direction of the court. In connection with the
renovation of the residence and the removal of this build-
ing, which for the time afforded adequate room for the
office, library and herbarium, a boiler pit was provided at
a safe distance, from which steam heat is supplied to the
office, residence, and museum buildings ; $1,204.07 having
been spent for this purpose in addition to subsequent
improvement charges.

Among the provisions in Mr. Shaw’s will, explained in
some detail in certain manuscript suggestions not made a

ae

eo eee Noe or ae CPS Ree SY pte ie, een OTe ag
YP Pima ak eae ee eo A a ee

ee 4

*

pea

ee ae PORE Ry Heal hg ot hey me OE

i. oF Smee ee ee Oe
Rh ee ae,

et
Oe

eerie ie
\

eo
int

eT oy

16 MISSOURI BOTANICAL GARDEN.

part of that document, was one looking to the education
of a limited number of garden pupils, to be lodged conve-
niently for their work. For this parpose a small building
at the southeastern corner of the Garden, found to be in
very bad repair, was renovated as far as possible and suit-
ably furnished; the cost of this being $1,933.25.

STORM AND FIRE LOSSES.

The foregoing represent the larger items of construction
and improvement that it has been found imperatively
necessary to make. Several hail and fire losses have been
met, from time to time, amounting to $3,988.97. In 1896
the Garden suffered severely from the tornado which did
such great damage generally in the City, and the Board were
compelled to spend $4,479.36 in making such repairs as
were possible, though a complete renovation of the Garden
would have necessitated almost entirely clearing and
replanting its wooded parts.

EXTENSION. :

In connection with the tornado damage attention was
directed to the need of general plans along which the
grounds might be harmoniously and adequately improved
and enlarged. The preparation of such plans was en-
trusted to Messrs. Olmsted, Olmsted & Eliot, whose final
report has been adopted for the future development of the
property at such times and in such degree as the Board
may find expedient. The total expenditure in connection
with these plans has amounted to $5,037.75. Acting on
the advice of the landscape architects, the Board, in 1897,
purchased, at a cost of $10,000.00, a narrow triangular
strip of land containing 2} acres, needed to square the
property out to Alfred Avenue.

In accordance with the plans which have been prepared,

. is » aes }
a = ft OF

IMPROVEMENT PLANS.

‘ ‘
REPORT OF THE OFFICERS OF THE BOARD. 17

the southern extension of the grounds as so enlarged,
limited by Tower Grove, Magnolia and Alfred Avenues,
has been graded, drained and supplied with water pipes, and
planted with a collection of about 1,400 species of plants,
forming a synopsis of the North American flora, which is
soon to be opened to the public. Thecost of this addition,
aside from the land purchased, has been $21,286.34, — of
which $9,045.99 was for grading, $3,116.66 for the water
supply, $2,900.00 for drains, and the remainder for plant-
ing and minor improvements.

‘GARDENING.

On the maintenance of the Garden, including ordinary
improvements, an average of $43,675.33 a year has been
spent. Of this, an average of $23,271.39 is for wages
of gardeners and laborers, repairs and supplies, fuel,
water, additions to the living plants, and other sim-
ilar gardening purposes. The annual expenditure on
the Garden office, including salaries of the Director and
Superintendent, averages $5,217.67. The Board have
acquiesced in the Director’s belief that it is impossible for
the Garden to stand still, and provision for maintenance
has therefore been on a basis permitting rather small but
continuous development from year to year.

By the addition of the North American Synopsis, the
area of the grounds has been enlarged from 44.7 to about
65 acres, or nearly one-half. In 1889 the plant houses cov-
ered 14,840 square feet of ground; they now occupy 30,740
square feet, an increase of 107%. Propagating frames have
been increased from 2,200 to 5,750 square feet, or 161%.
It has been estimated that not over 2,000 or 3,000 kinds of
plants were cultivated at the time when the Board took
charge of the Garden. In 1895, after an accurate system of
recording plant accessions had been introduced, the number
of species and varieties was inventoried at 3,921, exclusive

2

18 MISSOURI BOTANICAL GARDEN.

of, perhaps, 1,000 annuals. In 1898 a new inventory
showed the presence of 8,009 kinds; the inventory of 1903
showed that there were then 11,357, and the number now
reported is 15,976, — an increase of 431% over the larger
of the estimates for 1889. Several of the special collec-
tions of plants are among the largest of'their kind in the
world. This is especially true of the cacti, comprising 678
named species representing 17 genera, and of the agaves and
yuccas. The collections of palms and sago plants are also
unusually large. From almost nothing, the collection of
bromeliads, or plants of the pineapple family, has grown to
204 species. Orchids, of which originally there were few,
and though two-thirds of the collection were destroyed by
fire a few years since, have been recollected to the number
of 942 species representing 116 genera, and an expert dealer
in plants of this group has recently spoken of the Garden
collection of orchids as being the largest in the United
States. With this increase in the variety of living plants,
their scientific value has been constantly raised by the in-
corporation of specimens collected in their native homes
by the Director, who has visited Mexico and Central
America for this purpose, and by others, thus permitting
a definite record to be kept of their origin, — which is not
usually true of purchased plants.

One of the early provisions made by the Board was for
the adequate labeling of the plants,—a need specially
mentioned in Mr. Shaw’s will; and the entire large col-
lection is now provided with names as far as the state of
maturity of the specimens makes this possible. With the
greatly increased variety and the accurate naming of the
plants they are better cared for than formerly and
their decorative use has received greater attention. The
former promiscuous beds of flowers have given place to
well set instructive groups of attractive species. As an
example of this may be mentioned the recent solid planting
of the sunken garden in front of the gate with beds of tulips

EOE SE TPES fe ee

REPORT OF THE OFFICERS OF THE BOARD. 19

for the spring and of foliage plants in summer, producing
an unrivaled display of the best varieties of these plants.
A collection of decorative species that are hardy in this
vicinity occupies a separate section of the central part of
the grounds, and the names of these plants are frequently
noted by visitors for use in planting their own gardens.
For the convenience of teachers and classes, several
hundred instructive plants are arranged in _ botanical
sequence, and forage plants, savory herbs and medicinal
plants are similarly grown in separate groups. One of the
features of the Garden for several years past has been the
growth through the summer of chrysanthemum plants for
display in the fall. Notwithstanding the limited space
available for this purpose, the chrysanthemum show of the
Garden.has won recognition as a competitor in attractive-
ness with the florists’ show held in the city, and visitors
who have traveled largely characterized that of this year
as the finest they had ever seen — not even excepting those
of Japan.

VISITORS.

The number of visitors to the Garden varies so greatly
from year to year, especially as affected by fair or unpleas-
ant weather on the two Sunday afternoons on which the Gar-
den is opened by direction of Mr. Shaw’s will, that it is hard
to furnish comparative figures. Until the middle of 1898
only estimates were made: since then a regular count
has been kept. In 1904, the World’s Fair year, the vis-
itors numbered 316,747. The average from 1899 to 1903,
was 83,503. In 1905 there were 100,830. The visitors
on the open Sundays average a little over a fourth of the
yearly total. .

Though the number of visitors is evidently increasing,
it is a matter of regret that more of our citizens do not
avail themselves of the privileges given by Mr. Shaw’s
bequest, and that so few seem to know that, excepting

Re ae Si et OO eR SE ES
: tao ES ee

oP

a Re, Se Tea =

Sd

-

OSs far de ones She Fe

me r, =

FET ARE eee bee
¥i ‘cre bates |

20 MISSOURI BOTANICAL GARDEN.

Sundays and holidays, the Garden is always open to every-
body, and free of all charge.

LIBRARY.

The provision of a library and herbarium for purposes
of scientific research, in addition to the maintenance of a
beautiful and instructive garden, are among the designated
purposes of the founder of the Garden.

A small but good nucleus of a botanical and gardening ‘
library came into possession of the Trustees when they as-
sumed charge of the property. To this the special libraries
of the late Dr. George Engelmann and of the Director, and,
later, the collection of early books of the late Dr. E. L.
Sturtevant, have been added by gift. Further smaller
gifts have been received every year, and purchases have
been liberally made, a yearly average of $4,418.82 having
been spent on the likrary, including salaries, fuel, pur-
chases and binding. From somewhat less than 5,000
numbers, the library has thus been brought to a present
total of about 51,000 books and pamphlets, —a tenfold
increase. It is now justly ranked as the foremost botan-
ical library in the country, and compares favorably with
the great libraries of the Old World. Its present valuation
is $84,248.35.

HERBARIUM.

In the same manner, the Bernhardi and Riehl herbaria,
bought by Mr. Shaw and numbering about 60,000 speci-
mens, were almost immediately added to by the gift of the
invaluable Engelmann herbarium, of nearly 100,009 speci-
mens; and gifts, exchange and purchase have further
brought the number of. specimens of dried plants at the
Garden up to a present total of about 524,000, — eighty-
seven times the original number. On the herbarium, an
average of $2,531.91 has been spent yearly, for salaries,
fuel and purchases; its present valuation is $79,216.75.

REPORT OF THE OFFICERS OF THE BOARD. 21

INSTRUCTION IN GARDENING,

The requirement that instruction in gardening and hor-
ticulture should receive attention at the Garden, in addition
to the provision of a lodging house for pupils, led to the
entire renovation of the fruit orchard some years ago, at a
cost of $444.44, and two small vegetable houses have been
built for further experimental and educational use. The
annual expenditure on the gardening course averages
$930.34, |

The Director reports that of the 39 pupils thus far
enrolled, of whom 15 completed the course, ten are now
successful florists or gardeners, two have become land-
scape architects, three hold responsible park positions, two
are college horticulturists with teaching as well as practi-
cal duties, one is a surveyor, one is a government plant
experimenter, and one is a forester in the Philippine
Service.

‘

INSTRUCTION IN BOTANY.

Mr. Shaw’s provision for a close connection between the
School of Botany, which he had endowed in Washington
University, and the Garden has been of great assistance to
the undergraduate department of the University, and
through the Garden opportunities for work have been
offered to graduate students of whom five have received
the Master’s degree and six the degree of Doctor of Phi-
losophy with botany as a major study. The Board expect
to see a large increase in this utilization of the Garden
facilities commensurate with the very gratifying growth of
Washington University. Indirectly the Garden has been
of much use to the young men who have served as assist-
ants in its office, library or herbarium, or as teachers in
the school of botany, for with very few exceptions they
have gone to college, government or other positions of high

BGs ses od

PO Ree Wren Loy TB

rc * >. - A e 4 So ~ a.
eat 2 rig een Sein + i = Ss, 2
wt * : + 4 of Cas tt oe. ty A ake bt a
eat eRe Pe eae s ‘ 5 a i 4 4 ;
oe 7 es 3 iy et an hee bes
: ge oe ee ee ed oe ee £" oe ce Pe a e

‘ x
Pa Mees. Legh

+ ae

‘

a.

Seas Ma fe 183

4 a
¢

sg Rad

7

ie Wah ORT RS
ora

Sea pea
ee Hie Be

eo Bh i A ae Oe
BOY r

oy
pa
$2)
=
=
Bis
K
sa
.

\
a

J

Pere ee ee) ene. ee ae eee eee
aMe Pgatet hale ee

> PARE Re gh ee 3, coe Pe Be ee

2 ee ORE

stay ale Abate

he aD ee ES a ee DENIS Ee a SR se a
ha 3 é + oe by te nents * ¥

22 MISSOURI BOTANICAL GARDEN,

responsibility in botany or horticulture, for which their
service here gave excellent training.

RESEARCH AND PUBLICATION.

Repeated mention is made in Mr. Shaw’s will of his pur-
pose to have scientific research, no less than decorative
gardening and _ botanical and horticultural instruction,
included among the activities of the Garden. Details of
administration have been left necessarily to the Director in
consultation with a special Garden Committee of the Board,
but his purpose to develop the institution symmetrically
under the broad plan of its founder has always met with
the support of the Trustees, who have approved his policy
of not only allowing but expecting a part of the time of
capable employees to be spent ininvestigation. Though he
looks forward hopefully to the time when the revenue of the
Board shall permit the employment at the Garden of a
corps of highly trained and expert investigators giving a
large part of their time to such study, he has been able as
yet to devote to this work only a small part of the time
of otherwise indispensable employees, the salary apportion-
ment and incidental expenses for this purpose averaging
$1,000.83 per year, for the last 12 years. He therefore
points with pride —and the Board share his gratification —
to the fact that under existing conditions not a year passes
without the accomplishment of valuable research work at the
Garden, and that its publications win commendation for
their useful contents as well as for the attractive and con-
venient form in which they are brought out and the
liberal conditions on which they are furnished to botanical
libraries and investigators everywhere. The yearly cost of
publication, including reissues of some of the volumes from
electrotyped plates which are preserved, has averaged
$2,000.37, but the Garden’s Reports have given it a stand-
ing in the scientific world that it could have attained in no
other way, and by far the largest part of the gifts to the

A TANGLE OF MOONSEED.

ix ee Ae

REPORT OF THE OFFICERS OF THE BOARD. 23

library and herbarium — averaging $1,775.36 per year for
the past 13 years— are received in exchange from cor-
respondents to whom the Reports are sent.

EXCHANGES.

Though already planted and partly equipped, the Garden
was without scientific affiliations when its development was
entrusted to the Board. To bring it into exchange relations
with other establishments, the publication of these annual
volumes, devoted to administrative reports, scientific mono-
graphs, etc., was early decided on. Through the medium
of this publication the Garden now stands in exchange re-
lations with 859 institutions interested wholly or in part in
botany, gardening, horticulture or forestry. The exchange
of plants and seeds with other gardens has recently been

- greatly increased by the issuance of an annual exchange

seed list which is sent to the officers of such establish-
ments, who may thus readily ascertain and request such of
their desiderata as can be furnished. The first seed list
included 813 species, and the list just issued contains the
names of nearly 1,400 kinds of plants of which seeds were
saved for exchange purposes in 1905.

POLICY OF ADMINISTRATION.

Through its entire history under the management of the
Board the Garden has been treated as a public institution
inthe fullest and most liberal meaning of the term. None
of its Trustees receive any form of remuneration for their
services. It is not managed as a source of revenue. Such
of its publications as are sold are charged for at cost. Its
employees, who are always required and ready to render
any proper service in their power, are forbidden to accept
fees. The office staff devote a large part of their time to
examining into and reporting on questions put in connec-
tion with a large correspondence or asked by visitors.

opt ee. Pe cen hee

Wa cece

Pa ed Oe Pay here ars
ED et Aaah ptt oe PR PO ES

ae

5, ‘

ee che z oh Tie gig ei Rca Lg) Sect Ons : Pe

PARCEL RAT GI Ci Toy ee Apr eee Bate eae AS pk tit.
3 SO aE a Ee Do gO Ps

nes gi oe ey pe Ng me ee ee Wy res

byes

ue
PEs

6)

te

aes
es

ORT i en EN EE RE ee he me AO fe Te ay ne
2 seat % ‘ ? “a ee cw? ei é i
a ? \ a \ *

7

24 MISSOURI BOTANICAL GARDEN.

The library and herbarium are opened to any one capable
of using them, and books and specimens are freely loaned
to responsible students thus enabling an increasing number
of persons to make scientific use of the Garden facilities
each year without incurring the expense of a visit to St.
Louis. Above all, restrictions on visitors are limited to
the few simple rules necessary to preserve the property
and insure its pleasant and profitable use by all.

FINANCIAL REPORT FOR 1905.

We submit now for your consideration the financial re-
sults for the year ending December 31st, 1905.

Our rental receipts show an increase of only $521.34
over the previous year, a few vacancies in the early part of
the year interfering with an anticipated increase; but as
all properties. are now occupied we may hope for an increase
of at least $5,000.00 over last year.

We have disposed of 220 feet on Flora Boulevard for a
total sum of $17,510.00.

The Board deemed it wise to dispose of a piece of prop-
erty at the northeast corner of Seventh and Chestnut
Streets fronting 21 feet on Chestnut, for the sum of
$100,000.00, it having been under lease for 89 years at
$3,700.00 per year; we hope to invest the proceeds in
real estate yielding a greater income. , '

We have purchased a piece of property 70X70 feet on
the north side of Poplar Street between 12th and 13th
Streets, adjoining property already belonging to the Board
and improved with five small houses renting for $1,200.00
per year, for the sum of $10,000.00, for the purpose of
investment.

We have again been called upon by the City for heavy
special taxes for streets and sewers, amounting in all to
$47,151.47, distributed as follows:

REPORT OF THE OFFICERS OF THE BOARD. 25
_

Rock Spring District Sewer No.14 ....... +. § 7,309 24
Manchester Road Joint DistrictSewer . . . ». «. «© « 33,899 25
Rock Spring District Sewer No. 13 . . . »« « « » « 1,979 55
Alley’ In Clty Block Wo. 4047) 0. we oe ee 1,164 00

$ 44,352 04
Sundry streets reconstructed . . . . » « » we e 2,799 43

$ 47,151 47

These improvements were anticipated in our last report.
Other improvements of a similar character will necessitate
the following expenditures during the year 1906, and,
if paid in cash, a portion of the cost must be met by sales
of residence property: — |

Old Manchester Road Sit whe wiles fauce- 0. 0° 2 cp eee
Shaw Avenue a dae eo nec Re ae ee eae ee hne ral tie 16,000 00
Molen AYORMG 6 wet he me mae: le oy 8 4,000 00
RE BORG s+ 8) 0 ge ae ee ee 14,000 00
Vandeventer Avenue . «© 6 + + + 6 ee ee es 4,000 00

$ 58,000 00

The annual bequests provided for in Mr. Shaw’s will
have been carried out at a cost of $1,884.11, with the
exception of the Trustees’ annual banquet which was not
held.

No permanent improvements have been made at the
Garden, but a large number of additions have been made
to the herbarium, and much work has been done on the
library catalogue.

_ The following amounts have been credited to the Stock

Account: —

Library . . . . . e ° °. . . e . e . . . . . . $4,332 64
Herbarium . - - - + 6 6 + + we @ ee ew ee oe 5,805 25
RECEIPTS.

Bontsin ... ec + 0, + 0 0 85m eee en
Interest and dividends . . . . + + « « « 1,828 77
Garden pasturage above expenses, etc. . . - 403 04
Garden hand-book sales . . + + + « « « 198 25
Publication sales . . «. « + © » « #*¢ « 4 08

Total income collections .. . .j $123,177 08

ON MeL Ea ee Se NaN RR Mh, ee ee Pee MS SS ema

4 26 MISSOURI BOTANICAL GARDEN.
Ce
“iy Brought forward . . . . « « . $123,177 08
. Sales of realestate . . . . . . . « « «© $100,000 00
. sig under decree of Court . . 17,510 00
Be! Insurance for loss to buildings . . ... . 1,759 87
- Le “t BE ec ee 20 00
a Shaw School of Botany, rent ...... 2,850 00
: Bonds, stocks and certificates . . . . . . 40,000 00 162,189 87
* Total receipts .. . tee pte $285,316 95
Bs Cash on hand December 31, 1904 eee 1,638 28
% $286,955 23
DISBURSEMENTS.
Garden Account,
a= Tabor pay-rolt.. . ss.» . w+ » « QI 49880
: Students’ pay-roll ey ee ae 1,474 80
ee I eg ee a 1,683 00
ee ae a 1,839 53
: ob ee ee eo Ee a ioe bs 234 50
. mepAlIp and BUppNeS. 3. ow 6 ee sl 2,638 25
: Plants and seeds , 1,046 71 $28,205 68
Herbarium Account,
PRIBMIGR ae a a es 1,688 51
wuel. . os eer ee eee 128 54
Current expenditure oe ee ee ee eee 2,599 12 4,416 17
Library Account,
oe ee ee ee ee ee 2,893 92
BMG]. ¢ 6 0 pees gts? er besa e cots 146 68
cc; Current expeniditinne Cae ee 1,966 89 —-5,007 49

Office Account,
| ee ee eee 4,681 06

Fuel tee te Gh one eel a os 116 47

Current expenditure . . . . »« « « » « 943 52 5,741 05
Research Account,

Saigrich «.7 <4 Coma Seer Cs 774 96

Current experdttnre 463 35 1,238 31

Scholarship Account,

-
°
.
.
.

REECTCHION, ns ey ee 723 60
Gare-OF Podge 2s Gee 240 00
Poel: fo fe os) heey ee) 105 82
Current pS eae eee ee ae ee ee 101 15 1,170 57
‘Total maintenhnce . 4. 2. 6 sm 8 6 is $45,779 27
Garden Improvement,
if North American synopsis, plans. . .. . 1,479 60

Total amount expended on Garden , $47,258 87

REPORT OF THE OFFICERS OF THE BOARD. 27

Brought forward « «+ + + © + + $ 47,258 87
Publication Account,
Sixteenth annual volume ...-+ ++ + $ 2,000 00 ;
First “c he ewostns’. 3 oa 254 85 2,254 85
Property Expenses,
State, school, city and sprinkling tax . . . 86,397 57
Streets, sidewalks andsewers. . . + + + 47,151 47
Insurance Be i Gg te he ya e 7,501 56
OME oe en se ee a) 2, ES 8,625 O4
Sroyerientl ee + tt eS 1,456 06 101,131 70
Office Expenses, :
meg. ie ¢ ee ee ee 4,200 00
Ofsice rent <5 sia een 8 ee 900 00
Printing, advertising, telephone, etc. . . + 777 25 5,877 25
Bequests,
Annual Flower Sermon * .. + + + © * 200 00
Annual Flower Show . +--+ ++ «+ « « 423 00
Gardeners’ Annual Banquet ... + + + 405 30
Washington University, School of Botany . 855 81 1,884 11
Sundries,
Realestate ..- - . Wa ee ee ee ee
Bonds, stocks and certificates Cg Stee ee oe
Legalexpenses. . + + + San a 1,816 90
Repairs to buildings damaged py fire ce 1,759 87 }
Shaw School of Botany, rent, etc. . . + + 2,200 00
Commissions «© «oor 8 eee te eee 477 00 126,253 77
Total disbursements . . ATT $284,660 55
Cash balance December alist, 1905 mien 2,294 68
$286,955 23
, —<———
Respectfully submitted,
R. J. LACKLAND, President.
Attest:

A. D. CuNNINGHAM, Secretary.

SEVENTEENTH ANNUAL REPORT OF THE DIRECTOR.
SUBMITTED TO THE TRUSTEES JAN. 10, 1906.

To the Board of Trustees of the Missouri Botanical Garden:

The following report for the year 1905, on the Missouri
Botanical Garden and the School of Botany connected
therewith, is respectfully submitted in compliance with the
rules of the Board.

¢

GARDENING,

Decorative gardening, from the nature and limits of
the formal grounds devoted to this purpose, presents much
the same problems year after year, and they are neces-
sarily similarly met. Last year about 37,500 plants were
used in ornamental bedding, an increase of some 2,000 as
compared with the preceding year. The early planting of
the parterre with bulbs, begun in 1904, has been repeated,
and with equal success. Through the summer the same

tract was devoted to a collection of coleus in masses rep- |

resentative of choice varieties. These new features, and
the bulb and foliage bedding on either side of the main
entrance-way, met with general commendation.

Among the special greenhouse collections, the succu-
lents, palms, cycads, and orchids have attracted particu-
lar attention, and each of these has received material en-
largement. Out of doors, massed economic plants from
the tropics, asters, pinks and dahlias, were satisfactorily
grown in more than usual number and variety. For the
fortnight beginning with November 13, a notable feature
of the Garden was a collection of about 2,000 chrysan-
themum plants, representing 211 varieties, well grown as
standards, for massed bloom, or to single heads. These
were displayed under canvas, and many flattering com-

(28)

i
4
i
4
i
4

D WALK.

CLUDE

A SE

ar
Ped

He aaa eee

me dre Me
+ y ae ‘
a Le edy

SEVENTEENTH ANNUAL REPORT OF THE DIRECTOR. 29

ments have been heard on the exhibition, which was
viewed by about 25,000 people, — one-fourth of the visi-
tors for the entire year.

Accessions number 314, comprising 17,620 plants
or packets of seeds. Of these, 680 plants, valued at
$288.45, and 360 seed packets, valued at $22.65, were col-
lected, and 26,845 plants, valued at $1,879.15, were prop-
agated, by Garden employees; 7,745, representing 199 of
the 314 entries and valued at $645.55, were presented or
received in exchange for material or for Garden publi-
cations; and 9,875, representing 67 entries, were pur-
chased, the Secretary’s books showing an expenditure of
$1,046.71 for such purchases, including transportation,
duty or other charges. ;

The records show that 2,345 species or varieties were
added to the collection of living plants, while 576 were
lost or discarded, leaving a net gain for the year of 1,769,
and bringing the total of cultivated species up to 15,976, in
contrast with the 14,207 noted for the preceding year.*

In January the Gardenissued its first ‘exchange seed list,”’
which included 813 species or varieties, of which about 18 %
were collected from wild plants of this vicinity, —the
others being from Garden plants. The convenience of this
list for correspondents was indicated by the receipt of

greatly increased requests for seeds, so that the distribu- -

tion of exchange material has been far greater than usual ;
the total during the year amounting to 4,173 packets of
seeds, valued at $417.30, and 1,320 plants, valued at
$77.80. As in earlier years, surplus bedding plants,
together with many of those removed from the grounds at
the end of the season, including duplicates of ferns,
begonias, etc., not needed for the houses, have been given
to charities and schools. The latter distribution, of some
800 plants, was effected largely through the effort of Miss

* Rept. Mo. Bot. Gard. 163 14.

?

OS £ a ite SRT ie Eth awe Ins eee eT a —S ee a i er ae, ee ee es ee
cp hey ‘ < Fi | a

30 MISSOURI BOTANICAL GARDEN.

Mary C. McCulloch, Supervisor of Kindergartens in the
public schools of the city. About 300 surplus plants have
also been given to High Schools and Normal Schools, to be
used for purposes of instruction.

As in the preceding year, the summer of 1905 was
cool. The average mean daily temperature for the en-
tire year was 1.2° F. below the normal. The months
of January and February were much colder than usual,
necessitating a considerable increase in the expense of
maintaining the plant houses. The entire month of
March, on the other hand, was unusually warm, so that

DIAGRAM A.

proo° 8 saa

P75: . satin ~

o
. 1905 w >
z & & & = Zz = 4 & 5 fo) a
me ee eS eS ee ee
at ! |

MEAN TEMPERATURES.

operations usually performed in April were found to be
necessary in March, thus increasing the length of the
open season by about a month. The precipitation for the
year has amounted to 38.54 inches, in contrast with the
average of 37.14 inches given by the sheets issued by the
local Weather Bureau office, — from which the accompany-
ing diagrams A and B have been compiled. In the early
spring, in June and in all the cold months the precipitation
was considerably below the average; but in May, July and
August the average was a little exceeded and the Septem-
ber and October rainfall was very high. Through most of
the growing season it was rather uniformly distributed,

SEVENTEENTH ANNUAL REPORT OF THE DIRECTOR. 31

DIAGRAM B.

r—5-INs /
ee ASS . / \
pa bse \ Zee

UL

JUNE
AUG
r-SEPT
r-—-OCT=
P-NOV
—DEC

-—MAR;=
APR
—MAY

PRECIPITATION.

favoring the growth of most bedding plants but also
greatly increasing the labor necessary to keep the grounds
free from weeds. For several reasons, therefore, the cost
of gardening this year has been unusually high.

VISITORS.

The total number of visitors for the year reached
100,830, of whom 12,355 were recorded for the open
Sunday afternoon in June, and 15,208 for the open
Sunday afternoon in September, the remaining 73,267
being week-day visitors. Diagram C shows the distribution
of the latter by months. The contrasted dotted line, —
representing the average for the period covered by accu-
rate records (beginning with midsummer in 1898) with
exception of the aberrant years 1902* and 1904,f—
seems to correspond closely with the normal out-door

* Rept. Mo. Bot. Gard. 14317.
+ Rept. Mo. Bot. Gard. 16317.

|
{
|

32 MISSOURI BOTANICAL GARDEN.

DIAGRAM C,

}-25000

20000 / |
r-15000 / \
10000 2

Fe Rimes MO ca Da oe ; ‘ \

WAN. FEB. MAR. APR. MAY. JUNE. JULY. AUG. SEPT. OCT. NOV. D&G

VISITORS ON WEEK-DAYS.

activity of our citizens. The great increase in the
November visitors this season, due primarily to the chry-
santhemum exhibit already mentioned, is directly attribut-
able to the fact that the officers of the street-car company
placarded the cars in such a way as to attract public atten-
tion to the Garden. Special recognition is due to the
company for this effort, which resulted in mutual ad-
vantage. The relation of Sunday to week-day visitors,
who this year reached 27% of the total, is graphically
shown on diagram D.

@ DIAGRAM D.

60000
ee Se oe ee ae a as
3—3 S—B—B-—B-60-Bo0o 3 —— 8 —$ —8—S Ff S-$
c - - al am oan a - vd = = =- - -
me ae 14 Sy ]
bate 0000 oe %

A & 3 WY 9 i
<a y
en a 20000 et20% a

aff i / |

SUNDAY VISITORS,

SEVENTEENTH ANNUAL REPORT OF THE DIRECTOR. 33

Among the visitors for the year, were noted 431 children
brought from a neighboring city and 1,846 of our own
school children, accompanied by teachers and usually
guided by Garden employees. Many teachers also have
visited the Garden for material for class use. This in-
creasing use of the collections by teachers and pupils is
particularly gratifying; and, in view of the dearth of
nature-study material in a large city, its much larger
growth is to be looked for.

Of the little Handbook of the Garden, 795 copies have

been sold.
THE HERBARIUM.

The past year has been marked by unusual increase of the
herbarium. Early in the-season the herbarium of the late
Henry Eggert was secured by purchase. This collection
is especially rich in representatives of our local flora and
of the flora of the Southwest. Approximately half-of the
specimens are already mounted and incorporated. It has
also been found possible to purchase the collection, chiefly
fungi, of Professor S. M. Tracy, about two-thirds of which
have been mounted. Other especially large purchases are
the general herbarium of Professor Bruce Fink, — some
6,000 specimens; about 5,000 plants of the Spanish
peninsula collected by Dr. Michel Gandoger; 1,249 Texas
plants collected by Mr. Julien Reverchon; about 1,500
Californian plants from Mr. Leroy Abrams; about 1,200
plants of the same region from Mr. A. A. Heller and a
like number from Georgia collected by Mr. R. M. Harper;
approximately 1,000 Californian plants from Mr. A. D. E.
Elmer; a like number pertaining to the upper Rocky
Mountain region, from Professor Aven Nelson; about the
same number of Mexican plants from Dr. Edward Palmer;
and a similar number of African plants from Mr. G. Zenker.
In addition to these, the usual current collections offered
for sale have been purchased and many and valuable

exchange additions have been made to the herbarium,
8

34 MISSOURI] BOTANICAL GARDEN.

Through the courtesy of the Directors of several foreign
botanical gardens —notably those at Kew, Berlin and
Palermo —and Sir Thomas Hanbury, of Ventimiglia,
whose fine private garden was freely opened for the pur-
pose, I have been enabled personally to enrich our
herbarium by a valuable collection of specimens and pho-
tographs— including those of type plants — representative
of the genus Agave, to the field study and collection of
which I was also able to devote two months early in the
year.

The incorporated additions for 1905 amount to 34,535
specimens, of which 2,783, valued (unmounted) at $139.15,
were presented or received by way of exchange; 1,878,
valued at $93.90, were collected by employees of the
Garden; and 29,874 were bought, the expenditure for
specimens and material amounting to $2,599.12.

The herbarium, so far as now mounted, consists of : —

The Engelmann Herbarium (all groups) . . . . 97,859 specimens.

The General Herbarium : —
Higher plants.
The J. J. Bernhardi Herbarium 61,338
The Henry Eggert Herbarium* 18,894
: The J. H. Redfield Herbarium 16,447
The Sturtevant and Smith Her-

DOIG eT ty ee 7,446
The Gustav Jermy Herbarium . 4,789
The Chapman Herbarium* . . 3,536
The Sadie F. Price Herbarium . 2,912
Other specimens .. . . . 264,158 379,520 &
Thallophytes. e

The J. J. Bernhardi Herbarium* 610
The Gustav Jermy Herbarium . 1,659
The S. M. Tracy Herbarium* . 4,312
The Wm. Trelease Herbarium . 11,000

Other specimens. . . . . . 28,885 46,466 ‘F
Makingatotalof... . 523,845 Ug
Valuediat. . 2.96 «a. $78,576 75t

* So far as yet incorporated.
+ This valuation, at the rate of $15.00 per hundred mounded sheets.

Ka
oe

ee ee eee ne ee Ce Son
AN SEAR INUIT AW usa GARDEN.

SEVENTEENTH ANNUAL REPORT OF THE DIRECTOR. 35

Details concerning the larger exsiccatae of the herbarium
are given in my last report,* and need not be repeated
here, as only small current additions are to be noted.

In addition to the herbarium and shelved or incorpo-
rated exsiccatae, the following material should be noted as
of the same general character : —

-

OR ioe NT eee Mii

sath

Wood specimens of various sizes . . 800 valued at $ 55 00
Wood veneers. '
Spurr’s cabinet of veneers from the

Jesup collection . . . . . . . 6541 veneers,
Hough, American Woods, 10 fascicles 750 *
Michel, Feine Holzschnitte, 2 285 = “6
Nordlinger, Querschnitte, 11 “. Binge =.“
Sykyta;DasHolz, ...1 ied
Tokyo Imperial Museum,
Useful Woods, © 3... 3..% 255 af
Together . .. .25 “ 2,956 «¢ valued at $175 00
Microscope slides.
Penhallow, N. A. Coniferae. . . . 268 slides,
Other alias. 3... ockdieee sae. *f
Together... + ws even eta we gee «¢ valued at $410 00
Total VAIURMOR <sth aos aes + tf ROO

THE LIBRARY.

As in 1904, the absence of a special appropriation for
the purchase of books has prevented the addition to the
library of large sets, but no difficulty has been experienced
in providing books needed for work in hand; considerable
further progress has been made in binding serials and the
like, and the exchange relations of the Garden have added
the customary large representation of the current publica-
tions of learned bodies interested in botany. The librarian
has also found time to increase largely the number of in-
corporated authors’ separates, by cataloguing and shelving
material already on hand in addition to current accessions.

* Rept. Mo. Bot. Gard. 16:21.

36 MISSOURI BOTANICAL GARDEN.

The additions for the year number 829 books, valued at
$1,280.50, and 3,859 pamphlets, valued at $523.15, pre-
sented or received in return for Garden publications; and
475 books and 219 pamphlets bought, the expenditure for
purchases and binding amounting to $1,577.14.

The card index has been enlarged by the addition of
97,676 cards, of which 79,039 were written by employees ;
16,146 were purchased; and 2,491 were presented. Among
the additions are many cards referring to floral ecology
and to seedlings, the former subject being now represented
by some 33,000, and the latter by about 28,000 cards,
while the references to plant illustrations number about
190,000.

The distinct serial publications now received at the
library number 1,237, of which 93 are bought and 1,144,
issued by 859 institutions or publishers, are received in
exchange for the Reports of the Garden. The total here
reported is 13 smaller than that reported for 1904, as the
result of a careful revision to date, —a considerable num-
ber of publications having been discontinued, although a
number of additions have been made to the exchange list.*

As now constituted, the library contains :—

Pamphlets... . . + - 29,188
BOORS. i a we Be a FR
51,197 valuedat . . . . $78,533 99
Manuscript volumes .. . 77 ee oS eo 910 00
‘i. ee i i eo eree
Index cards .. ..» » » 480,486 Se a eon 4,804 36
Total valuation of library .. . . . . » $84,248 35

During the year just closed, a special provision by the
Board has enabled the librarian to begin the preparation
of alist of biological serials contained in the libraries of
St. Louis. Satisfactory progress has been made on the
manuscript of this list, which is expected to be completed

* Rept. Mo. Bot. Gard. 16 3 23.

SEVENTEENTH ANNUAL REPORT OF THE DIRECTOR. 37

in the early part of the new year. It is believed that the
publication of this list, when effected, will prove of value
to the increasing number of persons who are carrying on
scientific work in and near the city.

By direction of the Board, a third issue of the first
Report of the Garden was made, early in the year.

RESEARCH AND THE USE OF FACILITIES.

Such time as could be spared from their immediate
duties has been given to investigation by a number of
Garden employees, by whom scientific papers have been
published or are in course of preparation. By direction of
the Board I represented the Garden in June at an Inter-
national Botanical Congress held in Vienna, to which I had
the honor of being further accredited as the representa-
tive of the National Academy of Sciences. In connection
with this European trip a considerable amount of research
work was done,—to which reference has already been
made.

As in earlier years, a number of botanists have visited
the Garden for the purpose of using the library and collec-
tions. It is a matter of regret that more persons do not
avail themselves of its facilities in this manner, — particu-
larly since the presence of investigators is most stimulating
to the Garden staff. In compensation for this, however, —
and indeed, partly accountable for it, —transcripts und
photographs have been made in the library, notes taken in
the herbarium, and books and specimens loaned in consider-
able numbers to further the work of botanists who were
unable to bring their studies here; the passage of each
year marks an increase in this use of the equipment of the

institution.

THE HENRY SHAW SCHOOL OF BOTANY.

The teaching staff of the School of Botany has not
been changed since my last report. The stated University

38 MISSOURI BOTANICAL GARDEN.

courses in which botany figures, each of which is based on
three exercises a week unle-s otherwise stated, have under-
gone certain desirable changes,* and they are now offered
as follows: —

1-2. Biology: A lecture and laboratory course devoted to the consid-
eration of life processes as exemplified in the animal and plant kingdoms.
Two lectures and four hours of laboratory work each week throughout the
year. Open to all classes. aA

1-2. Botany: Elementary Morphology and Organography. Labora-
tory work, with supplementary lectures and quizzes, dealing with the
form and structure of plants, with special reference to their life pro-
cesses, including ecological adaptations, and to Systematic Botany.

3-4. Cytological Technique. Laboratory work in the cultivation and
examination of bacteria and the methods of histological investigation,
combined with a study of the principal types of plant anatomy —
specially adapted to the needs of students intending to study medicine,

5-6. Plant Physiology, including Ecology. Laboratory work, with
supplementary lectures.

7-8. Systematic Botany. Laboratory and field work, with supple-

mentary lectures, dealing with the comparative structure and the classi-
fication of plants.

$-10. Plant Pathology and Applied Mycology. Laboratory work.

In addition to these electives, special courses are
arranged for graduate or other advanced students, as need
is found for them.

I am pleased to report a still further increase in the
number of students taking work in the School of Botany,
the enrollment for the first term of 1905-6 being as fol-
lows: —** Biology 1,’’ 11; ‘* Botany 1,’’ 13; « Botany
3,”’ 10; ** Botany 5,’’ 7,—a total of 41 students, of
whom the eleven first noted give about equal time to botany
and zoology, and the others are taking one full botanical
course each.

At the 1905 Commencement of Washington University
the degree of Master was conferred on Miss Laura L.
Eames, Miss Winifred Ashby and Mr. Dean H. Rose, for
work in botany. There are now enrolled in the University

* See Rept. Mo. Bot. Gard. 14:24. 15:32.

SEVENTEENTH ANNUAL REPORT OF THE DIRECTOR. 39

one candidate for the Master’s degree, and three for the
Doctor’s degree, with botany as a major study.

GARDEN PUPILS.

The course in gardening and the teaching in it have re-
mained unchanged through the past year.

On the result of duly announced competitive examina-
tion, the free scholarships reported a year ago* were
awarded in March to Arno Nehrling of Gotha, Florida,
and Eugene Smyth, of Topeka, Kansas. The latter ten-
dered his resignation in July, to undertake preparation
for college work, and the scholarship so vacated was then
assigned to H. L. Ochs, of St. Louis, who had passed a
satisfactory examination in the spring though he did not
then win a scholarship. In November, Miss Eda A.
Sutermeister, who for some four years has been engaged
in practical landscape work which was held to be an ade-
quate equivalent for a portion of the required manual work
not done with her prior resident study, was admitted to
examination and given the customary Garden certificate.

THE GARDEN STAFF.

No changes in the office force are to be reported, though
the number of temporary assistants has been greater than
in other recent years because of library work on the list of
serials and the indexing already referred to, and of revisional
work in the herbarium.

SPECIAL TESTAMENTARY PROVISIONS.
Three special events provided for by the will of Mr.
Shaw have taken place in 1905, as follows: —
The annual flower sermon was preached in Christ Church

‘Cathedral, St. Louis, on the morning of May 14, by the

Right Reverend Lewis Burton, Bishop of Lexington,
Kentucky.

* Rept. Mo. Bot. Gard. 16: °7.

eS ae ee ee ee ey ae ere Pee

40 MISSOURI BOTANICAL GARDEN.

The Gardeners’ banquet was given at the Mercantile
Club, St. Louis, on the evening of November 8. The
Director of the Garden. presided, under the provisions of
Mr. Shaw’s will. About ninety guests were present,
among them the gardeners and office staff of the institution,
representatives of the Board of Trustees, the staff of the
Mississippi Valley Laboratory of the United States Depart-
ment of Agriculture which has an office at the Garden, and
a large representation of the floricultural, nursery and
market gardening .interests of St. Louis and vicinity, as
well as a number of gentlemen interested in the growth of
plants for public utility. Speeches were made by Mr.
George E. McClure, of the Missouri Botanical Garden, who
responded to the toast «The Gardener;’’ Mr. J. F. Am-
mann, of Edwardsville, Illinois, President of the St. Louis
Florists’ Club, who responded to the toast «« The Florist ;”’
Mr. E. A. Riehl, of Alton, Illinois, who responded to the

_ toast ‘‘ The Nurseryman;’’ Mr. L. A. Goodman, of Kan-

sas City, Missouri, President of the American Pomological
Society, who responded to the toast «* The Pomologist ;’’
Mr. L. C. Davis, of Old Orchard, Missouri, whose toast
was ‘* The Market Gardener ;’’ Mr. Gerard Swope, of St.
Louis, who spoke on «+ Public Playgrounds;’’ Mr. W. J.
Stevens, of St. Louis, who spoke on ‘*School and Vacant
Lot Gardens;’’ Mr. H. ©. Irish, of the Missouri Botanical
Garden, Secretary of the National Council of Horticulture,
who spoke on *‘ Federated Interests ;’? and Mr. Henry T.
Kent, President of the Civic Improvement League of
St. Louis, who spoke on « General Civic Work.”’

The sum of $500.00 authorized for annual use in the
form of premiums for a flower-show was placed at the
disposal of the St. Louis Florists’ Club. Their awards
for authorized premiums at their exhibition, held from
November 8 to 11 inclusive, were duly approved and paid.

Very respectfully,

WILLIAM TRELEASE,
Director.

SCIENTIFIC PAPERS.

STUDIES ON THE LIGNIN AND CELLULOSE OF WOOD.*

BY PERLEY SPAULDING.
INTRODUCTION.

In a recent paper Potter f gives the results of some in-
vestigations upon the cellulose and lignin of the xylem of
tree stems. By microchemical methods he found that cel-
lulose occurs as a distinct lining layer in the walls of the
wood fibers of perfectly healthy trees grown in various
parts of England. Trees of Quercus, Fagus, Aesculus,
Salix, Ulmus, Alnus, Betula, and Fraxinus were found
by him to have cellulose occurring in this manner through
all parts of the stem; that is, in both heart and sap wood
at varying distances from the bark. He tested the action
of boiling water upon the lignin of the cell walls of wood
by cutting microscopic sections and placing them in water
in a ‘‘ boiling tube’’ and boiling for about two hours per
day on consecutive days, the sections remaining in the
water all of the intervening time. The lignin began to
leave the walls in four to six hours of boiling in the sec-
tions of Fraxinus and similar results were obtained with
other woods, twelve hours being the longest period of boil-
ing which is definitely mentioned by him. Simple soaking
of small fragments of wood in told water was then tried
and it was found that a substance was obtained which had

* A thesis presented to the Faculty of Washington University, in
candidacy for the degree of Doctor of Philosophy, April, 1906. — Pub-
lished by permission of the Secretary of Agriculture.

+ Potter. Annals of Botany. 183 121-140. (1904).

(41)

42 MISSOURI BOTANICAL GARDEN.

been extracted from the wood and which gave lignin re-
actions with phloroglucin and thallin sulphate, thus indi-
cating a partial delignification of the treated wood. He
was led to conclude that, ‘‘ It is found that a gelatinous
thickening layer which reacts at once to the various color
tests for cellulose occurs very commonly, though very
irregularly, in the fiber walls of the xylem as a normal con-
dition in a great number of perfectly healthy trees, in all
localities and situations. It may have a very partial dis-
tribution, or may occur very generally and conspicuously
through the stem, and may be present only in parts of the
same annual ring. Sometimes this innermost layer is
represented only by a thin lining, at others by avery broad
band which appears swollen and occupies a large part of
the lumen.”’

Also, ** The presence of this unlignified layer in the
wood fibers probably represents a stage of arrested devel-
opment. Its general prevalence having been overlooked,
the conclusion is’ inevitable that the occurence of cellu-
lose which has been attributed to the action of fungi must
to some extent be ascribed to conditions already present
and the effect of any method of sterilization must also be
taken into account. The delignification cannot be attrib-
uted to an enzyme secreted by fungi.”’

AIMS OF THE PRESENT INVESTIGATIONS.

This paper with its rather striking results and conclu-
sions led to the investigations here detailed. Tests were
made: first,to determine the prevalence of cellulose in the
stems of the trees of America; second, to confirm or dis-
prove the results of Potter’s work in testing the effect of
boiling water upon the lignin of wood; third, to test the
relative solubility of the lignin of the spring and summer
wood of the annual ring of growth; fourth, to come to
some understanding of the conditions existing in wood

STUDIES ON THE LIGNIN AND CELLULOSE OF WOOD, 43 * pe

when used in testing the action of fungi upon its com-
ponent parts. a

MATERIALS USED.

The present experiments were performed with freshly s
cut wood taken from the stems of trees which, with but 5; ‘s
few exceptions, were above six inches in diameter. Two
sets of experiments were carried on: the first, of a prelimi- 8

~ nary nature, included but a few species of wood; while in Saad
3 the second, woods of about forty of the more common and
a important timber trees were tested. Still later a number a

of other species were examined for the presence of cellulose uy

_ without performing the boiling experiments. es.

The species of wood used in the preliminary experiments laa

were;* Populus tremuloidus from New Mexico ; Pinus pal- ae

ustris from Texas; Sassafras Sassafras, Taxodium dis- a
tichum and Nyssa aquatica from Arkansas; Acer Sacchar-
um, Quercus rubra, Hicoria ovata and Picea rubens from
Vermont; Fraxinus Americana and Liriodendron Tulip-
ifera from New York; and Larix Americana from

Michigan. : Re:
Further and more thorough tests were made later with a a
much larger number of woods, including most of the im- a
ae portant timber species of the country. ‘As will be per- vesleiiiaaia
i ceived, those species which were utilized in the first

experiment were also included in the later tests. The
woods used in these last experiments are: Sassafras ee
Sassafras, Taxodium distichum, Nyssa aquatica and
Liquidambar Styracifiua from Arkansas ; Sequoia sem- ‘e
pervirens, Sequoia Wellingtonia, Pinus Lambertiana, s

tit
Pinus ponderosa, Pinus flexilis and Pseudotsuga mucronata 7
from California; Catalpa speciosa from Kansas; Tilia
Americana, Populus balsamifera, Larix Americana, a
_ * In naming the trees Professor Sargent’s recently published Mandal s
of the Trees of North America is followed.
y a a
f * i
¢ ras “3
f epi

44 MISSOURI BOTANICAL GARDEN.

Abies balsamea, Pinus Strobus, Thuja occidentalis, Fraxi-
nus nigra and Fagus Americana from Michigan; Pinus
Taeda and Quercus rubra from Mississippi; Acer saccha-
vinum, Cercis Canadensis, Pinus echinata, Cornus jlor-
ida, Carpinus Caroliniana and Platanus occidentalis
from Missouri; -Juglans cinerea, Juglans nigra, Quercus
Prinus, Ulmus Americana, Pinus divaricata and Hicoria
ovata from New Jersey; Tsuga Canadensis, Picea
rubens and Liriodendron Tulipifera from New York;
Tsuga heterophylla from Oregon; Juniperus Virginiana
and Nyssa aquatica from Tennessee; Pinus palustris,
Ilex opaca, Magnolia glauca and Quercus Phellos from
Texas; Acer Pennsylvanicum, Acer rubrum, Prunus se-
rotina, Betula lutea, Acer Saccharum, Ostrya Virginiana
and Salix sp. from Vermont; and Libocedrus decurrens
from Washington.

Attention should be given to the fact that much of the
material used by Potter in his work was taken from
branches and not from the main stems of the trees.
Whether there is any difference between the branches and
the stem with respect to the amount of unlignified cellulose
present is as yet not definitely known; but the possbility
of such a difference should be kept in mind.

CELLULOSE TESTS.

The examination for cellulose was at first performed
simply with the microscopic sections after they were cut
ready for use, but it was soon perceived that this method
would not give accurate results. The samples of wood
were then very carefully examined in all parts. If no
cellulose was then found another search was made to make
sure that it had not been overlooked at first. Thus the
specimens which had no cellulose were really subjected to
a much more rigid examination than were those which
proved to have it.

STUDIES ON THE LIGNIN AND CELLULOSE OF woop. 45

Cellulose is said to occur in two ways in the fibers of
wood in tree stems:* much more commonly as an inner,
lining layer sometimes completely filling the cell lumen;
and much more rarely as a layer situated between other
lignified ones. Personal examinations have shown it to
occur, with few exceptions, as an inner, supernumerary
layer, but it has also been found that in exceptional cases
nearly the whole secondary layer of the wall, which is
usually completely lignified, gives a cellulose reaction
although the reaction is more marked as we pass from the
middle lamella to the cell lumen.

In the first lot of samples used for the preliminary tests
it was found that cellulose occurred only in the wood of
Populus tremuloides from New Mexico and in the wood
of several trees of Larix Americana from the same local-
ity in Michigan. In both cases the cellulose occurred only
in some of the wood fibers and was present in the form of
a lining, supernumerary layer of the wall.

In the more extended examinations made later cellulose
was found in the wood of Populus balsamifera from the
same locality as were the above mentioned trees of Larix
Americana from Michigan. It may be added that a tree
of Populus tremuloides from Vermont also showed large
quantities of cellulose in the wood fibers. The occurrence
of cellulose so markedly in all of the specimens of Populus
wood available led to the suspicion that the species of the
genus Populus might be characterized by this phenomenon
wherever they might grow. Accordingly an effort has
been made to obtain, as far as possible, specimens of all
of the native species of this country. Of the eleven
species which are listed by Sargent f in his latest publica-
tion the following have been examined: Populus tremu-
loides, Populus balsamifera, Populus deltoidea, Populus
trichocarpa and Populus acuminata. Every specimen

* Sanio. Bot. Zeitung. 1868 ; 101-111. ¢ Sargent. 1. ¢.

46 MISSOURI BOTANICAL GARDEN.

examined has had cellulose very abundantly and in thick
layers such as are shown to occur in the wood of Populus
tremuloides in plate 1 figure 1, and of Populus balsamifera
in plate 2, figure 2. Consequently there seem to be
good indications that the genus Populus,is characterized
by the occurrence of cellulose in an extra layer.

A number of species of Salix from widely scattered
localities have also been examined. Salix alba was found
to have an inner lining layer of cellulose in some of the -
fibers of the wood. Salix fragilis had a very pronounced
layer in stems about four years old. Salix sp. from Cali-
fornia had cellulose in nearly all of the wood fibers.
Salix fluviatilis from South Dakota had a very pronounced
cellulose lining in many of the fibers. Salix viminalis has
been found to have a thin layer of cellulose in the fibers.

Beside the above species of Populus and Salix which
have been personally found to have unlignified cellulose
there are a number of cases which other workers have
noted. Sanio®* found it to occur inwood of Populus pyr-
amidalis. Sablon ¢ found it in woodof Salix sp.

The sections of Populus and Salix were found to have
the cellulose in the form of a thick and very distinct in-
ner layer of the fiber wall which seemed to be somewhat
loosely attached to the secondary lignified one. In some
cases it filled the entire cell cavity, but this was not com-_
monly true. The cellulose layer was quite sharply delim-
ited from the lignified one. Cellulose was never present
in the cells of the oldest wood of the annual rings, but
was always situated in the more open early wood. It
was found only in the wood fibers and only in those
fibers which do not join or abut upon the vessels and
tracheids.

In the Larix sections cellulose was present only in cer-
tain of the.annual rings and also only in the wood fibers.

* Sanio. l. c. + Sablon. Rev. gén. d. Bot. 16: 360-363. (1904).

Se et in wD Tee . SS OR ee ee ee ee ee
Le he Pete oe een Set NE AY vA Yaa” 5
Be | t eS) Poe ra Waren ties tam

STUDIES ON THE LIGNIN AND CELLULOSE OF woop. 47

It consisted of an inner, lining, layer which was not as
thick as was that of the Populus species.

Later and more protracted examination of the other
woods showed cellulose to be present in small quantities
in a single ring of the wood of Acer Pennsylvanicum. It
was also noted in wood of Acer saccharinum, Catalpa
speciosa, Nyssa aquatica, Pinus Taeda, Platanus occt-
dentalis and Ulmus Americana. It was present in all of
these only in small quantities.

BOILING TESTS.

The boiling tests were performed as follows: transverse
and longitudinal sections of the wood were cut, placed in
distilled water in sterile test-tubes, and boiled in an Arnold
sterilizer for two or three hours per day in a manner sim-
ilar to that commonly used in the discontinuous steriliza-
tion of culture media. Attention is here drawn to the fact
that the temperature reached in tbe Arnold sterilizer is

‘about 100 degrees Centigrade. The boiling was continued

until a total of twenty hours had been attained. The tubes
were then placed in an autoclave and steamed at a pressure
of about sixteen pounds and a temperature of about 120
degrees Centigrade.. This steaming in the autoclave was
continued until forty hours of boiling had been accom-
plished.

After fifteen hours of boiling i in the Arnold sterilizer a
faint appearance of delignification was perceived in the
sections of Sassafras Sassafras. 'Twenty hours of boiling
made it very pronounced. See plate 2, figures 3 and 4.
After seven hours in the autoclave, or a total of twenty-
seven hours of boiling, the sections of Picea rubens also
showed a light blue reaction in the inner lignified layer of
the fiber walls when treated with chlor-iodide of zinc.
See plate 2, figure 1. With these two exceptions the
forty hours of boiling seemed to have no effect upon the

aha fee eS
i ha

.

48 MISSOURI BOTANICAL GARDEN.

lignin of the sections. Potter does not state at what tem-
perature his experiments were carried on but it is presum-
ably at about 100 degrees Centigrade. If his results were
applicable to woods in general the delignification in the
present experiment should have been apparent even sooner
than it actually was; and it should have taken place in all
or nearly all of the woods instead of but two out of the
twelve which were used. But we find none of the woods
were delignified in the time stated by him to be effective
with the woods tested by him. Apparently then the woods
used in these experiments are more thoroughly lignified or
else hold their lignin in combination more firmly than was
the case with the woods used by Potter.

The later boiling tests were performed wholly with the
autoclave at a temperature of about 120 degrees Centi-
grade. In general it may be said that many of the woods
gave evidence of delignification after fifteen hours of treat-
ment, and nearly all, after eighteen hours of steaming.
Sequoia Wellingtonia, Betula lutea and Juglans nigra
gave no results so far as could be detected, but it is barely
possible that these woods are dark colored enough so that
the reaction was not seen even if obtained.

SOLUBILITY OF LIGNIN IN SPRING AND SUMMER WOOD.

An experiment was next tried to test the relative solu-
bility of the lignin of the early and late wood of the annual
rings. For this purpose a wood was chosen which showed
a decided difference between the two parts of the ring.
Pinus palustris seemed to be very good, having a decided
difference in the thickness of the fiber walls of the spring
and summer wood, and also being entirely free of unlignified
cellulose. Microscopic sections were cut from fresh wood
and boiled in the manner above indicated. After a very
long period of boiling, a record of which was unfortunately
not kept, it was found that the spring wood seemed to be

STUDIES ON THE LIGNIN AND CELLULOSE OF woop. 49

wholly delignified, but the summer wood was still lignified,
and so far as could be noted showed very little effect from
the boiling. The various cellulose and lignin tests were
used and invariably indicated the same results. See plate
rE: figure 2. Thus it seems to be proved that the lignin of
the spring wood is held in combination less firmly than is
that of the summer wood. This is probably due to the
greater thickness of the walls in the summer wood, as the
difference in thickness between the two parts of the ring
was very marked as will be readily perceived on referring
to plate 1, figure 2. The results of this test seem to furnish
a clue as to the reason why certain of the wood rotting
fungi disintegrate the walls of the spring wood while the
summer wood of the same ring is apparently unaffected.

Later tests with the various woods enumerated above
have shown that what is true of the wood of Pinus palus-
tris is also true of many others. The degree of different-
iation between the two parts of the annual ring seems to
be the controlling factor in the relative solubility of the
lignin of those parts. In other words, those woods which
have a decided difference between the spring and summer
wood of the annual ring are almost sure to have the lignin
extracted from the former quicker than it is from the lat-
ter: and those woods which have but slight differences be-
tween the two parts of the ring are apt to have the lignin
extracted from all or nearly all of the ring at the same
time.

ENVIRONMENT AND LIGNIFICATION.

Potter’s results may be accepted as true for the woods
used by him but they cannot be taken as significant with
respect to woods grown under all conditions. The above
experiments have shown that there is a decided difference
between the results obtained by him and those of the
present tests. As is well known, there is a very great dif-
ference in the environment of trees grown under different

4

UTE eR ee

ee das, : ae

ce
are

ra.

ap

es

aber 2s

ok = fe

Sat aiee

on
oe
eon

ae ae
ius

Bing Dk grit Bs

»
eae

Fo Efe Se ae mS ae Sa eT

50 MISSOURI BOTANICAL GARDEN.

conditions and in different climates. All workers with
wood recognize the fact that its strength is a very variable

yenty and for that reason it is unsatisfactory for use in,

avy structures, since the amount of stress which it can
withstand is more variable than is that of steel and iron.
There is a decided difference between a certain species
grown on dry hills and the same species grown in moist
lowlands. Much more then must there be a difference
between wood of the same species when grown in essentially
different climates. May not, in the same way, the degree
of lignification of the walls be affected by external condi-
tions? Potter * says, ‘local conditions of soil and climate
seem in some cases, to retard the complete development of
the xylem, and thus render such trees constitutionally weak
and very liable to attack.’’ Von Schrenkt has also inti-
mated something of a similar nature in connection with
Fraxinus Americana when growing on the western border
of its range. He says, ‘In the present instance it would
seem, that there might be some relation between the greater
susceptibility on the part of the ash near its western limit
and its generally weaker development at thislimit. It will
be an interesting point to determine for instance whether
the rate with which branch wounds or stubs heal in Ohio
and Pennsylvania is greater than in Missouri and Kansas.
That the rate of growth is slower in the Western States we
know.’’ In the same way may not the firmness with which
the lignin is held in combination with the other constitu-
ents of the wall bealso affected by the external conditions?
This certainly does vary as is shown by the preceding
experiment upon the spring and summer wood.

ARRESTED DEVELOPMENT AND RESERVE STORAGE.

The occurrence of cellulose in the form of an inner,

* Potter, 2. ¢:
t Von Schrenk. U. 8. Dept. Agric. Bureau of Plant Seiensy: Bull.
$2: 11. (1908).

2 isd J
ee

¢

STUDIES ON THE LIGNIN AND CELLULOSE OF woop. 5l

supernumerary, lining layer of the fiber walls has been held
by several boganists to be due to arrested development of
the walls. It has seemed to the writer that there are
certain facts in connection with cellulose thus situated
which point directly against this theory. First, cellulose
as above mentioned forms an extra layer of the wall. In
other words the wall is completely formed and has, besides
the usual layers, an extra one consisting of cellulose.
Measurements of the fiber walls of Populus balsamifera
show a difference of five microns in favor of the walls
having the cellulose layer, over those which do not have ~
that layer. Second, the wall with the cellulose layer is
thicker than that without the cellulose, as is shown by the -
above measurements. It may also be stated that the dif-
ference in most of the woods was striking enough so that
the observer could easily perceive without measuring that
there was a decided difference between the two kinds of
walls. Third, those fibers which are situated near vessels
or tracheids do not usually have the cellulose layer. That
is, those fibers which commonly do not have starch depos-
ited within them are the ones which seem to be more apt to
have the cellulose. This might be connected with the con-
version of the starch into cellulose in such of the fibers as
do not have the starch. Sablon* has proved that in willow
wood the cellulose is plentiful when the starch has disap-
peared almost completely, and there thus seems to be a
direct conversion of the starch into the cellulose. _ Fourth,
the tendency is for the cellulose to be present in the spring
wood and not in the summer wood of the annual ring
where it would seem that arrested development would be
most apt to affect the formation of the wall. There has
been noted no indication of the final lignification of this
cellulose layer, although Sanio f has stated that he noted such

* Sablon. Revue générale de Botanique. 16: 362. (1904).
+ Sanio. 1. c.

eet at «Pet tant Man OR RE ee a ee ee Mer enr ye teae Meee ee Scape ee Mirae ra Ot bon Af cae Se nM ee = Baad, Bea heer a Pagan Oa
Reon Shag ee . lg a ‘At Ra ,, opt 84 pris eh ee Pe Ea Se i beads enti
4 4 am ¥ P Py ‘

52 MISSOURI BOTANICAL GARDEN.

indications in certain stems which were especially favorable
for the purpose. There certainly is still much doubt
whether all the stems having this cellulose layer do finally
become wholly lignified, and this seems especially true of
the Salicaceae.

The exceptional cases mentioned above where the cellu-
lose seemed to extend into the secondary layer (which is
usually lignified), are not considered in this connection.
Indeed they may very well be due to arrested development
and are here so considered.

DELIGNIFICATION BY FUNGI.

Potter states that ‘‘ the conclusion is inevitable that the
occurrence of cellulose which has been attributed to the
action of the fungi must to some extent be ascribed to
conditions already present,’’ implying that the cellulose
which is found as a supernumerary lining layer of the
wood fibers is the same cellulose which is found in woody
tissues which are badly rotted by certain of the wood rot-
ting fungi. He very evidently had in mind certain experi-
ments of Ward* and Biffenf in testing the action of
Stereum hirsutum and Bulgaria polymorpha upon wood
when grown in pure cultures upon it. In both of these
cases it happens that the delignification was apparently not
very extensive and affected but a portion of the wall. In
making the above statement Potter was at least partly
correct in applying it to such cases, where the delignifica-
tion was rather slight and affected but a portion of the wall.
The statement should have been limited to such cases, as
it is not applicable to those where the delignification has
taken place throughout the whole wall as is the case in the
last stages of disease caused by many of the wood rotting

fungi, for the cellulose of normal wood occurs in such a

® Ward. Phil. Trans. Royal Soc. 189. Series B. (1897).
+ Biffen. Annals of Botany. 15:119-133. (1901.)

STUDIES ON THE LIGNIN AND CELLULOSE OF woop. 53

manner that it can have little or no effect upon the final .
result where the delignification is extensive, as is shown
quite conclusively a little later in this paper.

The cellulose which is so often found in wood undergo-
ing the last stages of decay is the cellulose skeleton or
framework of the secondary, lignified layer which is uni-
versally present in old wood fiber walls, whether they have
the supernumerary cellulose layer or not. That this is so
is proved by a careful examination of the various stages of
rot caused by any one of the numerous wood rotting fungi
which delignify woody cell walls. The published work and
drawings of Robert Hartig * alone are sufficient to prove this,
as it has been found by the writer that cultures from rot-
ted wood in the middle of a decayed stem of a living tree,
if the exterior layers of wood are still alive, invariably
yield but a single fungus, thus proving that the attacking
fungus occurs in a pure state at least in most cases. If
there is any cellulose layer present in the wood before it is
attacked it is probably the first part of the cell wall to be
dissolved and it is fairly certain that it does not exist in the
last stages of decay.

Two instances of trees attacked by wood rotting fungi
which fully bear out the last statement have recently come
to the writer’s attention.

A tree of Carpinus Caroliniana was found badly de-
cayed by Polyporus gilvus,t which in this case seemed to
be a wound parasite. At the edge of the affected wood
was a narrow border of dark, infiltrated cells separating
the healthy from the diseased wood. In the healthy wood
just outside this infiltrated zone the wood fibers in many
cases had an inner, lining layer of cellulose. This layer
disappeared in the infiltrated zone while in the very badly

* Hartig. Wichtige Krankheiten der Waldbéume. 1-127. (1874);
Die Zersetzungserscheinungen des Holzes der Nadelholzbiume und der
Eiche. 1-151. (1878). :

+ Engler & Prantl’s Pflanzenfamilien is followed in naming the fungi. ©

54 MISSOURI BOTANICAL GARDEN.

rotted tissues the cell walls consisted only of cellulose, the
lignin appearing in greater and greater quantity as the
healthy wood was approached. This single instance alone
seems to prove that the cellulose layer of the healthy wood
does not have any connection with the cellulose left by the
action of the fungus upon the cell wall.

Still another instance was noted which seemed to con-
firm and make this point doubly certain. Sections of
trunks of trees of Populus tremuloides rotted by Fomes
igniarius were examined carefully for the occurrence of
cellulose. It was found that the wood fibers of the healthy
wood in most cases had an inner, supernumerary layer
which turned blue with chlor-iodide of zinc. The rotted
tissues were separated from the healthy ones by a narrow
infiltrated zone. Just inside this zone of infiltrated cells
the affected fibers were of a very uniform thickness and
the walls were decidedly thinner than those of the healthy
wood where the cellulose was present. In the rotted tis-
sues the cell cavities were also larger than in the unaffected
cells. In other words the cellulose layer is more or less
completely dissolved from the interior of the fibers, and
the secondary layer is delignified, while the cellulose skel-
eton is left until the last.

These two instances seem to show that there can be no
doubt regarding the solution of the supernumerary cellu-
lose layer during the early stages of decay caused by some
of the fungi. It is believed that this will prove to be true
of most if not all of them. Hartig very evidently under-
stood this and for this reason gave no further attention to
the occurrence of cellulose as an extra layer in the wood
fibers of oak wood than to mention its occurrence. *

Actual measurements made in the wood fibers of the
healthy and rotted wood of the same sections show beyond

* Hartig. Die Zersetzungserscheinungen des Holzes der Nadelholz-
baiume und der Kiche. 94, (1878).

>, STUDIES ON THE LIGNIN AND CELLULOSE OF WOOD. 55

any doubt that the thickness of the healthy wall is greater
by a decided difference than is that of the cellulose fiber
walls which are left in the badly rotted tissues. Such
measurements have shown a difference of about a third of
the total thickness of the original healthy wall between the
two series. It has been stated distinctly by practically all
workers with this group of fungi that delignification is a
very common phenomenon as a result of the action of the
mycelium of these fungi, which grows in the wood and dis-
integrates it to a greater or less extent as the case may be.

That cellulose is left by many of the fungi in the last
stages of decay of the wood in which they are growing is
but a very natural phenomenon to one who is familiar with
the structure of woody cell walls. Chemical analyses have
shown that cellulose is the basis of the structure of the wall
and makes up a large percentage of it by dry weight. The
methods of analysis are, to be sure, not satisfactory for
exact quantitative results since each method gives results
varying from the others by a small margin, yet the figures
obtained may be used provided we keep in mind the fact
that they are but approximate. Czapek * has given a sum-
mary of what has been done in this connection by differ-
ent chemists. The following table shows the results ob-
tained by two of the best methods of analysis.

LANGE. SCHULZE.

Beech ...- =. . + 54,0-53.0-53.0 51.0-50.5-50.0
MPS ee ees cee ce 51.0-50.0-50.6 48,0-48.2-49.0
tie Sa 55.0-56.0-56.0 52.0-52.0-52.5

The figures indicate the percentage of cellulose found
in the woods named by the two methods of analysis and
also show the amount of variation in the results obtained.
Analyses of pine wood showed that the percentage of cel-
lulose varied from 47.5 to 53.5 per cent, and it was found
that the sap wood was richer in cellulose than was the

* Czapek. Biochemie der Pflanzen. 1: 563-564. ( 1905.)

A

i Sy ip alpen ie an Se ea

ee
oo) ier

[Eee

56 MISSOURI BOTANICAL GARDEN.

heart wood. In general it may be said that from 45 to 60
per cent of wood is cellulose. Besides all this, pure cul-
tures of several of the more common of the wood rotting
fungi have been grown upon blocks of wood, which so far
as could be determined, had no unlignified cellulose in the
cell walls, and delignification has resulted while similar
control blocks have shown no such delignification.

EFFECT OF STERILIZATION UPON WOOD BLOCKS,

When it is considered for how long a time the thin micro-
scopic sections of the various woods were submitted to boiling
at extreme temperatures before any decided effect could be
detected one can hardly say that the effect of boiling at 100
degrees Centigrade for the time necessary for sterilization can
be appreciable in cell walls in the interior of relatively large
blocks of wood such as are commonly utilized for cultures
of the wood rotting fungi. It seems that the effect upon
the lignin of a block of wood which the investigator is sure
has no unlignified cellulose originally, is so slight that it
may be neglected. This is still more true when we remem-
ber that simple soaking has a more or less solvent effect
upon the lignin in any method of culture work. There
can be no doubt that when finely divided wood is soaked in
water or is boiled, lignin is extracted in sufficient quantity
for lignin reactions to be obtained in the filtrate from the
extracted material. But this is very different from the
detection of that extraction by micro-chemical examina-
tion of the treated walls themselves; for it must be con-
fessed that micro-chemistry furnishes methods which are
far from accurate for quantitative analysis. .

In testing the action of fungi upon wood we must keep
in mind the solvent action of boiling and of continual
soaking in water and not be too hasty in drawing conclusions
from meager results ‘which might be confused with such
action of boiling or soaking. Control tubes of wet wood

STUDIES ON THE LIGNIN AND CELLULOSE OF woop. 57

and careful examination of the material used should obvi-
ate much of this difficulty however.

ENZYMES OF FUNGI.

Potter also states that, ‘‘ the delignification cannot be
entirely attributed to an enzyme secreted by fungi.’’ As
above stated it has been shown that the original thickness
of wood fiber walls is very materially reduced by some
agent at the very time when the fungus is attacking the
tissues in question. It is also shown that the cellulose which
may or may not be present in the walls defore they are at-
tacked certainly is not the same unlignified cellulose which
is present after the destruction of the wood by the fungus.
It seems certain then that fungi do delignify woody cell
walls.

There can be no doubt that enzymes, or some substance
exhibiting the characteristics of enzymes, have been proved
to exist in some of the fungi and indeed in some of the
wood rotting ones. Such investigations as have already
been published in this connection would seem to prove be-
yond any reasonable doubt that enzymes are secreted by
fungi and that they are of the most varied character.
Diastase seems to he proved to be present in many of the
different fungi and in the wood rotting ones. The disap-
pearance of starch in the early stages of attack of some of
the wood rotting fungi upon wood seems to be generally
attributed to the secretion of diastase by the fungi. Why
should the disappearance of lignin from the fiber walls,
leaving cellulose in the last stages of decay, be attributed
to any other cause than the secretion of a delignifying
enzyme until we have proof of some other agent which is
capable of such action? ‘To be sure, the separation of an
enzyme or a juice showing enzymic action from the fungus
being studied is the exact proof of the existence of an
enzyme; yet when we consider that enzymes are accepted

58 MISSOURI BOTANICAL GARDEN.

by practically all botanists as being present in fungi we can
hardly come to but one conclusion regarding the existence
of a delignifying enzyme.

EXPLANATION OF PLATES.

Plate 1.— 1, Portion of transverse section of wood of Populus tremu-
loides showing several wood fibers with a third, inner, cellulose layer.
This layer is colorless, as the section is stained with phloroglucin which
is a test for lignin. This supernumerary layer is characteristic of the
species of the genus Populus so far as they have yet been examined, five
of the endemic North American species from the various parts of the
country having been tested. 2, Portionof a transverse section of wood of
Pinus palustris showing one annual ring of growth. The thicker walled
cells form the summer wood of the ring, while the thin walled ones make
up the spring wood. This section has been boiled until the lignin was
extracted from the spring wood, and was then stained with phloroglucin,
atest forlignin. The red walls have reacted with the chemical show-
ing that lignin is still present in the summer fibers. The yellow cells
have retained their normal color to a greater or less extent and gave no
reaction with the phloroglucin, thus showing that they have been de-
lignified quite completely. The section shows the relative solubility of
the lignin in the two parts of the annual ring.

Plate 2.—1, Portion of a transverse section of wood of Picea rubens
boiled for 27 hours. This has been stained with chlor-iodide of zinc,
a test for cellulose. The cellulose has been colored blue while the other
parts of the walls are of their normal color. This shows the delignify-
ing action of boiling water upon the lignin of the fibers. 2, Transverse
section of wood of Populus balsamifera cut from a freshly felled tree and
stained with chlor-iodide of zinc. The thick supernumerary layer of
cellulose is colored blue. 38, Transverse section of wood of Sassafras
Sassafras boiled 20 hours and stained with chlor-iodide of zinc. This
is drawn with a smaller magnification than are the other figures. It
shows the frequency of the cellulose layer, in the fibers of the wood.
’ Note that there is no cellulose present in the last cells formed in the ring,
which arein a vertical line near the left of the figure. 4, Transverse
section of wood of Sassafras Sassafras drawn on the same scale as the
_ other figures. This was boiled and treated with chlor-iodide of zinc
‘and shows the character of the delignified layer in the fibers. Most
of the fibers adjacent to the vessels and tracheids do not have the
cellulose layer.

re

2
o
a
“d
4

hier

SAME CMe Meee Te ee eta pet at eM mag: oe ae | é * Seb. tre Si 1. eee ea ee ee ee ee ee ee ee ee

REPT, MO. BOT. GARD., VOL. 17. PLATE 1.

LIGNIN anp CELLULOSE.

REPT. MoO. Bot. GARD., VOL. 17. PLATE 2.

LIGNIN AND CELLULOSE

STUDIES UPON SOME CHROMOGENIC FUNGI WHICH
DISCOLOR WOOD.*

BY GEORGE GRANT HEDGCOCK,

GENERAL REMARKS.

The discoloration of unpainted woodwork exposed to
moisture and to oxidation by the air is a familiar phenom-
enon. Of course much of this discoloration is due to the
dirt and soot which accumulate on the surface of boards

in addition to the chemical changes which take place

through weathering. There is also a class of stains of an
entirely different nature and more striking to the eye.
These occur on freshly sawn lumber in piles. It was in
the investigation of the most important of these lumber
stains in the study of the western yellow pine, Pinus
ponderosa,t under Dr. Hermann von Schrenk of the Mis-
sissippi Valley Laboratory, of the United States Depart-
ment of Agriculture, that the studies upon which this
paper is based were begun. In addition to the blue stain
in pine lumber, brown, black, pink, purple and yellow
stains and blotches were noted, not only on pine boards,
but also on gum, poplar and other kinds. These were
found to be caused by a number of fungi, some of which
were quite different from that which causes the blue stain
of the western yellow pine, viz: Ceratostomella pilifera
(Fr.) Winter. Again the bluing of wood was found to be
the result of the action of more than one species of
Ceratostomella. |

* A thesis presented to the Faculty of Washington University, in
edndidacy for the degree of Doctor of Philosophy, April, 1906. Pub-
lished by. permission of the Secretary of Agriculture. — Type cultures
of all of the species described as new in this paper have been divided into
parts deposited respectively in the United States National Herbarium, the
Physiological and Pathological Herbarium of the United States Depart-
ment of Agriculture, and the Missouri Botanical Garden Herbarium.

+ Von Schrenk, H. Bull. Pl. Ind. 36: 1-27. (1903). (5

9)

60 MISSOURI BOTANICAL GARDEN.

The work of investigation has been undertaken primarily
on account of the value of the knowledge which might
accrue through the study of the chromogenic fungi or
bacteria which are concerned in the color reactions which
take place in stained lumber; yet, on the other hand, the
economic value of the work has not been lost sight of,
owing to the loss of hundreds of thousands of dollars to
the lumber industry each year due to the lowering of the
grade of lumber in piles through the rapid action of many
tiny wood staining fungi.

In the study of the fungus flora of the lumber pile a
large number of forms have been observed not all of which
discolor the substratum upon which they grow. Attention
has been given not only to those which penetrate wood
deeply and stain it, but also to those which discolor it only
superficially.

In the isolation and culture of the fungi herein de-
scribed, recourse was had to the most careful bacterio-
logical methods, such as are described by Dr. Erwin F.
Smith * in his excellent monograph on ‘ Bacteria in rela-
tion to plant diseases.’’ In a number of instances new
conidial stages of fungi have been discovered, and in every
instance the relation of the new form to the older known
forms has been repeatedly established by starting with the
newly discovered stage and following the fungus through
all its known stages and back again to the beginning. The
saprophytic nature of the forms studied made it possible to
start with single colonies on a poured plate of agar medium
and make transfer cultures to test tubes of sterile wood or
other media. Each fungus was grown upon a number of
kinds of wood, as well as upon potato, rice, bean, sweet,
potato, and other similar media in tubes, in addition to
cultures upon agar media made from wood and other veg-
etable decoctions.

In order to better establish varietal and specific charac-

* Smith, E. F. Bacteria in relation to plant diseases. 1. (1905).

eae ag? te,
ED Se Pa Le Re A Re ee Eee ee ee ae oe OUR?
Fal ots Oe AP. By SP ng ee ; ae

CHROMOGENIC FUNGI WHICH DISCOLOR WwoopD. 61

ters, parallel cultures of related fungi have been grown on
sets of the same media in conditions as nearly identical as
can obtain in an ordinarily well-equipped bacteriological
laboratory. In order to make drawings of the delicate coni-
dial stages of some species it was often found necessary to
grow the colonies in Petri dishes of thin glass, and study
them in the open in the dish, without placing a cover glass
over the very fragile hyphae.

It is thought best to arrange the results under the fol-
lowing groups or heads : —

WOOD-STAINING FUNGI.
I. Wood-bluing fungi.
1. Ceratostomella.

II. Wood-blackening and»wood-browning fungi. .
1. Graphium.
2. Hormodendron.
3. Hormiscium.
4. Other wood-blackening fungi.
III. Wood-reddening fungi.
1. Penicillium.
2. Fusarium.

In conclusion of these introductory remarks, grateful
mention should be made of the valuable assistance to the
work of investigation accruing through the use of the
library and other facilities of the Missouri Botanical Gar-
den, kindly tendered by Dr. William Trelease, the Direc-
tor. Acknowledgment also should be made of the helpful
co-operation of Dr. Hermann von Schrenk, Mr. Perley
Spaulding and Miss Laura L. Eames, of the Mississippi
Valley Laboratory, and to Dr. A. D. Hopkins, of the Bu-
reau of Entomology of the United States Department of
Acriculture, for securing material for study and granting
other similar favors.

I. WOOD-BLUING FUNGI.
The blue stain in pine wood has been known in Europe for

62 MISSOURI BOTANICAL GARDEN.

many years. Hartig * and Frank f both refer to it in their
publications on plant diseases. Hartig ascribes the cause
of bluing to Ceratostoma piliferum (Fr.) Fuckel, which is
now placed under the genus Ceratostomella according to

Winter. This species was first described by Fries,

who placed it under the genus Sphaeria, where it remained
with other species until Fuckel placed it under the genus
Ceratostoma. Winter,t in the revision of the genus Cera-
tostoma, placed all species with colorless ascospores under
the new genus Ceratostomella.

Von Schrenk § in a bulletin of the Bureau of Plant In-
dustry, U. S. Department of Agriculture, fully describes
both the fungus Ceratostomella pilifera and its effect and
mode of entrance into the wood of Pinus ponderosa.

A Jarge number of species of Ceratostomella are de-
scribed by Saccardo,{ many of which occur on wood, but
no reference is made to the staining effect of any of these
species. There is little doubt that a number of these
are wood-staining fungi, but from the results of our inves-
tigations it appears that the common bluing fungus is C.
pilifera.

1. CERATOSTOMELLA.

Winter in Michelia characterizes the genus Ceratosto-
mella as follows: ‘* Perithecia superficial or partly im-
mersed, usually tough, leathery or carbonaceous, glabrous
or invested with filaments, with prominent well developed
beak; asci without paraphyses, 8-spored; ascospores con-
tinuous, globose, ovoid or oblong, hyaline; spermogonia
and conidia present in some instances.”’

Although Winter mentions that conidia are present ina
few instances, a careful search through the literature on
the subject reveals that the number of species which have

* Hartig, R. Lehrbuch der Pflanzenkrankheiten. 75, 106. (1900).
; Frank, A.B. Krankheiten der Pflanzen. 1: 107. (1895).

¢ Saccarado, P. A. Michelia. 1: 370. § lc.

{| Saccardo,P. A. Sylloge Fungorum. 1-17.

CHROMOGENIC FUNGI WHICH DISCOLOR Woop. 63

s

conidia in their descriptions is very limited indeed, only
two being noted, and where such descriptions have been
given, no mention is made of the connection between the
conidial and the perfect or perithecial stages as being
established by cultural methods. For instance, Ceratos-
tomellaalbocoronata (Ellis) Sacc. has a conidial stage where
the conidia are from 2- to 3-septate, and of abnormally
large size. So far as our investigation has proceeded, all
species of Ceratostomella have colorless conidia, that are
one-celled and of about the same measurements as the

‘ascospores. The presence of asci in the perithecia has
been the most difficult point to prove. Cultures must be

taken at the stage just preceding full maturity of the ascus,
else it apparently dissolves and frees the ascospores at ma-
turity, allowing their free ejection through the long, nar-
rew beak of the perithecium, but destroying the evidence
of the presence of a sac. In an examination of the ejected
ascospores they are frequently found adhering to each
other in fours, side by side, in the same position they
occupied in the ascus.

The conidial stage of Ceratostomeiia is very important,
owing to the immense number of conidia borne on the
mycelium in its earlier growth. These are readily dissem-

inated by the wind and are probably carried by insects

which penetrate the wood and bark of trees, like most of
the ambrosia and bark beetles. At the stage in which the
conidia form a mucilaginous mass, they adhere readily to
any insect that may pass over them. In the laboratory a
number of species of mites which feed on fungi carried
spores on their bodies from a colony in an agar plate to a
sterile portion of the surface of the medium and started
new colonies of the fungus. Bark beetles were placed in
a dish with the conidial stage of Ceratostomella, and after
allowing them to remain a short time were transferred to
sterile agar plates which were inoculated with spores
from the insects. It is probable that some species of in-

ce

cai alia ection Sai
™= os "¢ ype e

APOE LER NERS 5 RIVE NT God Rae yee eg ME ey RR eae eee ae
PEE Sa hghv OSES WE AE eID ena 7 . anata ae
ae : ‘

64 MISSOURI BOTANICAL GARDEN.

sects feed on the conidial stage of Ceratostomedla, especially
one or more species of ambrosia beetles and a number of
mites infesting their channels in the wood; but proof is
yet lacking on this point. The constant occurrence of this
fungus in the channels of a number of wood boring beetles
indieates that the conidia or the ascospores must be carried
in some manner by these insects. Hopkins* describes
some of these beetles in a bulletin of the U. S. Department
of Agriculture, and von Schrenkf describes how C. pilifera
follows the channels of wood-boring beetles in Pinus
ponderosa. The writer has made similar observations on
Ceratosmella in its penetration of Pinus ponderosa, P.
Arizonica, P. echinata and P. Virginiana.

Although the genus Ceratostoma is very closely related
to Ceratostomella none of the species of Ceratostoma have
been found in connection with the blue stain of wood. In
the course of the present investigation stained wood has
been collected from a large number of localities, including

wood from trees and shrubs of the following genera: Abies,

Acer, Fagus, Fraxinus, Liquidambar, Liriodendron,
Pinus, Rubus, Ulmus, Vitis and Wistaria. A number of
species and varieties of Ceratostomella have been identified,
cultivated and proved to be wood-bluing fungi. Ceratos-
tomella pilifera has been found far more prevalent than
other species. As the conidial stage of this species was
first discovered by the writer and the description has never
been published, it will now be given. It is taken fromthe
results obtained from a very large number of pure cultures
grown and studied over a period of nearly four years.

Ceratostomella pilifera (Fr.) Wint.

This fungus has been isolated in cultures taken from
stained wood from a number of species of Pinus taken
from nearly every region of the United States. It pene-

* Hopkins, A. D. Bull. Div. Ent. U. S; Dept. Agr. 32: 9, 10.
+ Von Schrenk, H. J. ¢. :

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD.) 65

trates rapidly the sap wood of dead and dying trees, giving
it a peculiar blue tint whose color is hard to explain, since
the mycelium of the fungus in the cells of the wood is a
dark brown. It has also been found in the sapwood of
species of Abies, Quercus, Fraxinus, and other genera.
Ceratostomella pilifera grows readily upon a large num-
ber of culture media. Pine decoction agar made from the
bark and sapwood was found an excellent medium for its
cultivation. Two distinct fruiting forms were observed in
all the varieties, the conidial and the perfect or perithecial.

MYCELIUM.

When either the conidia or ascospores are sown In pine
agar plates, germination takes place in a few hours, and in
two days colonies with a white hyaline mycelium develop.
The filaments areseptate, and, asa rule, branch alternately.
In a day or so upright hyphae, either simple or branched,
are sent out, upon which are borne branching whorls of
conidia. Inthe course of a week or so, portions of the
mycelium in the older region of the colonies develop with
thicker walls and assume a brown color. From these dark
colored filaments the perithecia originate.

The first growth of the mycelium is usually sparse,
but under favorable conditions a secondary profuse
growth with a fluffy white appearance develops, which
bears conidia, and the formation of perithecia is retarded.
Such a growth is formed on some of the richer media
made from pine and other plant decoctions. Under these
conditions the development of conidia is enormous, and
the number of perithecia is decreased somewhat. Cultures
on the sapwood of trees in test tubes bear both stages of
the fungus abundantly. On the heartwood, however, but
few conidia are formed, the mycelium is almost invisible to
the eye, and perithecia are either sparse or entirely absent.
The filaments of the fungus never penetrate deeply into

5

t
a
s
9
28
3
4

;

+
*

Mee. ee erat ees eee

waren ne eA

3 _

~~ as ms

‘ist Sal te Ai i Blt
ee ae hy 7.
aa /
eee,

are

} *

mt

ey

Bt

oe

Saks

ee ~

enix

aa:

a

s

ae

AG \

7:

BS

ag

ryt

ei

#

2

3

= oe

e

rt ss ER ck Ie fil Ok oats Lagi ok MR US) on Six ot ca

Ce ee ee tee se toy Pegt P See

66 MISSOURI BOTANICAL GARDEN.

the tissues of heartwood of pine, oak, ash, or similar
woods in artificial cultures, which fact coincides with the
observation that in nature, for some reason, the same
thing happens, the older discolored heartwood being
usually free from attacks by the mycelium.

ConIDIA.

The conidia are borne on erect or-slanting hyphae which
branch alternately from the mycelium. They develop
terminally in branched or simple moniliform short chains
which are often in whorls (pl. 4,f.5). They are de-
tached by the least movement, and in moist air fall’
together in agglutinated masses about the terminus of the
hyphae, in a form which superficially resembles Cephalo-
sporium (pl. 4, f. 6). These masses, however, are not so
regular as the masses of conidia in the heads formed by
Cephalosporium. The conidia bear some resemblance
to those of Ovularia. The Cephalosporium-like clus-
ters are always present in agar cultures after a few
days’ growth. The conidia vary from obovate to elliptical
or cylindrical. They measure from 8 to 12 in length,
and from 2 to 4m in width. They are hyaline when
young, but often become vacuolate or guttulate when old.
They are unicellular and borne on septate hyphae which
are from 3p to 5y in diameter. Only hyaline filaments
bear conidia. In germination the conidia send out terminal
germ tubes from either end (pl. 4, f. 7).

PERITHECIA.

The first indication of the formation of perithecia on the
mycelium is the union of two or more filaments and the
formation of a knotted mass. The adjacent mycelium
changes first to a light brown color, and later to a very
dark brown. In the center of the mass the young peri-
thecium develops first as a globular black body, without a
beak or neck.

The walls of the perithecium are usually formed by one

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 67

layer of cells united in. a rudimentary tissue. From the
upper side of the young perithecium, after it has grown to
about its normal size, there is projected a long tube or beak
composed of parallel hyphae which terminate with colorless
ends in a rounded tip. When the perithecium is mature it
contains a large number of irregularly obovate or elliptical
asci, each containing eight spores, usually arranged in
fours. The beaked ostiolum, when the ascospores are
ejected, splits at the end into a number of bristles which
spread out and form a supporting crown for the ejected
ascospores (pl. 3, f. 8). The perithecium without the
neck measures 160u in height and 180, in diameter, aver-
age, and the neck about 1 mm. in height and 20y in diam-
eter, average. The ascospores average 5.5u by 2.5p.

It is quite evident that the sac containing the ascospores
is dissolved at maturity, since it is rarely found with the
mature ejected spores, but must be looked for in perithecia
just approaching this stage. If the spores are ejected in
open air they usually collect in a globular drop at the
terminus of the neck, and when in this position the drop
resembles very much the heads of some species of
Graphium when covered with the mucus drop characteris-
tic of the latter genus. If the spores are discharged in
water they form worm-like sticky masses not readily mis-
cible with water.

Ceratostomella Schrenkiana n. sp.

The sapwood of pine lumber made from Pinus echinata
in Arkansas and other southern localities is stained by a
number of fungi, among which are species of Ceratosto-
mella, Hormodendron, Graphium, Hormiscium and Peni-
cillium. A species of Ceratostomella which greatly re-
sembles C’. pilifera has been frequently collected on the
wood of the southern pines by Dr. von Schrenk and others,
from various localities in Missouri and Arkansas.

This species, when grown side by side with cultures of
C. piliferafrom the Black Hills pine, attains the same

eee, ere ae
fn *

See

68 MISSOURI BOTANICAL GARDEN.

size in so far as the colonies and perithecia are concerned ;
but on the other hand, both the conidia and ascospores are
constantly much smaller. In nature, the perithecia that
mature are often surrounded by several closely adhering
dark bodies, which are either sclerotia or abortive peri-
thecia, apparently the latter. This feature has never been
noted with other species of Ceratostomella. In view of
the differences, it is thought best to separate this form
from C. pilifera and it is named C. Schrenkiana, in
honor of Doctor von Schrenk, who first collected the
fungus and noted its peculiarities.

The following cultural characters are taken from a large
number of natural and artificial cultures : —

MYCELIUM.
Conidia or asco-
v " . p spores, if sown on agar
VV RAY plate cultures, germi-
- Te fz nate in a few hours,

and after a day have
formed conidia, if the

temperature is favora-
\ ( ble. The most favora-
A! V ble temperature for
W tA growth is 80° to 90°
NU F. The filaments and
hyphae remain white
ae for several days, and
often become massed
in a furry outgrowth in

ments unite into upright
As clusters or false heads.
Pc LNA «(See figure). After a

~~ few days’ growth, por-

which strands of fila-—

C. SCHRENKIANA, X 50.
Dendroid fruiting tuft.

tions of the mycelium
lying next the agar

become pigmented with a brown color, and rapidly develop

Ss Ra ar)

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 6Y

perithecia. The filaments of the mycelium measure 5p to
7m in diameter.

CONIDIA.

The conidia are borne in clusters similar to those of
C. pilifera (pl. 4,f. 1,2). They are unicellular, obovate
to elliptical or cylindrical, and measure 3p to 7p by lp to
Qu, averaging 5u by 1.84. They are hyaline and are
usually neither guttulate nor granular, but when old may
have one or two guttules.

PERITHECIA.

Perithecia are formed on the mycelium (pl. 4, f. 4), as
in C. pilifera. The shape is spherical. They are coal
black in color, and measure 120u to 200u in diameter, with
a smooth, beaked ostiolum, measuring .8mm. to 1.2mm.
in length by 10% to 254 ia diameter, surmounted at ma-
turity by a row of short, hyaline, spreading bristles, which
support the. ejected ascospores ina globular mass. ‘The
bristles measure about 10u to 15y by 2u (pl.3,f.6). The
ascospores are hyaline, pointed, elliptical, often slighily
curved, and measure 2.54 to 4p by 1p to 1.5p, averaging
3.54 by lw (pl. 4, f. 3).

This species resembles, in its gross measurements, the
description of C. echinellaE. & E. It differs, however,
_ in that the perithecia are borne superficially, and are not
glandular-pubescent or thickened at the tips. The mature
perithecia are not gregarious, although they bear numerous
small bodies at their base, under certain conditions.

Ceratostomella echinella E. & E.

The last species of Ceratostomella identified and studied
before publication was collected by Dr. von Schrenk at Kir-
byville, Texas, on the wood of the red beech, Fagus atropu-
nicea (Marsh.) Sudworth. This was growing on freshly
cut heart and sap-wood, staining the wood either a bluish or

Aly Re pies eens ek th

70 MISSOURI BOTANICAL GARDEN.

a brownish color. The fungus was isolated and grown in
pure cultures on the wood of the red gum and on pine de-
coction agar, as well as on tubes of rice, potato, etc., the
following cultural characters being from both natural and
artificial cultures. The description differs but slightly
from that given by Ellis and Everhart* and has been
emended by the addition of the conidial stage.

MyYcELium.

The colonies resulting from either ascospores or conidia

are white at first with a hyaline mycelium which later’

becomes pigmented in certain of the larger filaments, assum-
ing a dark brown color. The filaments measure from 4
to 7 in the mature forms. Conidia appear after 36
hours, and perithecia in four or five days.

CONIDIA.

The conidia are borne in short branching moniliform
chains in small clusters, falling together in irregular masses
as they mature (pl. 6,f.1). They are unicellular, hya-
line, becoming guttulate when old, obovate to elliptical,
and measure from 4p to 6.54 by 2u to 3.5p, averaging 6
by 3z.

PERITHECIA.

The perithecia are brown at first, but when they are ma-
ture they become black and carbonaceous and are covered
sparsely with short, glandular hairs. They are globose,
or sometimes slightly flattened, measuring 50p to 100p in
diameter. The hairs measure 104 to 324 in length by
1.5mto 2, Each terminatesina gland or spherical enlarge-
ment measuring 24 to 34 in diameter. The perithecium is
tipped by a long slender beak, black at the base, brown

* Ellis, J. B. and Everhart, B. M. North American Pyrenomycetes.
p- 195. (1892).

¢
CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 71

or gray at the apex, but smooth throughout measuring
1 mm. to 1.7 mm. in length by 15» to 25 in diameter
and surmounted when mature by a terminal fringe of
bristles 15 to 254 in length and tapering from 4p at the
base to lp at the apex, being hyaline throughout (pl. 3, f.
3). The asci are clavate(?), bearing biseriate spores
which are hyaline, curved or straight, elliptic to cylindri-

cal, and which are ejected in the open air ina globular mass,

at the terminus of the beak.

Since writing the above, co-type material of C’. echinella
inthe herbarium of the Missouri Botanical Garden has been
examined. ‘The perithecia are probably as described by
Ellis and Everhart,* but they have persistent asci which
contain brown two-celled ascospores, placing the fungus in
the specimen under another genus. This indicates that at
least a mistake was made in a portion of the material sent
out as type specimens if not in the description.

Ceratostomella capillifera n. sp.

The wood of Liquidambar Styraciflua L. is an excellent
pabulum for a large number of fungi, and lumber made
from this tree, when in piles, soon becomes thoroughly
stained either blue or black by the growth of a number of
species of wood-discoloring fungi. Among these was
found quite commonly a species of Ceratostomella which
is closely related to C. pilifera, but differs in the length
of the beaked ostielum, the length of the terminal bristles
and the shape and size of the conidia and ascospores.
The following characters obtain with pure cultures : —

MYCELIUM.
Cultures of either conidia or ascospores sown on pine-

decoation agar germinate quickly, forming a diffuse, white,
hyaline, septate mycelium, which in less than two days

* Ellis, J. B. and Everhart, B. M. J. c.

:
;
5.

72 MISSOURI BOTANICAL GARDEN.

bears hyaline conidia abundantly. In about a week, black,
long-beaked pycnidia appear on certain portions of the
mycelium, which assumes a brown color, the development
being similar to that described for C. pilifera. The fila-
ments of the hyaline portion of the mycelium measure
from 2u to 8 in diameter, and the older brown-colored
filaments, in wood, from 3p to 6y,

CONIDIA.

The conidia are borne like those of C. pluriannulata,
and are similar in shape (pl. 6, f. 2,3). They are unicel-
lular, hyaline, elliptical to cylindrical, and measure 4 to
8» in length, and 1.5y to 2u in width averaging 64 by 1.8y.

PERITHECIA,

The perithecia are spherical, of a black color, with
black, long-beaked ostiola, terminating in long, wavy
hyaline filaments. The body of the perithecium is usually
nearly covered with dark brown hyphae or filaments. It
averages 200m in diameter, and the neck 1.5 mm. in length
and 254 in width. The terminal filaments are long and
slender and measure 804 by 1p, average (pl. 3, f. 1).

The ascospores are elliptical to reniform in shape and
average 4.54 by 1.5uin size. It will be noted that al-
though the perithecia of this species are larger than
those of C. pilifera the conidia are somewhat smaller.
Both ascospores and conidia usually germinate terminally,
some few exceptions having been noted.

The name C. capillifera is given to this species on
account of the long minute bristles which project from the
ostiolum.

Ceratostomella pluriannulata n. sp.

A block of blued sapwood from Quercus rubra was col-
lected in 1905 by Perley Spaulding and turned over to the

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 73

author for investigation. A species of Ceratostomella was
found among the cells of the wood and fruiting quite
abundantly on the surface. The characters of the fungus
agree in part with C’. pilifera, to which it is closely related.

‘It has ascospores which are constantly of a slightly differ-

ent shape, and is often characterized by more than one
whorl of bristles on the beaked ostiolum. The conidia are
considerably smaller, and the hyphae on which they are
borne are more frequently branched, thus giving rise to
larger clusters. Owing to the frequent presence of more
than one ring of bristles on the ostiolum, the name C.
pluriannulata is here given to the species, with the fol-
lowing cultural characters : —

MYCELIUM.

The growth of the mycelium from either conidia or
ascospores resembles that of C. pilifera. The spores
germinate in pine-decoction agar in 2 to 3 hours, and in 2 to
3 days a white, floccose mycelium bearing conidia appears.
In from 5 to 10 days perithecia appear, the development
of which is like that of C’. pilifera.

ConrIDIA.

The hyaline unicellular conidia are borne on branching -
septate, hyaline, upright hyphae in simple or compound
whorls (pl. 5, f.1). In the older portions of the colony,
the conidia fall off and adhere in masses forming clusters
resembling Cephalosporium. These clusters vary in size
from 10u to 100m in diameter. The larger clusters are
formed by the adherence of several spore clusters into one
mass. The conidia measure 54 to 8u in length and 2p to
3u in width, averaging 6p by 1.5p.

PERITHECIA.

The perithecia are borne superficially on the wood, or on
the mycelium, there being no definite stroma in this or in

74 MISSOURI BOTANICAL GARDEN.

the other species of Ceratostomella studied. The peri-
thecia are globose, varying from 904 to 200m in diameter,
and averaging 1204. The lengthened beak measures from
.9 to 2 mm. in length, and 104 to 30m in diameter, aver-
aging 1.5 mm. by 254. Some of the necks, in addition to
the ring of bristles ordinarily found at the terminus of the
beaked ostiolum, have one or two whorls at some distance
below the end (pl. 8, f.7). This character has never
been noted in cultures of C. pilifera. The terminal
bristles are tapering and average 20 in length by 2y in
width. The asci have not been measured. The ascospores
are reniform in shape, hyaline, and measure 4p to 5y in
length, and 1.54 to 1.7 in width, averaging 4.54 by 1.54
(pl. &, £..2).

This species penetrates pine wood when inoculations are
made, nearly as readily as C’. pilifera. The mycelium in
the wood cells follows the medullary rays in pine wood,
but in red gum wood it penetrates also numerous tracheary
vessels, and sometimes the wood fibres. It is most abun-
dant, however, in the medullary rays of the latter wood.

Ceratostomella minor, n. sp.

During the year 1905 some specimens of pine wood
stained with a dark blue color, from Pinus Arizonica Eng.,
were received from Dr. A. D. Hopkins. From these there
were isolated two species of wood-staining fungi; one
a species of Ceratostomella, the other a Graphium. The
Ceratostomella proved to be smaller than the species from
Pinus Virginiana, and differed in other points. The fun-
gus gained entrance through the galleries of wood beetles,
the stain radiating from cavities made by these insects.
When grown in pure culture the fungus possessed conidia
and ascospores. As the specific characters do not agree
with any previous description of Ceratostomella, the name

CHROMOGENIC FUNGI WHICH DISCOLOR woop. 175

C’. minor is assigned, with the following cultural charac-

ters: — :
MYCELIUM.
Cultures of eitner conidia or ascospores on pine agar

media germinated in a few hours, and in two days the col-
ovies began to form conidia. The mycelium is white and

sparse. The hyaline filaments of the conidial stage measure

in width 1.5y to 2.6, averaging 2.34. The brown or black
filaments in agar cultures and in wood measure in diameter
from 2u to 4p, averaging 3.54. These are often rugose,
the roughenings being coarser than those in C’. exigua.
In some portions of the mycelium on wood there are
filaments that contain cells that are unequal in diameter,

being enlarged at one end. Like C. pilifera this species
follows chiefly the medullary rays of pine wood.

CONIDIA.

The formation of the conidia and the form of fruiting
do not differ essentially from those of C. pilifera and
C. exigua. The unicellular, hyaline conidia measure from
4 to 5.54 in length, and from 1.84 to 2m in width, aver-
aging 4.54 by 2u. They areoval to elliptical in shape, and
fall off at the slightest touch, collecting in rounded masses
on the hyphae when in moist air.

PERITHECIA.

The perithecia mature in about three weeks in artificial
cultures. They are spherical, black rugose bodies, vary-
ing from 40 to 70» in diameter, averaging 524, witha
beaked ostiolum 120y to 160uin length, averaging 135p,
and from 6u to 124 in diameter (pl. 5, f.6). The beak
ends in a whorl of short, thick bristles (pl. 8, f. 4). The
asci are round to oval, hyaline, with eight spores in each,
usually arranged in fours (pl. 5, f. 7). The spores meas-
ure 3.1uto 4.2uin length and .9 to 1.94 in width, averag-

76 MISSOURI BOTANICAL GARDEN.

ing 3.54 by 1.54. They are hyaline and germinate termi-
nally.

The perithecia are smaller than those of C’. exigua, the
description of which follows; the ascospores differ in shape,
and are smaller, and the terminal bristles of the beak have
a different shape, being short and thick as compared with
the slender ones of C. exigua.

Ceratostomella exiqgua n. sp.

The wood of dead and dying trees of Pinus Virginiana
Mill., and one or two other species, is often colored a blue-
black color, even more intense than that produced by Cer-
atostomella pilifera in wood of the western yellow pine.
Specimens were first obtained for study from Dr. A. D.
Hopkins of the United States Department of Agriculture.
It was at first thought that the stain was caused by C. pili-
fera, but pure cultures of the fungus proved that it was
different ina number of particulars. The scanty mycelium,
the numerous diminutive perithecia and tbe smaller coni-
dia, as well as the more intense color of the stain, were
prominent points of difference which were considered of
specific importance. These characteristics remained quite
constant through a series of cultures on various kinds of
wood.

The life history of the fungus obtained from a series of
pure cultures is as follows: —

MYCELIUM.

Cultures grown on pine decoction agar from both conidia
and ascospores give in a short time colonies with a white
mycelium much more limited in growth than that of C. pili-
Sera. The conidial stage appears after 24 hours, and is
of the same type as that of C. pilifera, but with the con-
idia in lesser numbers. In a few days the perithecia ap-
pear on a mycelium which assumes a very dark brown,

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 77

*

almost black, color. The development of the fungus on
wood is slower than that of C. pilifera, staining pine, oak,
elm and ash with a blue-black discoloration contrasting
sharply with the unstained portions. The mycelium pen-
etrates pine wood through the medullary rays, and in some
cases enters the wood fibres. The hyphae of the hyaline
mycelium bearing conidia average 2.5y in diameter, while
the dark brown filaments in wood measure from 2p to
6u, averaging 4 in diameter. On the surface of wood
in the vicinity of perithecia some filaments have swollen
cells. Many of the filaments in wood cells have fine trans-
parent roughenings all over the surface.

ConrpIA.

The conidia, like those of C. pilifera, are borne in
whorls of single spores and short moniliform chains inter-
mingled (pl. 6, f. 4,5). In older cultures they drop
together in clusters resembling somewhat those of Cepha-
losporium. They are oval to elliptical in shape, one-celled,
hyaline, and vary in size from 3.54 to 4.5, in length, and
1.64 to 2.2uin width, averaging 4u by 2y (pl. 6, f. 6).

PERITHECIA.

In artificial as well as in natural growths, the perithecia
are very abundant. They are formed in large numbers
under favorable conditions after 10 to 14 days. The pro-
cess of formation is similar to that of C. pilifera. The
perithecia are globose. They are black, brittle, with a
roughened surface. They measure 604 to 80, averag-
ing 73 in diameter, with a beaked ostiolum 150p to 2004
in length, and 8u to 18m in diameter, averaging 1804 by
14u (pl. 3, f. 2). The asci are hyaline, irregularly
oval to elliptical, and contain 8 spores, often arranged in
fours. The ascospores are hyaline, elliptical, sometimes
slightly curved, and measure 2.1luto 2.84 in length, and
8 to 1.1uin width, averaging 2.54 by lp (pl. 6, f. 7).
The sacs usually disappear before the spores are ejected.

ee eee
Ste Eg

%
wi
4

”

re at aa aie i : he as ¥
ae ee Ee ETS ee Tee Bh Lag! FO eB

sl,

es > Oe, a Pa ee ee Sees

4s 0°.7"

a es a
RES ao oS at oe .

Mago eres Sree
aa ‘rs e a: Ai es
: ° skal

sa Ra ee ee
a Pace.

ae

oa
4

ag
a
Be
:. 4,
Z
:

aot tie, We
r

nla tea eo

be NG eesti

OP. Tae eee
F REP a ed

78 . MISSOURI BOTANICAL GARDEN.

This species resembles the description of C. microspora
E. and E. but differs in the shape of the ascospores, the
size of the perithecia, and the length of the beaks. It has,
in addition, a conidial stage, which serves to further set it
apart as a new species. On account of the minuteness of
its perithecia and spores it is named C’. ewigua.

Ceratostomella moniliformis n. sp.

Another species of Ceratostomella was found by Dr.
von Schrenk growing on gum wood in Texas, near Kirby-
ville, occurring on Liquidambar Styraciflua. This species
stains the wood of gum a brown color, and is the most
rapid growing form studied. It is distinct from the other
species of Ceratostomella in the form of the conidial
clusters, and in the early color of the mycelium. The peri-
thecia are covered with short spines, otherwise resembling
very much those of C. pilifera. The name Ceratostomella
moniliformis is given to the fungus, with the following
cultural characters :—

MYCELIUM.

Cultures of either conidia or ascospores sown in agar

plates germinate quickly, andin 24 hours produce the con-.

idial stage. In two or three days the perithecia are pro-
duced. The colony is hyaline at first, but in a few hours
begins to turn gray, and finally becomes black. The my-
celium is coarsely granular, the filaments measuring 24 to
8 in diameter.

CONIDIA.

The conidia are found on simple or branching, upright
hyphae, both in simple moniliform chains and in terminal
clusters, the latter resembling the form occurring with
some species of Graphium (pl. 5, f. 3, 4,5). They
are formed by the abstriction of the ends of the hyphae,

and often fall together as fast as they mature, in rounded :

.
e

Pea ae a ni bee Rake Si Ried a hearer) 2" ed oO Re eg Se See
ay 4 ng a et BE LOGE mE SRP OS ta aa

iy Saka: Sa

oo ee
Seats

aes,

CHROMOGENIC FUNGI WHICH DISCOLOR WooD. 79

masses, surrounded by either water or mucilage. They
are unicellular, cylindrical in shape, hyaline, and measure
64 to 8u by 1.8% to 2.2u, being of different shape
and dimensions from the ascospores, and are thus easily
distinguished from them under the microscope. The con-
idia in mass as they become old have a gray color.

PERITHECIA.

The perithecia are formed by the union of two or more
filaments of the mycelium. First, a gray mass of irregular
cells appears at the pointof union. This rapidly develops
into a globular perithecium, from the top of which there is
then rapidly thrown out a beaked ostiolum, consisting of
parallel filaments. The perithecia are brown to black in
color, 90% to 180 in diameter, covered with sparse, con-
ical spines, 12% to 16 long, with a diameter of 6y at
the base. The beaked ostiolum is brown to black through-
out its length, striate, and is surmounted by short, rather
thick, hyaline bristles, measuring 124 to 184 by 2u

(pl. 8, f. 5). When the perithecia are mature, the asco-

spores are ejected in water in long, slimy, gray masses.
In nature they are ejected in an irregular gray mass sup-
ported by the terminal filaments of the beak. The asciare
fugacious, hyaline, oval, and average 20uby 10u. The
ascospores are hyaline, not guttulate, at least at first, oval,
often flattened on one side, measuring 4¢ to 54 by 3u to
4p.

This species of Ceratostomella is so different in color

from the other species studied as to suggest that it may be

a Ceratostoma. The gray color of the mycelium on media
in which the mycelium of other species remains hyaline,
the constant tint of gray in mature conidia and ascospores,
and the peculiar moniliform chains of conidia very distinct-
ly set this form apart at least as avery distinct species.
But since the color of the spores shows only in mass, it is

Bf

5
; 3
3
3

s

80 - MISSOURI BOTANICAL GARDEN.

thought best to place the species at present under Ceratos-
tomella.

II. WOOD-BLACKENING AND WOOD-BROWNING FUNGI.
1. GRAPHIUM.

Graphium has been associated with the decay of wood and
other organic matter ever since the original species were
named by Tode* and Corda,f and placed under Stilbum
by Corda. It has been found also to be associated with
the staining of lumber, especially of pine and gum boards.
It has been isolated from black or brown stained portions
of wood taken from Pinus, Populus, Liriodendron,
Liquidambar, Quercus, Acer and Wistaria. Species of
this genus, like those of Ceratostomella, enter most kinds of
wood through the medullary rays, being usually confined
to the sapwood, with the exception that they may grow
upon decayed heartwood that other fungi have previously
attacked. Inthe study of anumber of species of Graphium
not all species stained wood alike: some scarcely stained
the wood at all, others, like the one from Liguidambar,
are quite effective as wood stainers.

The stain given by Graphium varies from a dirty gray
to a dark brown, or, rarely, a black color.

The genus Graphium was first described by Corda. It
is described as follows, using a free translation: — ‘* Stroma
cylindrical, clavate, or capitate, brownish, rather rigid,
the upper hyphae paler, lax, and bearing the conidia,
which are elliptical or oblong, hyaline, often involved in
mucus at first.”” No mention is made either by Saccardo,t
or Engler and Prantl { of any conidial stage other than
that found on the stroma or head.

* Tode. Fungi Meckl. 1:10. + Corda. Icon. Fung. 1:8.
t Saccardo, P. A. Michelia. 2:32; Syll. Fung. 43 564, 609.
¢{ Engler, A. and Prantl, K. Die Natiiriichen Pflanzenfamilien. 1 : 493.

4

Sieh oa oe
Se Tonge
oe ;

CHROMOGENIC FUNGI WHICH DISCOLOR woop. 81

Boulanger * has proven by the use of careful cultural
methods that Sporotrichum chlorinum Link var. grisea
Boul. is astage of Graphium eumorphum Sacc. This dis-
covery is now verified by the present investigation, and
Sporotrichum-like stages of other species of Graphium are
now published for the first time. Boulanger also discov-
ered a Sporotrichum stage of Chaetomium cuniculorum,
and obtained fruits of the latter fungus in connection with
Graphium, indicating that Chaetomium cuniculorum is
the perfect stage of the other two fungi. We have not
been able to verify the latter relationship.

Graphium giganteum (Pk.) Sace. is given by Durand t
as a conidial stage of Holwaya gigantea (Pk.) Durand.

Zopf t observed a moniliform conidial stage of Chae-
tomium which failed to germinate. He makes no mention
of finding any related forms of Sporotrichum.

Some of the species we have studied contain both Sporo-
trichum forms and moniliform fruits, the latter resembling
very much the branching clusters of the conidial stage of
Ceratostomella.

From the study of several species of Graphium the fol-
lowing cultural characters are found to be common :—

MYCELIUM.

The mycelium in a culture from either of the two spore-
forms found in the species studied is from the first hyaline
and septate. On rich vegetable-decoction agar, like that
made from green pine sapwood, bean pods, potato, etc., a
fluffy white mycelium is usually produced, varying in
density with the medium and the species. The first form
of conidia is borne on erect hyphae, which may be either
simple or branched. These hyphae or conidiophores

* Boulanger. Rev. Gén. de Bot. 1895 97-102.
+ Durand, E J. Bull. Torr. Bot. Club, 28% 351. (1901).
t Zopf, W. Ueber Pilzfarbstoffe. III. Bot. Zeit. 47; 86-89. (1889).

6

82 MISSOURI BOTANICAL GARDEN.

branch alternately from the mycelium. These first conidia
are borne in clusters of three types, two of which are
simple clusters of the type of those of Sporotrichum but
not borne on prostrate hyphae, the other being a form
which is like most of the conidia described under Cera-
tostomella. In this latter form the conidia are borne in
the short branching chains. These conidia are all ephem-
eral and do not hold their vitality as long as those which are
produced later upon the heads or stromata. In order to
distinguish them from the latter, they are now for the first
time called secondary conidia; and those of the heads or
stromata are now called primary conidia. As the colony
grows older, some portions of the mycelium become more
or less pigmented, and the darker filaments grow together
in masses from which erect columns of parallel hyphae
spring forth. These bear heads composed of branched
hyphae, which produce the primary conidia terminally by

abstriction. As fast as the conidia are formed they drop —

away into a mucous matrix which forms over the surface of
the head. .

SECONDARY CONIDIA.

The secondary conidia of the Sporotrichum-type are of
two forms and are borne on simple erect or branched
hyphae. The first form of this type are attached to the
hyphae in spreading, open clusters, which do not fall to-
gether in masses until some days after they mature (pl. 7,
f.3). The second form are produced in close clusters
which fall together as rapidly as they mature (pl. 9, f. 9).
The third form of fruiting, as mentioned in the previous
topic, more nearly resembles the branching chains of the
conidia of many species of Ceratostomella (pl. 9, f. 3).
The spores in all three forms are oval, elliptical, or cylin-
drical, and correspond closely in form and measurement to
the primary conidia of the same species. They are unicel-
lular and formed by abstriction either from the terminus of
the hyphae or from short, terminal or lateral branches.

stalks is composed of a num-

CHROMOGENIC FUNGI WHICH DISCOLOR woop. 83 tae

PRIMARY CONIDIA. ; é

After the primary conidia have been detached and sep-

arated from the mucus there is in many species nothing to | a
distinguish them from the secondary conidia. In fact, the ..
real difference usually in the two forms is whether they ; .

are borne on a single hypha, or on hyphae united in a

stroma. The agglutinated cluster on a single stalk of one

hypha éorteapands morphologically to the mucous-covered
head (pl. 7, f. 2,5). The latter are more persistent, and
retain the conidia for a much longer time. A secondary :
growth of the stromata has been “noted in some species | :
after the first spores have been formed, and have dropped :
away, in which there are formed branches from the end of
an old stalk, growing out |
from the apex. (See figure ).
Each of these subsidiary

ber of parallel filaments, or
hyphae, which in turn
branch at the outer ends and
form masses of conidia like
the original head, only small-
er. Such growths take
place on rich, starchy media
like cooked potatoes. On
cultures on rice and potato
an abortive type of stromata
is often formed in the shape
of somewhat flat, branching
stalks, which grow much tall-
er than the fruiting form,
but bear no conidia. This
abortive form resembles very
much the original description
of the sterile fungus Anthina Fr., which is probably a
form of Graphium.

G. EUMORPHUM, X 50. .
Proliferation of stroma. a

84 MISSOURI BOTANICAL GARDEN.

In the gross appearance of artificial cultures of Graph-
ium and of Ceratostomella there is often a striking resem-
blance or mimicry. The secondary conidial stage of the
former resembles superficially the conidial stage of the lat-
ter, and the mucous head of the former that of the perithe-
cial stage of the latter at the time the ascospores are ejected
ina viscid drop. The mucus surrounding the conidia of
Graphium is soluble in water; on the other hand, the jelly-
like substance exuded from the perithecia of Ceratostomella
is very difficultly soluble in water. Both are well adapted
for adhering to the bodies of insects with which they come
in contact.

PEeERFECT OR CHAETOMIUM STAGE.

The Chaetomium stage of Graphium observed by Bou-
langer has not been found in connection with the species
studied; but asthe cultural work on G'raphium has not yet
been concluded, it is hoped that by the use of the proper
culture media some species may finally be stimulated to pro-
duce perfect forms, either of Chaetomium or other genera.

In the consideration of the species of Graphium studied,
it is thought best to separate them into the following groups
based on the form of fruiting in the secondary conidial
stage : —

A. Species of Graphium with a secondary conidial stage with conidia
resembling those of Sporotrichum.
B. Species of Graphium with a secondary conidial stage unlike Sporo-
trichum,
1. Species with secondary conidia borne continuously and terminally,
falling at once in clusters.
2. Species with secondary conidia borne either in simple clusters or
in clusters of short branched chains.

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 85

A. Sprecies oF GRAPHIUM WITH A SECONDARY CONIDIAL
STAGE WITH CONIDIA RESEMBLING THOSE OF SPORO-
TRICHUM.

Graphium ambrosiigerum n. sp.

A species of Graphium was found growing in the gal-
leries of wood-boring beetles in the wood of Pinus Arizon-
ica Eng. This material was furnished by Dr. A. D. Hop-
kins in connection with the study of the relation between
ambrosia fungi and ambrosia beetles, a subject which is still
under consideration in connection with the study of such
beetles by Dr. Hopkins. Pieces were taken from the
stained wood in the interior of the block near a gallery,
and placed in test tubes of pine sapwood, which had been
sterilized after being moistened. From the mycelium in
the stained wood there grew out a white growth, bearing at
first the secondary then the primary conidia of Graphium.
Secondary conidia were found in some of the galleries
when the material was first examined, but were not recog-
nized as such at the time. The following are the cultural
characters of the fungus :—

MYCELIUM AND SECONDARY CONIDIA.

When cultures are made with either the primary or sec-
ondary conidia on pine decoction agar, the spores generally
germinate ina few hours, and in two or more days colonies
may be noted with a white, sparse mycelium. At this
stage of growth, secondary conidia appear of the Sporotri-
chum-type, borne on erect, simple hyphae (pl. 8, f. 5}.
These are hyaline, and oval in shape. After a few days,
portions of the mycelium assume a brownish color, and
from certain matted masses of filaments or stromata there
grow out tufts of erect, parallel hyphae (pl. 8, f. 6).

Re ae gee,

ah f ‘
a as pie I See

PND Ee geglet

TA aay inc hae ban ge ain Valk paeiaci tall Pete N SR RY Meee ke ete Boy ee
Ree Rae = J Fo EOE ys Sela Rag alone = PCIE Caste GET OLAS NS phe ice , ‘i fi :
Oxy 3 = Sale mess ' \ f ;

86 MISSOURI BOTANICAL GARDEN.

These tufts of hyphae elongate into a long, round stalk,
terminating in a bunch of either verticillate or alternate
branches.

PRIMARY CONIDIA.

The primary conidia are formed on the upper portion of
the stroma or head by an abstriction of the ends of the
hyphae (pl. 8, f. 7). They are formed continuously, and
as soon as mature, fall away into the mucous mass which
; surrounds the head. The mass is hyaline at first, chang-
ing, as it grows older, to a straw color, and finally toa
dark brown. The stalk is usually hyaline in the first
stages of its growth, but later becomes almost black at the
base, shading off to a light brown near the apex, with a
striated appearance. No chlamydospores or sclerotia are
4 formed. At first the stroma is simple, but by the addition
Be of branches from the base becomes gregarious (pl. 8,f. 4).

The following are the measurements of the mycelial
structures: the hyaline mycelium measures from 1.5 to
2.54 in diameter, averaging 24. The colored portion
averages in diameter about 4u. The secondary conidia
3 measure 3.54 to 4 in length, and 1.2 to 1.5u in width,
Po averaging 3.7¢ by 1.34. The primary conidia average 5u
a by 3. The stalks average in length from 500p to 900u,
: The head, including the mucous drop, measures on an
a average 200u. These do not agree with any previous de-
: scription of species of Graphium and the fungus is here
named Graphium ambrostigerum, owing to the relation of
ae the conidial stage to ambrosia beetles.

Cultures were made on a number of kinds of wood. The
fungus grew readily in the sapwood of pine, gum, oak and
ash, bearing in most cases numerous heads. Secondary
conidia are usually formed in abundance. Cultures on the
heartwood of these trees produce little or no growth of the
fungus. Light colored wood is stained a color varying
from a dirty gray to a light brown, or almost black.

CHROMOGENIC FUNGI WHICH DISCOLOR woop. 87

The mycelium penetrates pine and other similar woods
through the medullary rays, dissolving the starch. In
gum wood, Liguidambar Styraciflua Linn., it penetrates
in addition many of the vessels of the wood. As compared
with Ceratosomella the penetration is not so rapid, neither
isthe stain so intense; but on the other hand it may. in
time stain wood as deeply.

Graphium eumorphum Sace.

A species of Graphium has been found frequenting the
old wood of Rubus strigosus and related species. This
was isolated and grown first-on wood of several kinds which
it was found to stain lightly, then upon agar media. An
examination of the cultural characters of the fungus place
it readily under Graphium eumorphum Sacc. The follow-
ing description gives the principal characters, from both
natural and artificial cultures :— .

MYCELIUM.

Cultures made from both the primary and the secondary
conidia preduce a sparse, white mycelium, which soon
turns a gray-green, and whose filaments measure from lp
to 2u in diameter. These soon bear numerous conidia of
the Sporotrichum-type (pl. 7, f. 2), and about a week
later the stromata or stked heads of Graphium appear
(pl. 7, £. 1). The heads are white, changing to a green
or greenish yellow, and measure with the mucilage drop
from 30p to 100u. The stalks vary from a yellow to a
dark olive or a brown color, and measure from 3800p to
500p in length, and from 10, to 40» in diameter.

SECONDARY CONIDIA.

The secondary conidia are obovate to elliptical in shape,

and average 7.84 by 3.4m (pl. 7, f. 3). They are borne, '

88 MISSOURI BOTANICAL GARDEN.

y unlike Sporotrichum, on simple erect not prostrate hyphae.
This stage nevertheless is apparently identical with Sporo-
trichum vellereum Sacc. and Speg. var. grisea Boul.

a - 2

PRIMARY CONIDIA.

TT? ey ee, Oe eee Se ree

The primary conidia are borne on branching hyphae in
stalked heads (pl. 7, f. 5), and average 7.74 by 3.4.
They are of a light green or yellow tint, and are identical
in shape and appearance with the secondary conidia.

ee ees

Graphium atrovirens n. sp.

C% In addition to Graphium smaragdinum on the red gum
wood a new species was found present in a number of
instances. This discolored the wood with a brown stain,

; similar to the latter species, and at first was not distin-

| guished from it, having the same color and appearance in

mass; but after a more careful study of the colonies on
agar, different forms of secondary conidia were found
pres“nt in two types of colonies, making it necessary to
separate this form from G. smaragdinum and giving good

* reasons to call it a different species. It is now named

| Graphium atrovirens with the following as its cultural
characters :—

MYCELIUM.

The growth of the colonies on agar plates is fluffy, as

compared with that of G. smaragdinum. This is due to the

formation of strands in the mycelium by the union of a

‘a number of parallel filaments, these growing upward and
often forming large branching masses, similar to those

formed by the mycelium of some species of Cerastosto-

mella. The filaments of the mycelium are hyaline at first,

later changing to dark green or olive, in the vicinity of the

stromata or heads. They measure 3u to 4” in diameter.

Those which unite to form the stroma usually measure but

FOOT Ra, a EPP PRP ME MOR eS Ree is ee AN Oa wy ae gt ge. cig oe 7 ron ee ee

CHROMOGENIC FUNGI WHICH DISCOLOR woop. 89

2u in diameter. The color of the stroma is dark green to
black when mature, but in growing colonies all variations
between hyaline and dark green may be found. The
stalks measure from 1.5 mm, to 3 mm. in height, and 8u
to 80u in diameter, being proportionately more slender
than those of every other species of Graphium described
inthis paper (pl. 8,f. 1). In this species, as in others, all
the forms of gradation between a head with a stalk com-
posed of one filament, and a stalk with many filaments may
be found. The difference between those bearing secondary
conidia and those bearing primary conidia consists mainly
in the absence of color in hyphae bearing the former.

SECONDARY CONIDIA.

The secondary conidia are borne in simple open clusters,
of the type of the conidia of Sporotrichum. They are
borne either terminally or along the sides of lengthened
hyphae (pl. 8, f. 2, 3). They are obovate to elliptical in
shape, measuring 42 to 5.54 by 1.6 to Qu. They are
hyaline, becoming guttulate when old. The clusters re-
main open for a number of days after they are mature, but
under very moist conditions fall together in rounded masses
on the hyphae. The conidia of G'. smaragdinum fall to-
gether as soon as mature, the clusters being much closer in
their formation (pl. 9, f. 9).

PRIMARY CONIDIA.

The primary conidia are borne terminally on the ends of
the branching hyphae in the stroma which form large,
mucilaginous heads. These, without the sheath of mucus,
are flattened oval in shape, (pl. 8, f. 1), white at first,
later gray in color. They measure 40, to 600, in their
greatest diameter, the latter being the measure of the
head including the mucilaginous sheath. The primary
conidia are hyaline, obovate to elliptical, and measure

: a es e's, Pr " yr

8
ee:
=
a

"

ee Ee Sr ee ee, Oe ee |

FP de es

ee eee

Ps .
er, ae oe SD Oe ee

ae Bee

ex ee

oe Se ee oD

~

—_— |
jules
4 i -
e
i
$ ag
ae
3
%
t : ,
:
-
a
4
>
by :

Lelie era

Ee TT ee Ce

90 MISSOURI. BOTANICAL GARDEN.

8.52 to 4.54 by 1.44 to 24. The stalks in this species
bearing the primary conidia are sometimes much branch-
ed, but this feature is dependent more upon the culture
medium and physical conditions than upon the inherent
specific tendencies. One difference between the stalks of
this species and those of G. smaragdinum is that in the
case of the latter they tend to be gregarious, and are more
apt to be swollen at the base.

B. Species oF GRAPHIUM WITH A SECONDARY CONIDIAL
STAGE UNLIKE SPOROTRICHUM.

1. Species with secondary conidia borne continuously and
terminally, falling at once into clusters.

Under this subdivision will fall at least two species,
Graphium smaragdinum and G. rigidum. <A third species
is still under culture but the work has not progressed sufti-
ciently for identification.

It is a very difficult matter to determine the manner in
which the secondary conidia are borne in these species, be-
cause as rapidly as the conidia are formed by abstriction
from the end of simple or branched hyphae they break off
and cling closely to the ends of the hyphae, being enveloped
in a thin coating of water or soluble mucilage which hides
the outline of the hyphae and conidia. It is probable that
there is sometimes more than one conidium attached di-
rectly to the end of each hyphae branch and that these
secondary conidia are but modified forms of those de-
scribed in division A. The conidial masses formed are
similiar in appearance to those of G. ambrosiigerum,
G. ewmorphum and G. atrovirense

It is also quite probable that in all species of Graphium
the primary conidia are formed in the head precisely as the
secondary conidia are formed in the open and that the two

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 91

forms of conidia are morphological equivalents. In most
of the species studied many gradations were found between
a head on a stalk of a single colored filament and heads on
stalks of two to many filaments.

Graphium smaragdinum (A. & 8.) Sace.

The wood of the red gum, Liquidambar Styraciflua L.,
when freshly sawn and piled up, is very rapidly stained by

the combined effects of a number of fungi. The most im-

portant among these are Ceratostomella, Graphium, Hor-
modendron, and Dematium. On the surface, especially on
the sapwood, the fruits of Graphium appear more fre-
quently than those of the other fungi, especially when the
boards are placed in a moist chamber. Gtraphiwm was ob-
tained from gum boards taken from a number of localities,
and in most instances a species was found which re-
sembled Graphium rigidum but differed from it in a num-
ber of its cultural characters. From the following char-
acters, taken from both natural and artificial cultures,
it corresponds more nearly to Graphium smaragdinum
(A. & S.) Sacc., and it is assigned to this species.

_ MYCELIUM.

Primary and secondary conidia, sown in agar plate cul-
tures, germinate in a few hours, and small, white colonies
are visible in a day. The mycelium is abundant, floccose ;
the filaments are septate, hyaline at first, later changing
to dark green, measuring from 2» to 4 in diameter. In
two days secondary conidia appear, borne terminally on
simple hyphae, (pl. 9, f. 9). In less than a week, on dark
colored portions of the mycelium, the stalks and heads or
stromata of the Graphium stage appear (pl. 9,f.8). These
are hyaline at first, changing as they age to dark green,
then to black at the base, shading off to a light gray-green

he WEE a bf. yt oY, vn at
fn dae ag Ni illics SA Se

92 MISSOURI BOTANICAL GARDEN.

at the top. The stalks vary in height from 1 to 2mm.,
averaging 1.5 mm., with anaverage diameter of 25u. The
heads, when enveloped in mucus, are globular and measure
from 40 to 600u in diameter; without the sheaths they
are flattened oval or fungiform, and measure 30 to 200.
They are colorless at first, but later change to gray or
green. When a stalk is broken off before maturity, it
often sends out small branches, either of single hyphae or
of clusters of hyphae, bearing conidia.

SECONDARY CONIDIA.

The secondary conidia are usually elliptical in shape, and
measure 3.6u by 1.8y, average. They are hyaline, color-
less, and are borne both terminally in close clusters and
on simple hyphae, and in a short time fall off and adhere
together, thus forming miniature heads (pl. 9, f. 9).

PRIMARY CONIDIA.

The primary conidia are elliptical in shape, and are borne
terminally on branched hyphae (pl. 9,f. 10). They meas-
ure 3.8u by 1.64, average. They are hyaline and colorless.
In some cases, where the upper portion of the stalk or
head has been broken off new stalks sprout out from the
region of the apex in the old one. Branches are also
formed without injury.

The fungus penetrates deeply into red gum boards, giv-

_ing-them a dirty appearance, rendering them unfit for

making the best grade of boxes, crates, etc. It permeates
the medullary rays and many of the larger and some of the
smaller vessels of the wood. The mycelium in the wood is
generally of a color varying from gray to brown or dark
green.

Graphium rigidum (Pers.) Sacc.

A species of Graphium was found discoloring the Sap-

os

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 93

wood of lumber made from Quercus rubra L. in Indiana.
This, when isolated in pure cultures, grew readily on the
sapwood of pine, oak, gum and ash, and on vegetable agar
media, also on potato, bean, and other vegetables in test
tubes. The fungus corresponds quite closely to the de-
scription of Graphium rigidum (Pers.) Sacc., with the
exception that it has in addition a conidial stage, the coni-
dial clusters taking the closed form. The following
emended description of this species is now given from
artificial cultures on wood and agar media: —

MYCELIUM.

Cultures of both primary and secondary conidia sown in
ordinary vegetable agar media germinate in a few hours,
and in two days begin to bear an abundant crop of
secondary conidia in clusters of a type similar to those
of G. smaragdinum, which collect as rapidly as they de-
velop into rounded, mucilaginous clusters (pl. 7, f. 7 7
The filaments of the mycelium are hyaline at first, but in a
few days form a brown pigment in the older portion, and
throw up stalked heads bearing the primary conidia. These
vary from a stalk consisting of two or three filaments with
a small globular or flattened oval head, to one consisting of
many, with a much larger head (pl. 7, f. 6).

SECONDARY CONIDIA.

Secondary conidia are borne on both simple and branch-
ing hyphae by abstriction of the ends, and as fast as they
are formed adhere together in masses resembling Cephalo-
sportum. Theyare colorless, thin -walled, and are usually
free from guttules and granules. They measure 3 to
4.5u by lw to 1.5u (pl. 7, f. 8).

94 MISSOURI BOTANICAL GARDEN.

PRIMARY CONIDIA.

Primary conidia are elliptical, resembling in appearance
and form the secondary conidia. They are hyaline, and
measure 3.54 by 1.5, av-
erage, in size (pl. 7, f. 9,
10). They are borne term-
inally on alternately branch-
ed, hyaline hyphae, and fall
together in large, spherical,
mucilaginous heads. The
heads vary in color from a
white to a dingy yellow.
They measure, with the
mucilaginous drop, 20 to
00m in diameter. The
stalks vary from gray to
brown or black, and measure
fron 1 mm. to 2 mm. in
length, and 10n to 40m in
diameter. They are usually
solitary at first, but later may
* become gregarious, or even

i branched, especially in older
G. Blampum, X60. cultures where a secondary
Stroma, with conidiophores.
growth has taken place.

In the study of this species several instances were noted
where hyphae bearing the secondary conidia branched
directly from filaments connected with the heads bearing
the primary conidia, thus giving additional proof of the
relation between the two forms of fruiting. (See figure).

2. Species with secondary conidia borne either in simple
clusters or in clusters of short branched chains.

Graphium aureum nD. sp.

The first species of Graphium differing in conidial forms

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 95

from the previous species was found on discolored boards
made from the white pine, Pinus Strobus L. This stained
the wood a light brown color, less intense than that made
by any other species of Graphium investigated, with the
exception of G. ewmorphum. The constant presence of
short chains in the formation of the secondary conidia was
a character that needed the most careful verification, since
it was possible that an admixture of Ceratostomella with
the Graphium had taken place. This form of conidia was
accepted as a fact only after careful study for six months,
with repeated dilution culturesof both primary and second-
ary conidia in numerous agar poured plates and test tube
cultures. ‘

Another difficulty lay in the resemblance of the fungus
to species of Stilbuwm, the stroma in its earliest stages .
being identical with species of this genus, owing to the ab-
sence of color. The same difficulty is to be found with
others of the species studied. This indicates that the sep-
aration of Graphium from. Stilbum on a color basis is not
a good one, and that there is no good dividing line between
the two genera. Should it be found that species of Stil-
bum have secondary conidia similar to those of Graphium,
a better basis for generic separation would be found in the
types of such conidia.

The following are the cultural characters of this species
which is now named Graphium aureum:—

MYcELIUM.

In colonies grown on agar plates either from primary or
secondary conidia there appear in two or three days second-
ary conidia borne singly or in branched chains from simple,
erect conidiophores. These conidia are like those of Sporo-
trichum and of Ceratostomella. The mycelium is hyaline
at first, changing to yellow, then to light brown. The fila-
ments measure from 2# to 3u in diameter.

96 ' MISSOURI BOTANICAL GARDEN.

SECONDARY CONIDIA.

The secondary conidia are borne at first in loose, open
clusters which, after standing for a few days in moist air,
fall together in rounded masses, adherent to the terminus
of the conidiophores (pl. 9, f. 6). They are hyaline,
obovate to club-shaped, and are not guttulate except when
old. They measure 4y to 8» by ly to 2x, averaging Sy
by 1.8z.

PRIMARY CONIDIA.

The stalks or stromata bearing the heads of primary coni-
dia are white at first, changing to a yellow color, and often
to a dark brown at the base (pl. 9, f. 5). They measure
504 to 7504 by 10uto 90u, and are composed of parallel
filaments measuring 1“ to 2» in diameter. These bear
conidia terminally which are held together by mucilage in
a flattened, oval head, which is white at first, then creamy
yellow, finally a light brown when old and dry. The pri-
mary conidia are hyaline, obovate, and measure 4p to 5u
by 1p to 2u (pl. 9, f. 7).

In cultures on rice and potato tubes tall Anthina-like,
sterile, branching stalks, or stromatal outgrowths of a
light brown color from 1 mm. to 4 mm. in height, are
formed. These are very plainly abortive fruiting organs.

Graphium album (Corda) Sace.

A species of Graphium was found growing on the wood
of thebeech, Fagus atropunicea (Marsh.) Sudworth. This
discolored the wood a brown color. It was found to be a
species resembling G‘. aureum, but sufficiently distinct to
be separated from the latter species. A culture was sent
to Mrs. Flora Patterson, Mycologist of the Bureau of Plant
Industry, Washington, D. C., who identified the fungus as
probably Graphium album (Corda) Sacc., a species which

Fi

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 97

is imperfectly described. The fungus is now assigned to
this species with the following emended description taken
from cultural characters: —

MYCELIUM.

The growth of colonies from both primary and secon-
dary conidia is much like that of G. aureum. The colon-
ies remain hyaline for some time; even the stalked heads
show an absence of color for several days after their ap-
pearance. Finally the mycelium assumes a gray color, often
with a green tinge, and at the same time the stromata as-
sume a darker color. The filaments measure 2y to 3p in
diameter.

SECONDARY CONIDIA.

The secondary conidia are of the type of branching
chains, resembling Ceratostomella (pl. 9, f.2, 3). They
are obovate to cylindrical, hyaline, and measure 4p to 6
by lz to 2u. The conidial clusters are open at first, but in
moist air fall together after a few days, forming rounded
masses on the ends of the hyphae.

PRIMARY CONIDIA.

The stalks bearing the heads are hyaline at first, chang-
ing to a gray, finally to brown or black at the base." They
are often much swollen at the base, and measure from
.3mm., to 2mm. in length by 30¢ to 300 in diameter. The
heads are large and showy, often much flattened, or even
recurved at the edges, and measure 20, to 600, in width
(pl. 9, f.1). Their color is white at first, changing later
to a creamy yellow color. The primary conidia are borne

continuously on branched hyphae from the ends of the

filaments in the head, falling off rapidly as they are
formed, into the mucous mass which surrounds the head.
They are hyaline, and measure 3z to 5y by 1p to 1.5, (pl.
Ot, 4).

7

re al ~ + met ee
; ‘ . %

ce ae:

eas Saas, ies ied

i Bist: Fev

ea
pare

y

Fs
Re
.

-

2 ale ia ie i

ie!

ae , ps ae %
i as
a eee

weg
bat
he halite wy

io

Be Pe Se Se
SN ae Mey tm Nie ee

a

>
Py ieee 4

1 nites
re

¥ . 5 .
Ee OY LP Bae Ome ent aie Raa eee PTR

ie
=

ee
heal
a
E
ta
hs
Rae
\

ee Ae eee a ee ee Poses!
? . % +.¥ 7 ee as cts
"ba ae; Ss Se <A.
« ree ee he

*

ee OEE A RE a Ps es

Pe CR AGT

98 MISSOURI BOTANICAL GARDEN.

2. HoRMODENDRON.

An intense black color in irregular blotches often oc-
curs on new pine, elm, ash, gum, oak, and other kinds of
lumber, when sawn under a moist condition, during warm
weather. A large number of such spots were examined,
and the fungi present isolated for investigation. Species
of Hormodendron, Hormiscium and Alternaria were found
present. Two species of Hormodendron were identified,
viz: HH, cladosporioides (Fres.) Sacc. and a species not
agreeing with any description previously published, and
which is here given the name /. griseum.

Inoculations of these two species were made upon pine,
elm, oak, ash and gum wood. Both species were found to
produce superficial black stains, but neither penetrates
deeply into hard woods. On the other hand, either species
penetrates deeply boards made from the red gum tree,
Liquidambar Styraciflua L. Under the microscope the
conidial clusters of the two species closely resemble each
other, but in cultures the gross appearance of the colonies
is constantly quite different, the former varying from a
yellow-green to a dark olive, or almost black, the latter
from a light gray to a color approaching black; other
differences were noted in the length of the conidiophores
and of the conidial chains.

Hormodendron cladosporioides (Fres.) Sace.

The species of /ormodendron most commonly found on
boards is described here from artificial cultures, and
assumed to be H. cladosporioides. Two other species oc-
curring on wood, which have been previously described,
but at alater date, are HZ. atrum Bon. and H. elatum
Hartz. From the description of these two latter species
it would be very difficult to separate the one from the

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 99

other, or either of them from WH. cladosporioides. The
description of our fungus might fit any of the three, but
as the last named species is the older, its name will hold
valid.

The following cultural characters were obtained from a
large number of cultures on agar media, most of which
were made from pine-decoction alone, or from the juices of
vegetables combined with a pine-decoction :—

MYCELIUM.

Conidia sown on agar plates germinate and produce
small, gray colonies visible to the eve in 24 hours. In
three to four days the colonies begin to send up erect
sporophores, bearing clusters of short septate branches,
which are often branched again and from the ends of
which branched chains of olive-colored conidia are formed
(pl. 10, f. 1). The younger portions of the colony
assume a yellow-green color, which rapidly changes to an
olive, and finally to an almost black color, while the sur-
face has a velvety appearance. The filaments are coarsely
granular in the older portion, and are often constricted at
the point of septation. They measure 2 to 8uin diameter.
The sporophores measure 100 to 400y in length, and 3u
to 4u in diameter; the branches are 1- to 2- rarely 3- sep-
tate, and measure 6p to 15y by 3y to dp.

CONIDIA.

The conidia are borne in simple chains of 2 to 6, rarely
more; they are not granular or guttate except when old,
and vary in shape from a blunt to a pointed oval. They
measure 34 to Tu by 2u to 4,. When mature, both the
hyphal branches and the conidia are of a dark olive color.

The work of Planchon* indicates tbat Hormodendron

* Planchon, Louis. Influence de divers milieux chimiques sur quel-
ques champignons du groupe des Dématiées. Annales des Sciences
Naturelles. Botanique. 113 1-256. pl. 1-4. (1900).

FRE ee PLC MAR Cp ae NT aT Migs EE ad Se BRIG R GA NR ch PY
+ , ve Oss Sav t ah ' Lon grea : ‘ ; ard
. ; 7

100 MISSOURI BOTANICAL GARDEN.

cladosporioides (Fres.) Sacc. and Cladosporium herbarum
Link are identical. In someof our cultures forms inter-
mediate between the two, and identical with the latter, were
present (pl. 10, f. 1), but in no case were septate conidia
noted, such as are given in the description of C. herbarum,
the septation being confined to the hyphae which bear the
conidia. If, however, the modified hyphal branches be
mistaken for conidia, it is easy to assume them to be the
septate conidia of Cladosporium. Time was not had to
apply all the culture methods and media of Planchon to
our species. Enough of the forms described by him in his
study of these two fungi have been found in our cultures
to verify his work; but the question still remains as to
whether he is correct in assuming the fungus to bea Olad-
osporium. This hinges upon whether the conidia-like
branches bearing the conidia may themselves function as
conidia; but even this might not settle the question, since
it is a well-known fact that in most of our imperfect fungi
any portion of the young mycelium when broken up may
be able to reproduce the fungus. This being the case, we
will for the present use the name Hormodendron here to
designate our plant.

Hormodendron griseum n. sp.

The other species of Hormodendron studied was found
growing on the wood of Liguidambar Styraciflua L., Pinus
echinata Mill., and Fraxinus Americana L. The colonies
of this species on both wood and agar media were of greater
height, and of a different color when mature, as compared
with those of H. cladosporioides. The mycelium is often
hyaline, or nearly so in many portions, giving the colony
a gray appearance, even though the conidial clusters are of
a dark olive color. These two species of Hormodendron
were grown carefully for nearly a year in parallel sets of
cultures, with no reversion of the one type to the other,

CHROMOGENIC FUNGI WHICH DISCOLOR woop. 101

which was looked for. \ The following description of
H. griseum is made from cultures grown on the same
media from which the characters of H. cladosporioides

‘previously given were obtained :—

MYCELIUM.

The conidia of this species germinate readily, and simi-
larly to those of the preceding species, but the colony is a
day or two longer in producing conidia than that of the
former species. The habit of fruiting is the same, but the
conidia are borne in longer chains. The mycelium is
coarsely granular, and thick-walled in the older filaments,
which measure from 3 to 10u in diameter. The sporo-
phores or fruiting hyphae measure 200 to 800y in length,
and 34 to 4% in diameter. The hyphal branches are in
two series, and are from 1- to 3- septate, measuring 6.
to 14 by 3z to 4uz.

ConrIpDIA.

The conidia are borne in branching chains of 2 to 10,
and are formed from the ends of the hyphae by abstriction
(pl. 10, f. 2). They are of a sooty color, pointed oval,
and measure 3 to 6u by 2y to 4p.

3. HormIscrum.

Hormiscium gelatinosum nu. sp.

In the study of the black stain of pine, elm and red gum
boards at least one species of Hormiscium was found caus-
ing a very intense black stain, especially in sapwood. It
is with some hesitancy that this fungus is named Hormis-
cium in view of the great resemblance of the mycelium on
agar plate cultures to some of the forms of Dematium pul-
lulans de Bary, found by Planchon* in his study of the

* Planchon, Louis. J. ¢c.

102 MISSOURI BOTANICAL GARDEN.

latter fungus, and related fungi. The finding of chaias, or
round conidia borne in chains, on short hyphae (pl. 11, f.
7), and the identification of the fungus as Hormiscium,
by Mrs. Flora Patterson, has led to the placing of the spe-
cies under the genus Hormiscitum. This form differs from
other species of Hormiscium in its gelatinous nature and
dimorphous conidia, and is now called Hormiscium gelati-
nosum, with the following cultural characters : —

The fungus grows readily on many kinds of media, es-
pecially upon agar media from vegetable decoctions, upon
various substances containing starch, and upon the sap-
wood of pine, oak, gum, and other trees. It makes only a
scanty growth on the heartwood. On potato and rice
tubes it covers the surface with a black, shiny growth of
slimy, matted filaments.

MYCELIUM.

When conidia of either form are sown on agar plates
the colonies which result resemble very much those of some ~
yeasts and bacteria. The growth of the colony is slower
than that of most fungi. The form of the colony is vari-
able. Where the conditions are not favorable the colony
is rounded, cream-colored, mucous, growing with a rough-
ened surface and a wavy edge. In more favorable condi-
tions, on better media, etc., the edge of the colony becomes
fimbriate or toruloid (pl. 11, f. 4). The shape of the
colony thus apparently depends directly upon the condi-
tions under whichit grows. The filaments of the mycelium
are fewer in relative proportion to the bulk of the colony in
the younger forms than in the older forms. In older col-
onies, near the edge, filaments predominate. They are
toruloid in form, the cells are very irregular in shape, vary-
ing from long, cylindrical, almost hyaline cells (pl. 11, f.
7), to beaded forms (pl. 11, f. 8) of a dark olive color.
The former vary in diameter from 2» to 82; the latter,

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 103

from 52 to 10z. The latter form are of the nature of
chlamydospores, having thick walls, and from one to two
cells. These sometimes germinate, and apparently may
function as resting spores.

CONIDIA.

Two forms of conidia are borne on the mycelium, one of
these being the hyaline form (pl. 11, f. 5), which buds
both «from the cells of the mycelium and from other conidia
during the earlier growth of the colony. These are deeply
imbedded in mucilage, and germinate readily, if not too
old, in agar plate cultures. They are ephemeral, losing
their power of germination ina few weeks. They are
elliptical in shape, and measure 8. to 12, by 34 to dz,
and are rarely septate. Among the hyaline conidia, as the
colony grows older, brown forms of the same shape soon
occur (pl. 11, f. 6), and finally become very abundant.
These are thick walled, and do not appear to germinate as
readily as the hyaline type. They probably function as
resting spores. They are sometimes septate, but more often
have one cell. They measure 10, to 14, by 5p to 62.
Another form of conidia of the true type of Hormiscium
(pl. 11, f. 7) is found borne from the filaments in the
older portions of the extremities of the colony, and on the
filaments in cultures upon wood. These are globose in
form, of a dark olive color, not rugose, and are borne from
short hyphae in chains of two to several. They measure
from 7to 12, in diameter. These chains hold together
tenaciously, and do not break apart like those of related
genera.

4, OTHER WoOOD-BLACKENING Funai.

A number of other fungi were noted, which add to the
stains which occur superficially on lumber, especially on the
sapwood. Several of these were identified, which ap-
parently discolor the wood only by the color present either

cic ese eas

eT wheke soe See a ae hia oP, Ps Ee AL lth ae I CR aiid skh TR,
tier: . : BAY idk ? ' 4

104 MISSOURI BOTANICAL GARDEN.

in the mycelium or the fruiting bodies. None of them are
considered as important in their effects upon the value of
the lumber, asthe mycelium does not penetrate deeply into
the wood. Those identified were: Alternaria tenuis Nees,
Stachybotrys aliernans Bon., Chaetomium sp., Aspergillus
niger, Stemonites sp., and Gliocladium sp.

III. WOOD-REDDENING FUNGI.
1. PEeniciuurum.

The fact that Penicillium might stain wood was brought
to our attention during the year 1903. Some cultures of
Ceratostomella exigua were contaminated with a lemon-
colored fungus, which, under certain conditions, stained the
wood an orange red color. Upon isolation it was found to
be a Penicillium, which was dimorphous, having a fertile
mycelium with gray-green colored fruiting clusters, and a
sterile mycelium which assumed either a lemon color or an
orange red color, varying with the reaction of the medium
upon which it was grown, the former color being present
on acid media, and the latter on alkaline.

This led to the study of Penicillium from various
sources, and from different kinds of wood. Some were
found to stain pine sapwood, under certain conditions, an
orange red color, others a crimson red, and still others, a
color intermediate between the two. At least three species
of Penicillium were found which were proven to stain
wood. These were often intermingled, sometimes with
each other, and at some times with Fusarium. This com-
plicated the problem, and the difficulty was further in-
creased by the fact that it is hardly possible to name cor-
rectly the color-producing species of Penicillium without
along series of cultures. The investigations of Dierck *

* Dierck, F. Essai de révision du genre Penicillium. Ann. de la Soc.
de Brux. 1901.

Be
et

CHROMOGENIC FUNGI WHICH DISCOLOR woop. 105

and that of Thom * have shown that even when the form

characters of two forms of Penicillium are alike, there
may be distinct differences in the color of the fungus on
certain different kinds of media. The former has made a
revision of the genus Penicillium, based chiefly on physio-
logical characters, but also taking into consideration the
strictly morphological. The latter in studying the species
of Penicillium which were found concerned in the ripen-
ing of the various kinds of cheese found great difficulty in
naming them, owing to the fact that what may be mor-
phologically the same species may not have the same effect
on the chemical properties of freshly made cheese, owing
to varying action of different strains or races of Penicillium
on the organic compounds present.

Penicillium aureum Corda.

Of the two species of Penicillium studied, the most
common was identified finally as Penicillium aureum Corda.
It is dimorphous, as has already been mentioned, and is
one of the common species found on pine wood. It was
also found in cultures from gum, oak, and other woods.
A culture sent to Mrs. Flora Patterson was identified as
P. aureum, thus verifying our observation.

This species grows readily on the fresh sapwood of our
most common forest trees, upon bouillon, and various
vegetable decoction agar media, upon potato, starch, and
other similar substances. The following cultural characters
are taken from both natural and artificial cultures.

.

MYCELIUM.

Cultures upon agar media develop rapidly from conidia,
the mycelium usually becoming visible within a day as a

* Thom, C. Some suggestions from the study of dairy fungi. Jour.
Myc. 11: 117-124. (1905).

106 MISSOURI BOTANICAL GARDEN.

small, white patch on the surface. In two or three days,
sporophores are thrown up from the prostrate mycelium.
These bear branching hyphae which form terminally by
abstriction, simple, continuous chains of conidia. This
portion of the colony is usually gray to blue green in color,
varying with the medium. The filaments of the mycelium
in rich agar media often bear swollen cells, resembling
chlamydospores in shape (pl. 11, f. 2), but which have
not been observed to form the thicker wall, characteristic
of such spores. Under the same condition, the mycelium
becomes fluffy, and many filaments unite together in
strands, forming Coremium clusters. The filaments ordi-
narily measure 3u to 4u in diameter, and contain elongated
cells. On the other hand, the swollen cells are spherical,
with a diameter of 4y to 8y.

Upon older cultures, especially upon wood, there are
formed rounded tufts or clumps of curled and distorted
hyphae and filaments, which are either lemon-yellow or
orange-red in color, varying with the acidity or alkalinity of
the culture medium. The color is due to the secretion of
a pigment in the form of granules on the surface of the fil-
ament, evidently by exudation (pl. 11, f. 3). This pig-
ment is readily soluble in slightly acid or alkaline water,

-in hot alcohol, and in some other solvents. It is red when

alkaline and yellow when acid. The fungus, as a rule,
when grown artificially, overcomes a moderate amount of
alkalinity by the formation of an acid, and as a result,
most cultures assume the lemon-yellow color when old. In
cultures on wood, especially sapwood, the pigment is car-
ried into the wood cavities by absorption, thus staining the
wood. The color, although bright at first, fades as the
culture grows old and dries out.

CONIDIA.

The conidia are borne on conidiophores which vary in
length from 1004 to 5004, with a diameter of 3p, or slight-

ee, Se
meee coats ieee pee rae Maas » Pelee

CHROMOGENIC FUNGI WHICH DISCOLOR WOOD. 107

ly more. These bear terminally a whorl of branched hy-
phae. The hyphal branches of either the first or second
series average 124 by 2. The conidial cluster is loose, and
spreads out widely, averaging 100p to 350p by 354 to 70u
(pl. 11, f. 1). The conidia are pointed oval in shape,
varying from 3 to 4uin length, by 1.54 to 2p in diameter
They are borne in simple chains of 40 to 80. They have
a blue-green color, and contain a pigment soluble in hot
alcohol, which does not change its blue-green color either
by the addition of an acid or alkali.

Penicillium roseum Link.

A species of Penicillium which resembles Penicillium
roseum Link was first found mingled with P. aureum.
This fungus secretes a crimsonred pigment which discolors
sapwood. The work of investigation is not completed,
and it is thought best not to give further details concern-
ing this fungus at present. A third species has not been

identified.

2. FUSARIUM.

The species of Fusarium, identified as Pusarium roseum
Link, was isolated from pink, red and violet blotches found
on new pine lumber taken from lumber piles near Ashland,
Wis., in 1905. This grew readily under artificial culture, and
upon inoculation was found to be the cause of the blotches

upon the lumber. The boards from which the original
cultures were taken were sawn from trees of Pinus Strobus
L., and P. resinosa Ait. Similar stains have been found
on pine boards from the southern United States made from
Pinus palustris Mill., and Pinus echinata Mill., also on
pieces of wood from Pinus Virginiana Mill.

This fungus, like other species of Fusarium that have
been studied, varies in form, color, and babit of growth,
depending upon the substratum upon which it grows. The

108 MISSOURI BOTANICAL GARDEN.

fruiting bodies are usually borne at or above the surface of
the a upon which it is cultivated. The following is
a résumé of the cultural characters of the fungus :—

MYCELIUM.

The spores of the fungus germinate in a very short time,
usually a few hours, and in thirty-six hours small white
colonies are visible upon the surface of agar media. The
first mycelium that forms is rather sparse, but thickens as
the colony grows older. Two types of conidia are borne
on the fertile hyphae, which are formed in a short time
after germination of the spores. These are the micro-
conidia (pl. 12, f. 1, 2,3), which are found very numerously
in all parts of the mycelium, except in the newest growth,
and the macroconidia (pl. 12, f. 4, 5), which are found
sparsely only in the older portions of the mycelium.
Chlamydospores are formed abundantly i in all the older
portions of the colony about the time the growth of the
mycellum is checked by a limitation of nutrition, being
formed usually with or a little later than the macroconidia.
The anastomosing of filaments is quite frequent (pl. 12,
£..8).

MICROCONIDIA.

Microconidia are formed in great numbers on various
vegetable media, such as agar media made with plant decoc-
tions. They are usually borne on hyphae which branch
alternately from the mycelium. They are formed by the
abstriction of the ends of hyphae, and where they are not
disturbed by air currents or some sudden shock, they
adhere to each other side by side in clusters which assume
a form of fruiting resembling Cephalosporium (pl. 12,
f. 3). These microconidia are at first one-celled. A sep-
tum is often formed before the spore becomes detached
from the hypha, and later many of the one-celled conidia
that have fallen off form septa. The two-celled form

CHROMOGENIC FUNGI WHICH DISCOLOR woop. 109

slightly resembles the conidia of Cephalothecium (pl. 12,
f. 3).

The microconidia are oval to elliptical in shape, varying
in length from 8y to 14.54, and in diameter from 3p to 6p.
They are hyaline, colorless, thin walled, and are usually
uninucleate. Many are slightly curved.

MAcROCONIDIA.

The macroconidia are of the Fusarium type, and vary
considerably in shape and size. They are usually formed
by abstriction of the ends of short, swollen, branching
hyphae, and are from two- to four-celled, straight or
curved, with tapering, rather blunt ends, varying in length
from 19 to 30u, and in diameter from 3.54 to 64. Both
microconidia and macroconidia germinate readily on all
ordinary agar culture media, the time varying with the
medium. The macroconidia are rarely found attached to
the hyphae when fully mature.

CHLAMYDOSPORES.

Both terminal and intercalary chlamydospores are
formed singly or several in series (pl. 12, f.6,8,9). They
are formed by the enlargement of the cells of the mycelial
filaments. Their development is more gradual than that
of the conidia, and the cell wall is much thicker. In shape
they are spherical, or slightly oval. The cell contents are
granular, and vary in color from yellow to dark brown.
The diameter of the spores averages 124. When the
macroconidia start to germinate, and are hindered by a sud-
den drying out, or other adverse conditions, it is not unus-
ual for one of the cells to form a chlamydospore (pl. 12,
f. 5); these are rarely mature, and have never been
observed to germinate.

SCLEROTIA.

Sclerotia are produced in cultures on boiled potato and

110 MISSOURI BOTANICAL GARDEN.

rice. They vary in color from dark green to dark brown.
As they grow older they become rather hard and dense,
but bear no definite fruiting bodies.

PERITHECIA.

In all of our cultures no perfect form of fruiting has

been found, although the fungus has been grown on all the

common culture media. If such form exists, our failure
to find it is due to the fact that natural conditions were
not imitated closely enough to produce it.

THE FORMATION OF COLOR.

’ On pine sapwood, a red or purple color is produced,
which in old cultures fades to a dirty brown. A number
of species of Fusarium are now being grown on pine wood
for comparison with this, and the results will be published
ata later date. Musarium roseum secretes a soluble pig-
ment which is taken up by the adjacent wood cells, stain-
ing them lightly with a red or purple color, which varies
with the acidity or alkalinity of the wood. In natural
wood, which is slightly acid, the color varies from a pink
to a lilac. In cultures where there is a profuse growth of
the mycelium the colored chlamydospores and mycelium
modify the appearance of the wood, giving it a dull color
in contrast with the original brighter one.

CAUSES OF COLOR IN WOOD STAINED BY THE FUNGI
INVESTIGATED.

1. CERATOSTOMELLA.

The cause of the blue-gray or blue-black stain in wood
penetrated by the mycelium of Ceratostomedla lies in the
color of the filaments of the fungus, which exudes no stain
from its mycelium. The dark brown pigment in the walls

Ay.

ee eee ee ee a ae oe 4 Age ‘«

seed

CHROMOGENIC FUNGI WHICH DISCOLOR woop. 111

of the filaments is insoluble in alcohol, ether, chloroform,

benzol, alkalis and acids. The brown color of the fungus
apparently contains traces of a blue pigment whose color is
transmitted by the wood cells of pine more readily than the
brown color.

2. GRAPHIUM.

The mycelium of Graphium imparts a dingy gray or
brown, or even a black color to the wood it enters. This is
caused by the color of the mycelium, and is due to no solu-
ble pigment. The cells of the wood remain unstained.

3. HoRMODENDRON AND HorMIScIUM.

The cells of wood penetrated by these wood-blackeners
are not stained, the color being due to the presence of the
mycelium of the fungi. It is not yet known whether the
color is one that may be extracted.

4, PENICILLIUM.

The species of Penicillium discoloring wood form a
soluble red or yellow pigment which is taken up by the cell
walls of the wood, giving them a red or yellow stain which
fades when the wood dries out, but is increased in intensity
when it is moistened again.

5. FusARIUM.

The wood stained by Fusarium is discolored both by a
soluble pigment which is secreted by the fungus and taken
up by the wood cells and by the presence of colored hyphae
and chlamydospores.

The investigation of color production by these fungi will
be carried on in the future, and it is hoped that the nature
and composition of some of these pigments may be discov-
ered, and that other facts may be found worthy of publica-
tion.

112 MISSOURI BOTANICAL GARDEN.

KEY TO THE SPECIES OF FUNGI DESCRIBED IN THIS PAPER.

A. Colonies at first white, changing color later. First fruiting stages,
clusters of hyaline conidia borne on upright hyphae which usually col-
lect together in rounded masses under moist conditions.

a. Staining wood bluish, black or brown; with beaked perithecia.
CERATOSTOMELLA.
1. Conidia borne in short, branching moniliform chains on upright
hyphae.
* Beaked ostiolum more than twice the height of the perithe-
cium, ;
+ Fringe to ostiolum terminal.
§ Terminal filaments short and often thickened.
Perithecium smooth or sparsely hirsute. C. pilifera.
Perithecium often with outgrowths.C. Schrenkiana.

e Perithecium with glandular hairs. . Q. echinella.
1 §§ Terminal filaments long and slender.
Sat C. capillifera.
t Fringe to ostiolum often supplemented by additional rings
7 beneath. C. pluriannulata.
Bs t ** Beaked ostiolum only once or twice the height of the peri-
é thecium.
a Terminal filaments short and thickened. C. minor.
ee - Terminal filaments lengthened and slender. _ C. exigua.
ae 2. Conidia borne continuously, either singly or in moniliform chains.
. Perithecium with conical spines. C. moniliformis.
4 b. Staining wood gray to brown or black; mycelium bearing stromata
ie or stalked heads. GRAPHIUM.
F 1, Secondary conidia resembling those of Sporot~ichum.
Bee Stroma finally brown to black, heads white to brown.
G. ambrosiigerum.
Stroma finally brown or olive, heads greenish. G. eumorphwm.

Stroma finally dark green to black, heads gray. G. atrovirens.
2. Secondary conidia differing from those of Sporotrichum.
_ * Conidia formed continuously, falling together as rapidly as

formed.
Stroma finally dark green to black, heads white to gray or
green. G. smaragdinum,
Stroma finally gray to black, heads white to creamy yellow.
; G. rigidum.
** Conidia in simple whorls or in whorls of*branching moniliform
ih chains,
: Stroma finally yellow to brown, heads yellow to light
iy . brown. G. aureum.
) z Stroma finally brown to olive, heads white to yellow.
G. album.

CHROMOGENIC FUNGI WHICH DISCOLOR woop. 113

B. Colonies at first gray, later brown or black, conidia dark colored,
borne in branched chains in clusters from upright hyphae,
HORMODENDRON.
Velvety colonies, yellow-green at first, then olive to black.
H. cladosporioides.
Furry colonies, light to dark gray, finally almost black. H. griseum.
C. Colonies yeast-like at first, mucilaginous, finally turning brown or

black. HorMIscIum.
Mycelium cream-colored, then brown, conidia dimorphous, chlamy.
dospores present, HI, gelatinosum.

D. Colonies gray, green, yellow or red, conidia hyaline or slightly
colored, borne in clusters of simple chains from branched upright
hyphae. PENICILLIUM.

Mycelium dimorphous, lemon-yellow to orange in one form, gray or
blue-green in the other; staining wood yellow or red. P. aureum.

Mycelium white or pink to red; staining wood crimson.
; P. roseum (2)

E. Colonies white, often changing to red or lilac, microconidia one- to
two-celled, macroconidia several celled, fusiform or sickle-shaped.

FusARIUM.
Staining wood pink, red, or purple, or even brown in old cultures,
F. roseum,

EXPLANATION OF PLATES.

The illustrations are from camera lucida drawings by the
author except for pl. 6, f. 7, 8, drawn by Miss Laura
L. Eames. Throughout, perithecia and their tips, and
stromata, are enlarged 50 diameters; and mycelial struc-
tures, conidiophores, conidia, and ascospores are enlarged
500 diameters.

Plate 3. — Ostiola of Ceratostomella. 1. C. capillifera. 2. C. exigua.
3. C. echinella. 4. C. minor. 5. C. moniliformis. 6. C. Schrenkiana.
7. C. pluriannulata. 8. C. pilifera.

Plate 4.— Ceratostomella Schrenkiana. 1, 2. Conidial clusters. 3.
Ascospores. 4. Young perithecium, with conidiophores from some of
the older hyphae.— C. pilifera. 5,6. Conidial clusters. 7. Germinat-
ing conidia. — Cephalosporium-like aggregations are shown in figures 1,
5, and 6.

Plate 5. — Ceratostomella pluriannulata. 1. A conidial cluster. 2.
An ascus and ascospores.— C. moniliformis. 8. Conidial clusters. 4.
A moniliform chain of conidia formed like arthrospores, two conidia
germinating. 5. A. Cephalosporium-like aggregation of conidia termi-

8

114 MISSOURI BOTANICAL GARDEN.

nating two moniliform chains. — C. minor. 6. Perithecium almost
mature. 7. An ascus and ascospores. |

Plate 6. — Ceratostomella echinella. 1. A conidial cluster. — C. capil-
lifera. 2. A conidial cluster. 3. Cephalosporium-like aggregations of
conidia. — C. exigua. 4, 5. Conidial clusters, one having become
massed. _ 6. Conidia. 7, Ascospores.

Plate 7. — Graphium eumorphum, 1. Stroma with heads. 2. Clusters
of secondary conidia. 3. Secondary conidia. 4. Parallel hyphae com-
posing a portion of a stromatic stalk. 5. Branches of hyphae bearing
primary conidia from head. — Graphium rigidum. 6, Stroma with head.
7. Conidial masses on surface of an agar culture, secondary conidia.

'. g, Primary conidia. 10. Germination of primary conidia.

’ plate 8. — Graphium atrovirens. 1. Stromata with heads. 2. Clusters
of secondary conidia. 3. Compound cluster, or side clusters of second-
ary conidia. — Graphium ambrosiigerum. 4, Stroma with heads. 5. A
cluster of secondary conidia. 6. Stroma, origin of stalk bearing head.
7. Filament bearing primary conidia from head.

Plate 9. — Graphium album. 1. Stromata with heads. 2%. Clusters of
secondary conidia. 3. Compound cluster of secondary conidia. 4.

.Primary conidia.— Graphium aureum. 5. Stromata with heads. 6.
Conidial clusters and masses of secondary conidia. 7. Primary conidia.
— Graphium smaragdinum. 8. Stromata with heads. 9. Clusters of
secondary conidia. 10. Filaments bearing primary conidia from head.

Plate 10. — Hormodendron cladosporioides. 1. Conidial clusters. — H.
griseum. 2. Conidial clusters, the smaller showing a Cladosporium

“type.

Plate 11. — Penicillium aureum. 1. A portion of a conidial cluster with
two complete chains of conidia, the remainder being incomplete and the
conidiophore drawn in two pieces. 2. A portion of the mycelium from a
culture of pine-decoction agar showing swollen cells suggestive of
chlamydospore formation. 3. A filament rugose with a deposit of a
colored granular exudation. — Hormiscium gelatinosum. 4. A colony on
agar showing the toruloid form, (enlarged 6 times). 5. A hypha with
hyaline yeast-like conidia. 6. Brown form of loosely attached conidia.
7,8. Conidia in chains borne on short hyphae, the Hormiscium type. *

Plate 12.— Fusarium roseum. 1, 2. Microconidia, Verticillium-like
forms. 8. Microconidia aggregated in Cephalosporium-like masses and
at the same time becoming uniseptate. 4. Microconidia one- to two-
celled, macroconidia three- to four-celled. 5. Formation of abortive
chlamydospores caused by the arrested germination of macroconidia.
6, 7, 8. Chlamydospore formation. .

PLATE 3.

Rept. Mo. Bot. GARD., VOL. 17.

CERATOSTOMELLA.

REpT. MO. BOT. GARD., VOL. 17. PLATE 4.

he eI j
ee

eee |
SS
|

CERATOSTOMELLA.

REPT. MO. BOT. GARD., VOL. 17, PLATE 5.

CERATOSTOMELLA,

REPT. MO. Bor. GARD , VOL, 17. PLATE 6.

CERATOSTOMELLA.

REPT. MO. BOT. GARD., VOL. 17. PLATE 7.

GRAPHIUM.

PLATE 8.

REpT. MO. BOT. GARD., VOL. 17.

GRAPHIUM.

PLATE 9.

REPt. MO. BOT. GARD., VOL. 17.

GRAPHIUM.

PLATE 10.

REptT. MO. BOT. GARD., VOL. 17.

HORMODENDRON.

PLATE ll.

REPT. MO. BOT. GARD., VOL. 17.

PENICILLIUM AND HORMISCIUM.

ee
is ye

REPT. MO. BOT. GARD., VOL. 17.

FUSARIUM ROSEUM.

ZONATION IN ARTIFICIAL CULTURES OF CEPHALOTHECIUM
AND OTHER FUNGI.

BY GEORGE GRANT HEDGCOCK.

The results given in this paper are from a continuation
of the investigation, mention of which was made in a brief
paper to the American Mycological Society at the Phila-
delphia meeting, December, 1904, and are published by
permission of the Secretary of Agriculture.

Certain fungi like Cephalothecium, Penicillium, Mucor
etc., often exhibit in artificial cultures a distinct zonation
of the mycelium due to denser masses of spores being
formed daily on certain portions of the mycelium, often
giving an agar plate culture the ringed appearance of a
target. Last year it was proven by a set of experiments
that daily variations of temperature are not the cause of
this zonation. At the same time it was shown that they
were not produced under conditions of total darkness, but
occurred only in cultures grown in the light.

In order to determine what colors of light affect spore
formation, cultures of . Cephalothecium, Penicillium,
Mucor, and Hormodendron were grown on agar plates
under five conditions of light. Double bell jars of ordinary
glass were filled with solutions as follows: for orange light,
potassium bichromate; for red, a cochineal solution; for
blue, a mixture of ammonium and copper carbonate; for
green, a mixture of * anilin gelb’’ and ‘* licht griin;’’ for
ordinary light, a double bell jar without a solution; and
for darkness, a tight, black-lined box. No spectrum anal-
ysis of the rays of light transmitted by some of these solu-
tions has yet been made.

The cultures grown under red and orange light, and in
darkness, exhibited a uniform dense spore formation over
the whole surface of the mycelium in every instance.

The cultures grown under blue light and in ordinary

light, in every instance exhibited distinct daily rings of
(115)

116 MISSOURI BOTANICAL GARDEN.

growth of alternating denser spore formation of a texture
similar to that of the former set of cultures; but between
the denser concentric rings were regions where spores were
only sparsely formed.

Cultures under green light exhibited rings of growth of
less distinctness, indicating that it is the blue rays of light
that affect spore formation. Careful observation estab-
lished that the rings of sparse spore formation are formed
in the daytime, and the denser at night, proving that the
blue rays of light inhibit spore formation in these fungi.

Cultures of Cephalothecium.taken from decaying sugar
bect roots and from rotting apples were much more sensi-
tive to blue light than the same fungus taken from the
fruits of Rosa rugosa. The species from all these sources
was one and the same, if mycological characters were to
decide the question; yet, the difference in growth on vari-
ous media under blue light suggests that we have two dis-
tinct physiological forms of Cephalothecium roseum. The
form occurring on apples and sugar beets grows more
often in the dark in nature, in stored apples and beets; the
one on the fruits of the rose is found growing on the rose
fruits in open light. May it not be that the latter has
become more accustomed to strong light, and is less af-
fected by the blue rays?

There is another type of zonation in artificial cultures of
some fungi which is not due to light stimuli, but probably
to variations in the amount of food taken in by the my-
celium, and to possible resting periods in spore formation.
This type has been observed by the writer in a number of
species of fungi; among these is Hypocrea. Thomas
Milburn mentions this type of ring formation in his studies
of Hypocrea rufa and other fungi.* In Hypocrea the
concentric rings are formed, not daily, but in a much
' longer period of time.

* See Centralbl. fiir Bakt. Par. u. Infek, 18: 129-138.

ZONATION IN ARTIFICIAL CULTURES. 117

‘EXPLANATION OF PLATES.

Plate 13.— 1. Penicillium album, grown in the light. 2. Cephalothecium
roseum from sugar beet roots, grown in light. — Natural size,

Plate 14.—1. Cephalothecium roseum, from apple, grown in dark-
ness. 2. Same, grown in light.

Plate 15.—1. Cephalothecium roseum, from Rosa rugosa fruits,
grown in darkness. 2. Same, grown in orange light. — Colonies enlarged.

Plate 16.— 1. Cephalothecium roseum, from Rosa, grown in blue
light. 2. Same, grown in natural light. — Colonies enlarged.

REPT. MoO. BOT GARD., VOL. 17. PLATE 13.

PENICILLIUM anp CEPHALOTHECIUM.

REPT. MO. BOT. GARD., VOL. 17. PLATE 14.

CEPHALOTHECIUM ROSEUM.

REPT. MO. Bot. GARD., VOL. 17. PLATE 15.

CEPHALOTHECIUM ROSEUM.

tEPT. MO. BOT. GARD., VOL. 17. PLATE 16,

CEPHALOTHECIUM ROSEUM.

SOME NEW TEXAS PLANTS.

BY B. F. BUSH.

TRACYANTHUS ANGUSTIFOLIUS TEXANUS D.. Var.

Foliage glaucous. Stems 7-15 dm. tall, simple, with
compound racemes at the top. Leaf-blades narrowly
linear, 2-6 dm. long, shorter above, flat; racemes ovate in
outline, compound, 10-20 cm. long; pedicels slender,
1-2 cm. long; perianth decidedly yellowish; sepals and
petals oblong, narrowed at the base, the sepals somewhat
broader, all 4-5 mm. long, acutish at the apex; capsules
not seen. ;

Sandy swamps, eastern Texas. Spring.

Differs very greatly in appearance from 7’. angustifolius
(Michx.) Small, in the large compound panicles, the more
robust size, and in the decidedly yellowish flowers.

Specimens examined: Texas: The only specimens seen
are those of the type, collected at Swan, Smith County,
by J. Reverchon 2782, May 16, 1902.

ALLIUM HYACINTHOIDES N. sp.

Bulbs usually several together in bunches, 2-3 cm. high,
their coats prominently fibrous-reticulated. Scapes terete,
2-3 dm. tall; leaves basal, channeled, 3-4.5 mm. wide,
much shorter than the scape; bracts of the umbel 3, ovate,
acuminate; umbel erect, many-flowered, rarely bulblet-
bearing; pedicels filiform, 12-15 mm. long; flowers pink,
very sweet-scented, 6-9 mm. long; perianth segments
oblong, obtuse, thin, with a reddish midvein, twice longer
than the stamens; filaments but little dilated below; cap-
sule 83-3.5 mm. high, 4—4.5 mm. wide, its valves not crested.

(119)

| al a yee TF, to es as ee ee a eee
PA Pia Pee tae ee En Bm Pe eg Ca eee Tener SOME pe cr tag gee ale aes
3

120 MISSOURI BOTANICAL GARDEN. .

Calcareous limestone prairies, central Texas. Early
spring.

A. hyacinthoides is quite readily distinguished in the
field by its very fragrant attractive pink flowers, very
early blooming, and habit of growing in bunches. From
A. mutabile Michx., to which it is related, and with which
it has been confused, it differs in being very fragrant, in
having shorter stems, rather longer obtuse sepals which
are twice longer than the stamens, smaller capsules,
shorter pedicels, different habitat, and an earlier season of
blooming. A. mutabile Michx. grows in sandy woods
about Dallas, Texas, and blooms from two to four weeks
later.

Specimens examined: Twxas: Dallas, Bush 270, April
14, 1900; Reverchon 2183, April 5, 1900; Reverchon
4025, April 1, 1880.

PSORALEA SUBULATA n. sp.

Acaulescent or nearly so, 1.5-3 dm. tall, from a shallow
seated, oval, farinaceous bulb. Stems densely pubescent
with long, shaggy white hairs, many-leaved and many-
flowered. Petioles 1-2 dm. long, much stouter and longer
than the peduncles, commonly considerably exceeding them
and mostly from the base of the stem; leaflets 5, digitate,
short-stalked, broadly obovate, with a cuneate base, obtuse
and rounded at the apex, cuspidate-tipped, very pubescent
beneath with long, shaggy white hairs, sparsely so above,
3.5-6 em. long, and 2-2.75cm. wide; peduncles from the
base of the stem, 5-10 cm. long, commonly reclining or
recurving; spikes many, oblong, dense, 5-8 cm. long,
2.5—-4 cm. thick, very many flowered; bracts lanceolate,
shorter than the calyx-tube, about 7.5 mm. or less long ;
calyx- teeth subulate, 6-10 mm. long; pod oblong, 4—5mm.
long, nearly smooth, tipped with a long, densely hairy sub-
ulate beak, which is 12-15 mm. long.

SOME NEW TEXAS PLANTS. 121

Dry sandy soil along streams, eastern and central
Texas. Summer.

P. subulata is allied perhaps more nearly to P. hypogaea
Nutt., butis abundantly distinct from that species in the
larger differently shaped leaflets, larger and longer spikes,
and different habitat, P. hypogaea Nutt. growing on the
rich prairies about Dallas. It is easily distinguished from
P. esculenta Pursh, by its nearly acaulescent habit, dif-
ferently shaped leaflets, larger and stouter petioles, differ-
ent habitat, and the long subulate beak of the pod.

Specimens examined: Texas; Dallas County, Bush 697,
May 10, 1900, type: Hempstead, Hall 1350, June 28,
1872.

PsoORALEA PALUSTRIS 0. sp.

Perennial from a deep elongated tuber. Stems 6-9 dm.
tall, slender, erect, simple, unbranched, strigose-hirsute ;
stipules linear-subulate; leaflets 3, the blades lanceolate to
narrowly oblong, strigose-hirsute, 4—7 cm. long, and 1-1.5
cm. wide, rugose-veiny, acutish; peduncles 5-10 cm. long,
as long as or longer than the subtending leaves, equalling
or exceeding the racemes in length; racemes 4—5 cm. long,
not elongating in fruit; pedicels of the flowers 3-6 mm.
long; tube of the calyx campanulate, thickly glandular-
dotted; corolla 6-9 mm. long, deep purple; bracts linear-
aristate ; pods suborbicular, apiculate by the style, 4-5 mm.
long, strongly transversely wrinkled, scarcely margined,
thickly glandular-dotted.

Sandy swamps, eastern Texas. Summer.

This species differs from P. pedunculata (Muhl.) Vail,
in the taller, unbranched stems, more strigose foliage,
larger and longer leaves, longer pedicels of the flowers,
lax racemes, shorter peduncles, and especially in the ©
marshy habitat, and in the large deep purple flowers which
are produced much later in the season, P. pedunculata

122 MISSOURI BOTANICAL GARDEN.

(Muhl.) Vail flowering more than a month earlier in the
same locality.

Specimens ecamined: Texas: Lindale, Reverchon 3175,
May 15, 1902, type; 3016, June 9, 1902; Big Sandy,
Reverchon 2658, May 19, 1901.

TRAGIA NIGRICANS N. sp.

Perennial, bright green, smooth and glabrous, turning
blackish in drying. Stem slender, erect, 2.5-4.5 cm. tall,
simple or few-branched ; leaves thickish, oblong or oblong-
lanceolate, deeply dentate or laciniate, sometimes lobed,
acute at the apex and base, short-petioled, or the upper-
most sessile, glabrous beneath, sparingly pubescent on the
midrib above, and copiously ciliate with long white hairs,
2.5-7.5 cm. long, and .5-2.5 cm. wide; flowers in lateral,
short, spike-like racemes from the upper axils; staminate
flowers with a 3-lobed calyx and 3 stamens; pistillate flow-
ers one or two at the base of the racemes, and with a 6-
lobed calyx; capsule short-pedicelled, somewhat depressed,
8-10 mm. in diameter, densely pubescent with long white
hairs; seeds subglobose, 3 mm. long, smooth.

Rocky woods on the Upper Hondo, central Texas.
Summer.

This very distinct and peculiar 7ragia is not referable
to any species known to me, but is most nearly related to
T.urens L., from which it differs in being smooth, and
in having few-flowered racemes, 3-lobed staminate calyxes,
and larger leaves.

Specimens examined: Texas: The only specimens seen
are those of the type, collected on the Upper Hondo, by
J. Reverchon 1594, June, 1885.

LOBELIA PUBERULA PAUCIFLORA D0. Var.

Stems 6-10 dm. tall, simple, very slender, densely white-

SOME NEW TEXAS PLANTS. 123

pubescent. Leaves elongated, oblong-lanceolate, entire or
glandular-toothed, rather few on the stems; flowers few,
3 to 9, scattered along the stem, 2.5-5 cm. apart; sepals
densely pubescent and bristly with long white hairs; co-
rolla azure blue, about 3 cm. long; tube more than twice
as long as the sepals; hypanthium densely hispid with long
bright white hairs.

Sandy swamps near the coast, Florida to Texas. Summer
and autumn.

This is evidently related to ZL. puberula Michx., but
differs conspicuously from that species in the taller, more
slender, simple stems, longer leaves, and especially i in the
fewer flowers, which are much larger.

Specimens examined: ‘TEXAS: Swan, Reverchon 3206,
September 17, 1902, type. Fiorma: Without locality
and date, Chapman. Lovuistana: Without locality and
date, Hale.

XANTHIUM BUBALOCARPON Nl. sp.

Stems very stout, much branched and spreading, about
one meter tall. Upper part of the stem and the petioles
very rough, covered with stout white papillae; petioles
5-15 cm. long, stout; leaf blades broadly triangular,
8-20 cm. wide, 3- to 5-lobed, dentate, very thick, rough-scab-
rous on both surfaces ; burs pubescent, ovoid, clustered in the
axils of the upper leaves, and in terminal racemose clusters,
2.5-4 em. long, including the short beaks, 2-3 cm. wide,
including the prickles, the short hispid beaks 6-8 mm.
long, nearly straight, hooked at the apex, equalling or a
little shorter than the very dense subulate, slightly uncinate
prickles, which are about one-third as long as the diameter
of the bur, and hispid nearly to the middle with yellowish-
brown hairs, the prickles and the pubescence usually
entirely concealing the body of the bur.

Prairies and barrens, western Arkansas and eastern

bg Aaa acest aaa Co vay itr ear

124 MISSOURI BOTANICAL GARDEN.

Texas, abundant, where it is called Buffalo Bur, whence
the specific name. Autumn.

This very distinct species is more nearly related to Y.
speciosum Kearney, but is easily distinguished from that
species by the much larger burs which are of an entirely
different shape.

Specimens examined: Texas: Dallas County, Bush
1185, September 29, 1900, type. Arxansas: Fulton,
Bush 2408, April 22, 1905, seedlings, with burs still
attached.

ANTENNARIA GREENEI n. sp.

Floccose-woolly, stoloniferous, fertile and sterile plants
in separate patches. Floweringstems stout, 1-2.5cm. tall;
basal leaves cuneate-spatulate, obtuse, cuspidate, distinctly
3-ribbed, long-petioled, silvery beneath, dark green above,
becoming nearly smooth on the upper surface, 5-6 cm.
long, including the long slender petiole, 1.5-2.5 cm. wide,
those of the stolons cuneate-spatulate, acuminate, apex
cuspidate, tapering from about the middle to each end,
thickly covered beneath with silvery tomentum, appressed-
pubescent above, strongly 3-ribbed, 8-13 cm. long, includ-
ing the long slender petiole which is as long as the blade or
longer, and 2.5-4 cm. wide; stem leaves sessile, oblong,
obtuse, or the upper acute, cuspidate; heads 6 to 9 in
corymbose clusters, 6-8 mm. broad; involucre about 6 mm.
high, the outermost bracts obtuse, the inner lanceolate,
acute, all brownish and greenish, and pubescent at base,
above with white tips; achenes minutely glandular. Stami-
nate plants about 1 dm. tall. Basal leaves similar to those
of the fertile plant; stem leaves mostly linear-lanceolate ;
heads smaller, 6-7 mm. broad, in a capitate cluster; bracts
oblong-spatulate, brownish and greenish and pubescent at
base, the upper dilated portion white, erose-dentate all
around the upper portion.

SOME NEW TEXAS PLANTS. 125

Damp, sandy woods along the Trinity River, central
Texas. Early spring.

The species here proposed is most nearly related to A.
occidentalis Greene, but is distinguished from that species
by the exceedingly long-petioled leaves, which are about
twice the length of those of that species, and much more
pointed at apex.’

Specimens examined: Texas: Dallas County, Bush
644, May 7, 1900, type; Reverchon 330, April 1, 1900,
type; March, 1880, type.

Smtpaium REVERCHONI n. sp.

Stems apparently tall, very hispid with long white hairs

which arise from papillae, much branched, the branches

elongated. Leaves alternate; blades lanceolate to linear-
lanceolate, 5-10 cm. long, dentate-serrate or the uppermost
nearly entire, both faces pilose with long white papillate
hairs like those of the stem; lower leaves long-petioled,
ovate-lanceolate, acute, sinuate-dentate or entire, with
dense short pilose hairs which arise from papillae ; petioles
slender, terete, with long white papillate hairs; heads
showy, very large; involucres campanulate, 2.5 cm. high;
bracts lanceolate, acute, ciliate, densely papillose-hispid on
the back, the outer ones 3 cm. long, the inner shorter,
obtuse; rays about 20, 3-4 cm. long, hispid on the nerves,
and ciliate; achenes not seen.

Dry sandy soil, eastern Texas. Spring.

Clearly related to 8. Gatesit Mohr, but is abundantly dis-
tinct in the larger heads, longer papillose-hispid hairs,
different type of leaves and different root-leaves.

Specimens examined: Texas: Troupe, Reverchon 3342,
May 8, 1902, type; Lindale, Reverchon, June 9, 1903.

eee ee

ASCIDIA IN GASTERIA AND AGAVE,

BY J. ARTHUR HARRIS.

In the summer of 1905 Mr. C. H. Thompson, in charge
of the collection of succulents at the Missouri Botanical
Garden, called my attention to a peculiarity of two of
several young plants of Gasteria grown from seed received

2. G. BRACHYPHYLLA, X 1.

1. G. BRACHYPHYLLA, X 1. 3. G. BRACHYPHYLLA, X 1.

from the botanical garden at Palermo under the name G.
brachyphylla.

The general appearance of the abnormal plants is shown
in figures 1 and 2, while for the sake of comparison a nor-
mal specimen from the same lot is shown in figure 3. Fig-
ure 1 represents the anomaly atits best. The four earlier

(126)

ie,

ASCIDIA IN GASTERIA AND AGAVE. 127

leaves are of the normal form, while the remainder of the
plant is composed of an ovate organ of the same texture
as the leaf, but continuous throughout except for a small,
irregular opening at the apex through which a younger leaf
is visible. The first impression was that the structure in
question consisted of an ascidium formed by the fusion of
two leaves as has been described for Crassula arbores-
cens,* but Mr. Thompson, upon

looking through the rather large

series of young plants then grow-

ing, found two interesting forms,

one of which is represented in

figure 4, which were cultivated

under the names G'. verrucosa

and G. trigona. These seem to

show a folding inof one or both

margins of the leaf in a way

which might indicate that the eg SRE ah +
open calyptra originates from a single leaf developing
from a primordium extending entirely around the vegetative
point of the axis. In these two cases the tip of the leaf
exhibits two apiculae and the same is apparently true for
the plant represented in figure 1, though here the irregular-
ity of the opening at the apex precludes a final decision.

A transverse section near the
base of the ascidium shows it to
be composed of a layer of tissue
slightly thicker on one side, sur-
rounding two leaves of normal
form, but not of the normal 5. G. BRacHYPHYLLLA, X 1.
distichous arrangement, as may be seen from the diagram,
figure 5. This anomaly of the leaf arrangement cannot
be regurded as having any special significance since it is
frequently to be observed in otherwise normal individuals

* Morren, C. Bull. Acad. Roy. Sci. Belg. 19°: 458-462. pl. 1852.

‘

e me, Lr Teen BS ee eee
cS cont tins iin

128 MISSOURI BOTANICAL GARDEN.

of Gasteria. It seems probable that the ascidium
originated from a continuous ring of meristematic
tissue.

The second specimen, figure 2, was seen by the writer
only in its present condition, but according to Mr. Thomp-
son it was earlier exactly comparable with the one de-
scribed above, and reached its present condition by the
rupturing of the calyptra.

A comparable anomaly was seen in one of the leaves of
a young plant of Agave Americana variegata grow-
ing at the Garden. Figure 6 renders any description

unnecessary. Inthe her-

barium of the Garden is

a photograph of another

young plant belonging to

a different species of the

6, A, AMERICANA, X 3. section Americanae of the

same genus, in which precisely the same condition
prevails.

Foliar ascidia are familiar to biologists in several in-
sectivorous plants. The great interest attaching to these
and the curious appearance presented by teratological
ascidia have caused the latter to receive particular at-
tention among plant anomalies and more than one
monographic treatment has appeared. In 1838 Mor-
ren* suggested a classification of all the foliar ascidia
known to him. He again publishedt on this sub-
ject in 1852, presenting an elaboration of his classi-
fication of teratological forms and citing all cases known
to him. In 1863 Kickx t modified and extended the
classification proposed by Morren. From this paper

* Morren, C. Bull. Acad. Roy. Sci. Brux. 5430-442, 582-586.
1838.— Ann. Sci. Nat. II. 11: 119-128. 1839.

t Morren, C. Bull. Acad. Roy. Sci. Belg. 195 3 444-462. 1852.

¢ Kickx, J. Bull. Acad. Roy. Sci. Belg. II. 16: 625-632. 1863.

ASCIDIA IN GASTERIA AND AGAVE. 129

the treatment in the work of Masters* is largely taken,
though some differences are to be noted. Penzig’s t
lists are extensive, but his arrangement is purely a sys-
tematic one.

Since Masters’s work no general treatment has appeared
but several individual cases have been described and some
important papers bearing directly or indirectly on the sub-
ject have been published. Of these, one by Penzig ¢ and
three by De Candolle § should be especially mentioned.

None of the ascidia described in these papers are at all
comparable with those here discussed except, perhaps,
those of Tulipa, Polygonatum, Convallaria, Lilium,
Acanthophippium and Vanilla. .

The calyptriform ascidium of Tulipa Gesneriana was
first mentioned by Hincks || who described it as ‘‘ a speci-
men of Tulipa Gesneriana, in which the leaf on the stem,
folding around it, had cohered by its edges, so as to com-
pletely enclose the flower-bud, which, as it enlarged, car-
ried up the upper part of the leaf, like the calyptra of a
moss.’’ Saint-Pierre J describes a similar anomaly.

Hincks also mentions a plant of Convallaria multiflora
‘¢in which the two lowermost leaves cohered by their edges
into a sort of a bag, which considerably obstructed the
growth of the stem.”’

Kicks merely mentions a tubiform monophyllous asci-
dium in Lilium lancifolium. He also observed ** a fusi-
form-cylindrical ascidium of nearly two decimeters in
length on the leaf of Acanthophippium bicolor. - James ft

* Masters, M.T. Vegetable Teratology. London. 1869.

+ Penzig, O. Pflanzen-Teratologie. Genua. 1890-’94.

¢ Penzig, O. Malpighia. 16 167-175. pl. 4-6. 1902.

§ De Candolle, C. Bull. Soc. Bot. Genéve. 8 3 61-69. 1897; 9¢ 1-51.
1899. — Bull. Herb. Boissier. II. 2: 753-762. pl. 8-9. 1902.

|| Hincks, W. Rept. Brit. Ass. Adv. Sci. 7: 120. 1839.

{ Saint-Pierre, G. Bull. Soc. Bot. Fr. 1363. 1854.

** Kickx, J. Bull. Acad. Roy. Sci. Belg. 18!3 591. 1851.

tt James. Gard. Chron. n. s. 153799. 1881.

130 MISSOURI BOTANICAL GARDEN.

found an ascidium in the leaf of Masdevallia Lindeni but
no description is given. Jacob de Cordemoy * describes
an anomaly in Vanilla planifolia which may be exactly
comparable with those seen in Agave and Grasteria. The
branches were terminated by an erect, slightly flattened,
hollow, fusiform body of eight centimeters in length and
one centimeter in thickness. A histological examination
showed this organ to be foliar in nature, but the description
of the solid stalk of the fusiform body does not permit of
a final decision as to the comparability of the tubiform
ascidia under consideration.

The condition in Polygonatum multiflorum is of in-
terest in this connection. The first record is that of
Morren f of a monophyllous, tubiform ascidium. Later
he describes t a specimen in the collection of Hincks
as a series of three ascidia through the center of
which passed the axis bearing normal leaves and flow-
ers. ‘* Qu’on s’imagine done un cornet foliacé qui
se termine par deux oreillettes latérales et opposées,
servant d’étui & deux autres cornets ayant aussi chacun
deux oreillettes opposées, et au-dessus d’eux un bouquet de
feuilles et de fleurs, et on aura la représentation de cette
belle monstruosité, de cette curieuse anomalie.’? Morren
thought it remarkable that the ascidia should have been
formed by the fusion of two leaves but this point was, he
considered, established by the presence of the two auricles
which formed about the terminal fourth of the total length
of the ascidia. To give rise to such ascidia the position
of the three pairs of leaves must have changed from
alternate to nearly opposite.

* Jacob de Cordemoy,H. Rev. Gén. Bot. 11 : 258-267. jig. 54-58.
1899.

t Morren, C. Bull. Acad. Roy. Sci. Brux. 534387, 439, 586. 1838, —
Ann. Sci. Nat. Il. 11: 124, 126. 1839.

¢t Morren,C. Bull. Acad. Roy. Belg. 5 ¢ 584-585. 1838; 19° 3 456, 458.
1852.

ASCIDIA IN GASTERIA AND AGAVE. 131

Puel and Saint-Pierre * also observed an ascidium in
this form. The fourth leaf was transformed into a sack
with a small opening at the tip through which the stem ex-
tended. The fifth leaf was normal while the sixth also had
the margins partly fused. The veins of the leaf forming
the main ascidium are described as of equal size.

Itis now quite impossible to draw any conclusions con-
cerning the ascidia described as diphyllous in Convallaria
and Polygonatum, but the normal arrangement of the
leaves in these forms and the condition found in the other
Monocotyledons in which ascidia have been found suggests
that possibly these anomalies are monophyllous in origin.

Leaves showing a more or less complete longitudinal di-
vision into two component parts are among the commonly
described foliar anomalies. There are two processes by
which such structures might arise: — by the fusion of two
adjacent primordia or by the longitudinal division of
one already formed. The numerous cases which have been
described have been referred, by implication at least,
to one of these categories. Paxf and Kleint have sug-
gested criteria for determing the probable method of de-
velopment of any individual case, butsince the necessary data
are wanting a correct interpretation of most of the pub-
lished descriptions is quite impossible. But for our present
purpose it is sufficient to note that either the production
of two primordia in the region where only one normally
occurs or the broadening of the single primordium by
division increases the total width of the meristematic
region from which a foliar organ is developing. If this
region comes to extend entirely around the axis, tubiform
ascidia such as those described in this paper may result.

A broadening of the leaf accompanied by longitudinal

* Puel and Saint-Pierre. Bull. Soc. Bot. Fr. 1: 62-63. 1854.
+ Pax, F. Allg. Morph. Pfl. 92. 1890.
} Klein, J. Jahrb. Wiss. Bot. 243 425-498. pl. 13-18. 1892.

4
.
i)
a
5
<¥
%
i
Ҥ
Be
¥ ¥
%
e

132 MISSOURI BOTANICAL GARDEN,

division for a greater or less distance or the more or less
complete fusion of two original primordia has been observed
in Musa Ensete by Meyran,* in Tulipa Glesneriana by
Lange,f in Ophrys aranifera by Saint-Pierre t in Orchis
maculata by Clos,§ in Strelitzia Reginae by Massalongo, |
in Richardia aethiopica, Colocasia undulata, Lolium per-
enne and Potamogeton plantagineus by Braun and_per-
haps other cases from the teratolgical literature might
be referred to these categories. I have seen the same
phenomenon in a specimen of Agave sp. growing in the
Missouri Botanical Gardens and in the herbarium isa leaf of
A, picta showing the same anomaly. While such divided
leaves may not be described as intermediate between the
normally developed leaf and the tubiform ascidia — which
have as yet been observed only in the Monocotyledons — it
seems possible that there is some similarity in their mode of
development.

* Meyran, O. Bull. Soc. Bot. Lyon. II. 8 ¢ 123-124. 1885.
+ Lange. Bot. Tids. 7: 209. 1873-74.
. } Saint-Pierre, G. Bull. Soc. Bot. Fr. 23 ; xxxvi-xl. 1876.
§ Clos, D. Bull. Soc. Bot. Fr. 36 ccx. 1889.
|| Massalongo, C. Nuovo Giorn. Bot. Ital. 203 283. pl. 76. 1888.
{ Braun, A. Sitzb. Ges. Naturf. Freunde, Berlin. 1871: 7.

Fe oe Se AE at Ye

we at Se .. kee oe © ae. Uses! ree
PAIRS PUNE ae AES 2 ARR GER EEE SC eR Ay MOE RS
a ; ‘ °

PROLIFICATION OF THE FRUIT IN CAPSICUM AND PASSIFLORA.

BY J. ARTHUR HARRIS.

The appeurance of a recent paper on the influence of
cold in the production of cleistogamy and teratological
phenomena in Solanum Melongena and in Capsicum
annuum, by Dr. G. Mottareale,* has induced me to put
into print notes and drawings made some years ago at the
Missouri Botanical Garden. These observations were made
upon a very large series of fruits from plants grown in the
open and are consequently interesting for comparison
with those studied by Mottareale. Since they refer chiefly
to an anomaly which he was not able to study extensively
it seems best to present the notes in full with most of the
figures as originally prepared. Observations on prolifi-
cation of the fruit of Passiflora are also added.

CAPSICUM.

The material studied by Mottareale was grown under the
following conditions. In three large greenhouses the prin-
cipal plants cultivated were Phaseolus vulgaris, Cucumis
sativus, Cucurbita Pepo, Solanum Melongena, Capsicum
annuum, C’. grossum, and Lycopersicum esculentum. Dur-
ing the early part of the winter the heating plant had not
yet been installed, but owing to the mild climate the
plants made a vigorous vegetative growth and were begin-
ning to produce normal flowers when, about the middle of
December, two or three nights of cold wrought havoc with
them. The egg plants and the peppers lost their flowers
and most of their leaves, and the more tender apical shoots
blackened and dried. But the vitality of the plants was

\
* Mottareale,G. Ann. R. Scuola Sup. Agr., Portiei. 6, 1904.
(133)

134 MISSOURI BOTANICAL GARDEN.

not entirely destroyed and when provision for a suitable
temperature was made, the vegetative organs gradually re-
gained their former luxuriance. But the sexual organs
were very teratological in nature, and these anomalies of
flower and fruit Mottareale considers the result of the con-
ditions of temperature to which the plants were subjected.

Of the anomalies of the calyx, corolla and stamens, only
pistillody of the stamens need be mentioned here. In
Capsicum adesmy of the carpels was noted in several cases
but we may confine our attention to the internal struc-
ture of the fruit. Mottareale saw diaphysis of the
fruit —the production of another imperfect fruit from the
center of the thalamus, and ecblastesis— the production
of accessory fruits from the axils of the carpels. Both
anomalies were seen in some fruits. He found the seeds
either normal, abortive, more or less foliaceous, or tending
to assume the form of a carpel.

The evidence for referring these anomalies to the influ-
ence of the cold seems unsatisfactory. He himself tried
for two successive years to reproduce the result but with-
out success. Furthermore, anomalies resembling those he
describes are not infrequently seen in Lycopersicum, So-
lanum and Capsicum.* Confining our attention strictly
to the fruit we note the occurrence of pistillody of the
stamens in all three genera, the occurrence of supernu-
merary whorls of carpels in Lycopersicum, the rupturing
of the fruit wall by the placentae in Solanum Melongena,
and the production of carpel-like bodies inside the fruit of
Capsicum.

Terracciano f observed two types of prolification in
C. annuum. In one form the walls of the fruit, ter-
minated by a hollow tubiform style open at the top,
showed five gibbous evaginations. Internally the fruit

* Penzig, O. Pflanzen-Teratologie. 2: 169-174. 1894. :
+ Terracciano, N. Nuovo Giorn. Bot. Ital. 103 28-34. pl Z. 1878.

PROLIFICATION OF FRUIT. : 135

produced no seeds but contained five smaller fruits, like-
wise without seeds and with short styles and almost obso-
lete stigmas, disposed upon the apex of the thalamus. In

the other type the fruits were quite normal externally and

bore numerous seeds but produced also a single included
fruit at the top of the axis. According to Penzig, Borbds
describes central prolification in Capsicum. Halsted *
saw, in the same genus, asmall included fruit which origin-
ated from the apex of the axis, and Viviand-Morel f figures
an almost identical case, while Raymondaud f{ also figures
an internal fruit. Penzig § reports central prolification as
especially common.

In 1902 a considerable plot was devoted to the cultiva-
tion of capsicums at the Missouri Botanical Garden.
Fruit-like bodies were noticed occasionally during the
summer in those gathered for table use, and late in the
autumn the entire crop was gathered and the fruits were
dissected for the purpose of determining the frequency of
the anomaly. Other material was secured for the purpose
of comparison during the present year, but since it yielded
only corroborative evidence and was not large enough in
amount to permit of statistical discussion the following
statements are based on the series collected in 1902.

From the literature cited above it seems that the carpel-
or fruit-like structures may originate in three ways: — by
central prolification or diaphysis, by prolification from the
axils of the carpels or ecblastesis, and by the transforma-
tion of the primordium of an ovule into a carpellary body.
All of these phenomena have been seen in other genera of
plants and are to be regarded as representing valid cate-
gories, but it proved impossible to divide the anomalies

* Halsted, B. D. Bull. Torr. Bot, Club. 183151. fig. 1891.— Pop.
Sci. Mo. 422 321-322. fig. 16. 1893.

+ Viviand-Morel, J. V. Lyon Horticole. 943582. fig. 1902.

¢ Raymondaud, E. Rev. Sci. Limousin. 12 3369-372, 1904.

§ Penzig, O. l.c.

Ps c ae |
FP tye oe

Fae Cpe Oe OR Se ye ee ae ae

136 MISSOURI BOTANICAL GARDEN,
t

found in over three hundred abnormal fruits examined into
these classes. It is quite impossible to determine whether
a given carpel-like body was produced from a bud in the
axil of a carpel or indeed from an adventitious bud on any
portion of the carpel, or from a primordium which nor-
mally would have developed into an ovule. It is no easier
to ascertain whether the accessory organ was produced
from the apex of the torus or from a neighboring ovule
primordium. The nature of the Capsicum fruit precludes

1. CAPSICUM ANNUUM GROSSUM, x $.

the determination of some points which may be studied in
other forms. ,

The margins of the included carpel-like bodies were in
all cases completely closed, and not open as has been de-
scribed in some cases of carpels produced by prolification.
The smaller ones showed no cavity whatever but were
solid throughout. In form they varied from an irregu-
larly contorted body through an almost perfectly farmed
sterile fruit, comparable in shape with the one in which it

PROLIFICATION OF FRUIT. 137 . a
: : gh

was produced, to linear structures from a few millimeters
to two esutimeters or more in length, terminated by a

.

minvte style.

Histologically the similarity of the carpel-like bodies to
the wall of the fruit is very great. Both show the exceed-
ingly heavy epidermis and the large-celled parenchymatous
ground tissue. The bodies show one or more vascular
bundles similar to those of the wall of the fruit. When
ae: fresh sections are examined, numerous chromoplasts are

2
Z

2. CAPSICUM ANNUM GROSSUM X $.

seen in both but they are generally more abundant in the “3
wall of the fruit. : | ‘
Since the appearance of the teratological papers of a
de Vries, one of the chief objects of teratological studies B
is the determination of the frequency of occurrence of
the various anomalies. The fruits examined by the a
me | writer belong to twenty named varieties but here only |
ae sixteen which have been described by Irish in his Revision

4

i;
4
s
ze
—

138 MISSOURI BOTANICAL GARDEN.

of the Genus Capsicum * or in a later paper f, will be
included.

In the accompanying table the garden varieties are
grouped under the botanical varieties recognized by Ivish.
The number of normal fruits and the number which con-
tained carpel-like bodies, as well as the percentage which
the latter are of the total number examined are given for
each variety. No attempt is made to classify these bodies
according to form or origin.

TABLE A. DISTRIBUTION OF ANOMALIES IN CAPSICUM.

VARIETY. Normal. |Abnormal. Fhe 9
Capsicum annuum acuminatum,
CARN os, ae tks wlateas wt oe age 105 0 ie
Long Cayenne ...... . 200 1 5
Long Yellow Cayenne ... . 250 0 ee
Yellow Nepal Chilli. ... . 176 1 5
Capsicum annuum longum.
COUNTY Bates fe eS es 200 0
Long Red 6 6 ew te ce 165 0
Capsicum annuum grossum.
WEOUMURGG ib se fe eves 917 167 15.4
Abra Ss 5 ies oo age rece 40 0 we eke
Colger DAO a ae 73 6 7.5
Bell Sat rai Dae oak ear ee ee 89 19 17.5
ECOLGOR MIE cle bat oe te GON 150 33 18.
Sguash, or Tomato Shaped . . 133 ll 7.6
Capsicum annuum abbreviatum.
ROCA ee eee a 250 0 EF
Hed Were a oe ke ee 223 0 eae
Burpee’s Mikado... . «ss « 258 88 25.4
Capsicum annuum cerasiforme.
Cherry FE ee ee 198 16 7.4
pi Sa eg er ote 3,427 342 9.

* Irish, H.C. Rept. Mo. Bot. Gard. 953-110. pl. 8-28. 1899.
t Irish, H.C. Proc. Soc. Prom. Ag. Sci. 23 3 63-64. 1902.

PROLIFICATION OF FRUIT. 139

Some suggestive points are to be gathered from this
table. The slender-fruited varieties classified under C’.
annuum acuminatum and C. annuum longum show only
two cases of prolification although nearly 1,100 fruits were
dissected. Of the 2,671 fruits examined from the ten
garden varities of C’. annuum grossum, C. annuum abbre-
viatum and OC. annuum cerasiforme, 340, or 12.7%,
show prolification in some form. But that the relative
length and diameter of the fruit is not the only factor
involved is rendered probable by the three exceptions
among these ten short-fruited varieties. The absence of
abnormal fruits in Ruby King may be a chance result due
to the small series of material examined, but this ex-
planation cannot be offered for Etna and Red Wrinkled.

The chief conclusions to be drawn from the literature
and from the present series of observations are the fol-
lowing.

The included bodies are doubtless formed by the three
processes heretofore recognized as giving rise to intra-
carpellary fruits, namely diaphysis, ecblastesis, and pistil-
lody of the ovule primordium, but the fruit of Capsicum is
not suitable for a determination of the number of cases to
be referred to each category.

In form the bodies vary greatly, but two main types —
rounded or irregularly contorted, and linear—are to be
recognized. The more or less isodiametric form is almost
invariably found at the apex of the torus or at the base of
the seed-bearing region. The linear form may occupy any
position.

Histologically the bodies are very similar to the wall of
the fruit.

Prolification of the fruit, using the term to cover all
three phenomena mentioned above, is very common in
several garden varieties of Capsicum. It is practically
wanting in the slender-fruited forms, although some var-
ieties form exceptions to this rule.

140 MISSOURI BOTANICAL GARDEN.

The foregoing facts preclude any conclusions concerning
the influence of environmental conditions, until careful
statistical studies have been made. These must take into
consideration ancestry as well as environment.

_ PASSIFLORA.

A number of observations have been published on ab-
normal fruits in Passiflora, but since all the work has been
done independently, I take this opportunity of bringing
together the literature and of presenting further observa-
Z tions made upon a much larger series of material than has
3 been examined heretofore.

Clos* describes the prolification of the fruit of P.
gracilis. He found two types. In one there were three
pedicels each with three styles and in the other a single
pedicel terminated by three styles. Salterf describes
an abnormality in the ovules of P. palmata and P. caeru-
lea. Lindberg { records proliferous fruits in P. gracilis.
His descriptions agree with those of other authors.
Bernoulli § records a case of prolification in P. serratistt-
pula in which a slightly abnormal flower was produced in
an apparently normal fruit. He also states that in the
: section Granadilla, a short, filamentous projection is fre-
A quently to be seen at the bottom of a fruit-cavity. Mas-
f ters || describes a fruit of P. quadrangularis ‘* wherein
: | small flower-buds were found springing from the placentas
intermixed with the seeds.’’ Hildebrand { notes fruits with

hE ea ORME, ORE OS
¥ A ‘ é

* Clos,D.. Mém, Acad. Imp. Sci. Inscript. Belles- Lettres, Toulouse,
V. 83113. 1859.

t+ Salter, S.I. Trans. Linn. Soc. 243 148-150. pl. 24. 1863.

¢ Lindberg, S. O. Ofvers. Finska Vetensk.-Soc. Férh. 103 15-16.
1867.

§ Bernoulli,G. Bot. Zeit. 27 321-23, 1869.

|| Masters, M. T. Trans. Linn. Soc. 273607. 1871.

{ Hildebrand, F. Bot. Centralbl. 9: 401-404. pl. 7. 1882.

RRR CCT nS Re Te eee ee
al ca te bp ’ . é t ge , ‘ * phe on re Petit Se pry

Che OF ta oO re
ie saat oe r

PROLIFICATION OF FRUIT. 141

fourcarpels. He figures and describes in considerable detail.

prolification in fruits of P. gracilis in the Botanical Gar-
den at Freiburg. He finds the produced axis bearing an in-
definite number of carpels, sometimes over sixteen, having
more or less completely developed stigmas and producing
on their edges, which cohere in various degrees, a limited
number of ovules. In several cases the main axis bore a
lateral branch, which in turn produced a number of carpels
of diverse degrees of development. Wigand * describes
the structure of the enclosed shoot of P. gracilis in detail.
Miiller t describes briefly prolification of the fruit of P.
alata. In several fruits of this species he found a per-
fectly developed flower bud. The fruits contained an
abundance of good seeds. A plant raised from these had
produced, up to the time of the writing of his paper, only
normal fruits.

In the summer of 1902, vigorous specimens of P. gracilis
were growing in the Missouri Botanical Garden in part in
the Cycad house and in part over a lattice in the open air.
My observations are based upon the examination of 167
proliferous fruits secured by dissecting 4,240 fruits. The
material was collected in four lots, the first. taken from
the vines growing in the greenhouse and the last three
from those on the outside. Lots one and two were gath-
ered early in October, the third was taken October 25, and
the fourth lot about five days later. The distribution of
the material was as follows: —

Lot I. 558 normal, 39 proliferous—6.5 per cent.
Lot II. 1402 normal, 56 proliferous — 3.9 per cent.
Lot III. 1239 normal, 38 proliferous — 2.9 per cent.
Lot IV. 874 normal, 34 proliferous — 3.7 per cent.

—_

Total. 4073 normal, 167 proliferous aan per cent.
!

* Wigand, A. Bot. Hefte. Forsch. a.d. Bot. Gart. Marburg. 2: 125-
126. 1885. :
+ Miller, F. Flora. 78 3332-883. fig. 1890.

Se

142 MISSOURI BOTANICAL GARDEN.

As pointed out by earlier writers, the prolification con-

sists in the production from the base of the fruit of astalk,
or sometimes two or more stalks bearing a variable number
of carpels closely packed together as may be beautifully
seen in a cross section of the fruit. The margins of the
carpels are fused in various degrees or free. A few ovules
are sometimes produced. For other details the accompany-
ing figures may be consulted.

Of 125 proliferous fruits dissected, 100 had only a
single central stalk while 25 had more. In several cases
small bud-like or filamentous processes were seen near
the base of the stalk bear-
ing the accessory carpels,
but when these produced no
differentiated carpels they
were not counted. The de-
termination of the number
of the included carpels is
sometimes rather difficult
owing to the varying degrees
of fusion of the members of
the outer whorls and the
small size of those of the
inner ones. The best cri-
terion seems to be the num-
ber of styles. These seem to be independent of the fusion
of the carpels and they are usually well developed even in
the smallest members of the inner whorls. The number
of carpels counted in the 25 fruits containing more than
one carpel-bearing pedicel was: — 22, 22, 33, 85, 38, 42,
44, 46, 46, 46, 48, 49, 49, 51, 51, 55, 57, 64, 64, 66,
71, 72, 83, 90, 90. The number of carpels in the 100
fruits containing only a single carpel-bearing pedicel is
shown in the accompanying table. The smallness of the
series and the fact that the material was derived from two
habitats renders any biometric discussion unprofitable.

)

3. PASSIFLORA GRACILIS, X 1.

PROLIFICATION OF FRUIT. 143

TABLE B. NUMBER OF CARPELS IN ABNORMAL FRUITS OF

PASSIFLORA.

N F N F N F N F N F
3 3 18 3 27 i 36 3 46 2
4 18 19 2 28 1 37 4 47 ii
5 3 21 4 29 4 38 2 48 1
6 8 22 2 30 1 39 2 55 1
7 3 23 a 31 3 41 3 57 1

13 1 24 2 32 1 42 1

14 1 25 2 33 1 43 2

15 3 26 3 34 4 45 2

A rupturing of the wall of the fruit in the abnormal ex-
amples was very frequent, amounting to about 30% in one
of the lots collected in the open. A very noticeable differ-
ence between the material growing in the greenhouse and
in the open was the very much smaller size of the included
group of carpels in the former. In only about 5% of the
examples collected in the Cycad house was the wall of the
fruit ruptured by the included body, which was usually
quite small and simple, while in the material grown outside
the inner group was usually large and very complicated in
structure, being composed of a great number of irregularly
arranged carpels.

It is interesting to note that prolification of the fruit
has been seen by three observers in another member of
the Passifloraceae, Carica Papaya. Méiiller* figures a
sterile fruit containing a large central fruit with numer-
ous ovules. According to Houard,f Prain{ figures a
specimen containing numerous secondary fruits. Urbina§
discusses an example in which the cavity in addition to the
normal black seeds was filled by six conical, conduplicate

* Miller, F. J. c.

+ Houard, C. Rev. Gén. Bot, 18 131. 1906.

¢ Prain, D. Journ. As. Soc. Bengal. 18953 196-198. pl. 4-6.
§ Urbina, M. La Naturaleza. 83359-8360. pl. 29. 1900.

144 MISSOURI BOTANICAL GARDEN.

carpellary bodies of 8 to 10 centimeters in length and 4 to
6 centimeters in diameter, bearing abortive ovules. The
center of the fruit was occupied by a cylindrical body of
about 14 centimeters inlength and 4 to 5 centimeters in
diameter, bearing five branches near the base. Urbina
considers this fruit to be composed of three whorls, each
of five carpels. The lowermost forms the normal fruit.
The second is represented by five included carpels, of
which one is divided longitudinally. The third whorl,
which is much more rudimentary, appears as the central
column with its basal branches.

Diaphysis and ecblastesis of the fruit, while not so com-
mon as many other anomalies, need not be considered fur-
ther here. The transformation of the primordium of an
ovule into a carpel-like body is still rarer. The only in-
stances known to me are the Chetranthus Cheiri de-
scribed by DeCandolle* and Masters, t the Barbarossa
grape figured by Masters,t the Sinapis arvensis described
by Baillon,§ and the Dianthus figured by Berkeley.|| A
careful search through the older discussions of the evidence
of teratology on the homologies of the ovule would, per-
haps, disclose other cases.

Such anomalies as the present ones are interesting in
connection with Sach’s theory of ‘ material and form,”’
to which Goebel is inclined to attach much significance.

* DeCandolle, A. P. and A. DeCandolle. Nouv. Mém. Soc. Helv. Sci.
Nat. 5. pl. 5. 1841.

t Masters,,M. T. Veg. Ter. 182. jig. 94-95. 1869.

t Masters, M. T. J. c. 182-183. fig. 96-97.

§ Baillon, H. Adansonia. 8: 351-353, pl. 72. 1862-'63.

|| Berkeley, M. J. Gard. Chron. 1850: 612. yigures.

PROLIFICATION OF FRUIT. 145

EXPLANATION OF TEXT FIGURES.

Allof the figures are diagrammatic.

Figure 1.— Capsicum annuum grossum. a—c, ‘* Monstrous.’? d—e,
«Golden King.” f, Bell.”? Ina and c the carpel-like bodies (1), seen
at the apex of the torus surrounded by seeds (2), probably originated by
central prolification. In b the body from a is seen in longitudinal sec-
tion, showing the thickness of the walls. Inc is also to be seen & care
pel-like body (3), probably originating by pistillody of anovule. Slight
adesmy of the carpels is represented, from above, in d, and in longitudi-
nal section in the two halves of the same fruit, in e. The torus of this
fruit was abnormally large and almost the entire cavity of the fruit was
filled with large, irregularly rounded fruit-like bodies (1). The origin
of these bodies is uncertain. Inf an irregularly rounded fruit-like
structure (1) is seen at the apex of the torus surrounded by seeds.

Figure 2.— Capsicum annuum grossum. g—h, “ Bell.’” i—l, “ Golden
King.”’?’ m—n, ‘‘ Monstrous.’’ o—p, “ Bell.’? In g half of the fruit wall
is removed to show the torus bearing seeds (2), and a large flattened
carpel-like body (1), originating by ecblastesis or by pistillody of an
ovule. In h the same condition prevails but here the body is linear. In
j seven linear bodies (1), originating by ecblastesis or by pistillody of
ovules, are seen. j—k represent various forms of carpel-like bodies
removed from the fruits, sometimes with a block of the torus.

Figure 8, — Central prolification of the fruits in Passiflora gracilis. a,
Fruit with two carpel-bearing pedicels. b, Carpel-bearing pedicel
removed from fruit, c, Fruit with single included pedicel. d, Diagram
of cross section of fruit, with two carpel-bearing pedicels, such as 1s
shown in a, showing irregular arrangement of included carpels.

10

ee cere Bee)”

FASCIATION IN OXALIS CRENATA AND EXPERIMENTAL
PRODUCTION OF FASCIATIONS.

BY HENRI HUS.

Owalis crenata is largely grown in South America,
more particularly in Peru, where it is known as ‘¢ Oka,”’
and in Bolivia where it flourishes on the Puna, a plateau
12,000 feet above sea level. It bears tubers (pl. 17, +.
1) three to four inches long and about an inch in diameter.
They contain much starch and are used as food after hay-
ing been exposed to the sun for two or three days‘and then
boiled for about twenty minutes in water containing some
cooking soda.

During four years more or less I had the opportunity to
observe this plant in the experimental grounds at Berkeley,
California, where it was grown from tubers received from
the Botanic Garden at Sydney, N.S. W. Each year a
number of fasciations of the stem made their appearance,
but only in the latter part of the season. Some of the
plants of the 1903 crop were chosen for purposes of illu-
stration. Not only did they show the more normal type
of fasciation, 7. e., a fasciated mainstem and branches, the
majority of which are normal (pl. 18), but there also oc-
curred instances where the entire plant was fasciated in the
most fantastic manner and even bore tubers at various
places on the aérial stem ( pl. 19). Many of the tubers
were likewise fasciated.

From Dr. Otto Kuntze it was learned that this fasciated
condition is the one most frequently met with in Bolivia,
and all plants of Oxalis crenata seen by him at Cochabam-
ba were of this character. * There are no means of knowing

* Penzig in his Pflanzen-Teratologie mentions a fasciation of Oxalis
crenata a8 enumerated by Crépin in Recueil de faits tératologiques, 2,
(Bull. de la Soc. Roy. de Botanique de Belgique. 4 276-278. 1865),
but which was not found when the paper in question was referred to.

(147)

148 MISSOURI’ BOTANICAL GARDEN.

with exactitude what was the nature of the ancestors of our
plants in the Sydney Botanical Garden, but for the present
purpose the data gathered at Berkeley are sufficient.

The specimens of Oxalis crenata grown at Berkeley con-
firm in the most striking manner the views on the trans-
missibility of fasciations which at the present time are
more generally held than formerly. Up to ten or fifteen
years ago the concensus of opinion was that fasciation was
an individual phenomenon and that its appearance depended
entirely on external conditions. But a more careful study
of the phenomena and especially the observations of de
Vries * have shown that fasciation is a character which may
be inherited, though to a varying degree. In some individ-
uals it may seem to have disappeared, but only to reappear
in their descendants. The latent condition of the charac-
ter is also shown by trees which for some years produce
fasciated branches and then normal ones, to be again fol-
lowed by fasciated limbs. It has been shown by de Vries
that in such cases the manifestation of fasciation is due to
favorable external conditions. Unfavorable conditions
have a tendency to prevent it. At Berkeley, plants of
thus diversiloba, the poison oak, which had shown fas-
ciation in the growth of three successive years, ceased to
produce fasciations after being transplanted. Numerous
similar instances might be quoted. As Goebel f points
out and de Vriest again recalls, wherever external in-
fluences cause the appearance of fasciations or anomalous
structures in general the tendency to this must be present
in a latent condition. If this tendency is absent all efforts
to bring these structures about are futile.

That the fasciated character may remain dormant is well

* deVries, Hugo. Die Mutationstheorie. 23541. Leipzig. 1901-1903,
— Here other references will be found.

t+ Goebel K. Organographie der Pflanzen. 13158, Jena. 1898.

t Loe. cit. 551.

FASCIATION IN OXALIS CRENATA. 149

shown by our Oxalis tubers. Morphologically normal
tubers, when planted, produced, in some cases at least,
faseiated stems, Still another instance is yielded by the
tuber illustrated in pl. 17, f. 2, which has the following
hir ory: — The plants, after having been photographed,
v re laid away on a shelf, tubers and all. Three months
auter, one of the tubers, slightly fasciated, was found to
have sprouted and to have produced two stems, the one
normal, the other fasciated. Artificial conditions, such as
nutrition, irrigation, or the cutting away of parts were here
entirely excluded. Yet the character is present, in the one
case in an active, in the other, probably, in a latent condi-
tion.

Though it is well known that the members of some
groups seem to have a greater tendency to fasciate than
others, this tendency must occur very generally through
the plant kingdom. Both Monocotyledons ( Asphodeline,
Lilium, Asparagus etc.) and Dicotyledons ( Verbascum,
Acacia, Rosa, Sisymbrium etc.), though the former to
a lesser extent, yield numerous instances. But is this
tendency present in all members of the plant world?
De Vries* believes it not necessary toassumethis. ‘*Aller-
dings scheint die Anlage zu Verbinderungen im Pflanzen-
reich aiisserst allgemein verbreitet zu sein, doch wohl
nicht so allgemein, dass sie nicht gewissen Gruppen fehlen
kénnte.’’ Time and experiment will have to decide. Var-
ious methods are now known by which fasciation may be
brought about. Heavy manuring may cause normal de-
scendants of fasciated plants to fasciate ( Crepis biennis
Jasciata).f| The stems of biennials with a tendency to
fasciation, if produced in the first year are mostly normal.
But if they are produced during the second year, fine fasci-
ated specimens are ordinarily the result (Aster Tripolium,

* loc. cit. 551.
¢ de Vries, H. loc. cit, 561.

cami iia als

150 MISSOURI BOTANICAL GARDEN.

Picris hieraciotdes etc.*) The time of sowing also
seems to be of great influence. Goebel t calls attention to
the fact that by leading the sap rapidly and with great in-
tensity to a lateral bud, which ordinarily contains but little
sap, fasciation may be produced. Thus he explains the
frequent occurrence of fasciation in suckers and water-
sprouts. Similarly, fasciation may be produced in annuals
(Phaseolus multiflorus, Vicia faba) by cutting off the main
stem directly above the cotyledons. Lopriore t was able to
produce root fasciation both by cutting off and by splitting
the root tip, but did not succeed in producing it by sub-
jecting roots to prolonged pressure. On the other hand,
Sorauer § quotes an instance of a stem of Tecoma radicans
which had become fasciated through appression to a wall,
and of which the parts reaching above the wall likewise
showed fasciation.

Experiments, mainly carried on at the Missouri Botanical
Garden, have shown that at or just previous to the flower-
ing period, plants, especially those with an indeterminate
inflorescence, may be caused to fasciate by the use of the
following method:—About the time of the appearance of
the first flower the plant is kept as dry as possible, only
enough water being given to prevent wilting. As a result
the flowering period will be comparatively short, and, in an
indeterminate inflorescence, the buds near the end of the
spike remain undeveloped. If at this time the plants are
daily abundantly irrigated, occasionally with manure-water,
numerous fasciations will make their appearance. But it
must be remembered that this result is usually reached
only with plants which throughout their existence have

* Ibid. 549.

+ Goebel, K. Joc. cit. 164.

I Lopriore, G. I caratteri anatomici delle radici nastriformi.
Rome. 1902.

§ Sorauer, P. Handbuch der Pflanzenkranheiten. 1:334. Berlin.
1906.

FASCIATION IN OXALIS CRENATA. 151

been well nourished and well cared for generally. For no
apparent reason one plant will fasciate, while the next one,
belonging to the same species, remains normal. This,
however, affects in no way the value of the experiment,
since this is undertaken only with a view to determining
whether in any species fasciations can be produced. Dur-
ing 1905 fasciations were obtained by this method in An-
tirrhinum majus, Actinomeris squarrosa, Solanum Lyco-
persicum ‘¢ Magnus, ”’ Lythrum virgatum, Oenothera La-
marckiana and Collomia grandiflora. Experiments with
Solanum Pseudo-capsicum, Capsicum annuum, Sola-
num nigrum and Abutilon Avicennae yielded no fascia-
tions.

It would seem that the cause of fasciation must be
ascribed to the rapid introduction of sap into buds which
would not normally produce them, as is suggested by Goe-
bel. This is further borne out by observations on numer-
ous plants, both cultivated and wild, growing at Berkeley,
California. It was noted that within a week after a heavy
rainfall at an unusual time (September, 1904), an ab-
normally large number of teratological cases were to
be met with. They were not limited to fasciations,
though these were most frequent, but included torsions,
petalody of the stamens and phyllody of the pistil,
frondiparous diaphysis — the prolongation of the axis
through the center of the flower when the prolongation
bears the character of a leafy shoot,” and fasciculations.
One of the most interesting cases of the latter was met
with in Picris echioides which showed also a slight chloro-
sis of the flowers, which were stalked. Contrary to expec-
tation, prolification, i. e., the production of secondary
heads, which so often goes hand in hand with this phenom-
enon, was not observed.

* This may be frequently observed in the peloric variety of Digitalis
purpurea.

152 MISSOURI BOTANICAL GARDEN.

Among the most familiar instances of fasciation is that
of the Asparagus. Interesting in this connection is the ex-
perience of an asparagus grower of California, who claims
that the greater percentage of fasciated asparagus shoots
is to be met with among those which are the first to pierce
the soil, and especially so after a cold winter. This again
would tend to confirm Goebel’s view.

EXPLANATION OF PLATES.

Plate 17.— Ovalis crenata. Above, normal tubers, x 1/,. Below,
fasciated tuber bearing a normal and fasciated shoot; two views, Slightly
reduced.

Plate 18.— Oxalis crenata. A fasciated plant, X !/,.

Plate 19.— Oxalis crenata. Three fasciated shoots, one of them, bear-
ing a number of aerial tubers, x 1/,, the other two X15.

Rept. Mo. Bot. GARD., VOL. 17. PLATE 17.

OXALIS CRENATA.

PLATE 18.

Mo. Bot. GARD., VOL. 17.

REPT.

OXALIS CRENATA.

REPT. Mo. BoT. GARD., VOL. 17. PLATE 19.

OXALIS CRENATA.

CONSTRICTION OF TWIGS BY THE BAG WORM AND INCI-
DENT EVIDENCE OF GROWTH PRESSURE.* ©

BY HERMANN VON SCHRENKE.

INTRODUCTION.

For a number of years many branches of arbor vitae
(Thuja occidentalis) in the Missouri Botanical Garden
showed signs of disease and gradual dying from the tips of
the branches inward. A careful examination of the twigs
showed a condition represented on plate 20, figs. 1, 2 and
3. Ata short distance from the end of the twig a swell-
ing was usually found, which in the case of twigs which
were still comparatively green was very small, but in many
cases reached a size of four and five times the diameter of
the twig below the swelling. Thepartof the branch above
the swelling was usually very much larger in diameter than
‘the branch below the swelling. Where the swelling was at
all large the parts of the beanah towards the outside of the
tree were usually found either in a dead or dying condition.
In every case the diseased branches showed marked ten-
dencies towards the formation of secondary buds and
branches, so that frequently the tip of a branch had the
appearance of a small broom (pl. 20, fig. 3). After look-
ing at a number of trees the swelling was finally found to
be due to the girdling action of the bands of the common
bag worm (Thyridopteryx ephemeraeformis Haworth),
ey 20, fig. 2). A great many instances of the swellings
with the bag worm still in position were found in the vi-
cinity of St. “Louis and in other parts of the country; and
in all of the older swellings where no bag was visible, the
band of the worm was found still encircling the branch.

* Published by permission of the Secretary of Agriculture. The main
results here detailed were communicated to the Society for Plant Mor-
‘ phology and Physiology, at its Ann Arbor Meeting, in December, 1905.

(153)

154 MISSOURI BOTANICAL GARDEN.

There seems to be no doubt therefore that the band of the
bag worm was responsible for the swellings, serving as a kind
of ligature which prevented the passage of elaborated ma-
terials from the outer part of the branch towards the
trunk, and acting just as a wire would when wound
around a growing twig. In the course of time the part of
the branch above the band increased in size until such time
as the passage of the elaborated material was prevented
entirely, resulting in the final death of the branch.

After attention had been called to the occurrence of the
bags on arbor vitae and the girdling effect which they ex-
erted, asearch of the various trees in the Garden and in
the country around St. Louis showed that this action of
the bag worm on arbor vitae was an extremely common
one; in fact it was almost impossible to find a tree of ar-
bor vitae which did not show one or more of the swellings.

LIFE HISTORY OF THE BAG WORM.

In order to understand the manner in which the insect
kills the branch it will be necessary to refer briefly to its
life history. The common bag worm ( Thyridopteryx
ephemeraeformis Haworth*) hatches from eggs which have
remained on the trees in the bags over winter. In the
vicinity of St. Louis the larvae appear on shade trees in
the early spring just as the buds are opening. The larvae
feed on the young leaves, and very shortly after their ap-
pearance begin to spin small bags which they carry about
with them. As the insect grows older practically nothing
of the larva is visible except the anterior portion of its
body, the rest of its body being encased in the bag.
Towards the latter part of the summer, from the beginning
of August on, the larvae attach the bag to some twig near
the outer part of the tree; they do this by spinning a band

* For a good description of this insect and its habits, see Felt, E. P.
Insects injurious to elm trees. (5th Annual Report, Fisheries, Game &
Forest Commission of New York, p. 359, 1903).

CONSTRICTION OF TWIGS BY THE BAG WORM. 150

around the branch in such manner that when the bag is
finally fastened it is suspended from the branch, as shown
in plate 21. The band varies in width and in thickness.
It is closely adapted to any irregularity in the bark of the
particular twig, and is so wound around the twig that it
can be moved laterally only with difficulty. It is in no
way pasted or glued to the bark, as can easily be shown
by cutting the band at any point, which will cause the
cocoon to drop. The insect will attach its bands, almost
without fail, only on one-year-old twigs. Whether this is
due to the fact that it is upon these twigs that it has been
feeding, or whether it is due to the fact that these twigs
are usually smallest, I have not been able to determine.
That the size of the twig has something to do ‘vith the
weaving of the band is shown by the following experiment :
In order to test the strength of these bands, the writer de-
sired to get bands of as great a length as possible. A
number of the larvae which were found ready to attach
themselves to twigs were taken and placed on twigs having
a diameter of about three-eighths of an inch; in no case
did these transplanted larvae weave a band around a twig of
this size. When they found that they could not get away
from the large twig they made a band which they glued to
the lower side of the twig, thereby suspending the cocoon;
when the branch was one-fourth inch thick or less they
invariably made a band completely around the twig. The
relation between the size of the twig and the number of
bands is shown in detail in Table I. After the band has
been woven, the cocoon hangs on the tree over winter and
into the next spring, in fact until about the end of June of
the following year. Some six weeks or more after the leaves
have fully opened one can pass under trees which have been
covered with the bag worms during the winter and find
hundreds of them lying on the ground under the trees.
The falling of the bags is brought about by the bursting of
the bands by the twigs as they increase in diameter. That

156 MISSOURI BOTANICAL GARDEN.

this is the cause of the bursting of the bands was determ-
ined repeatedly by noting that in no case’ would the bands
burst until a considerable formation of wood took place in
the twig to which the bags were attached. The bursting of
the bands took place at almost any point in their circum-
ference except at the point where the bag was attached to
the band; in other words, the weakest point in the bands
usually was at some point away from the bag itself. This
was a fact of considerable interest, the force of which will
be referred to again further on.

FORMATION OF THE SWELLINGS.

By the end of July practically all of the bands of the
previous year have dropped from the trees; here and there
however, one finds abag which has not fallen. Upon ex-
amination it will be found that the band of this particular
bag still passes intact around its twig; and should this bag
remain upon the twig until September, one will find that a
swelling, such as the ones referred to above, has begun
to form. It would appear from this that any particular
band which remains throughout the summer has been strong
enough to resist the forces exerted by the growing twig as
it expanded in the formation of new wood, and that the
band then acted as a ligature. All of the conditions for
the formation of the swelling were thereby brought about.

As the bag worm occurs on other trees besides the
arbor vitae, an examination was made of trees in the
Garden and in the streets of St. Louis for the purpose of
finding out whether the girdling action noted for the
arbor vitae trees also took place on other trees. After
a few days of searching it was found that practically every
tree species upon which the bag worm occurred was girdled
now and then very much as were the arbor vitae trees. On
some species the girdling action appeared to bevery common,
particularly on the sycamore and soft maple, but nowhere
did it occur as frequently as with the arbor vitae. Thus

CONSTRICTION OF TWIGS BY THE BAG WoRM. 157

far the constrictions have been found on the following
trees: soft maple, sycamore, willow, poplar, sassafras, red
gum, white oak, locust, apple, cherry, hemlock, bald cypress,
Virginia pine, Deodar cedar, larch, Juniperus virginiana,
Juniperus occidentalis, Thuja occidentalis, and Juniperus
chinensis. It is very probable that similar swellings will be
found on other species of trees where the bag worm occurs.

The effect of the girdling action of the band differed
very materially on different kinds oftrees. Two types may
be distinguished. In one case the band evidently has
stopped the passage of plastic materials through the bark
absolutely. As a result of this an accumulation
of these substances has taken place on the outer side
of the band and the cambium layer at this point has
formed wood cells and bark cells to an enormous de-
gree, as a result of which the outer part of the twig has
grown very much in diameter, particularly so immediately
beyond the band. The portion of the twig towards the
tree has practically stopped growing entirely; cases of this
kind are shown on pl. 20 and pl. 24. |

In the second case the pressure exerted by the band was
evidently not sufficient to entirely stop the passage of plas-
tic materials, or only temporarily so. In this case, a slight
swelling of the portion of the branch outside of the band
will be formed after a month or so, following the begin-
ning of wood development. At the end of the first year
a condition such as is shown on pl. 22, fig. 1 is found.
This represents a two-year-old maple twig. It will be noted
that the outer part of the twig is somewhat larger than the
part towards the trunk, but that there is also a considerable
increase in diameter of the twig on the side of the band
towards the trunk; this is shown likewise for the willow,
oak, red gum, sycamore, etc. (pl. 23). After another
year’s growth a still further increase in the diameter of the
outer part of the twig has taken place (pl. 22, fig. 3), but
the inner portion of the twig, towards the trunk, has like-

158 MISSOURI BOTANICAL GARDEN.

wise increased in diameter. Two distinct flat surfaces are
thus formed where the swellings above and below the band
face one another. In the course of time, in certain cases,
two years after the band has been on the twig, these two
surfaces unite completely around the twig (pl. 22, fig. 2),
and there is to all intents and purposes a union between
the two swellings which resembles in many respects the
union which takes place between two pieces of callous tissue
after they meet from opposite sides of a wound. After
the two surfaces have united, the development of the
branch goes on as if there had been no girdling; and the
band of the bag worm is enclosed within the swelling and
may remain there for an indefinite period. Plate 22, fig.
4 shows a branch of a maple tree ten years old. There
are no signs visible on the outside, of any disturbance due
to the insect except the old cocoon, which is still hanging
from the lower side of the branch, with not only the slen-
der filament, but also a large portion of the upper part of
the cocoon enveloped by the wood of the maple. In the
ease of the sycamore, maple and other rapidly growing
trees the disturbance caused by the bag worm is generally
only a temporary one, and in the many hundred cases
found on these trees, there is not one instance where death

followed the failure of a twig to break the band.

The formation of the single swelling on the outside of
the band is confined singularly enough almost wholly to
the coniferous trees, in some of which it is extremely
striking. The most remarkable cases of excessive swelling
with subsequent death were found on a young tree of the
Deodar cedar on the grounds of Tulane University, in New
Orleans (pl. 24, fig. 5, 7, 8). A similar large swelling was
found on a tree of Pinus virginiana growing near Wash-
ington, D. C. (pl. 24, fig. 1). The only case of this kind
among dicotyledonous trees was found in the locust
(Robinia pseudacacia). Four trees of this species were
found in a park in New York City with some ten swellings ;

CONSTRICTION OF TWIGS BY THE BAG WORM. 159

of these one is shown on plate 23, fig. 2. In every case
the outer part of the branch was dead, and no growth had
taken place in the twig between the constriction and the
tree. The marked inability of the conifers to adapt them-
selves to these unusual circumstances is very striking.

STRUCTURAL CHANGES CAUSED BY THE BANDS.

The changes which take place in the branch of a tree as
aresult of the pressure exerted by the bands of the bag
worm are of two distinct types, one of which may be
illustrated by the arbor vitae and the other by the maple
or sycamore. Inthe arbor vitae and other conifers, growth
starts under the band in June following the year in which
the band is attached to the tree. Where the expanding
force of the tree is great enough to burst the band, or
where the band is weak, no change takes place which dif-
fers in any way from the normal development of the par-
ticular twig. Where the band happens to be a strong one
which cannot be burst by the growing twig, changes of a
very radical nature make their appearance a month or more
after the beginning of the growth period of the twig. A
longitudinal section made through such a
twig after a year’s growth of the arbor
vitae is shown in fig. 1. It will be noted
that the cambium layer formed a number
of rows of wood cells immediately under
the band, but that after a very brief
period the formation of wood in the
twig, beginning at the band and below it,
stopped entirely, while the part towards
the outside of the band continued to grow
vigorously, producing the peculiar swell-
ing above referred to. The most striking change in the
immediate vicinity of ‘the band consists in the enormous
development of bark. Immediately above and below the
constriction the bark is two or three times as thick as it is
normally (fig. 1). The wood cells formed immediately

1. ARBOR VITAE.

160 MISSOURI BOTANICAL GARDEN.

beyond the constriction continue to grow normally. The
new wood formed above the band differs but little from
the normal type; the tracheids are somewhat shorter than
the normal ones, but the chief difference lies in their
manner of arrangement, for a complete turning takes
place in their position. Instead of growing longitudinally

‘they grow at right angles to the long axis of the twig. In

other words, in the region of the swelling all wood fibers
formed during the first year under the constriction extend
around the twig instead of longitudinally. In the arbor
vitae the normal direction of the wood fiber is rarely re- .
established except in such twigs as are growing vigorously,
because before a resumption of normal growth is possible
the twigs usually die. Where the development of the swell-
ing is very large, a decided buckling of the cells of the
dick frequently takes place. .

An examination made of a two-year-old constriction in
the latter part of the summer, shows the presence of very
large amounts of starch in the bark, medullary rays and
pith ring in the part above the constriction, while there
is practically none in the part below it. An inch or more
below the constriction no starch whatever occurs.

The changes which take place in hardwoods differ very
materially from those just described for the arbor vitae.
In the spring or early summer following the constriction,
both the wood and bark continue to
form under the band, both above and
below it. Figure 2 represents an
early stage in a maple constriction.
After a month or more the wood cells
<7j appear as if constricted immediately
under the band. A diversion of the
wood cells from their normal longitu-
dinal direction to one at right angles
takes place in the sycamore, maple
2. SOFT MAPLE. and other hard woods very much as it

CONSTRICTION OF TWIGS BY THE BAG WoRM. 161

does in the arbor vite, but it occurs on both sides of the
band. Asa result of the growth both
above and below the band, two lips or
rings of tissue are formed (fig.2—4).
which at first are separated, but
which soon meet above the band and
become closely appressed (fig 3). At
the end of the first year the two lips
have usually joined, and a continu-
ous ring of wood fiber has once more
been formed. As soon as the union
has taken place between the two lips 3, SOFT MAPLE.
the wood fibers once more assume their normal longi-
tudinal position (fig. 4).

The parts immediately above
and below the band show an abund-
ant supply of starch throughout
the summer and late into the fall,
_ a from which it is evident that the

FE] flow of nutrient substances from
the outer part of the twig to the
lower part is never entirely stopped
by the constriction.

The wood cells formed under
the band during the process of
healing differ considerably from
the normal wood tissue. The first wood cells formed
under the band in June, following the year when the
band is first attached to the twig, are normal in every
respect. Although the band is fastened firmly around
the twig, it is probable that it at first exerts very little
pressure upon the growing cambium cells; it is probable
likewise, that the band stretches to some extent during
the first month or two following the beginning of the
growth period of the twig. At a certain point, how-

ever, a time is reached where the band no longer stretches,
11

MAPLE.

162 MISSOURI BOTANICAL GARDEN.

and from then on the band exerts an increasing amount
of pressure upon the underlying cambium cells. A word
should be said at this point as to the structure of the
band. It is composed of a series of fine strands of silk
which are densely interwoven, forming a very compact
mass; the individual fibers appear to be pasted together
with a cementing material; they are closely adapted to the
irregularities of the bark, so that the pressure which the
band as a whole exerts is a very even and regular one.
After the pressure has begun to act a marked change in
the character of the wood cells formed under the band is
noticeable, which gradually becomes more marked as the
amount of wood formed below the band increases. After
about six weeks all further formation of wood tissue under
the band ceases. On plate 25 are shown sections of nor-
mal maple wood (fig. 1) and of a portion of the wood
formed under a band in a maple twig (fig. 2); likewise
sections of wood of the red gum from normal (fig. 3) and
constricted regions (fig. 4); and plate 26, figs. 1 and 2,
shows two photomicrographs of a cross section of maple
and red gum twigs immediately under the band. It will
be noted that as the pressure increases the number of
vessels decreases proportionately, and at a period when the
growth has almost ceased, the formation of vessels has
almost stopped. Under the increased pressure the wood
cells take on an entirely different appearance; instead of
the comparatively thin-walled, large-lumened tracheids,
one finds thick-walled, very small-lumened tracheids form-
ing, which continue to be formed up to the very time when
all growth ceases. The vessels which form at the period
of greatest pressure show a tendency to become round and
thick-walled, as wellas smallerin diameter. The medullary
rays show a decided tendency to buckling. Instead of
growing out in a straight line radially, they bend towards
one side or the other as if pulled laterally. This bending
is more striking in some instances than others. It is of

Tene st ee
Yate

CONSTRICTION OF TWIGS BY THE BAG WORM. 163

further interest to note that the very last wood cells formed
before all growth ceases still have all the characteristics of
wood fiber; they are smaller, to be sure, but there is no
mistaking them for anything but typical wood cells. Im-
mediately between the last wood cells formed and the band,
will be found the very much compressed and shriveled
remnants of the original bark, which in the majority of
cases has been so thoroughly compressed that it appears as
a dark, opaque line, in which no structure can be dis-
cerned.

The influence which pressure has upon the development
of wood tissue was first investigated by Sachs (8) and
DeVries(9). Sachs, in an attempt to explain the cause of
the formation of spring and fall wood in trees, assumed
that the cambium is acted upon by a definite pressure exerted
by the bark, and that this pressure increases from the
spring to the fall, equalizing itself during the winter; he
took the formation of cracks in the bark in the spring as
an evidence of the growth energy of the woody cylinder
overcoming the bark pressure acting against it. DeVries,
in an endeavor to explain the causes of spring and fall
wood, made a series of slits in the bark in the fall, as a
result of which he noted the formation of spring wood.
This he explains by imagining that the bark pressure was
destroyed because of his making slits, and consequently
large-lumened cells were formed. He likewise placed a
number of ligatures on branches inthe spring and observed
the formation of summer wood under these ligatures. He
concludes from these experiments that the radial diameter
of the wood elements is determined by the bark pressure
exerted during their formation; and furthermore that the
number and size of the vessels in any one wood ring are
determined by the bark pressure which happens to be
experienced at the time of formation of this particular
ring, higher pressures resulting in fewer vessels which are
of smaller diameter.

164 MISSOURI BOTANICAL GARDEN.

Krabbe (5) disagrees with both Sachs and DeVries as to
the existence of bark pressure. In a series of extensive
experiments he shows that the difference in the bark pres-
sure during the spring and summer is so small that it could
not possibly account for the difference in structure of
spring and summer wood. In another series of experi-
ments Krabbe (6) increased the bark pressure by means of
- wooden cylinders formed into a chain, to which weights
were attached; by increasing the pressure on the growing
tree he found that one can increase the bark pressure two
or three times without influencing the growth in thickness
of the branch. For conifers he found that a pressure of
from three to five atmospheres slowed up the growth and
changed the radial diameter of the cells, while a pressure
of ten atmospheres stopped the growth entirely. For
broad-leaved trees he found that a pressure of from five to
seven atmospheres reduces growth; but he failed to deter-
mine the upper limit of growth. In discussing the mor-
phological changes which the wood cells undergo as the
pressure increases, he notes in particular a reduction
in the number of cells formed as the pressure increases,
and the gradual change of the vessels from an ellip-
tical cross-section to circular. Another point which
he notes is that the cambial cells continued to form
wood cells of a normal character, irrespective of any
pressure which he was able to bring to bear upon
them. The newly formed wood cells, which under ordi-
nary circumstances are capable of expanding until they
reach a size very much larger than when first formed, are
materially influenced by the pressure, and as the pressure
increases, the stretching of the newly formed wood cells
may cease altogether, thereby giving rise toa much smaller
Jumen than is normally the case. Although Krabbe states
nothing as to the shifting of the medullary rays, he shows
(in two figures) a shifting similar to that found in the

CONSTRICTION OF TWIGS BY THE BAG WORM. 165

sycamore and maple under the pressure of the bag worm
bands.

Kiister (7), in an investigation on the union of apposed
branches, states that the first effect of the mutual pressure
of two such organs, aside from the changed direction of
the medullary rays, consists in a flattening of the surfaces _
apposed, due to growth reduction. An increased growth
of the primary bark at the edges takes place. He does not
agree with Krabbe as to the activity of the cambium, find-
ing that the growth of the cambium does not remain nor-
mal up to the point of union, but that growth stops entirely,
because of too great pressure. Before it stops growing,
Kiister finds that the cambium forms a more or less well
developed layer of modified woody tissue, which Krabbe
does not mention at all. This modified layer of woody
tissue differs from the normal wood in that it is composed
of parenchymatous cells, and is to be considered as a par-
encbymatous tissue. It consists of libriform vessels and
parenchymatous cells, which are more or less thickened,
pitted and lignified. For the sycamore, Kiister states that
‘it is surprising to see the extraordinary development
under the influence of pressure; parenchymatous tissue
develops, which starts sharply defined from the normal
wood. Outside of this normal wood one finds a thick layer
of radially elongated, lignified and very much pitted
parenchymatous cells. True wood cells continue to be
formed only here and there, and in a cross-section the
groups of these wood cells enter into the parenchymatous
mass in a wedgelike manner. Finally the formation of
even these isolated groups of wood cells etops entirely. . .
The pressure is not the physical cause of this growth,
but acts only as a stimulus, which affects the protoplasm,
and the real cause must be sought in the nature of the
latter itself.’’ Kiister discusses the apparent discrepancy
of his own and Krabbe’s results, and sees an explanation in
the different manner in which the pressure was applied; in

eee ee oe ee

Pet Lie ee ee

166 MISSOURI BOTANICAL GARDEN,

one case with rollers acting for a short period and rapidly
applied, while in the case of Kiister’s unions it started as
a slight pressure and, increasing to a maximum, continued
for a long period of years until a pressure of 11-17 atmos-
pheres was reached. Kiister furthermore calls attention to
the change in direction of growth of the cambial cells, which
normally form woody cells with their long axes parallel to
the axis of the stem, but which under the influence of the
pressure change their direction 90 degrees, many of them
being curved and bent into **C”’ and‘*S”’ forms. He
regards these as originating from bent and displaced cam-
bial cells; their formation being a physical process, and
their displacement probably due to the pressure coming
only from one side, which gives these cells an opportunity
for lateral development.

In the case of the bands of the bag worm, the pressure
applied was very similar to that obtained in the investiga-
tions of Krabbe, with the exception that the force was ap-
plied gradually, and in that manner was more like the
pressure obtained by Kiister when he caused two branches
to join by being pressed together at one side. From the
description of the changes which take place in hardwood
trees particularly, under the influence of the constriction
caused by the bands of the bag worm, it will be noted that
these changes are almost identical with those described by
Krabbe. In no case is there any evidence of the formation
of the parenchymatous tissue spoken of by Kiister. Ex-
amination of branches of maple, sycamore, red gum, wil-
low, and other hardwoods, showed that in every case, in
spite of the enormous pressures that were exerted by the
band towards the latter part of the growing season, there
was no cessation on the part of the cambium in the for-
mation of typical wood cells, though the last cells formed
were very much smaller, just as were those found by
Krabbe under his rollers (pl. 25 and 26). I am inclined
to agree with Krabbe that the pressure does not in any

CONSTRICTION OF TWIGS BY THE BAG WORM. 167

way influence the division of the mother cells of the cam-
bium, but that any change which takes place occurs after
the wood cells have been formed. The cambial cells
divide even under great pressure, though when the pressure
reaches a certain point it is probable that the rate of divi-
sion is considerably reduced. The formation of new cells
evidently stops abruptly; in other words, the conditions
for the development of new cells from the cambium are
such that the latter is able to form these new cells irre-
spective of the growing pressure up to a certain point, and
when this point has been reached growth stops. The cam-
bium appears to be more or less uninfluenced in the char-
acter of its cell divisions by pressure exerted upon it, and
is able to continue its activity up to a maximum point.

- Concerning the discrepancy between the findings of
Kiister and those of Krabbe and myself, I am inclined to
ascribe the formation of the parenchymatous tissue in the
branches investigated by Kiister to the unequal pressure to
which the apposed surfaces were exposed. I have found
atissue which resembles that described by Kiister under
similar conditions in stems of maple and other trees in the
region immediately below twining vines, particularly where
the latter caused a constriction in the bark of the trunk
around which they were growing.

The change in direction of growth of the cambial cells
resulting in a shifting of the axis of the wood cells to an
angle of 90 degrees from that usually occupied by them,
has already been referred to above.

Summing up the changes which occur in the region of
wood under the band, it may be said that with an increase
of pressure smaller and smaller cells are formed, because of
the inability of the newly formed wood cells to expand to
a normal size; the cambium continues forming wood cells
of the usual kind but at a decreased rate as the pressure
increases. Where the pressure is reduced, at the edges of
the band, a shifting of the cambial cells takes place, result-

168 MISSOURI BOTANICAL GARDEN.

ing in the turning of the long axis of the wood fibres at an
angle of 90 degrees to their usual position.

It has been stated that as soon as the pressure exerted by
the band becomes great enough to retard the expansion of
newly formed wood cells, an increased growth becomes evi-
dent above the band in conifers, and both above and below
the band in hard woods. The cambial tissue in the region
not acted upon by the pressure at first continues to grow
normally, but very soon increases its rate of cell division.
Under the changed condition of tension a change of direc-
tion of the cambial layer takes place ; instead of remaining
a sheet parallel to the long axis of the branch, the cam-
bium both above and below the band is curved outward.
The wood and bark cells produced, because of this
altered position of the cambium, instead of growing in a
radial direction, as they do normally, are turned so that
their axes gradually approach a direction at an angle of 90
degrees from their normal position. With this changed
position in the direction of growth of the wood and bark
cells comes an increased development of both, so that after
awhile the branch is very much thicker immediately above
and below the band than it is at some distance from it.
The swellings appear as the lips already referred to,

The tissue composing the two lips is, to all intents and
purposes, wood tissue. The tracheids are normal in char-
acter, although many of them are somewhat shorter than
those formed under normal conditions; there is, in other
words, no development of what might be termed callous
tissue or anything resembling the tissue known as * wound
wood.’’ The development of the lips, in many respects,
however, resembles the development of callous tissue after
a ring of bark has been removed from a twig. Inthe latter
case, the stimulus, as a result of which an increased devel-
opment of tissue takes place at the margins of the cut,
consists in the interruption of the strains normally active in
the growing region of the twig. By cutting away a certain

CONSTRICTION OF TWIGS BY THE BAG woRM.- 169

layer of bark the cambium is free to develop without the re-
straining pressure exercised by the bark, and as a result
grows rapidly in the direction of the least strain. This
growth continues until the callous surfaces formed on the
upper and lower sides of the wound unite and re-establish the
normal pressure conditions. In ‘the case of the constric-
tions caused by the bands of the bag worm, a similar dis-
turbance is brought about by arresting the development of
the cambial layer for a certain distance completely around
the twig, and thereby disturbing and probably rendering
inactive the normal bark pressures. To all intents and
purposes therefore the formation of the lips of the bag
worm constrictions may be said to be due to a similar stim-
ulus to that which results in callous formation after a
wound, though the formation of the lips is not to be re-
garded as a callous development, but rather as a new type
of healing of a different character.

It is a well-known fact that, where branches are girdled
by removing a portion of the bark down to the woody
cylinder, the development of callous tissue takes place from
the upper or outer edge of the wound. The transfer of
carbohydrates and nitrogenous substances takes place in a
downward direction from the leaves towards the trunk ( Cza-
pek,land2). Some writers, e. g., Fischer (3), have gone
so far as to maintain that the carbohydrates formed in the
leaves go downward only in the bark, and that in girdled
branches they cannot even be transferred across the wound
through the pith or wood parenchyma. Although very
little attention has been paid to the problem of nutrition in
the healing of the disturbance caused by the bands of the
bag worm, it is nevertheless suggestive to note the differ-
ence in the manner of healing between the coniferous and
hardwood trees. As soon as the pressure exerted by the
bands becomes great enough to obstruct the cells of the
bark, an evident congestion takes place in the region above
the band. In the coniferous trees the food supply is evi-'

4

170 MISSOURI BOTANICAL GARDEN.

dently entirely cut off, and as a result practically no
growth takes place in the twig on the side below the bands.
The portion of the twig above the band continues to grow
for a considerable period. The absolute cessation of the
passage of starch and other substances from the leaves
towards the trunk is well shown by the total absence of
starch in the region below the band in the fall of the year,
while in the region above the band the wood parenchyma
and bark cells are completely filled with starch. In the
hardwood trees the pressure of the band is evidently never
great enough to entirely obstruct the passage of elaborated
materials. No cases were found in the large number ex-
amined, with the exception of the locust, where the con-
striction caused by the band was great enough to prevent
the growth of the part below it. In the fall of the year
starch is found in the parenchymatous tissue both above
and below the band in almost equal quantity. The differ-
ence in the behavior of the conifers and the hard woods
is a suggestive one, as indicating the greater adaptability
of the latter, and their higher grade of physiological organ-
ization.

GROWTH ENERGY OF TWIGS AS INDICATED BY THE PRES-

SURE OF THE BANDS.

It has been stated that in the normal course of events
the twigs burst the bands of the bag worm shortly after
the beginning of the period of wood formation in the early
summer of the second year. Now and then the strength of
a band was such, however, that the twig failed to burst it,
and as a result the constrictions just described were formed.
The bursting of the bands has been explained as due to the
force exerted by the twig after it began to grow in
diameter. Putting it in another way, the bursting of the
band may be said to be due to the growth energy developed
by the growing twig. That plants do exert a certain
amount of growth energy is a familiar fact. It is well
known how the roots of trees will lift the flagstones in the

CONSTRICTION OF TWIGS BY THE BAG WORM. 171

streets of our cities; the lifting and pushing power of roots
and trunks is furthermore frequently noted in the forest,
where large boulders are pushed aside or lifted. Another
familiar instance of the energy exerted by the growth of
plant tissue is found in the roots of most higher plants,
which exert a certain amount of energy in pushing aside
soil particles. Although this growth energy may some-
times be very large, it frequently is not sufficient to over-
come the obstacles to which growing parts are subjected.
This is frequently illustrated by roots which grow in the
clefts of large boulders— they are frequently unable to
burst these, and as a result, are forced into a flattened form.
Many climbers exert a sufficient pressure upon the trunks
around which they are growing to arrest development.
Wires and other bandages frequently bring about girdling
of trees very similar to those described for the bag worm.

While it has been realized that the amount of energy
exerted by growing tissues may sometimes be very large,
very little is as yet known as to how large the amount of

energy developed by growing tissue actually is. Inthe course ~

of the investigation on the constrictions caused by the bag
worm bands, it was suggested that these bands might serve
to indicate something as to the extent of the growth energy
of young twigs. In the largest per cent of the cases the
twigs burst the bands; they must therefore exert a radial
pressure outward greater than that usually exerted by the
band in an opposite direction. A series of experiments
was therefore undertaken to determine how great the pres-
sures were which the bands exerted upon the twigs, with
the object of using the results as a guide to determine the
growth energy of the twig.

Of the investigations hitherto made to determine the ex-
tent of the growth energy of plant tissues, there is practi-
cally only one which needs consideration. Krabbe (6)
undertook some years ago to determine the extent of the
growth energy of various trees, by placing bands, consist-

172 MISSOURI BOTANICAL GARDEN.

ing of a series of rollers, on the growing branches, and
causing these bands to press with varying loads upon the
branch, by attaching weights. For the details of Krabbe’s
method, the reader is referred to his original paper.
Krabbe placed different weights on different branches, and
after having left them for a season, he made microscopic
examinations of the tissues immediately under the band.
He attempted by so doing to find what weight, or in other
words, what radial pressure, would stop the increase in
diameter of the branch. He holds that this weight would
be equivalent to the growth energy of the twig.’

As a result of his investigations Krabbe concludes that
the growth energy of the cambium ring and young wood
cells in coniferous trees is at least 10 atmospheres. For
hard woods he finds that the growth energy is at least 15
atmospheres. Krabbe was unable to determine the exact
value of the expansive force of the limbs and branches with
which he experimented. His results simply show that
with the greatest pressures exerted by-his rollers, viz: 10
atmospheres for conifers and 15 atmospheres for broad-
leaved trees, growth still took place under the bands. The
figures which Krabbe obtained are to-day in general accep-
tance as probably representing the approximate minimum
growth energy of branches. Friedrich (4) makes the state-
ment that ‘‘ the increase in diameter of the tree trunk takes
place with a development of energy equal to at least 10
atmospheres.”’

In order to arrive at some conclusion as to the amount
of energy required to burst the bands of the bag worm, it
was necessary to determine, first of all, the pressure which
the bands were capable of exerting radially. The radial
pressure exerted by a band placed around a cylinder is de-
termined by the formula

pbs io
y rx oe
in which P equals the radial pressure exerted by the band

CONSTRICTION OF TWIGS BY THE BAG WORM. 178

per square mm. of surface, S equals the strain which the
band is capable of exerting tangentially; or, in other
words, the strength of the band which resists tearing,
r equals the radius of the twig around which the band is
fastened, and w equals the width of the band. A num-
ber of microscopic examinations were made to determine
the variability of thickness in the band, which ought to be
included as a factor in the above formula. The variation
was found to be so small, however, that for the purpose of
the experiment the thickness was left out of consideration.
Several hundred twigs from which bag worms were sus-
pended were collected in the spring of the year from shade
trees in and near the Missouri Botanical Garden. The tree
upon which the bag worm is most commonly found in St.
Louis is the soft maple, and the bags found on this tree
were made the chief subject of study. Another reason for
go doing was that the bands found on the maple twigs were
usually of a sufficient size to permit being measured, while
those on other trees, and particularly on the conifers, were
usually so small that with the apparatus at hand it was im-
possible to make exact determinations. The method which
was used to determine the strength of the bands was as
follows: With a sharp knife the bands were cut as near to
their union with the bag as possible; the bag with the band
was then removed from the twig without difficulty, and a
second cut was made in such manner that the band was
separated entirely from the bag. A flat sheet was thus
obtained, varying from 6 to 12 mm. in length. After the
band had been thus removed, a careful measurement was
made of its width and of the diameter of the twig from
which it was taken. After numerous trials, an apparatus
was constructed consisting of a framework, from the top
of which a pair of pliers was suspended, with the jaws
pointing downward; a similar pair of,pliers with the jaws
pointing upward was arranged s0 that a scale pan was
attached to its handles. In both cases the pliers were so

174 MISSOURI BOTANICAL GARDEN.

arranged that when a pull was exerted on the lower pair
the handles of both pliers became more and more firmly
compressed, thereby increasing their hold upon the bands
placed between the jaws. When ready for the test the
bands were carefully straightened, and the ends were
placed between the jaws of the two pliers; only enough of
the band being caught by the pliers to give a firm hold.
_ Very fine shot was then gradually put into the scale pan
attached to the lower pair of pliers until the band broke;
the weight of the shot, plus that of the pan and lower pair
of pliers, constituted the breaking weight of the bands.

It was at first thought that the weakest point in the
bands would be where the bands were fastened to the bag ;
in order to determine whether this was the case, a number
of tests were made by attaching weights to the bags them-
selves, and increasing the weights until something gave
way. In every one of the preliminary tests the break in
the band occurred at some point away from the bag at-
tachment. In other words, it appeared that the weakest
part of the band was not at or near the bag attachment.
It was therefore considered to be perfectly fair to cut away
the bag and to enclose the ends of the bands between the
jaws of the pliers; the results obtained by breaking hun-
dreds of the bands bore out these preliminary conclusions.
It was found that in almost all cases the bands broke at
some point between the jaws of the two pairs of pliers and
not immediately at the edge of the jaws. It was further-
more noted that under most conditions where the bags
were attached to twigs on the growing tree the break in the
band occurred ina similar manner to that found when
weights were attached to the bags themselves. An exami-
nation of a large number of bands made during June showed
that with the increase in diameter of the twigs which
brought about the bursting of the bands on the trees, the
bursting of the band took place at a point away from the
bag insertion.

CONSTRICTION OF TWIGS BY THE BAG WORM. 133.

The results obtained were calculated according to the
formula given above; and, following Krabbe’s method, the
final figures were calculated to show thepressures in grams
per square millimeter of bark surface, which were then re-
duced to atmospheres of pressure. For this purpose 10
grams were used as equivalent to 1 atmosphere of pres-
sure per square millimeter. In the case of the maple some
400 bands were tested. It is proposed to test a great
many more bands in the future in order to have a larger
number of figures to work with.

Before drawing any conclusions as to the results of the
strength tests, it should be stated that the use of the bands
as an indication of the amount of growth energy exerted
by the twig must be taken with some caution. No claim
is made that the results to be discussed are in any way
absolute; the bands are an extremely variable quantity ;
they differ in width and thickness, and the error in meas-
urement is consequently to be considered. It was further-
more suggested that there might be a very decided differ-
ence in the strength of the bands, depending upon whether
they were wet or dry, and likewise upon their age. The
former point was considered of great importance, and
accordingly a large number of bands were broken after
they had been soaked in water for five days. It was
thought possible that as a result of wetting during rains in
the early summer, the bands might be considerably weak-
ened, and that it would be at such a time that they were
burst by the twigs. The results obtained from the tests
with soaked bands in no way differed, however, from those
obtained with dry bands, and it is accordingly the belief
that the soaking practically does not weaken the bands at
all, or at least very little. Concerning the effect of age,
most of the bands were taken and tested during May and
June, 7. e., after they had been exposed to the weathering
influences during the fall and winter. It is therefore be-

176 MISSOURI BOTANICAL GARDEN.

lieved that if the weather exposure weakens them, they
were tested at the time when they were weakest.

The first analysis of the measurements made deals with
the size of the bands and their relation to the size of the
twigs to which they were attached. It has already been
stated that attempts were made to cause the insects to weave
bands around large twigs, but in every case the insect
refused to do so. On Table I the bands taken from

RELATION OF DIAMETER OF MAPLE TWIGS TO WIDTH

OF BAND.
Width Diam Total
a Twig. io)? as | oe as) eae ae ee eee
mm, mm, width
1 8 2 1 6 2 14
1.5 2 13 6 2 23
2 tS ae bee 46 16 | 12 3 2 1 114
2.5 4 | 16 28 12 5 65
3 beep | 8 52 16 | 18 5 1 1 122
3.5 a: 4 4 11 6 3 1 30
4 4 5 10 1 5 1 1 1 28
4.5 1 2 8
5 2 + 1 1 8
5.5 1 1
6 1 1
6.5 0
7 2 2
7.5 1 1
8 1 1
Total bands per
CW ee a ea” BVA OB. 7671-869 65 | 43 | 10 6 3 413
TABLE I.

maple twigs are arranged according to their width and
according to the diameter of the twigs. It will be
noted that the largest number of bands occurred on
twigs 3 mm. in diameter, and that there is a regular
falling off towards both sides; in other words, the insect
seems to select twigs having a diameter of about 3 mm.

CONSTRICTION OF TWIGS BY THE BAG WORM. 177

Concerning the width of the bands, there is more variation.
Of the 413 bands, 122 were 3 mm. in width, 65 were 2.5
mm. and 114 were 2 mm., in decreasing numbers of smaller
and greater widths. The largest number occurring on any
particular diameter of twig were 3 mm. in width. Prefer-
ence for the twigs 3 mm. in diameter is furthermore shown
on Diagram 2, a, representing the relation of strength,
using the twigs of different widths as the variable, and
Diagram 2, b, where the bands of different widths are
used as thevariable. The final results obtained with bands
taken from maple twigs are plotted in Diagram 1; the
figures at the bottom of the table indicate the number of
atmospheres pressure exerted by any band upon the particu-
lar twig which it subtended. The figures at the left of the
table represent the numbers of individua] bands exerting
any particular number of atmospheres pressure. There
were, as may be seen from this, four bands which
exerted a pressure of 10 atmospheres and 16 which
exerted a pressure of 19 atmospheres each, etc. While
the number of bands which were broken was a rather small
one for the purpose of any definite statistical study,
enough were tested, nevertheless, to show a number of in-
teresting facts. The radial pressure exerted by the bands
varied from 4 atmospheres to 162 atmospheres in the case
of the maple. The greatest number of the bands varied
between 14 and 44 atmospheres. The strength which
some bands showed was surprising. On Table II a num-
ber of instances of extreme strength are shown for maple
and other trees. The deductions which may be made in a
tentative way from the results shown in Diagram 1 are as
follows: A great variation in the pressures exerted by the
bands of the bag worm on the twigs of maple trees exists.
Some of these pressures are very great, and it is probable
that bands like those which showed pressures of 102 to 162
atmospheres are the ones which are responsible for the
12

178 MISSOURI BOTANICAL GARDEN.

constrictions described above. Remembering that most of
the bands which occur on maple twigs are burst by the twig
every year (only 1.5-2 per cent are not burst), it seems
probable that the larger number of bands represented in
this table would have been burst during the growing period.
As many of these show that these bands exert radial pres-
sures running up to 40 atmospheres and more, it seems

EXCEPTIONALLY STRONG BANDS.

Tree on which bags | Widthof | Diam. of | Total Breaking Atmospherio
occurred. band. mm. | twig. mm. | weight. grams. pressure.

White Pine.... 1.5 1.5 935 83.1
3 2 2275 75.8
3 2 2472 82.4
1 2 650 65.0
2 1.5 1090 73.7

MADIG . 6 dt HA 3 3 8202 71.1
8 3 4324 96.8
1 2 668 133.6
2.5 3 8727 100.7
3 2 8632 121.1
5 4 1620 162.0

Sycamiore..... 3 2.5 1778 177.9

WAUOW 66a 1.5 1.5 1217 108.1
2.5 ls 2. 8272 130.8
8.5 3 4737 90.2
2 1.5 1632 108.8
2.5 1 1775 158.0
2 2 2130 106.5
1.5 1 900 120.0
2 2 2350 117.0
4 2 38675 91.9

TABLE II,

likely that the pressure which the twigs are capable of
exerting radially must be at least equivalent to these pres-
sures in order to burst the bands. It will be remembered
that the figures which Krabbe obtained, showing that the
pressure exerted by growing twigs of hard woods was 15
atmospheres or thereabouts, in no way indicated what the
upper limit might be. Judging from the structure of the
wood which he shows, cell division was still taking place

CONSTRICTION OF TWIGS BY THE BAG WORM. 179

with considerable activity under the bands weighted so as
to exert a pressure of 15 atmospheres, and Krabbe states
distinctly that the growth energy must therefore be equiv-
alent to at least 15 atmospheres. As a result of the few
tests herein described, I am inclined to the belief that the
pressures are probably very much higher than 15 atmos-
pheres. Owing to the variability of the bands it does not
seem practicable, particularly with the small number of
bands tested and in the absence of determinations as to
their thickness, to place any absolute value upon the
growth energy as determined by the bands. That the
growth energy exceeds 15 atmospheres seems very prob-
able, and I am inclined to the belief that it may reach the
extent of 30 to 40 atmospheres, and possibly more. Asa
matter of statistical interest, Diagram 2, a and b, is added.
The upper diagram (a) shows the relation between the
strength of bands of different width, andthe lower one (b)
the relation between the strength of bands found on twigs
of different diameters. Both of these tables show that the
bands of greatest strength are those that are attached to
twigs of 3 mm. in diameter.

BIBLIOGRAPHY.

1. Czapek, Fr. Uber die Leitungswege der organischen
Baustoffe im Pflanzenkérper. (Sitzungsber. d. Kais. Akad.
d. Wiss. zu Wien, Math. Naturw. Cl. 106:117. 1897).

2. Czapek, Fr. Zur Physiologie des Leptoms der Angi-
ospermen. (Ber. d. D. Bot. Ges. 15:124. 1897).

3. Fischer, A. Beitrige zur Physiologie der Holzge-
wichse. (Pringsh. Jahrb. f. Wiss. Bot. 22: 73. 1891).

4. Friedrich, Josef. Uber den Einfluss der Witterung
auf den Baumzuwachs. (Mitth. aus d. Forst. Versuchs-
wesen Oesterreichs. 22 :1897.— Ref., Bot. Zg. 55 : 368.
1897).

5. Krabbe,G. Uber die Beziehungen der Rindenspan-

Wie eS Oe ee, ee 4 ge

180 MISSOURI BOTANICAL GARDEN.

nung zur Bildung der Jahrringe und zur Ablenkung der
Markstrahlen. (Sitzungsber. der Kgl. Preuss. Akad. d.
Wiss. 51: 1093. 1882).

6. Krabbe, G. Uber das Wachstum des Verdickungs-
ringes und der jungen Holzzellen in seiner Abhingigkeit
von Druckwirkungen. (Sitzungsber. d. K. Preuss. Akad.
d. Wiss. 53:57, 69, 1884.— Also Separat aus Abh. d. K.
Preuss. Akad. d. Wiss. 1884).

7. Kiister, E. Uber Stammverwachsungen. (Prings.
Jahrb. f. Wiss. Bot. 83:487. 1899).

8. Sachs, J. Lehrbuch der Botanik. 409. 1868.

9. de Vries, H. Del’influence de la pression du liber
sur la structure des couches ligneuses annuelles. (Extrait
des Archives Néerlandaises. 1876).

EXPLANATION OF PLATES.

Plate 20. — Constrictions caused by the bands of the common bag worm
(Thyridopteryx ephemeraeformis Haworth) on branches of arbor vitae
(Thuja occidentalis). Natural size.

The two upper figures are living branches. The lowest one shows a
branch kiiled by the band of the bag worm.

Plate 21.— The bag worm on living one-year-old branches, as it appears
a month after the attachment of the bags, usually about the end of
September. The upper figure shows it on a sycamore branch, the lower
on a branch of the ash-leaved maple. Natural size.

Plate 22.— Constrictions caused by the bag worm on maple twigs of
various ages. 1. Twig two years old showing early stage of constric-
tion. 2. Twig 2 years old, showing beginning of union of two lips.
3. Twig 3 years old with healing almost complete. 4. Branch ten years
old, healing complete, and top of bag partially overgrown by the wood.
Natural size.

Plate 23. — Constrictions caused by the bag worm in various hardwood
trees. 1. On willow. 2. On locust. 8.On sycamore. 4. On red gum,
6. On sassafras. 6. On poplar. 7. On white oak. 8. Onapple. 4.

Plate 24.— Constrictions caused by the bag worm on various conifer-
ous trees. 1. On Pinus virginiana. 2. On Thuja occidentalis. 3. On
bald cypress. 4. On larch. 5. On Deodar cedar. 6. On hemlock. 7,
8. On Deodar cedar. X 4.

Plate 25.— Cross sections showing structure of normal wood, and
wood from the bag worm constrictions. 1. Normal wood of the soft
maple. 2. Wood of the soft maple taken froma point immediately under

,

pO emerald

eee c

CONSTRICTION OF TWIGS BY THE BAG WoRM. 181

the band of the bag worm where growth had stopped entirely. 3. Nor-
mal wood of the red gum. 4. Wood of the red gum taken from a point
immediately under the band of the bag worm where growth had stopped
entirely. >< 400.

Plate 26. — Photomicrographs showing the cross sections of wood
taken at points immediately below the bands of the bagworm. The
bands were situated at the top of both figures. Upper figure, from
a maple constriction. Lower figure, from a red gum constriction.

>< 50.

DIAGRAMS.

Diagram 1.— Showing the number of atmospheres pressure exerted
by the bands of the bag worm, and the frequency with which the vari-
ous pressures are exerted among 413 bands. The abscissae show the
number of atmospheres pressure; the ordinates indicate the number of
individual bands exerting any given pressure.

Diagram 2.—a,. The relation between the varying widths of bands
and the strength of the bands, expressed in terms of number of atmos-
pheres presssure. The abscissae represent the number of atmospheres;
the ordinates the number of individuals. —b. The relation between the
varying diameters of twigs and the strength of the bands found en-
circling them, expressed in terms of number of atmospheres pressure.
The abscissae represent the number of atmospheres; the ordinates, the
number of individuals.

TEXT FIGURES.

Figure 1. — Longitudinal section through an old swelling of the arbor
vitae ( Thuja occidentalis), showing how the upper or outer portion grew
in diameter, while the part below the band stopped growing entirely.
(P. 189).

Figures 2-4. — Longitudinal sections through constrictions of the soft
maple, caused by the band of the bag worm. The figures show succes-
sive stages of healing. Figures 2 and 3 show the top of the bag at the
left of the constriction. (P. 160, 161).

TEXT TABLES.

Table I.—Showing the relation of the diameter of maple twigs to the

width of the bands of the bag worm which were found upon them.

(P. 176).
Table II. — Showing a number of exceptionally strong bands, their
dimensions and the trees they were found on. (P. 178).

a

REptT. Mo. Bot, GARD., VOL. 17.

PLATE 20.

CONSTRICTIONS OF ARBOR VITAE.

REPT. MO. BOT. GARD., VOL. 17.

PLATE 21,

BAG WORMS— FIRST YEAR.

PLATE 22.

Rept. Mo. Bot, GARD., VOL. 17.

CONSTRICTIONS OF SOFT MAPLE.

REPT. MO. BOT. GARD., VOL. 17, PLATE 23.

CONSTRICTIONS OF HARDWOODS.

REP?T. Mo. Bot, GARD., VOL. 17. PLATE 24.

CONSTRICTIONS OF CONIFERS.

- i i eo ce ae

PLATE 26.

REPT. MO. BOT. GARD., VOL. 17.

WOOD BELOW BANDS.

DIAGRAM

Tt : i i
+ f a # +: ate . : 7
i i! zi
i tf ses ipes tEHE! Bi tt p
3 + ¢ : t *
: i i 3 tte 8 ; 3
i
se tS babs sit ee tec sists +H astieiiiiy
+ : i
+ $I i + spe fis ides:s3:¢: ttt
ite i ; in teats {
f : 3 258 Hapies
: i aie isis Biilieae
33 HN SHEE ests sh :
is
i? 3 +)
ri 3] 4 bees o sf
it
psiyitty
3 sss
of 53
: i
+
i 7
®
it : wn
Ee £2
Stet: < 14
i it 3
ai ee
att 7 MN
Gt OQ
abit A j Fat
e t < gi
a pa~ dee

+h
4ay,

ND STRENGTH OF BA

et

b. TWIG DIAMETERS A

WIDTH AND STRENGTH OF

a.

PRESSURE EXERTED BY BANDS.

i pe Da
ee
_ ane ‘pee
a itt Lit = +T ne L-|
igs ae
: ty as pa faauee
. ~
+ es bed
| 5 aa58 1m i tT *
tt = Tet ae
9 +t T a TTT v
iat anne ' aaa
a. Ty LES Ce OG
5 | ste
sannna
3 44H +4
whe gp nes gee
P mitt sat
riyat it
bebe teat et re
at
|:
iH
|
aie
ert
aass
I>
+r
sueas +4
n
an:
aeere Ht
q ap
fee
2
a rt i++
«as
te t+ 1 rT
4
.
Pri ae
++]
Ly } t
Seane seessqececnansoees
a
i+. 4 ese
TT sam
ae Pi Ly
wae : ageges 2 os see sgee
22 oe q + be +4 be deh = Ji
a +H ae Cy paenee gees ess eaeo ae ye
rs Lt i444 + 4 pes Lt fas
7 gue
rH TH tit 1 trea ttt auune
+ 4 4 t i
r ; * tae +
s a $44} ; + 4 Bos Sanne HE
: : zilete ii
ian 4. sane L +) +4 + ane - b = = =
~ — ee J
HH +44 gaan as sedan uanns tH ait +
: 4 on +] +4 gd tf
va ‘ 9
» > : a tT ee 4
. we 5 rstr
ir }' +! pit p e -4 anne an
we = Lad pene ym 4
a +4 3
= =e ee - itt % aS
scene is) ae sere: 53 Bs ~~ aan ame
oa ar Bes eeaeee }
en = === crs. HH sosgee
fa ae Ps
aa
ee + +r t
sae ' age 44 TT
TT n
+ |
To H
a 1 + |
= 1s ss ri
eeee se tt ae
TT
*
Seenes H
4 ap Sues P
THT 344 a"
m
> oe -
TEL
3 rer
1
ane
seosaese 5 :
is Take
Prt Tit L] s X 3
gaanpee segsane . eaepenases
s 1
oa t
al
cageee CS
Sgece dena *
seeecay
|. Sink geae
- = | ana
5 +4 r
neee Ltt i
- 2 a i +-} -|
C 4 4 Ky r
im +H r t <1
Tr gai rr
rr
ne
444
cups is :
r A4tt tet rr
a o.
++H4
ea Cet T enge
tT S :
pa at u
TT : Sn weni aq hnawe
thesgas HH + f
inal
Tht L — Lo
- +H HEF Ss +4
7 tT Riegen
na rT. am f
ib
rH + Ht tte sieeagesee
Tey suey Ect sasenee
im paeeers
’ +t ++} + punsens
iil
t TE jane pe wi
aay i jane ages
aunae
ae :
+ TE
T .e a T
1
as C - ine
an th +e ae 6
Fttt4 q
sheen | x mi
Pit Th jase ws a ry +
sseves seenes
. eae byt
reap ee + a8 { r
Eee Ld
asee tH Seesese
Tt 2
+ bios 4 feet | 4-44
a vt. sees +1
t 3 meeesead
3 “ 4 i bt
t TALE sagapeas
eae “Lis. r an a
pb baal. dated +4-+ 4 + sere ++ 4
+ sae : 4 So
: . i sas suaee
} b+ b+! f bm! oes
ig 4 age
saps br 4 +34 be hat ) - sees
ma eae
fi
14 eee
apes seunaannn 6h: a i
Saee ee i+ H+ peuees
Ci] sendhen 4 , baled trae. wapaas put age | bet
ttt! -+44 ttt 4 +4 ay geueece
L sgepuss
a neal pa keaws r —headie get =)
¥ 4 + 14 + PH ot bit t.
seayees f+ s +4 apagage unpese supuovs ee
i im TT! Lt. gees ] eugene
Le a be bs
4+ Tit tr enbese rrst
sgn aapee +
aereaes i, r+ be
~ Besse . TH
sane La ae
. aa a9
t } : si
T ap at ti itr TTI ena
t gua Is
Sense
yr seugecg. canoes
passes
Ht ehaue apes
eal Sesenes
iii.
on ++ Saggan Ho a: dupee
tt asceee danggas
aapas - : cups aan eeacee H
sees s
se agaues aapuse ekueee . «
sae go sesqease as H :
Pm: 4
isa ® " + s .
an tT THT ®
@ ® I +:
1 _ f nk Ps s
; ty TET [ty i CI Ton
= esaeu geseas segade
anaaes eps ands see +4 4 rH
an TT 3 t+ +44 -- tt So. +
tT. 5 a - ++ anak “4 Ht 4
SeeSEeeseces Pet H : :
Tr TT TE:
Y gauss . Hh ess cucceeha -gpecaae .
t : the tT trids ss Ht ob +
nos
f + sacuye
ttt Tr ITT Tit it Lor 1IT jae Rage se $ +f
Tr. Try Trt Tit T41 p T sgeees
t TTT thy tir Tit ss ‘
pes +44 t+ ft + ++ +4
‘ seus That Bestes
‘ ++ aseea
eee
1 t sales . + eae
itt pat TT gt ry ee a aan
seb isegecebeness secinses
oa _ ppt
ed + + +4
ion 7 age
enegks
= —
jee
epegse H 4 +H
> ayngss
= : ir :
re)
Bass sgtesae
<4 Ty
a paguere
5
Pe) 4+
s + ao
| f ties:
: i
. r T +f f
HW a= + t+ « 7
® eas pees
i saith suit FESO WS
sass ;
g Lt LiL Li Lit Tit s TI rr s ri . . d